This Article Abstract Full Text (PDF) All Versions of this Article: 139/3/1366    most recent pp.105.066944v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Web of Science (10) Citing Articles via Google Scholar Google Scholar Articles by Lievens, S. Articles by Holsters, M. Search for Related Content PubMed PubMed Citation Articles by Lievens, S. Articles by Holsters, M. Agricola Articles by Lievens, S. Articles by Holsters, M. Related Collections Legume Biology

PLANTS INTERACTING WITH OTHER ORGANISMS

Gibberellins Are Involved in Nodulation of Sesbania rostrata¹

Department of Plant Systems Biology, Flanders Interuniversity Institute for Biotechnology, Ghent University, B–9052 Gent, Belgium (S.L., S.G., J.D.H., W.C., R.M., M.H.); and Department of Agricultural Science, University of Bristol, Long Ashton, Bristol BS41 9AF, United Kingdom (P.H.)

	ABSTRACT

TOP ABSTRACT RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

 
Upon submergence, Azorhizobium caulinodans infects the semiaquatic legume Sesbania rostrata via the intercellular crack entry process, resulting in lateral root-based nodules. A gene encoding a gibberellin (GA) 20-oxidase, SrGA20ox1, involved in GA biosynthesis, was transiently up-regulated during lateral root base nodulation. Two SrGA20ox1 expression patterns were identified, one related to intercellular infection and a second observed in nodule meristem descendants. The infection-related expression pattern depended on bacterially produced nodulation (Nod) factors. Pharmacological studies demonstrated that GAs were involved in infection pocket and infection thread formation, two Nod factor-dependent events that initiate lateral root base nodulation, and that they were also needed for nodule primordium development. Moreover, GAs inhibited the root hair curling process. These results show that GAs are Nod factor downstream signals for nodulation in hydroponic growth.

The best known mode of invasion is the root hair curling (RHC) mechanism that is used by most crop legumes and the model legumes barrel medic (Medicago truncatula) and Lotus japonicus. Rhizobia induce growing root hairs to curl in the root zone I, just above the root meristem, whereby a rhizobial microcolony is entrapped. Local cell wall degradation and subsequent inward growth of the root hair plasma membrane result in the formation of an infection thread (IT) that guides the bacteria to the cortical cells. RHC is Nod factor dependent, and purified compatible Nod factors trigger several nodulation-related effects within the root hair, such as deformation, gene expression, Ca²⁺ spiking, membrane depolarization, and ion effluxes (Oldroyd and Downie, 2004). Several components of the Nod factor perception mechanism have been identified via map-based cloning of symbiotic genes (Oldroyd and Downie, 2004).

A different type of invasion, via crack entry at lateral root bases (LRBs), is observed during nodulation under flooded conditions on semiaquatic legumes, such as Sesbania rostrata (Goormachtig et al., 2004b). The rhizobia enter at the cracks that are generated by the protrusion of the lateral roots. Local induction of cell death creates space for bacterial colonization in intercellular infection pockets (IPs; D'Haeze et al., 2003). From these IPs, ITs guide the bacteria to the target cells for intracellular settlement. LRB invasion, as with RHC invasion, depends on proper Nod factor recognition (D'Haeze et al., 1998, 2003). The LRB mode of invasion is used also during stem nodulation at the bases of adventitious rootlets present on S. rostrata stems (Goormachtig et al., 1997). Developing adventitious root nodules have been very useful for transcript profiling because of their site-specific origin (Goormachtig et al., 1995; Lievens et al., 2001).

LRB nodulation is an adaptive trait to water tolerance (Goormachtig et al., 2004a) and, when S. rostrata roots are grown in aerated soils, rhizobial invasion occurs via the RHC process. The versatility in the invasion modes is mediated by ethylene (D'Haeze et al., 2003; Goormachtig et al., 2004b). Upon flooding, accumulating ethylene inhibits the RHC process, hampers the formation of zone I root hairs, promotes LRB invasion (Goormachtig et al., 2004a), and enables the switch from indeterminate (apical nodule meristem, elongated) to determinate (no meristem, round) nodules (Fernández-López et al., 1998).

Plant hormones control all developmental plant processes, including nodulation that is presumably initiated by a change in the cytokinin/auxin ratio within the root. Nod factor-induced inhibition of auxin transport would lead to the local accumulation of auxins needed to trigger a nodule primordium (Mathesius et al., 1998). Also, cytokinins may be involved because nonnodulating bacteria overproducing the isopentenyl transferase gene (Cooper and Long, 1994) as well as the exogenous application of cytokinin (Libbenga et al., 1973; Reli et al., 1994; Bauer et al., 1996) can provoke nodule-like structures on legume plants. Recently, in L. japonicus, cytokinins have been shown to accumulate within the nodule primordia by using the Arabidopsis (Arabidopsis thaliana) cytokinin-responsive promoter ARR5 (Lohar et al., 2004). Ethylene exerts a positional control on nodule primordium formation. Nodules mostly arise opposite protoxylem poles because ethylene, produced in the pericycle cells opposite phloem poles would prevent cell division at those sites (Heidstra et al., 1997; Penmetsa and Cook, 1997; Nukui et al., 2004).

Cytokinin and ethylene are also involved in the invasion process. The cytokinin-responsive promoter ARR5 is expressed in curling root hairs, and cytokinin-insensitive hairy roots of L. japonicus nodulate less well than control hairy roots (Lohar et al., 2004). Ethylene controls the RHC process in barrel medic, as demonstrated by the ethylene-insensitive sickle mutant that allows a higher number of successful infections (Penmetsa and Cook, 1997). Moreover, ethylene inhibits the Nod factor-induced Ca²⁺ response within the barrel medic root hairs (Oldroyd et al., 2001). On the other hand, ethylene is needed to initiate cell death for IP formation at LRBs of S. rostrata (D'Haeze et al., 2003).

Little is known about the role of GAs in the rhizobium-legume interaction. Elevated GA levels have been measured in nodules of lima bean (Phaseolus lunatus) and cowpea (Vigna unguiculata; Dobert et al., 1992). However, analysis of effects of exogenous GAs on root nodulation yielded conflicting results (Bishnoi and Krishnamoorthy, 1990; Zhang et al., 1997). Exogenous GA₃ induced the formation of nodule-like structures on the roots of L. japonicus, and this response was nitrogen sensitive (Kawaguchi et al., 1996) but originated from cell divisions in the pericycle and not from cortical cells. Although some free-living rhizobia have the capacity to produce GAs (Atzorn et al., 1988; Tully et al., 1998), they probably do not contribute significantly to the amount of GA within the nodule (Atzorn et al., 1988). Recently, different pea (Pisum sativum) lines mutated in GA synthesis or perception have been analyzed for their nodulation phenotypes (Ferguson et al., 2005). Nodulation in some mutants was hampered, resulting in fewer, nonfunctional nodules, a phenotype that could be complemented by exogenous application of GAs (Ferguson et al., 2005).

GAs comprise a large group of more than 130 diterpenoid carboxylic acids, of which most are precursors or inactivated forms, but some members, including GA₁, GA₃, GA₄, and GA₇, have an intrinsic growth-promoting activity. The biochemistry of GA biosynthesis can be subdivided into three main stages (Fig. 1; Hedden and Phillips, 2000; Yamaguchi and Kamiya, 2000). In the first stage, geranylgeranyl diphosphate is converted by two terpene cyclases to ent-kaurene, which, in the next stage, is oxidized by cytochrome P450 monooxygenases to yield GA₁₂ and GA₅₃ that, in the final stage, are converted to bioactive GAs by the 2-oxoglutarate-dependent dioxygenases, GA 20-oxidase (GA20ox) and GA 3-hydroxylases (GA3ox). GA20ox converts GA₁₂ and GA₅₃ to the C₁₉-GAs, GA₉ and GA₂₀, respectively, which are converted by GA3ox to the bioactive GA₄ and GA₁. The dioxygenases involved in GA synthesis are products of small multigene families, with each family member having a specific pattern of expression, as visualized with promoter--glucuronidase (GUS) reporters or in situ hybridization (Phillips et al., 1995; García-Martínez et al., 1997; Carrera et al., 1999; Rebers et al., 1999).

View larger version (11K): [in this window] [in a new window]   Figure 1. Overview of the GA biosynthetic pathway. The activities of the two main biosynthetic enzymes, GA20ox and GA3ox, are boxed.

	RESULTS

TOP ABSTRACT RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

 

SrGA20ox1 Codes for a Functional GA20ox

To obtain the full-length cDNA of the partial clone Srdd16, previously identified by differential display as an early induced tag in S. rostrata stem nodule development (Lievens et al., 2001), we used 5' RACE (see "Materials and Methods"). By excluding the polyadenylation tail, the cDNA sequence was 1,365 bp long with an open reading frame of 372 amino acids. BLAST searches (Altschul et al., 1997) revealed that the deduced protein exhibited significant homology with GA20ox. The highest homology was found with Ls20ox1 from lettuce (Lactuca sativa; Toyomasu et al., 1998), sharing 54% identity and 61% similarity at the amino acid level. According to the nomenclature by Coles et al. (1999), the full-length sequence was designated SrGA20ox1.

