FR2788660A1 - Production of plants enriched in nitrogenous compounds, by overexpression of the enod40 gene and growing in symbiosis with Rhizobium - Google Patents

Abstract

Use of a transgenic plant (A) that overexpresses the enod40 gene and lives symbiotically with Rhizobium to produce a plant enriched in nitrogen-containing products (I). Independent claims are also included for the following: (1) method for enriching plants in nitrogen by growing (A) in presence of Rhizobium; (2) plants enriched in (I) produced by the method of (1); and (3) plant materials derived from the plants of (2).

Description

<Desc / Clms Page number 1>

 The present invention relates to a new use of transgenic plants overexpressing the enod40 gene and to its applications.

 The early enod40 gene, expressed during organogenesis of nodules, has been described in CRESPI et al. (EMBO Journal, 1994, 13, 21, 5099-5112), which show that enod40 is expressed in nodule differentiation cells both in spontaneous nodules and in roots treated with Nod factors and codes for an RNA (riboregulator) involved in the control of plant growth. The results from this article indicate that the transcripts of the enod40 gene are associated with the development program for the organogenesis of the nodule; in fact, after infection of the roots with Rhizobium, enod40 transcripts are detectable from the start of nodule formation, that is to say long before the bacteria reach the internal cortex. In particular, the enod40 transcripts are detected not only in the dividing cortical cells but also in the cells of the pericycle of the vascular tissue of the root, adjacent to the point of infection.

géniques ayant un taux élevé d'expression d'lenod40. La division des cellules corti- cales est plus fréquente dans la partie supérieure de la racine, en dessous l'hypocotyle. A detailed description of the preparation of plants containing an enod40 transgene (vector pGMCSe40.pG3.3) is described in CHARON et al., PNAS, 1997,94, 8901-8906; this document describes in particular that plants (Medicago truncutala) overexpressing enod40 exhibit a division of cortical cells from their roots under conditions of nitrogen deprivation. Shortening of the roots is not observed and a significant increase in the number of divisions of cortical cells per cm of root is observed in trans-

genes with a high level of expression of lenod40. The division of cortical cells is more frequent in the upper part of the root, below the hypocotyle.

 The results described in this document indicate that the expression of enod40 leads to the induction of the division of cortical cells, which represents the initial event in the development of the nodule.

 This article also highlights the complexity of the nodule formation process. However, it is clear from this article that enod40 participates, at least at the start of this process, by initiating de-differentiation and division of cortical cells.

<Desc / Clms Page number 2>

 Other articles also relate to the expression & enod40 and to its role in the organogenesis of nodules. We can cite: - CRESPI et al., 1996, in Biology of Plant-Microbe Interactions (eds. G. Stacey, B. Mullin and P.M. Gresshoff) IS-MPMI, St Paul, Minnesota, pp.

363-368), - CRESPI et al., 1997, NATO ASI Series G: Ecological Sciences, 39, 55-58, - CHARON et al., 1997, in Recent Research Developments in Plant Physiology (Ed. Pandalai, SG), Research Signpost, 1, 117-124.

 These various publications show: that there is a relationship between the induction of the division of cortical cells by enod40 and the initiation of nodulation, because this response is blocked in the presence of nitrogen in the culture medium. that the overexpression of enod40 is capable of activating certain cortical cells, by causing their division in the roots of alfalfa under conditions of nitrogen deprivation. These dividing cells express in particular enodl2A and accumulate significant amounts of amyloplasts, which suggests that the action of enod40 could be linked to the regulation of the phytohormonal balance in the internal cortical root cells, necessary for the generation of the nodule.

 In a review article, J. Cohn et al, (Trends in plant science, 1998,3, 3,105-110) summarize the various known data on the organogenesis of nodules in legumes. It is recalled, in particular that the formation of nodules is an ordered process which involves the de-differentiation of the cortical root cells and their division. This process involves the Nod factors of the bacteria, which induce the expression of early nodulins in the plant, including enod40 which is recognized, as involved in the initiation of primordia.

 The entire literature therefore shows: - that enod40 is one of the first markers expressed during the initiation of the organogenesis of the nodule and is induced by the Nod factors of bacteria,

<Desc / Clms Page number 3>

 - that transgenic plants overexpressing enod40 induce a specific cellular response in the roots, de-differentiation and division of internal cortical cells.

 However, no use on an industrial scale of transgenic plants overexpressing enod40 has been found so far.

 The present invention therefore relates to the use of a transgenic plant overexpressing the enod40 gene and living in symbiosis with a Rhizobium for obtaining a plant enriched in nitrogen products.

 Unexpectedly, plants and in particular legumes overexpressing the enod40 gene see their growth and productivity capacities significantly improved, in the presence of Rhizobium, both in the laboratory and under natural conditions.