Alignment with amino acid sequences of GA20ox from diverse species revealed the presence of several characteristic sequence features in SrGA20ox1. The His and Asp residues involved in the binding of Fe²⁺ were present at conserved positions (–238, –240, and –294; Roach et al., 1995). The consensus sequence NYYPxCxxP was found in residues 221 to 229 and the conserved Arg and Ser residues that are involved in binding the 5-carboxyl of 2-oxoglutarate were recognized at positions 304 and 306, respectively (Valegård et al., 1998). The LPWKET motif that had been proposed to be important for GA substrate binding (Xu et al., 1995) was present in residues 140 to 145 in the SrGA20ox1 sequence.

To confirm that SrGA20ox1 encodes a functional GA20ox, the protein was produced in Escherichia coli. The cDNA fragment c8.2fl1 was inserted in sense orientation and in frame with the short N-terminal tag into a pET-3a vector (see "Materials and Methods"). Soluble extracts of E. coli cells producing the recombinant protein were assayed for GA20ox activity by incubation with [¹⁴C]GA₁₂, and the reaction products were purified by HPLC and identified by gas chromatography-mass spectrometry (data not shown). The major peak contained [¹⁴C]GA₉, [¹⁴C]GA₂₅, [¹⁴C]GA₁₅, and [¹⁴C]GA₂₄ in a ratio of approximately 4:1:1:0.1 based on the total ion intensities of the spectra (data not shown). These data demonstrate that SrGA20ox1 is an active GA20ox.

To determine the number of GA20ox genes in the S. rostrata genome, a probe spanning the open reading frame (c8.2fl1) was hybridized with genomic DNA digested with different restriction enzymes (Fig. 2A). Several bands were observed in each lane, suggesting that the SrGA20ox1 gene is part of a small gene family. When similar filters were hybridized with the more gene-specific differential display fragment (Srdd16, containing the 3' untranslated region and a short stretch of the open reading frame), a single band pattern was visible after high stringency washing, except for the lane with the EcoRI-digested genomic DNA, in which two bands occurred in accordance with the presence of an EcoRI restriction site in SrGA20ox1 (Fig. 2B).

View larger version (117K): [in this window] [in a new window]   Figure 2. DNA gel-blot analysis. The different restriction enzymes used are indicated above each lane. A, Digested genomic DNA hybridized with the full-length c8.2.fl1 probe. B, Identical blot hybridized with the specific Srdd16 fragment.

Reverse transcription (RT)-PCR analysis showed a very low background signal in uninfected, stem-located, adventitious root primordia (Fig. 3A). SrGA20ox1 transcript accumulation was induced 8 h after inoculation of the root primordia with Azorhizobium caulinodans and increased to maximal levels after 3 and 5 d. Later, expression gradually dropped to background levels. During LRB nodulation on hydroponic roots, the gene is induced approximately 4 to 8 h after inoculation (Fig. 3B).

View larger version (51K): [in this window] [in a new window]   Figure 3. SrGA20ox1 expression analysis determined by RT-PCR analysis. A, SrGA20ox1 expression levels in dormant adventitious root primordia before (–) and at different time points from 1 h (1h) to 20 d (20d) after infection with A. caulinodans ORS571. SrGA20ox1 expression levels in adventitious root primordia 8 h (8h) and 2 d (2d) after inoculation with a Nod factor-deficient strain, ORS571-V44 (V44), or with a surface polysaccharide-deficient strain ORS571-X15 (X15). As a constitutive control, an ubiquitin gene (Srubi1) was amplified. B, SrGA20ox1 expression levels in uninfected roots (–), in roots 30 min (30'), 1 (1h), 2 (2h), and 4 h (4h) after treatment with Nod factors, and in developing hydroponic root nodules, at 4 (4h), 8 (8h), and 12 h (12h) after infection with A. caulinodans ORS571. C, SrGA20ox1 RNA levels in seedlings (seedling), apices (apex), flowers (flower), leaves (leaf), and roots (root).

To test whether the bacterial signaling molecule alone is sufficient for GA20ox gene induction, roots were inoculated with 10^–8 M Nod factors and harvested after several time points in two independent experiments. RT-PCR analysis indicated that SrGA20ox1 transcripts started to accumulate 1 to 2 h after Nod factor addition and peaked after 8 h (Fig. 3B; data not shown).

In all plants in which several members of the GA20ox gene family have been identified, individual members show a differential tissue specificity (e.g. Phillips et al., 1995; García-Martínez et al., 1997; Carrera et al., 1999; Rebers et al., 1999). To test the specificity of SrGA20ox1 expression, RT-PCR analysis was applied on RNA samples taken from different plant tissues (Fig. 3C). In samples prepared from vegetative shoot apices, flowers, leaves, and roots, various levels of expression were observed.

SrGA20ox1 Transcripts Accumulate Around the Bacterial Invasion Track and in the Preinfection Zone

The spatio-temporal pattern of SrGA20ox1 expression during stem-located, adventitious root nodulation was examined by in situ hybridization. Sections (10 µm) of adventitious root nodules harvested at different phases of development were hybridized with an antisense probe derived from the gene-specific Srdd16 cDNA fragment. No expression could be detected in uninfected root primordia (data not shown). Consistent with the results of the RT-PCR analysis, the earliest time point at which SrGA20ox1 induction could be perceived was 8 h after inoculation with A. caulinodans (Fig. 4, A and D). At this early stage, no morphological changes occur in the root primordia, and the bacteria proliferate in the fissure that surrounds the base of the root primordium (Tsien et al., 1983; Duhoux, 1984; Goormachtig et al., 1997). SrGA20ox1 transcripts were highly abundant in the outermost cortical cell layers that enclose the epidermal crack (Fig. 4, A and D). The first cell divisions, which will give rise to the nodule primordium in the mid-cortex, usually become noticeable approximately 1 d postinoculation (dpi), and at the same time the rhizobia in the fissure start penetrating the intercellular spaces of the outer cortex. Approximately 2 dpi, large, heavily colonized IPs could be seen with strong SrGA20ox1 expression in the neighboring plant cells (Fig. 4, B and E). Then, azorhizobia move from the IPs toward the nodule primordium that develops deeper into the cortex through intercellular and intracellular ITs (Tsien et al., 1983; Duhoux, 1984; Goormachtig et al., 1997). In 3-d-old developing stem nodules, SrGA20ox1 transcripts were found in the cells neighboring deeper IPs and in cells containing the growing ITs (Fig. 4, C and F). The expression in the outer cell layers had mostly disappeared by this time.

View larger version (149K): [in this window] [in a new window]   Figure 4. In situ localization of SrGA20ox1. Signals are detected as white spots or dark spots using dark-field (A–C, G, and J) or bright-field (D–F, H, I, K, and L) optics, respectively. A to C, Longitudinal sections through an adventitious root primordium, 8 h after infection, a developing adventitious root nodule, 2 d after bacterial inoculation, and developing adventitious root nodule, 3 d after infection, respectively. D to F, Bright-field pictures of A to C, respectively. G, Longitudinal section through a developing adventitious root nodule, 4 d after infection. H, Bright field of G. I, Longitudinal section through a developing adventitious root nodule, 5 d after infection. J, Longitudinal section through an indeterminate root nodule (25 dpi). K, Bright-field picture of J. L, Enlargement of the meristematic and infection zone of K. f, Fixation zone; fi, fissure; i, infection zone; ic, infection center; ip, infection pocket; it, infection thread; m, meristem; np, nodule primordium; rm, root meristem; vb, vascular bundle. Bars = 100 µm.

The Expression That Colocalizes with the Invasion Track Depends on Locally Produced Nod Factors

To analyze whether local Nod factor perception or distant Nod factor signaling were responsible for SrGA20ox1 gene induction, in situ hybridizations were performed on developing nodules induced after coinoculation of ORS571-V44 and ORS571-X15 strains. The latter strain produces Nod factors but stays in outer-located IPs, and can complement the Nod factor-deficient strain ORS571-V44, giving rise to complementation nodules in which only ORS571-V44 resides in ITs and, later on, in the infected cells (D'Haeze et al., 1998). As shown in Figure 5, in the complementation nodules, transcripts were not detected around the IPs or the ITs. On the other hand, SrGA20ox1 transcripts were still seen in the preinfection zone of the developing nodule (Fig. 5, C and D, arrow). Because only ORS571-V44 resides in the ITs, these results demonstrate that SrGAox1 gene expression around the invasion track during wild-type invasion depends on Nod factors produced locally by bacteria within the ITs.

View larger version (174K): [in this window] [in a new window]   Figure 5. SrGA20ox1 in situ localization in ORS571-V44/ORS571-X15 complementation nodules. Signals are seen as white or black spots in respectively dark-field or bright-field optics. A and C, Longitudinal sections through a developing complementation nodule, 6 and 14 d after inoculation, respectively. B and D, Bright-field pictures of A and C, respectively. Abbreviations: See legend of Figure 4. Bars = 100 µm.

Because IP formation involves controlled death of a few cells (D'Haeze et al., 2003), we wondered whether SrGA20ox1 would be induced upon pathogen attack. Two different approaches were followed. First, S. rostrata leaves were infected with spores of the pathogenic fungus Botrytis cinerea. After 48 h of spore application, macroscopic lesions were visible (data not shown). -1,3-Glucanase gene expression was strongly induced from 4 h after inoculation on. However, no SrGA20ox1 induction was detected by RT-PCR (Fig. 6A).

View larger version (76K): [in this window] [in a new window]   Figure 6. SrGA20ox1 expression levels upon pathogen attack. A, SrGA20ox1 expression, measured by RT-PCR, in untreated leaves (–) and 4 (4h), 8 (8h), 17 (17h), 30 (30h), and 48 h (48h) after infection with B. cinerea. B, SrGA20ox1 expression level measured by RT-PCR analysis in adventitious root primordia, infected with a hrcR mutant (hrp–) or wild-type (WT) R. solanacearum strain, 8 h (8h) and 1 to 5 d (1d, 2d, 3d, 4d, and 5d) after inoculation. Srubi1 and Srglu2 expression levels serve as a constitutive and a positive control (Lievens et al., 2004), respectively.

Nod Factor-Induced Axillary Root Hair Outgrowth Is Mediated by GAs

Hydroponic S. rostrata roots carry bulge-like structures at the LRBs. Upon application of A. caulinodans or its purified Nod factors, these bulges grow out into deformed axillary root hairs (Mergaert et al., 1993), a specific effect of Nod factors (Mergaert et al., 1993). To determine whether GAs are also downstream signals of Nod factors, 5.10^–3 M chlormequat chloride (CCC), an inhibitor of GA synthesis, was added prior to Nod factor, and axillary root hair outgrowth was monitored for 5 d (see "Materials and Methods"). As shown in Figure 7C, no axillary root hairs could be detected. Moreover, GA₃ application (10^–5 M) provoked outgrowth of two types of root hairs (Fig. 7, D–F): straight ones, similar to those obtained with ethylene and hydrogen peroxide (H₂O₂; D'Haeze et al., 2003), and deformed ones, similar to the Nod factor-induced axillary root hairs but smaller in size (Fig. 7F). Based on this experiment, we conclude that GAs might indeed be downstream signals of Nod factors.

View larger version (64K): [in this window] [in a new window]   Figure 7. Axillary root hair outgrowth at LRBs of hydroponic roots, analyzed 5 d after treatment and stained with methylene blue for visualization of root hairs. A, Control. B and C, LRBs treated with A. caulinodans Nod factors and pretreated with 5.10–3 M CCC, respectively. D to F, LRBs treated with 10–5 M GA3. Bars = 500 µm.

The restricted, localized expression of a GA20ox1 gene during nodulation points to a specific role of GAs. The effect of inhibition of GA biosynthesis was tested by exogenous application of three different inhibitors, each one interfering with a distinct stage of the GA biosynthetic pathway (Rademacher, 1991). Various concentrations of daminozide, CCC, or paclobutrazol were applied to determine the optimal quantity that had a clear effect on nodulation without affecting plant health (see "Materials and Methods"). All three inhibitors inhibited the nodulation process. The results obtained with CCC are presented here (Fig. 8).

View larger version (125K): [in this window] [in a new window]   Figure 8. Involvement of GAs during S. rostrata nodulation. A, Graphic representation of the effect of CCC and GA3 on nodule number during LRB nodulation. The number of nodules counted on 10 plants is given. Lane 1, Control roots. Lanes 2, 3, 4, 5, and 6, Addition of 10–3 M CCC 2 d before inoculation, at 1 dpi, 2 dpi, 3 dpi, and 4 dpi, respectively. Lane 7, Addition of 10–3 M CCC and 10–5 M GA3 2 d before inoculation. Lane 8, Addition of 10–3 M CCC 2 d before inoculation and 10–5 M GA3 at 1 dpi. Lane 9, Addition of 10–5 M GA3 2 d before inoculation. Lane 10, Addition of 10–5 M GA3 at 1 dpi. B to D, GUS-stained LRBs infected with A. caulinodans (pRG960D-32), at 1 dpi, 2 dpi, and 3 dpi. E to G, GUS-stained LRBs infected with A. caulinodans (pRG960D-32) after addition of 10–3 M CCC 2 d prior to inoculation, 1 dpi, and 2 dpi, respectively. H and I, Toluidine blue-stained sections through E. Rectangles and triangles indicate dying cells and ITs containing cells with blue precipitate within the cytoplasm, respectively. J to L, Toluidine blue-stained sections through F. Arrows in L indicate irregular ITs. M, Toluidine blue-stained section through B. N, Graphic representation of the effect of CCC and GA3 on nodule number during RHC nodulation. Lane 1, Control roots. Lanes 2 and 3, Addition of 10–3 M CCC and 10–5 M GA3 4 d before inoculation, respectively. Abbreviations: See legend of Figure 4. Bars = 1 cm (B to G); 100 µm (H to M).

To confirm the specific effect of CCC on GA synthesis, 2 d before and 1 d after inoculation, GA₃ (10^–5 M) was added to samples that had been pretreated with CCC 2 d before inoculation, and nodules were counted at day 6. As shown in Figure 8A (compare lanes 7 and 8 with lane 2), a slightly higher nodule number (P < 0.05) was obtained, showing that GA₃ could partially complement CCC for the nodulation phenotype. GA₃ (10^–5 M) added alone 2 d before and 1 d after inoculation had no significant effect on nodule number (Fig. 8A, lanes 9 and 10).

To exclude the possibility that the effect of the GA biosynthesis inhibitors on root nodulation resulted from an inhibitory effect on the Nod factor-producing capacities of the azorhizobia, Nod factor production of wild-type A. caulinodans under normal growth conditions was compared with the production in the presence of the inhibitors by labeling the molecules in vivo with [¹⁴C]acetate. The final concentration for daminozide and CCC was 10^–2 M, and for paclobutrazol 10^–4 M, a slightly higher concentration than that applied in the nodulation assays. Extracts from the culture supernatant were analyzed by thin-layer chromatography. No difference could be observed between the patterns obtained for any of the samples, indicating that in the presence of the inhibitors, A. caulinodans produces normal Nod factors (data not shown). To test whether the nod gene-inducing capacity of S. rostrata is compromised by the addition of the inhibitors to the growth medium, S. rostrata seedlings were grown overnight on a lawn of an A. caulinodans ORS571 strain harboring a lacZ fusion in nod locus 1, on plates containing the inhibitors at the same concentrations as those in the first control experiment. For all three inhibitors, a blue halo, indicative of -galactosidase activity, was observed around the seedlings, revealing that nod gene activity was induced at levels similar to those in the control plates, to which no inhibitor had been supplemented (data not shown).

These experiments indicate that GA plays a role in the early stages of LRB nodulation. To analyze at which level the nodulation is blocked, the treatments were repeated with A. caulinodans (pRG960D-32) expressing uidA driven by a nodA promoter that enables staining of the bacteria via the GUS assay. Nodules and nodule-like structures were analyzed at 6 dpi. Addition of 5.10^–3 M CCC 2 d before inoculation completely blocked the rhizobial invasion and no blue staining was observed, indicating the absence of bacteria and IPs (data not shown). When 10^–3 M CCC was added 2 d before inoculation, no nodules were obtained, but rhizobial colonization was allowed and outer cortical colonization was more pronounced than that of control infection at 1 and 2 dpi (Fig. 8, B, C, and E). When CCC was added at 1 dpi, bumps occurred, in which the bacteria had proliferated partially (compare Fig. 8, F and D). Mostly functional nodules were obtained after CCC had been added 2 dpi (Fig. 8G).

Sections through a developing nodule treated with 10^–3 M CCC 2 d prior to infection showed large IPs in which the plant cell structures were still visible (Fig. 8H). Neighboring cells were often stained pink by toluidine blue and contained pinkish granular threads, indicating that the nuclei had disintegrated and that the cells were dying (rectangles in Fig. 8, H and I). Often outer cortical cells with initiated ITs were blue because of the presence of bacteria in the cytoplasm (triangles in Fig. 8, H and I) and showed signs of cell death (pinkish granular threads; Fig. 8I). No cell division and no sign of primordium formation were observed (Fig. 8, H and I). In contrast, nodule primordia were seen in sections through LRBs that were treated with CCC at 1 dpi (Fig. 8J), comparable to those of control infection at 1 dpi (Fig. 8M). Many more IPs than in the wild type were observed and new IPs were still being made (Fig. 8K), a process that does not occur at 6 dpi in control samples. The infection had proceeded further than in samples where CCC was added 2 d before inoculation and ITs were broader (Fig. 8L) than those of the wild type. No transition from globular primordium to a zonated structure took place and no meristem was observed (Fig. 8L).