 Indeed, it appears from the studies carried out by the inventors that the overexpression of enod40 allows, surprisingly in transgenic plants infected with a Rhizobium, earlier nodulation, which results in a significantly better nitrogen yield, in all of the organs of the plant.

In addition, there is an improved growth of the root, which leads to better use of nutrients by transgenic plants as well as better mycorrhization of legumes.

 The inventors have also found that the effect of the gene appears to be specific to the roots.

 In summary, the inventors have found that overexpression of the enod40 gene makes it possible to develop a higher potential for nodulation and root growth of legumes, cultivated under non-optimal conditions, without affecting the shoot / root metabolic balance.

 Thus, they consider that the use of leguminous plants overexpressing the enod40 gene in the field will make it possible to obtain a better agricultural yield from these plants.

 According to an advantageous embodiment of the invention, said transgenic plant is a legume.

<Desc / Clms Page number 4>

 According to an advantageous arrangement of this embodiment, said legume is Medicago truncatula.

 The present invention also relates to a process for enriching plants with nitrogen, characterized in that it comprises the cultivation of a transgenic plant overexpressing the enod40 gene, in the presence of a Rhizobium.

 The present invention also relates to plants enriched in nitrogen products, characterized in that they are capable of being obtained using the enrichment process as defined above.

 Said plants are preferably legumes, such as Medicago truncatula.

 The present invention further relates to plant material from said plants.

 By plant material is meant both seeds, flowers, leaves and possibly oils.

géniques portant la construction 35S-Mtenod-10 : . la figure 1 A représente des plantes F2 (Se40) surexprimant Mtenod40 comparées à des plantes de type sauvage 108R (contrôle) 18 jours après l'inoculation (j.a.i.) par Sinorhizobium meliloti Rm41. On remarque la longueur plus élevée de la racine primaire des plantes Se40. In addition to the foregoing provisions, the invention also comprises other provisions, which will emerge from the description which follows, which refers to examples of implementation of the process which is the subject of the present invention as well as to the accompanying drawings, which: - Figure 1 illustrates the nodulation of the M. truncatula trans-

genes carrying the construction 35S-Mtenod-10:. FIG. 1A represents plants F2 (Se40) overexpressing Mtenod40 compared to plants of wild type 108R (control) 18 days after inoculation (i) with Sinorhizobium meliloti Rm41. Note the longer length of the primary root of plants Se40.

 . Figures 1B and 1C correspond to an enlargement of Figure 1 (A) and respectively represent the roots of the control (Figure 1B) and transgenic (Figure 1C) plants carrying nodules (white arrows) at 18 d.i.i.

 . Figure 1D illustrates the number of nodules counted on different days after infection with Rm41 (20 to 40 plants were used for each point). The asterisks indicate the significant differences (P <0.01).

<Desc / Clms Page number 5>

 . Figure lE illustrates the evolution of the length of the root relative to the number of days after infection.

 - Figure 2 represents microscopic analyzes of infected roots at different stages of nodular morphogenesis:. Figures 2A and 2B represent the root tips of the control plants (Figure 2A) and Se40 (Figure 2B) 10 days after infection with a strain derived from Rm41 constitutively expressing lacZ. An increased number of infected filaments (arrows) and already developed nodular primordia were detected in Se40 plants, while few bacteria were detected in the roots of control plants. The lines represent 80 microns.

 . Figures 2C and 2D correspond to an enlargement of the root region shown in Figures 2A and 2B. The arrows (Figure 2D) indicate an infected filament and the stars indicate first divisions of cortical cells in response to bacteria. The lines represent 20 microns.

 . Figures 2E and 2F represent cross sections (8 m) through the root tip at 10 d. a.i. in a control plant (Figure 2E) and a Se40 plant (Figure 2F). The lines represent 50 microns.

 . FIG. 2G represents a longitudinal section (10 m) of the root tip of a Se40 transgenic plant at 10 d.i.i. We observe a very extensive proliferation of cortical cell layers. The line represents 20 microns.

 . Figure 2H illustrates a young nodule at 15 days. a.i. of a control plant root infected with a strain derived from Rm41 constitutively expressing lacZ. Bacteria have been detected from the invasion region to the central symbiotic region. The line represents 80 microns.

 . FIG. 21 represents young nodules and primordia of a developing nodule on a transgenic root Se40 at 12 d.a.i. inoculated with the same strain derived from Rm41. No difference was detected in the bacterial location, while the nodules are enveloped by a region of proliferating cortical cells. The line represents 70 microns.

 . FIG. 2J represents a longitudinal section (10 m) of a root tip region of a Se40 plant at 15 d. a.i. after inoculation with

<Desc / Clms Page number 6>

 same strain derived from Rm4l. We notice the presence of nodules gathered surrounded by cortical cells in division. The line represents 100 microns.