GAs Block the RHC Invasion

To analyze the effect of GA on the RHC invasion, S. rostrata plants were grown in Leonard jars, the experimental system that allows aerated root growth and nodulation via RHC in zone I (Goormachtig et al., 2004a). Four days prior to inoculation 10^–3 M CCC was added, and nodules were counted at 14 dpi. No effect on the nodule number was observed (Fig. 8N). On the other hand, no nodules were formed when 10^–5 M GA₃ was applied 4 d prior to inoculation. Microscopic analysis revealed root hair deformation, but no ITs within the root hairs (data not shown).

	DISCUSSION

TOP ABSTRACT RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

 

SrGA20ox1 Codes for an Active GA20ox and Is Induced upon Adventitious and LRB Nodulation

Differential display has been used to search for genes that are involved in adventitious root nodulation on S. rostrata stems (Lievens et al., 2001). One of the tags was homologous to genes coding for different GA20ox. The deduced amino acid sequence of the full-length cDNA clone SrGA20ox1 was approximately 50% identical to GA20ox sequences from other species, in agreement with the commonly observed degree of amino acid sequence identity of 50% to 60% between GA20ox of unrelated species (Hedden, 1999). The identity of the protein was confirmed by an in vitro assay. Soluble extracts from E. coli producing the full-length SrGA20ox1 protein displayed GA20ox activity, converting the substrate GA₁₂ mainly to GA₉.

As for other plants, also in S. rostrata, a small family of GA20ox genes is present. To reduce the possibility that more than one gene would be visualized during the expression analysis, primers and probes were used that cover the 3' untranslated region of the gene because probes of this region only recognized one band by DNA gel-blot analysis.

RT-PCR analysis showed that SrGA20ox1 transcript accumulation starts between 4 and 8 h after contact with the microsymbiont, whereas mitotic activity in the mid cortex or bacterial invasion of the outer cortex can be perceived only approximately 1 dpi (Goormachtig et al., 1997). In S. rostrata, this is one of the earliest induced genes yet identified (Lievens et al., 2001). SrGA20ox1 transcripts were also detected in shoot apices, leaves and to a lower extent in roots and flowers, indicating that the expression is not restricted to nodulation. Until now, only a few genes have been isolated that are uniquely expressed during nodulation, demonstrating recruitment of similar functions for different developmental processes.

Two Distinct Patterns of SrGA20ox1 Expression during LRB Nodule Development

GA20ox are multifunctional enzymes that catalyze oxidation on carbon-20 of C₂₀-GA precursors, some of the final steps in GA biosynthesis. After GA20ox-mediated synthesis of C₁₉-GAs, they are converted into bioactive GAs by the action of GA3ox (Hedden, 1999). Considerable evidence indicates that the stages catalyzed by GA20ox are important regulatory checkpoints. Experiments with GA response mutants demonstrated a feedback regulation of GA biosynthesis, involved in adjusting levels of bioactive GAs (Hedden and Kamiya, 1997) that act at the transcriptional level on GA20ox and GA3ox (Chiang et al., 1995; Phillips et al., 1995; Xu et al., 1995; Martin et al., 1996). Also, for SrGA20ox1, preliminary data showed that the basal expression observed in roots disappeared upon GA treatment (W. Capoen and M. Holsters, unpublished data). In many plant species, GA levels control growth through GA20ox activity that is a major determinant of GA production. For example, in bolting Arabidopsis plants, higher GA levels in the rapidly elongating shoots were accompanied by increased GA20ox gene expression (Xu et al., 1995). A similar observation was reported from spinach (Spinacia oleracea; Wu et al., 1996). In addition, transgenic Arabidopsis plants overexpressing GA20ox had elevated levels of bioactive GAs and were taller than control plants, indicating that GA20ox might be a rate-limiting step in the biosynthesis of GAs (Coles et al., 1999). Comparable results were obtained for potato (Solanum tuberosum) plants overexpressing GA20ox (Carrera et al., 2000). The specific expression patterns of SrGA20ox1 suggest roles for GA in LRB nodulation. SrGA20ox1 transcripts could not be detected in uninoculated root primordia, but throughout early nodule development were located transiently in cells associated with the invading bacteria, in cells adjacent to the epidermal fissure, IPs, or ITs. This induction faded out as the invasion front moved on. When a zonated developing nodule was established, transcripts accumulated in a number of cell layers at the transition between meristematic and infection zone. When the meristematic activity of the LRB nodules disappeared, SrGA20ox1 transcripts did not accumulate anymore. In conclusion, on the basis of SrGA20ox1 expression, GAs might be produced during intercellular invasion at LRBs and in differentiating cells derived from the nodule meristem.

GAs Act Downstream of Nod Factors during LRB Nodulation

Up-regulation of SrGA20ox1 transcripts upon azorhizobial invasion depends on Nod factor-producing bacteria. SrGA20ox1 transcript accumulation is induced when pure Nod factors are applied to hydroponic roots, suggesting that GAs might be direct or indirect downstream signals of Nod factors.

A. caulinodans provokes different Nod factor-related effects on S. rostrata roots (Mergaert et al., 1993), of which the outgrowth and deformation of axillary root hairs at LRBs has been used to analyze the role of H₂O₂ and ethylene in Nod factor signaling (D'Haeze et al., 2003). Inhibitors of GA biosynthesis inhibit the Nod factor-dependent root hair outgrowth, and GAs can partially mimic the Nod factor effect. Coinoculation between ORS571-V44, which does not produce Nod factors, and ORS571-X15, which has defective surface polysaccharides and is restricted to superficially located IPs, and subsequent analysis of the resulting complementation nodules addressed the question whether SrGA20ox1 expression is triggered by locally produced Nod factors or from a distance. ORS571-V44 can invade the plant when Nod factors are supplemented from outside to initiate the nodule formation. The Nod factors, provided by ORS571-X15, are not expected to travel within the plant tissue because they attach to cell walls (Goedhart et al., 2000). Because no SrGA20ox1 gene expression could be detected around IPs and ITs that contained ORS571-V44 bacteria, Nod factors are locally responsible for the invasion-related SrGA20ox1 expression pattern. Moreover, GAs were able to enhance the expression of another infection-related and Nod factor-depending tag coding for a peroxidase, Srprx1 (J. Den Herder, S. Goormachtig, and M. Holsters, unpublished data).

What could be the function of GAs during IP and IT formation? By using pharmacological approaches, we showed that GAs are needed to initiate intercellular invasion. Addition of inhibitors of GA synthesis 2 d prior to bacterial inoculation completely blocked nodulation, an effect that could be partially complemented by exogenously added GAs. No GUS staining that marks the presence of bacteria was observed when the roots were pretreated with 5.10^–3 M CCC. The bacterial invasion during LRB nodulation is initiated by IP formation, which involves local cell death (D'Haeze et al., 2003). In the aleurone of cereals, GA accelerates cell death after reserve remobilization by sensitizing the cells to H₂O₂ (Bethke et al., 1999; Fath et al., 2001). By analogy, GAs produced at the onset of LRB invasion might make the surrounding outer cortical cells more prone to death. Interestingly, the GA20ox1 gene was not switched on after infection of S. rostrata leaves with B. cinerea and after infection of adventitious root primordia with R. solanacearum, two pathogens that induce cell death and necrosis (Lievens et al., 2004).

Addition of CCC at a slightly lower concentration (10^–3 M) inhibited nodule formation, but bacterial colonization was allowed and even enhanced. Many more IPs were formed compared to wild-type. ITs were broad and had an irregular form, indicating that GAs might be an important signal to control IT structure. Both IT formation and pollen tube growth have common features (Szczyglowski and Amyot, 2003; Rodríguez-Llorente et al., 2004). It is interesting that GAs are needed for pollen tube elongation in Arabidopsis (Singh et al., 2002).

The enhanced formation of IPs and the release of bacteria within cells containing ITs observed at 10^–3 M CCC contrasted with the inhibition of IP formation at slightly higher concentrations (5.10^–3 M). Possibly, 10^–3 M CCC might enable enough GAs to be produced to make the tissue sensitive to cell death. Enhanced cell death and IP formation within this sensitive tissue may be a consequence of the inhibition of IT growth and proper bacterial invasion. Indeed, enhanced primary colonization has been observed during invasion by mutant bacteria that are impaired in triggering IT growth (Mathis et al., 2005). The plant feedback mechanisms that control bacterial invasion might simply not be switched on when invasion is impaired. Thus, these apparently opposite effects reflect that GA levels have to be strictly controlled and that different processes require different doses of GA.

GAs Are Essential for Induction of Cortical Cell Division and Differentiation of the Nodule Primordium

The pharmacological studies showed that GAs are needed for nodule primordium formation in the cortex and the establishment of a nodule meristem. Supplementation of 10^–3 M CCC allowed extensive colonization, but no cell division was initiated in the cortex. When the inhibitor was added 1 d after invasion, after a nodule primordium had been formed, the ITs could spread in the preformed primordium, but further development was arrested. By inhibiting GA synthesis after formation of an indeterminate meristem, nodule development was no longer affected and fewer, but functional, nodules were initiated, reflecting the imperfect synchronization of nodulation at all the LRB sites.

It is well known that GAs control various plant developmental processes by promoting cell division or cell elongation (Stuart et al., 1977; Toyomasu et al., 1998; van den Heuvel et al., 2001). Recently, in cucumber (Cucumis sativa) and tomato (Lycopersicon estivum) GAs have been shown to be needed to induce the wound meristem during tissue reunion in the hypocotyls (Asahina et al., 2002). Interestingly, induction of the nodule primordium and of a wound meristem has been proposed to have much in common (Brewin, 2004). The need for GAs for the nodule primordium formation and differentiation might support this hypothesis.

GAs Are Required for Differentiation of the Nodule Meristem Descendants

SrGA20ox1 transcripts accumulated in a narrow zone of nodule meristem descendants and this pattern was still observed in coinoculation nodules that were invaded by non-Nod factor-producing bacteria. Differentiating cells derived from the meristem undergo enlargement, a process that may be regulated by GAs (Huttly and Phillips, 1995; Jacobs, 1997; Kende and Zeevaart, 1997; Dolan and Davies, 2004). Expression of GA20ox genes at the periphery of the meristem has been used to propose a role for GA in the shoot apical meristem and leaf development (Barley and Waites, 2002; Vogler and Kuhlemeier, 2003). Proteins of the KNOTTED (KNOX) homeodomain produced within the shoot apical meristem repress GA20ox expression to maintain the meristem. Disappearance of KNOX proteins, concomitantly with the appearance of GA20ox, may be associated with differentiation of meristematic cells for the formation of leaf primordia, for instance (Sakamoto et al., 2001; Barley and Waites, 2002). Interestingly, a KNOX homolog from barrel medic is produced in the meristem of indeterminate nodules (Koltai et al., 2001). Thus, this KNOX protein could inhibit GA production to keep a meristem identity, while the presence of GA20ox would trigger differentiation into central tissue cells.

It has been a long-standing question whether a nodule is a modified root, stem, or an organ sui generis (Hirsch and LaRue, 1997). Several experiments support the hypothesis that lateral root programs have been recruited for nodule formation (Ferguson and Mathesius, 2003). However, also shoot-linked expression patterns for nodulins have been reported, and the organization of the nodule vascular bundles is similar to that in a shoot. Here, we show that the meristem organization during nodule formation might have features in common with shoot apical meristems.

GA Inhibits RHC Nodulation

The most widespread mode of invasion is via curling of zone I root hairs and this invasion process occurs on S. rostrata roots when they are grown aeroponically (Goormachtig et al., 2004a, 2004b). The main difference between the two modes of invasion is that in LRB nodulation the epidermis, where several control checkpoints are established, is circumvented (Goormachtig et al., 2004b). After the invasion of the root hair during RHC, or IP formation during LRB, the process for IT formation and initiation of cell division within the cortex is the same (Goormachtig et al., 2004b). In contrast to LRB, application of GA₃ to roots grown in Leonard jars inhibited nodulation. Microscopic analysis revealed Nod factor-dependent root hair deformations, but curling of the root hairs was hampered, which is a requirement for IT initiation. Thus, excessive amounts of GAs inhibit the epidermal response during RHC. A similar observation has been made for ethylene (Goormachtig et al., 2004a).

Recently, addition of GA₃ at low concentration (10^–9 M) in pea has been shown to enhance nodule formation, whereas at 10^–6 M or higher the process was inhibited. Our data might fit with these results because inhibition is observed when 10^–5 M GA₃ is added. It would be interesting to know at which stage, depending on the concentration range, GA₃ interferes with the pea nodulation: either at primordium formation or infection, or both (Ferguson et al., 2005).

Nodulation in aerated soils not only gives rise to the RHC process, but also to indeterminate nodules (Fernández-López et al., 1998). We were not able to detect SrGA20ox1 transcripts by in situ hybridization during the RHC invasion because of problems with tissue fragility of aerated roots (data not shown), but we detected the transcripts in the nodule meristem descendants, where it presumably plays the same role in cell differentiation.

CCC application did not inhibit the nodulation process, although for LRB nodulation GAs are needed for IT growth and primordium formation and differentiation, processes that are common to both nodulation types. The simplest reason might be the consequence of the inaccessibility of the roots to pharmalogical compounds in the experimental set up for RHC nodulation. On the other hand, the requirement for GA might depend on the physiology of the root tissue in which invasion occurs. Different physiological environments are present during both infection ways (hydroponic versus aerated) and nodules are formed at different locations within the root (LRB versus zone I root hairs above the tip). As a result, for nodule formation, different concentrations of GAs might be needed to elicit the same processes.

Our work demonstrates that GAs have a function during LRB nodulation, which is recruited for nodule formation under hydroponic conditions (Goormachtig et al., 2004a). GAs are downstream signals of Nod factors for intercellular invasion and for formation of IPs and ITs at LRBs, and are essential for nodule primordium formation and differentiation. Like ethylene, GAs play an important role in adaptations to water logging. Upon submergence shoot elongation of deepwater rice (Oryza sativa) depends on GAs, as does hyponastic growth of leaves of marsh dock (Rumex palustris; Vriezen et al., 2003; Cox et al., 2004). Therefore, it is not surprising that GAs are needed during LRB nodulation.

	MATERIALS AND METHODS

TOP ABSTRACT RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

 

Biological Material

Sesbania rostrata Brem seeds were surface sterilized, grown, and inoculated as described (Goormachtig et al., 1995; Fernández-López et al., 1998). For the inhibitor assays, plants were grown in 70-mL tubes containing sterile nitrogen-free Norris medium (pH 7.0; Vincent, 1970) and sealed from the exterior by an aluminum foil cap surrounding the hypocotyl. For the Nod factor experiment, 20 roots were grown together in a 1-L pot for each stage. The bacterial strains ORS571, ORS571-X15 (Goethals et al., 1994), and ORS571-V44 (Van den Eede et al., 1987) of Azorhizobium caulinodans were cultivated as previously described (Goormachtig et al., 1995). Purified Nod factors were obtained as described by Mergaert et al. (1997) and added at a final concentration of 10^–8 M (5.10^–9 M of each fraction pI and pII; see Mergaert et al., 1997). Ralstonia solanacearum and Botrytis cinerea infections were performed as described (Lievens et al., 2004).

Isolation of the Full-Length SrGA20ox1 Clone

The sequence of cDNA clone Srdd16 was used as a source of primer sequences for the isolation of the corresponding full-length sequence by using the Marathon cDNA Amplification Kit (Clontech). cDNA was synthesized from RNA extracted from root primordia harvested 2 dpi with A. caulinodans ORS571. Several rounds of nested PCR amplification with the anchor primers AP1 and AP2 provided by the manufacturer in combination with gene-specific antisense primers sh8 (5' GCAGCAGGAGCAGATATAACAGAAGC 3'), sh7 (5' GTGGTTTGGAGGATAGCAACCACTTGG 3'), and sh23 (5' CAGGCTCTGAGTTATTGTCATGGAAGGGG 3') were necessary to obtain the full-length sequence, which was designated SrGA20ox1. A fragment of this sequence that corresponded with the open reading frame, initiating with the start codon and ending between the stop codon and the polyadenylation sequence, was amplified from the same cDNA template with Vent polymerase (New England Biolabs) and sense primer sh29 (5' ATGGATTCAGGTTTGTGCTTAGTGTCTG 3') and antisense primer sh30 (5' GCAGCAGGAGCAGATATAACAGAAGC 3'), cloned in the pGEM-T vector (Promega), and designated pGEMTc8.2fl1.

DNA Gel-Blot Analysis

DNA gel-blot analysis was performed as described by Lievens et al. (2004). Labeled probes were generated from either the full-length cDNA insert of pGEMTc8.2fl1 or the differential display clone Srdd16.

Heterologous Expression in Escherichia coli and GA20ox Activity Assay

The insert from pGEMTc8.2fl1 was PCR amplified using primers containing a BamHI restriction site and inserted into the BamHI site of pET-3a (Novagen). Heterologous expression of the construct was carried out essentially as described by MacMillan et al. (1997). A 50-mL culture of E. coli containing the recombinant plasmid was induced at an OD₆₀₀ of 0.36 with isopropyl--D-thiogalactoside (1 mM final concentration) and then grown at 30°C for 5 h. Cells were pelleted for 5 min at 4,000g and resuspended in 1 mL of 100 mM Tris-HCl (pH 7.5 at 25°C) containing 4 mM dithiothreitol and lysozyme (2 mg). After incubation at room temperature for 15 min, 25 µg DNAse was added, followed by an incubation at room temperature for another 30 min. After centrifugation at 15,000g for 5 min, 90 µL of the supernatant was incubated overnight at 30°C with [¹⁴C]GA₁₂ (167 Bq, 30 pmol) and cofactors in a total volume of 100 µL. The samples were processed and the reaction products were separated by HPLC according to MacMillan et al. (1997). The products were identified by gas chromatography-mass spectrometry (MacMillan et al., 1997).