 - Figure 3 illustrates the expression of nodulin genes in Se40 transgenic plants after bacterial infection in the root region near the tip:. FIG. 3A illustrates the Northem blot analysis of the expression of the enod40 gene within the leaves (L) and the roots (R) of wild (control) and transgenic (Se40) plants.

de l'expression de Mtenodl2 et de Mtcal (enodl2 et CAI de lvfedieago truncatula), dans la région racinaire proche de la pointe au sein des plantes contrôle et transgéniques durant la nodulation. Dans tous les cas, le gène Msc27 est utilisé comme contrôle constitutif. . FIG. 3B corresponds to the semi-quantitative analysis by RT-PCR

of the expression of Mtenodl2 and Mtcal (enodl2 and CAI of lvfedieago truncatula), in the root region close to the tip within the control and transgenic plants during nodulation. In all cases, the Msc27 gene is used as a constitutive control.

 . FIG. 3C corresponds to an arbitrary quantification of the intensity of the hybridization signals obtained (A. U = arbitrary units).

 - Figures 4A - 4C represent the root region of a transgenic plant Se40.23 days after infection (i) with a mutant Rhizobium NodC strain (Figure 4A), showing different sizes of cortical cells, and with a mutant Rhizobium EPS strain ( Figures 4B and 4C), showing pooled empty nodules surrounded by cortical cells in extensive division. The line represents 50 m (Figure 4A) and 80 microns (Figures 4B and 4C).

 - Figures 4D - 4E represent the root region of a Se40 transgenic plant two days after treatment with the factor NodRmIV (S) (10-'M); they show a group of dividing cortical cells (hook in FIG. 4D, the line representing 80 microns). The use of a Nomarski optic makes it possible to clearly observe a deformation and a curvature of the root hairs (FIG. 4E, the line representing 20 m).

 - Figure 4F illustrates lateral root primordia with dividing cortical cells and divisions in the pericycle (zones of type (i)) two days after treatment with NodRmIV (S) 10-7 M. The line represents 20 m.

<Desc / Clms Page number 7>

 FIG. 4G represents the root region showing cells in anticline division in all the layers of the root cortex, with a high amount of root hairs (zones of type (ii)), two days after treatment with NodRmIV (S) 10- 7 M. The line represents 50 m.

 - Figure 4H shows an enlargement of the cluster shown in Figure 4C, by a hook. The cells of the internal cortex divide both anticlinically and periclinally (areas of type (iii)). The line represents 40 m.

 FIG. 41 represents the number of zones of type (iii) in the first cm of the root region close to the tips of the plants, two days after the treatment, with different concentrations of NodRmIV (S), of tetraacetyl-chitotetraose 10- 7 M and water, (average + SEM). The asterisks indicate statistically significant differences (P <0.05).

 - Figures 4J and 4K represent the main root length of control plants and transgenic plants Se40.23 days after treatments with different concentrations of chitotetraosis (Figure 4J) or NodRmIV (S) (Figure 4K) (mean + SEM). The asterisks indicate a statistically significant difference (P <0.01).

 - Figure 5 illustrates the control by nitrogen and ethylene of the action of enod40 in transgenic plants Se40 during nodulation:. FIG. 5A represents the number of nodules in the roots of control and transgenic plants at 18 d.a.i. The plants were grown on a nitrogen-free medium enriched with 0, 1, 2 or 5 mM potassium nitrate.

 . FIG. 5B represents plants carrying developing nodules (10 days a.i.) treated on days 0, 4, 8, 16 (T) with a solution of potassium nitrate 20 mM. The number of nodules is represented as a function of time.

 FIG. 6 represents the main root length of plants cultivated for 5 days on agar plates in a medium poor in nitrogen, enriched with: FIG. 6A: FIG. 6B: of nodules on the roots of the plants at 15 and 23 I have (Rm41) or on roots also enriched with ACC and AVG. In all cases, number of nodules: average + SEM. The root length is expressed by the mean + SEM and the asterisks indicate the significant differences

<Desc / Clms Page number 8>

 statistically significant (P <0.01).

 It should be understood, however, that these examples are given only by way of illustration of the subject of the invention, of which they do not in any way constitute a limitation.

 EXAMPLE: Behavior of Medicago truncatula transgenic plants overexpressing enod40.