RNA Analysis

RNA was prepared according to the protocol of Goormachtig et al. (1995) and the RT-PCR analysis was carried out as described (Corich et al., 1998). A 3'-sequence fragment of SrGA20ox1 was amplified with sl32 (5'-AGAGCCGACGAAGATACCCT-3') and sl106 (5'-GCCGTACAAAGTAGAATTAGGTTAAG-3') as sense and antisense primer, respectively. As a constitutive control, a ubiquitin cDNA fragment was amplified (Corich et al., 1998). Twenty PCR cycles were performed, and PCR products were detected radioactively with probes generated from the cDNA fragment Srdd16 and Srubi1 (Corich et al., 1998) by means of the T7 QuickPrime kit (GE-Biosciences). The membranes were analyzed with a PhosphorImager (GE-Biosciences).

GA Inhibitor Experiments

Daminozide (Sigma-Aldrich) and CCC (Sigma-Aldrich) were prepared as aqueous solutions. Paclobutrazol (Duchefa) was stored as a stock solution of 5.10^–2 M in methanol and diluted to lower concentrations in water. All chemicals were filter sterilized before addition to the liquid root medium. Pure methanol added to control plants at concentrations comparable to those present in the medium of roots treated with paclobutrazol did not affect nodulation. Media levels in the plant growth tubes were kept maximal by regular refilling, upon which the inhibitor concentrations were adjusted by adding the appropriate volumes of fresh stock solution.

To establish the optimal concentration for GA₃ and CCC, a range of concentrations was examined and the general fitness of the plant and the effect on axillary root outgrowth and nodulation were analyzed. For the antagonist, CCC, 10^–2 M, 5.10^–3 M, 10^–3 M, 10^–4 M, and 10^–6 M were tested. After 2 weeks, only 10^–2 M had a clearly negative effect on the health of plants compared to that of nontreated plants, including smaller stems, browning of leaves, and a general low fitness. Plants treated with 5.10^–3 M and 10^–3 M, but not with 10^–4 M and 10^–5 M, had clearly stunted shoots and dark-green leaves, typical effects of GA inhibition. For the root hair essay, 5.10^–3 M CCC was chosen because this concentration gave an effect on more than 95% of the plants whereas for 10^–3 M this percentage was a bit lower. For the interference on nodulation, the effects of 5.10^–3 M and 10^–3 M are described in the results. Addition of 10^–4 M of CCC had no effect on nodule number just as it had no typical GA antagonistic effect.

For GA₃, concentrations of 10^–4 M, 10^–5 M, and 10^–6 M were tested. After 2 weeks, neither concentration affected the general plant health as scored above. The typical GA effects, such as the elongated stems and light-green leaves, were observed. For the axillary root hair experiment, the 10^–5 M was chosen because it was the lowest concentration that affected more than 95% of the plants. A similar effect was obtained with 10^–6 M, but the efficiency was lower. Each experiment was done at least in triplicate. Because of the heterogeneity of the seeds of S. rostrata, the nodule number has a degree of variation, and, therefore, the sum of the nodules counted on 10 plants is given for each treatment.

Control experiments in which Nod factors were labeled in vivo with 2-[¹⁴C]acetate in the presence of GA inhibitor were carried out according to Mergaert et al. (1993). A. caulinodans ORS571 was grown overnight in the presence of the appropriate amount of GA inhibitor. The cultures were centrifuged and the pellet was resuspended in 1 mL water. Both pellet and culture supernatant samples were processed as described and analyzed by reversed-phase thin-layer chromatography. The results were visualized on a PhosphorImager (GE-Biosciences).

The nod gene induction was assayed as described by Goethals et al. (1989). Two-day-old S. rostrata seedlings were incubated in the dark at 37°C for 12 h on plates containing 80 µg mL^–1 5-bromo-4-chloro-3-indolyl--D-galactopyranoside and the appropriate amount of GA inhibitor and covered by a lawn of A. caulinodans ORS571 (pRG290-12::T20) in top agar supplemented with the GA inhibitor as well.

Microscopic Analyses

In situ hybridization was performed on 10-µm sections of paraffin-embedded tissue as described by Goormachtig et al. (1997). The plasmid pBlueSKSrdd16 was digested with BamHI and PstI to obtain templates to produce a ³⁵S-labeled antisense and sense probe with T7 and T3 RNA polymerase (GE-Biosciences), respectively. Hybridizations with the sense probe did not result in above-background signals (data not shown). Histochemical analysis and staining for GUS were performed according to Fernández-López et al. (1998) and D'Haeze et al. (2003).

Sequence Analysis

DNA was sequenced with universal SP6 and T7 primers. Sequence data were assembled and analyzed with the GCG Wisconsin Package (Accelrys). Percentage of identity and similarity between sequences was determined with the Gap program and alignments were produced with the PileUp program (GCG Wisonsin Package).

Sequence data from this article can be found in the GenBank/EMBL data libraries under accession number DQ090959.

	ACKNOWLEDGMENTS

 
The authors thank Christa Verplancke and Annick De Keyser for technical help and Martine De Cock for help in preparing the manuscript.

Received June 9, 2005; returned for revision August 19, 2005; accepted August 21, 2005.

	FOOTNOTES

 
¹ This work was supported by the Interuniversity Poles of Attraction Program-Belgian Science Policy (grant no. P5/13), the Research Foundation-Flanders (research fellowship to J.D.H.), and the Institute for the Promotion of Innovation by Science and Technology in Flanders (predoctoral fellowship to W.C.).

² These authors contributed equally to the paper.

³ Present address: Department of Biochemistry, Ghent University, B–9000 Gent, Belgium.

⁴ Present address: Laboratoire de Biologie Moléculaire, Station Nationale d'Essais de Semences, F–49071 Beaucouzé cedex, France.

The authors responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) are: Sofie Goormachtig (sofie.goormachtig{at}psb.ugent.be) and Marcelle Holsters (marcelle.holsters{at}psb.ugent.be).

Article, publication date, and citation information can be found at www.plantphysiol.org/cgi/doi/10.1104/pp.105.066944.

^* Corresponding author; e-mail marcelle.holsters{at}psb.ugent.be; fax 32–9–3313809.

	LITERATURE CITED

TOP ABSTRACT RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

 
Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 3389–3402[Abstract/Free Full Text]

Asahina M, Iwai H, Kikuchi A, Yamaguchi S, Kamiya Y, Kamada H, Satoh S (2002) Gibberellin produced in the cotyledon is required for cell division during tissue reunion in the cortex of cut cucumber and tomato hypocotyls. Plant Physiol 129: 201–210[Abstract/Free Full Text]

Atzorn R, Crozier A, Wheeler CT, Sandberg G (1988) Production of gibberellins and indole-3-acetic acid by Rhizobium phaseoli in relation to nodulation of Phaseolus vulgaris roots. Planta 175: 532–538[CrossRef]

Barley R, Waites R (2002) Plant meristems: the interplay of KNOX and gibberellins. Curr Biol 12: R696–R698[Medline]

Bauer P, Ratet P, Crespi MD, Schultze M, Kondorosi A (1996) Nod factors and cytokinins induce similar cortical cell division, amyloplast deposition and MsEnod12A expression pattern in alfalfa roots. Plant J 10: 91–105

Bethke PC, Lonsdale JE, Fath A, Jones RL (1999) Hormonally regulated programmed cell death in barley aleurone cells. Plant Cell 11: 1033–1045[Abstract/Free Full Text]

Bishnoi NR, Krishnamoorthy HN (1990) Effect of waterlogging and gibberellic acid on nodulation and nitrogen fixation on peanut. Plant Physiol Biochem 28: 663–666

Brewin NJ (2004) Plant cell wall remodelling in the Rhizobium-legume symbiosis. CRC Crit Rev Plant Sci 23: 293–316[CrossRef]

Carrera E, Bou J, García-Martínez JL, Prat S (2000) Changes in GA 20-oxidase gene expression strongly affect stem length, tuber induction and tuber yield of potato plants. Plant J 22: 247–256[CrossRef][Web of Science][Medline]

Carrera E, Jackson SD, Prat S (1999) Feedback control and diurnal regulation of gibberellin 20-oxidase transcript levels in potato. Plant Physiol 119: 765–773[Abstract/Free Full Text]

Chiang H-H, Hwang I, Goodman HM (1995) Isolation of the Arabidopsis GA4 locus. Plant Cell 7: 195–201[Abstract]

Coles JP, Phillips AL, Croker SJ, García-Lepe R, Lewis MJ, Hedden P (1999) Modification of gibberellin production and plant development in Arabidopsis by sense and antisense expression of gibberellin 20-oxidase genes. Plant J 17: 547–556[CrossRef][Web of Science][Medline]