1) Materials and methods - Plants used and cultivation of plants
Medicago truncatula 108R plants were previously transformed with Agrobacterium tumefaciens containing the binary vector pG3.3 which carries a 35S uidA-poly (A) 35S promoter cassette and a second 35S Mtenod40-poly (A) 35S promoter cassette (Charon and al., 1997). Medicago truncatula seeds
108R FI or F2 transgenic, as well as Medicago truncatula 108R control plants, were sterilized by immersion for one hour in a solution comprising INOV'CHLORE # (8.64 g / 1, Inov'chem SA, France) and 0.1% SDS, followed by careful washing with sterile water, overnight. The seeds are germinated vertically on 0.55% Kalys agar plates, for one day, and the seeds are transferred to magenta flasks or to square petri dishes containing vermiculite immobilized on a Gibson medium without nitrogen (Gibson and Nutman, 1960). For cultivation in petri dishes, wet vermiculite is added to half the plate, which is placed vertically. Seeds that have germinated are loaded onto the vermiculite and held almost vertically to allow root and nodular observations at different times. A week later, the roots are inoculated with S. meliloti or treated with Nod factors (see below).

 Culture conditions on vermiculite are required for nodulation tests, since the roots of M. truncatula 108R do not grow properly on agar plates for long periods (20-40 days); against, short-term hormonal treatments (5 days) are carried out by transferring the germinating seeds to new agar plates containing a medium poor in combined nitrogen, enriched with hormones at different concentrations, such as

<Desc / Clms Page number 9>

 as described in Bauer et al, 1997. The plants are grown in a culture chamber at 24 C, for a period of 16 hours, in the light.

- Nodulation tests
Strains of S. meliloti were grown overnight in TA medium with suitable antibiotics, centrifuged and resuspended in 10 mM MgSO 4, at an OD 600 of 0.4. The bacterial inoculum (300 µl) is applied around each seed. For the nodulation studies and the combined ACC / AVG treatments, the wild-type strain S. meliloti 41 (Rm41) was used. The early stages of nodulation were followed by a strain Rm41 carrying a fusion gene # -ALA :: IacZ, as described by Bauer et al., 1997. The strain S. meliloti 1021 of wild type (Rm1021) and its mutants NodC- , EPS-, Bac-Rm 1021 were used for the nodulation experiments (Bauer et al., 1997). No difference was detected between the nodulation behavior of the Rm1021 or Rm41 strains. On different days after inoculation or treatments, the roots of the plants were carefully washed to remove vermiculite, the nodules were counted and the primary root length was measured from the root point to the hypocotyle, for each plant. In each operating condition, 20 to 40 plants were used in at least three independent experiments.

 ACC and AVG treatments combined with bacterial inoculation were performed by applying a solution of ACC (1-aminocyclo-propane carboxylic acid) or AVG to semen aged one week. (aminoethoxyvinyl-glycine) (300 l at 10- 'M or 10-7 M), in the absence or concomitantly with the rhyzobial inoculum. The nodules are counted at 15 and 23 days. In each operating condition, 10 to 20 plants were used and at least 2 independent experiments were carried out.

 To study the nitrogen control of nodular initiation, the plants were cultivated in Magenta flasks containing the vermiculite immobilized on a Gibson medium without nitrogen, enriched in KNO at different concentrations (1 mM to 5 mM). 18 days after inoculation with Rhizobia, the roots are thoroughly washed and the nodules are counted. 20 to 40 plants have been used in at least

<Desc / Clms Page number 10>

 2 independent experiences. The nitrogen control of the nodulation was tested by cultivating the plants in vermiculite petri dishes (containing 10 ml of medium without nitrogen) and, 10 days after infection with S. meliloti, the roots are treated per 300 l of a 20 mM KNO solution. On days 1, 4.8 and 16, after this treatment, the roots are washed to determine the number of nodules. From 10 to 20 plants were used in at least 2 independent experiments.

- Root growth tests
Treatment with Nod factors
A purified NodRm-IV factor (CI 6: 2, S) (Schultze et al., 1992) at concentrations ranging from 10-7 to 10-10 M, tetraacetyl chitotetraose (Sigma) to 10-7 M and a aqueous controls are used. As for the rhizobial inoculation, a suspension (300 l, optical density at 600 nm: was applied around the end of the root of each seed cultivated in petri dishes with vermiculite.

Two days later, the roots of the plants, carefully washed to remove vermiculite, were fixed in FAA as described by McKhann and Hirsch, 1993.

The root tip area is examined under a microscope to detect cell divisions in each seed. In parallel, the primary root length was measured 23 days after treatment with factor Nod. In each operating condition, 10 to 20 plants were used in at least 2 independent experiments.

Hormonal treatments
After germination on the agar plates, the seeds are transferred to new agar plates containing a medium poor in nitrogen, ACC (1-aminocyclopropane carboxylic acid) at concentrations varying from 10-5 M to 10 -9 M. Five days later, the root lengths are measured. In each operating condition, 10 to 20 plants were used in at least 2 independent experiments.

- Analyzes of the expression of nodulin genes
Total RNA was prepared from the last 1.5 cm of the roots of 40 plants, excluding the root tips (1 mm), using the guanidium chloride method (Charon et al., 1997). To avoid genomic contamination

<Desc / Clms Page number 11>

of DNA, total RNA (10 g) was treated for 30 minutes at 37 C with 10 U of
DNase I, without RNase, in a volume of 24 μl and in the presence of 40 mM Tris-HCl (pH
7.5), 6 mM MgCl and 12 U of RNAguard (Pharmacia). After the heat inactivation of DNase I, the RNA is precipitated with ethanol and resuspended in water treated with DEPC.

DNase-treated RNA (1-2 g) is reverse transcribed by
50 U of MMLV H-superscript reverse transcriptase (Gibco BRL) in a reaction medium (30 l) containing 100 pmol of oligo (pdT) 12-18, a corresponding buffer (Gibco BRL), 10 mM DTT, dNTP 0 , 8 mM and RNAguard 14.4 U (Pharmacia) for 1 hour at 37 C. After the inactivation of the enzyme by heat, equal amounts of cDNA (in general, one tenth of the reaction medium) were used in
100 l of buffer 1 x Eurobio Taq (Eurobio, Les Ulis, France) containing MgCl2
1.5 mM, 0.25 mM dNTP, 200 pmol of 5 'and 3' gene primer and 1.5 U of Taq polymerase (Eurobio, Les Ulis, France). Amplification was carried out for 15-25 cycles as follows: 1 minute, 94 C; 1 minute, 55 C; 2 minutes, 72 C. Control of the RT-PCR is carried out by co-amplification of the MSc27 gene, expressed endogenously. One tenth of the PCR products are separated on a 1.2% TBE agarose gel, transferred to positively charged nylon membranes. The probes corresponding to MsCAl (Coba de la Pena et al., 1996), enodl2A (Bauer et al., 1994) and Msc27 (Gyôrgyey et al., 1991) were prepared using the DIG RNA labeling kit ( SP6 / T7) (Boehringer Mannheim, France). The hybridization and detection protocols by the CDP-star chemiluminescent substrate were carried out according to the manufacturer's instructions and the modifications described by Dessaux et al. (1995).

- Histological analysis of P-Gal activity
The roots of treated seeds of M. truncatula were harvested, immediately immobilized in FAA as described by Charon et al., (1997), and blanched for 15 minutes in a commercial solution (1.75% of active chlorine) to allow microscopic analysis. Each first centimeter of the root, close to the root tip, was screened using a magnification of 10 x 0.8 and in interference contrast (to allow good visualization of the cells) in order to

<Desc / Clms Page number 12>

 to detect groups of dividing cortical cells.

topyranoside) ou du magenta-D-Gal (5-bromo-6-chloro-3-indolyl-i-D-galactopyrano- side). Les tissus ont été blanchis comme décrit ci-dessus. The prefixation and the p-Gal stains were carried out according to Bauer et al. (1997), using XD-Gal (5-bromo-4-chloro-3-indolyl-PD-galac-

topyranoside) or magenta-D-Gal (5-bromo-6-chloro-3-indolyl-iD-galactopyranoside). The fabrics were laundered as described above.

 To make the semifine cuts, the roots were fixed and immobilized in a paraplast according to the procedure described by McKhann and Hirsch (1993), and the cuts were made with a micro-Biocut 2035 (Reicher and Jung).

For the observations, the sections have been dewaxed.

- Statistical analysis
For each operating condition in which the primary root length was measured, the average values of the measurements were given with a standard deviation (SEM). For each operating condition in which the number of nodules was counted, the mean value of the measurements was given with a standard deviation divided by the square root of the number of plants (SEM). The results, analyzed by a Student t test, are significant.

 2) RESULTS - Medicago truncatula transgenic plants overexpressing enod40 exhibit accelerated nodulation kinetics.

 Several transgenic Medicago truncatula plants overexpressing the Mtenod40 gene (construction comprising the 35S promoter cauliflower mosaic (plants Se40)) were obtained, in accordance with the technique described in Charon et al., 1997.

 Lots (20-40 plants) of F2 independent transgenic lines (in particular, from Se40 lines (Charon et al., 1997)), cultivated in the presence of vermiculite, were inoculated with Sinorhizobium meliloti Rm41 of wild type, and 18 days after infection (days or dpi, for days post infection), the plants were analyzed. Two phenotypes resulting from bacterial inoculation were detected in the transgenic plants: a higher number of nodules and an increased primary root growth (FIG. 1A). As shown in Figures 1B-1C, Se40 plants have more nodules compared to control plants (Figure 1B). In addition, the

<Desc / Clms Page number 13>

 majority of transgenic nodules appear in a region close to the root tip (Figure 1B).