Cooper JB, Long SR (1994) Morphogenetic rescue of Rhizobium meliloti nodulation mutants by trans-zeatin secretion. Plant Cell 6: 215–225[Abstract]

Corich V, Goormachtig S, Lievens S, Van Montagu M, Holsters M (1998) Patterns of ENOD40 gene expression in stem-borne nodules of Sesbania rostrata. Plant Mol Biol 37: 67–76[CrossRef][Web of Science][Medline]

Cox MCH, Benschop JJ, Vreeburg RAM, Wagemaker CAM, Moritz T, Peeters AJM, Voesenek LACJ (2004) The roles of ethylene, auxin, abscisic acid, and gibberellin in the hyponastic growth of submerged Rumex palustris petioles. Plant Physiol 136: 2948–2960[Abstract/Free Full Text]

D'Haeze W, De Rycke R, Mathis R, Goormachtig S, Pagnotta S, Verplancke C, Capoen W, Holsters M (2003) Reactive oxygen species and ethylene play a positive role in lateral root base nodulation of a semi-aquatic legume. Proc Natl Acad Sci USA 100: 11789–11794[Abstract/Free Full Text]

D'Haeze W, Gao M, De Rycke R, Van Montagu M, Engler G, Holsters M (1998) Roles for azorhizobial Nod factors and surface polysaccharides in intercellular invasion and nodule penetration, respectively. Mol Plant-Microbe Interact 11: 999–1008[CrossRef][Web of Science]

Dobert RC, Rood SB, Blevins DG (1992) Gibberellins and the legume-Rhizobium symbiosis: endogenous gibberellins of Lima bean (Phaseolus lunatus L.) stems and nodules. Plant Physiol 98: 221–224[Abstract/Free Full Text]

Dolan L, Davies J (2004) Cell expansion in roots. Curr Opin Plant Biol 7: 33–39[CrossRef][Web of Science][Medline]

Duhoux E (1984) Ontogénèse des nodules caulinaires du Sesbania rostrata (légumineuses). Can J Bot 62: 982–994

Fath A, Bethke PC, Jones RL (2001) Enzymes that scavenge reactive oxygen species are down-regulated prior to gibberellic acid-induced programmed cell death in barley aleurone. Plant Physiol 126: 156–166[Abstract/Free Full Text]

Ferguson BJ, Mathesius U (2003) Signaling interactions during nodule development. J Plant Growth Regul 22: 47–72[CrossRef]

Ferguson BJ, Ross JJ, Reid JB (2005) Nodulation phenotypes of gibberellin and brassinosteroid mutants of pea. Plant Physiol 138: 2396–2405[Abstract/Free Full Text]

Fernández-López M, Goormachtig S, Gao M, D'Haeze W, Van Montagu M, Holsters M (1998) Ethylene-mediated phenotypic plasticity in root nodule development on Sesbania rostrata. Proc Natl Acad Sci USA 95: 12724–12728[Abstract/Free Full Text]

García-Martínez JL, López-Diaz I, Sánchez-Beltrán MJ, Phillips AL, Ward DA, Gaskin P, Hedden P (1997) Isolation and transcript analysis of gibberellin 20-oxidase genes in pea and bean in relation to fruit development. Plant Mol Biol 33: 1073–1084[CrossRef][Web of Science][Medline]

Goedhart J, Hink MA, Visser AJWG, Bisseling T, Gadella TWJ Jr (2000) In vivo fluorescence correlation microscopy (FCM) reveals accumulation and immobilization of Nod factors in root hair cell walls. Plant J 21: 109–119[CrossRef][Web of Science][Medline]

Goethals K, Gao M, Tomekpe K, Van Montagu M, Holsters M (1989) Common nodABC genes in Nod locus 1 of Azorhizobium caulinodans: nucleotide sequence and plant-inducible expression. Mol Gen Genet 219: 289–298[Medline]

Goethals K, Leyman B, Van den Eede G, Van Montagu M, Holsters M (1994) An Azorhizobium caulinodans ORS571 locus involved in lipopolysaccharide production and nodule formation on Sesbania rostrata stems and roots. J Bacteriol 176: 92–99[Abstract/Free Full Text]

Goormachtig S, Alves-Ferreira M, Van Montagu M, Engler G, Holsters M (1997) Expression of cell cycle genes during Sesbania rostrata stem nodule development. Mol Plant-Microbe Interact 10: 316–325[Web of Science][Medline]

Goormachtig S, Capoen W, Holsters M (2004b) Rhizobium infection: lessons from the versatile nodulation behavior of water-tolerant legumes. Trends Plant Sci 9: 518–522[CrossRef][Web of Science][Medline]

Goormachtig S, Capoen W, James EK, Holsters M (2004a) Switch from intracellular to intercellular invasion during water stress-tolerant legume nodulation. Proc Natl Acad Sci USA 101: 6303–6308[Abstract/Free Full Text]

Goormachtig S, Valerio-Lepiniec M, Szczyglowski K, Van Montagu M, Holsters M, de Bruijn FJ (1995) Use of differential display to identify novel Sesbania rostrata genes enhanced by Azorhizobium caulinodans infection. Mol Plant-Microbe Interact 8: 816–824[Web of Science][Medline]

Hedden P (1999) Recent advances in gibberellin biosynthesis. J Exp Bot 50: 553–563[Abstract/Free Full Text]

Hedden P, Kamiya Y (1997) Gibberellin biosynthesis: enzymes, genes and their regulation. Annu Rev Plant Physiol Plant Mol Biol 48: 431–460[CrossRef][Web of Science]

Hedden P, Phillips AL (2000) Gibberellin metabolism: new insights revealed by the genes. Trends Plant Sci 5: 523–530[CrossRef][Web of Science][Medline]

Heidstra R, Yang WC, Yalcin Y, Peck S, Emons A, van Kammen A, Bisseling T (1997) Ethylene provides positional information on cortical cell division but is not involved in Nod factor-induced root hair tip growth in Rhizobium-legume interaction. Development 124: 1781–1787[Abstract]

Hirsch AM, LaRue TA (1997) Is the legume nodule a modified root or stem or an organ sui generis? CRC Crit Rev Plant Sci 16: 361–392

Huttly AK, Phillips AL (1995) Gibberellin-regulated plant genes. Physiol Plant 95: 310–317[CrossRef]

Jacobs T (1997) Why do plant cells divide? Plant Cell 9: 1021–1029[CrossRef][Web of Science][Medline]

Kawaguchi M, Imaizumi-Anraku H, Fukai S, Syõno K (1996) Unusual branching in the seedlings of Lotus japonicus—Gibberellins reveal the nitrogen-sensitive cell divisions within the pericycle on roots. Plant Cell Physiol 37: 461–470[Abstract/Free Full Text]

Kende H, Zeevaart JAD (1997) The five "classical" plant hormones. Plant Cell 9: 1197–1210[CrossRef][Web of Science][Medline]

Koltai H, Dhandaydham M, Opperman C, Thomas J, Bird D (2001) Overlapping plant signal transduction pathways induced by a parasitic nematode and a rhizobial endosymbiont. Mol Plant-Microbe Interact 14: 1168–1177[Web of Science][Medline]

Libbenga KR, Van Iren F, Bogers RJ, Schraag-Lamers MF (1973) The role of hormones and gradients in the initiation of cortex proliferation and nodule formation in Pisum sativum L. Planta 114: 29–39[CrossRef][Web of Science]

Lievens S, Goormachtig S, Holsters M (2001) A critical evaluation of differential display as a tool to identify genes involved in legume nodulation: looking back and looking forward. Nucleic Acids Res 17: 3459–3468

Lievens S, Goormachtig S, Holsters M (2004) Nodule-specific protease inhibitor gene: emerging patterns of gene expression in nodule development on Sesbania rostrata. J Exp Bot 55: 89–94[Abstract/Free Full Text]

Lohar DP, Schaff JE, Laskey JG, Kieber JJ, Bilyeu KD, Bird DM (2004) Cytokinins play opposite roles in lateral root formation, and nematode and Rhizobial symbioses. Plant J 38: 203–214[CrossRef][Web of Science][Medline]

MacMillan J, Ward DA, Phillips AL, Sánchez-Beltrán MJ, Gaskin P, Lange T, Hedden P (1997) Gibberellin biosynthesis from gibberellin A₁₂-aldehyde in endosperm and embryos of Marah macrocarpus. Plant Physiol 113: 1369–1377[Abstract]

Martin DN, Proebsting WM, Parks DT, Dougherty WG, Lange T, Lewis MJ, Gaskin P, Hedden P (1996) Feed-back regulation of gibberellin biosynthesis and gene expression in Pisum sativum L. Planta 200: 159–166[Web of Science][Medline]

Mathesius U, Schlaman HRM, Spaink HP, Sautter C, Rolfe BG, Djordjevic MA (1998) Auxin transport inhibition precedes root nodule formation in white clover roots and is regulated by flavonoids and derivatives of chitin oligosaccharides. Plant J 14: 23–34[CrossRef][Web of Science][Medline]