 To analyze the nodulation kinetics of plants inoculated with wild-type S. meliloti Rm41, the nodules were counted on different days after the bacterial infection. In fact, the significant increase in the number of nodules at 18 a.i.i. (observed for Se40 plants) is due to an acceleration in the kinetics of nodulation (Figure 1 D). Under the operating conditions, the first nodules appear at 4-6 days a.i. on enod40 transgenic roots, while they appear at 8-10 a.i.i. on the roots of the plants control. The gap remains until the transgenic and control plants reach a maximum number of nodules.

 Significant acceleration of nodulation is confirmed using a t-test on days 18 (P <0.01) and 23 (P <0.05). At day 32, the differences are no longer significant. The root phenotype is also due to a symbiotic interaction, since no significant difference in the length of the roots is observed in the absence of bacterial inoculation, whether in the presence or in the absence of combined nitrogen in the medium. (Charon et al., 1997). During symbiosis, the average root length of the transgenic plants is low, but significantly stimulated (Figure 1E).

 A microscopic analysis of the infected roots and developing nodules was carried out after infection with a strain derived from Rm41, constitutively expressing lacZ, which allows the detection of bacteria.

A few days after infection of the enod40 transgenic roots, a blue coloration, specific for P-gal, shows that the bacteria are mainly present around the piliferous zone of root growth and have already started to penetrate the root of the plant at the level absorbent filaments (Figure 2B). On the roots of the control plants, only a few infected bacteria and filaments could be observed (Figure 2A). Thus, it seems that the transgenic roots reveal several persistent rhizobial infections giving rise to the formation of nodules in regions located at the level of the root tip (Figure 2D). In addition, at the same stage of development, the primordia are larger in the transgenic roots, as confirmed by enlargement of the piliferous zone of root growth (Figure 2D

<Desc / Clms Page number 14>

 compared to Figure 2C). Transverse and longitudinal sections of the transgenic roots enod40, at the level of the nodulation zone, confirm the presence of an increased division in the cortex, the cells dividing both anticlinically and periclinally throughout the cortex (figures 2F and 2G), and are significantly different from the root sections of the control plants, at the same nodulation stage (FIG. 2E).

 In the root pericycle, however, no increase in the number of cell divisions was observed. At a more advanced stage of nodulation, bacteria were detected from the invasion region to the central symbiotic region (Figure 2H). In the transgenic enod40 roots, the mature nodules are similar to those induced in the control plants, although a region of extensive cortical cell division envelops the developing nodules (Figures 21 and 2J). This proliferation phenotype could not be detected in the control roots. On the contrary, the nodules appearing in the upper part of the transgenic roots are similar to those observed in the control plants (Figure 2B).

 Since this accelerated nodulation phenotype is localized in the root tip region, the expression of the nodulin genes in this region has been verified by RT-PCR studies.

 An analysis of the expression of Mtenod12 (Bauer et al., 1994) and MtCAl (Coba de la Pena et al., 1997) showed that these genes are induced earlier in the transgenic roots (6 jai; FIG. 2B), confirming at a molecular level, the acceleration of the kinetics of nodulation in the roots constitutively expressing enod40.

 Thus, the Se40 transgenic plants inoculated with S. meliloti exhibit specific proliferation responses associated with the expression of enod40: increased root growth and accelerated nodulation, accompanied by extensive cortical cell divisions in the region near the tip. root. Overexpression of enod40 also results in a disruption in the positioning of dividing cortical cells during early symbiosis.

<Desc / Clms Page number 15>

 The synergy of action observed by the inventors, between the overexpression of the enod40 gene and the symbiosis, in legumes results, due to a significantly increased nitrogen fixation, a significant acceleration of nodulation as well as 'an increase in root growth (in relation to the rate of division of the apical meristem). This notably results in a significantly improved crop yield.

 - The proliferation responses induced in transgenic plants overexpressing enod40 depend on the Nod factor.

 To study the formation of blocked nodules at different stages of development, transgenic roots were inoculated: - by a Bac S. meliloti mutant unable to differentiate into bacteroids, - by an EPS S. meliloti mutant unable to invade nodules by absorbent filaments (nodules not contaminated with bacteria), and - by a NodC S meliloti mutant incapable of producing the Nod factor signals.

 23 days after infection, microscopic analyzes of the roots infected with the NodC mutant revealed only a significant increase in the number of cortical cell divisions in the transgenic roots, compared to the roots of the control plants (FIG. 4A), similar to the phenotype observed in the absence of Rhizobium (Charon et al., 1997). On the other hand, transgenic plants infected either by EPS mutants or by Bac mutants exhibit accelerated nodulation kinetics. Indeed, empty nodules appear on the transgenic roots at 20 d.i.i. with the mutant EPS, whereas they do not appear until about a week later on the roots of the control plants. In addition, pooled nodules and extensive proliferation of cortical cells are observed, as in the case of a wild-type bacterial infection (Figures 4B and C). The roots of transgenic plants infected with EPS and Bac are also longer, suggesting that the two responses to the proliferation of transgenic plants Se40 are linked to the action of the factors Nod.

<Desc / Clms Page number 16>

 To confirm this result, the effect of the purified Nod factors was tested on transgenic plants. Two days after treatment with NodRmIV (S), microscopic analysis of control plants and transgenic plants Se40 revealed root responses to factor Nod in the piliferous zone of root growth, including the deformation and curvature of the root hairs (Figure 4E ), as well as cortical cell divisions (Figure 4D). Detailed analysis of this region revealed three types of areas of dividing cells, which are distinguished by the location and orientation of the divisions. These areas are either (i) dividing cells in the pericycle and the cortex (lateral lateral putative primordia, Figure 4F), or (ii) dividing cells anticlines in all the cellular layers of the cortex, with a high quantity of hairs root (Figure 4G), or (iii) groups of cells in the internal cortex that divide both anticlinically and periclinally (putative nodular primordia, indicated by a black hook in Figures 4D and 4H). After a treatment of the transgenic roots and the roots of the control plants with a Nod factor at different concentrations, a significant increase in the number of zones of type (iii) in the first cm of the region close to the root tip was detected in the plants. transgenic overexpressing enod40 (Figure 41). No difference was found in the number of the other two types of areas. In addition, the main root length of the Se40 transgenic roots is higher than that of the control plants with treatment with factor Nod 10-'M (Figure 4K). The significance of the length difference was confirmed using a t-test (P <0.05). No significant difference was found after treatment with chitotetraosis (Figure 4J).

 These results show that the different responses observed after infection with Rhizobium in the roots of transgenic plants are due to a synergy between the factor Nod and the actions of enod40.

- Control of the action of enod40 by nitrogen and ethylene
The limited supply of a nitrogen source is essential for initiation and nodular development. It has previously been shown that overexpression of enod40 in nitrogen-deprived plants leads to increased de-differentiation and division of cortical root cells, this effect not being observed when plant growth is not limited by amount of combined nitrogen (Charon et al.,

<Desc / Clms Page number 17>

1997).

 To test whether nitrogen limitation is also required for the accelerated nodulation phenotype observed, Se40 transgenic plants were cultivated in a medium containing respectively 1 mM, 2 mM or 5 mM of potassium nitrate. The plants were then inoculated with S. meliloti for 8 days after germination, and the nodules were counted 18 days after. As can be seen from FIG. 5A, the accelerated nodulation induced by the action of enod40 is repressed by nitrogen from a concentration of 1 mM in potassium nitrate (number of nodules reduced by 50%). To test whether the action of enod40 can be regulated by nitrogen once the nodules are formed, a solution of nitrate was applied to the transgenic plants at 10 d.i.i. (plants with developing nodules). On different days after the nitrogen treatment, a count of nodules indicated that the addition of potassium nitrate blocks nodulation both in the control plants and in the Se40 transgenic plants (FIG. 5B). Eight days later, a higher number of nodules is detected, probably due to the metabolism by the plant, of added nitrogen (Figure 5B). Thus, a constitutive expression (independent of an addition of nitrogen) of enod40 does not allow plants to overcome the control, by nitrogen, of nodular development.

 As mentioned above, the inoculated Se40 transgenic plants exhibit an increased number of nodules as well as deregulation of the positioning of dividing cortical cells (Figure 2F), which could indicate a weakening of ethylene control in deregulated plants in relation to the expression of enod40. Thus, seeds were cultivated on a solidified agar medium containing little nitrogen, enriched with ACC at different concentrations, and observed five days later. For control plants and Se40 plants, AAC induced several responses, including inhibition of root growth, elongation of the hypocotyl and the formation of a hypocotyl hook. No difference was observed with respect to primary root length after treatments with different ACC concentrations (Figure 6A). Inhibition of root elongation was observed in all cases, indicating that the Medicago control and transgenic enod40 plants are sensitive to the hormone. In addition, the plants were treated by

<Desc / Clms Page number 18>

 ACC (1-aminocyclopropane carboxylic acid, immediate precursor of ethylene) or by AVG (aminoethoxyvinyl-glycine, an ACC synthase inhibitor) in the absence or concomitantly of inoculation with S. meliloti. The experiments confirmed the downregulation of ethylene-induced nodulation in M. truncatula plants, since a significant reduction in the number of nodules was observed at 15 a.i. as well as 23 days a.i.i. after treatment with 10-5 M or 10-7 M ACC, in combination with a bacterial infection (Figure 6B). Despite the repression of nodular formation by ACC in Se40 transgenic plants, their differential nodulation kinetics was maintained. On the other hand, the ethylene inhibitor AVG (10-7 M) was applied to the roots in combination with a bacterial infection. The control plants treated with AVG showed an acceleration of their kinetics of nodulation, as for the transgenic plants Se40 (FIG. 6B). However, microscopic analyzes of their roots have revealed extensive cortical cell divisions in only a few cases.