Mathis R, Van Gijsegem F, De Rycke R, D'Haeze W, Van Maelsaeke E, Anthonio E, Van Montagu M, Holsters M, Vereecke D (2005) Lipopolysaccharides as a communication signal for progression of legume endosymbiosis. Proc Natl Acad Sci USA 102: 2655–2660[Abstract/Free Full Text]

Mergaert P, Ferro M, D'Haeze W, Van Montagu M, Holsters M, Promé J-C (1997) Nod factors of Azorhizobium caulinodans strain ORS571 can be glycosylated with an arabinosyl group, a fucosyl group, or both. Mol Plant-Microbe Interact 10: 683–687[Web of Science][Medline]

Mergaert P, Van Montagu M, Promé J-C, Holsters M (1993) Three unusual modifications, a D-arabinosyl, an N-methyl, and a carbamoyl group, are present on the Nod factors of Azorhizobium caulinodans strain ORS571. Proc Natl Acad Sci USA 90: 1551–1555[Abstract/Free Full Text]

Nukui N, Ezura H, Minamisawa K (2004) Transgenic Lotus japonicus with an ethylene receptor gene Cm-ERS1/H70A enhances formation of infection threads and nodule primordia. Plant Cell Physiol 45: 427–435[Abstract/Free Full Text]

Oldroyd GED, Downie JA (2004) Calcium, kinases and nodulation signalling in legumes. Nat Rev Mol Cell Biol 5: 566–576[CrossRef][Web of Science][Medline]

Oldroyd GED, Engstrom EM, Long SR (2001) Ethylene inhibits the Nod factor signal transduction pathway of Medicago truncatula. Plant Cell 13: 1835–1849[Abstract/Free Full Text]

Penmetsa RV, Cook DR (1997) A legume ethylene-insensitive mutant hyperinfected by its rhizobial symbiont. Science 275: 527–530[Abstract/Free Full Text]

Phillips AL, Ward DA, Uknes S, Appleford NEJ, Lange T, Huttly AK, Gaskin P, Graebe JE, Hedden P (1995) Isolation and expression of three gibberellin 20-oxidase cDNA clones from Arabidopsis. Plant Physiol 108: 1049–1057[Abstract]

Rademacher W (1991) Biochemical effects of plant growth retardants. In HW Gausman, ed, Plant Biochemical Regulators: Books in Soils, Plants, and the Environment. Marcel Dekker, New York, pp 169–200

Rebers M, Kaneta T, Kawaide H, Yamaguchi S, Yang Y-Y, Imai R, Sekimoto H, Kamiya Y (1999) Regulation of gibberellin biosynthesis genes during flower and early fruit development of tomato. Plant J 17: 241–250[CrossRef][Web of Science][Medline]

Reli B, Perret X, Estrada-García MT, Kopcinska J, Golinowski W, Krishnan HB, Pueppke SG, Broughton WJ (1994) Nod factors of Rhizobium are a key to the legume door. Mol Microbiol 13: 171–178[Web of Science][Medline]

Roach PL, Clifton IJ, Fülöp V, Harlos K, Barton GJ, Hajdu J, Andersson I, Schofield CJ, Baldwin JE (1995) Crystal structure of isopenicillin N synthase is the first from a new structural family of enzymes. Nature 375: 700–704[CrossRef][Medline]

Rodríguez-Llorente ID, Pérez-Hormaeche J, El Mounadi K, Dary M, Caviedes MA, Cosson V, Kondorosi A, Ratet P, Palomares AJ (2004) From pollen tubes to infection threads: recruitment of Medicago floral pectic genes for symbiosis. Plant J 39: 587–598[CrossRef][Web of Science][Medline]

Sakamoto T, Kamiya N, Ueguchi-Tanaka M, Iwahori S, Matsuoka M (2001) KNOX homeodomain protein directly suppresses the expression of a gibberellin biosynthetic gene in the tobacco shoot apical meristem. Genes Dev 15: 581–590[Abstract/Free Full Text]

Singh DP, Jermakow AM, Swain SM (2002) Gibberellins are required for seed development and pollen tube growth in Arabidopsis. Plant Cell 14: 3133–3147[Abstract/Free Full Text]

Stuart DA, Durnam DJ, Jones RL (1977) Cell elongation and cell division in elongating lettuce hypocotyl sections. Planta 135: 249–255[CrossRef]

Szczyglowski K, Amyot L (2003) Symbiosis: inventiveness by recruitment? Plant Physiol 131: 935–940[Free Full Text]

Toyomasu T, Kawaide H, Mitsuhashi W, Inoue Y, Kamiya Y (1998) Phytochrome regulates gibberellin biosynthesis during germination of photoblastic lettuce seeds. Plant Physiol 118: 1517–1523[Abstract/Free Full Text]

Tsien HC, Dreyfus BL, Schmidt EL (1983) Initial stages in the morphogenesis of nitrogen-fixing stem nodules of Sesbania rostrata. J Bacteriol 156: 888–897[Abstract/Free Full Text]

Tully RE, van Berkum P, Lovins KW, Keister DL (1998) Identification and sequencing of a cytochrome P450 gene cluster from Bradyrhizobium japonicum. Biochim Biophys Acta 1398: 243–255[Medline]

Valegård K, Terwisscha van Scheltinga AC, Lloyd MD, Hara T, Ramaswamy S, Perrakis A, Thompson A, Lee H-J, Baldwin JE, Schofield CJ, et al (1998) Structure of a cephalosporin synthase. Nature 394: 805–809[CrossRef][Medline]

Van den Eede G, Dreyfus B, Goethals K, Van Montagu M, Holsters M (1987) Identification and cloning of nodulation genes from the stem-nodulating bacterium ORS571. Mol Gen Genet 206: 291–299[CrossRef][Web of Science]

van den Heuvel KJPT, Barendse GWM, Wullems GJ (2001) Effect of gibberellic acid on cell division and cell elongation in anthers of the gibberellin deficient gib-1 mutant of tomato. Plant Biol 3: 124–131[CrossRef]

Vincent JM (1970) A Manual for the Practical Study of the Root-Nodule Bacteria, IBP Handbook No 15. Blackwell Scientific Publications, Oxford

Vogler H, Kuhlemeier C (2003) Simple hormones but complex signalling. Curr Opin Plant Biol 6: 51–56[CrossRef][Web of Science][Medline]

Vriezen WH, Zhou Z, Van Der Straeten D (2003) Regulation of submergence-induced enhanced shoot elongation in Oryza sativa L. Ann Bot (Lond) 91: 263–270[Abstract/Free Full Text]

Wu K, Li L, Gage DA, Zeevaart JAD (1996) Molecular cloning and photoperiod-regulated expression of gibberellin 20-oxidase from the long-day plant spinach. Plant Physiol 110: 547–554[Abstract]

Xu Y-L, Li L, Wu K, Peeters AJM, Gage DA, Zeevaart JAD (1995) The GA5 locus of Arabidopsis thaliana encodes a multifunctional gibberellin 20-oxidase: molecular cloning and functional expression. Proc Natl Acad Sci USA 92: 6640–6644[Abstract/Free Full Text]

Yamaguchi S, Kamiya Y (2000) Gibberellin biosynthesis: its regulation by endogenous and environmental signals. Plant Cell Physiol 41: 251–257

Zhang F, Pan B, Smith DL (1997) Application of gibberellic acid to the surface of soybean seed (Glycine max (L.) Merr.) and symbiotic nodulation, plant development, final grain and protein yield under short season conditions. Plant Soil 188: 329–335[CrossRef]

This article has been cited by other articles:

				W. Capoen, S. Goormachtig, and M. Holsters Water-tolerant legume nodulation J. Exp. Bot., November 20, 2009; (2009) erp326v1. [Abstract] [Full Text] [PDF]

				W. Capoen, J. Den Herder, J. Sun, C. Verplancke, A. De Keyser, R. De Rycke, S. Goormachtig, G. Oldroyd, and M. Holsters Calcium Spiking Patterns and the Role of the Calcium/Calmodulin-Dependent Kinase CCaMK in Lateral Root Base Nodulation of Sesbania rostrata PLANT CELL, May 1, 2009; 21(5): 1526 - 1540. [Abstract] [Full Text] [PDF]

				W. Capoen, J. Den Herder, S. Rombauts, J. De Gussem, A. De Keyser, M. Holsters, and S. Goormachtig Comparative Transcriptome Analysis Reveals Common and Specific Tags for Root Hair and Crack-Entry Invasion in Sesbania rostrata Plant Physiology, August 1, 2007; 144(4): 1878 - 1889. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 139/3/1366    most recent pp.105.066944v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Web of Science (10) Citing Articles via Google Scholar Google Scholar Articles by Lievens, S. Articles by Holsters, M. Search for Related Content PubMed PubMed Citation Articles by Lievens, S. Articles by Holsters, M. Agricola Articles by Lievens, S. Articles by Holsters, M. Related Collections Legume Biology