 Thus, even if the nodulation of the Se40 plants is controlled by ethylene and nitrogen and the roots are also sensitive to ACC, the action of enod40 can be partially imitated by treatment of the infected roots with an inhibitor of ethylene.

REFERENCES - Asad et al., Protoplasma, 1994, 183, 10-23.

- Bauer, P. et al., Plant Physiol., 1994, 105, 585-592.

- Bauer, P. et al., Plant J. 1996, 10, 91-105.

- Bauer, P. et al., Mol. Plant-Microbe Interact., 1997, 10, 39-49.

- Charon, C. et al., Proc. Natl. Acad. Sci. USA, 1997,94, 8901-8906.

- Coba de la Pena, T. et al., Plant J., 1997, 11, 407-420.

- Cohn, J., et al., Trends in Plant Science, 1998,3, 3, 105-110.

- Cooper, J.B., et al., Plant Cell, 1994,6, 215-225.

- Crespi, M. D., et al., EMBO J., 1994, 13, 5099-5112.

- Dehio, C., et al., Plant J., 1992, 2, 117-128.

<Desc / Clms Page number 19>

 - Dénarié, J., et al., Cell, 1993,74, 951-954.

 - Dessaux, Y., et al., Cell. Mol. Biol., 1995, 41, 933-943.

 - Fang, Y., et al., Plant Physiol., 1998, 116, 53-68.

 - Gibson, A.H., et al., Ann. Bot., 1960, 24, 420-433.

 - Gyorgyey, J., et al., Plant Mol. Biol., 1991, 16, 999-1007.

 - Heidstra, R., et al., Development, 1997,124, 1781-1787.

 - Hirsch, A.M., et al., Proc. Natl. Acad. Sci. USA, 1989,86, 1244-1248.

 - Hirsch, A.M., et al., Plant Mol. Biol., 1994, 26, 5-9.

- Journet, E. P., Plant J., 1994, 6, 241-249.

- Elmayan, T., et al., Plant J., 1996,9, 787-797.

- Kouchi, H., Mol. Gen. Genet, 1993, 238, 106-119.

- Lee, K.H., et al., Plant Physiol., 1992,100, 1759-1763.

- Lerouge, P., et al., Nature, 1990, 344, 781-784.

- Ligero, F., et al., Plant Physiol., 1991.97, 1221-1225.

- McKhann, H.I., Methods in Plant Molecular Biology and Biotechnology (Glick, - B. R. and Thompson, J.E., eds.). Boca Raton FL: Press, 1993, 179-205.

- Mathesius, U., Plant J., 1998, 14, 23-34.

- Minami, E., et al., Plant J., 1996, 10, 23-32.

- Mylona, P., et al., Plant Cell, 1995,7, 869-885.

- Penmetsa, R. V. et al., Science, 1997, 275, 527-530.

- Peters, N. K., et al., Plant Physiol, 1989, 91, 690-693.

- van de Sande, K., et al., Science, 1996, 273, 370-373.

- Schultze, M., et al., Proc. Natl. Acad. Sci. USA, 1992, 89, 192-196.

- Smit, G., et al., Plant Mol. Biol., 1995, 29, 869-873.

- Thimann, K. V., Proc. Natl. Acad. Sci. USA, 1936, 22, 511-514.

- Yang, W.C., et al., Plant J., 1993, 3, 573-585.

- Zaat, S.A.J., et al., Planta, 1989, 177, 141-150.

 As is apparent from the above, the invention is in no way limited to those of its modes of implementation, embodiment and application which have just been described more explicitly; on the contrary, it embraces all

<Desc / Clms Page number 20>

 variations which may come to the mind of the technician in the field, without departing from the scope, or the scope, of the present invention.

Claims (7)

1) Use of a transgenic plant overexpressing the enod40 gene and living in symbiosis with a Rhizobium to obtain a plant enriched in nitrogen products.  2) Use according to claim 1, characterized in that said plant is a legume.  3) Use according to claim 1 or claim 2, characterized in that said legume is Medicago truncatula.  4) Process for enriching plants with nitrogen, characterized in that it comprises the cultivation of a transgenic plant overexpressing the enod40 gene, in the presence of a Rhizobium.  5) Plants enriched in nitrogen products, characterized in that they are capable of being obtained using the method according to claim 4.  6) Plants according to claim 5, characterized in that they are, preferably, legumes.  7) Plant material, characterized in that it comes from plants according to claim 5 or claim 6.