The broad bean nodulin VfENOD18 is a member of a novel family of plant proteins with homologies to the bacterial MJ0577 superfamily.

Full-length transcript sequences were isolated from broad bean root nodules, which encode a novel nodulin designated VfENOD18. The corresponding transcripts were detected in early and in late stages of nodule development and were localized exclusively in the nitrogen-fixing zone III. The VfENOD18 sequence is not only homologous to a number of ESTs from various mono- and dicotyledonous plants, but also to the ATP-binding protein MJ0577 from Methanococcus jannaschii and to a range of bacterial proteins that belong to the MJ0577 superfamily. Hence, VfENOD18 is a member of a ubiquitous family of plant proteins that might function as ATP-binding proteins or ATPases. On the genomic level, VfENOD18 genes can be divided into two groups on the basis of differences in their 5' UTRs. One group lacks the 5' UTR region including the ATG initiation codon, whereas the second group contained the complete 5' UTR region. Further upstream of this VfENOD18 gene, a retrotransposon sequence was identified. The -14/-964 VfENOD18 promoter fragment was devoid of complete organ-specific elements known from other nodulin gene promoters. Nevertheless, this region was able to mediate full promoter activity in the central region of transgenic Vicia hirsuta root nodules.

The promoter of the Vicia faba L. gene VfEnod12 encoding an early nodulin is active in cortical cells and nodule primordia of transgenic hairy roots of Vicia hirsuta as well as in the prefixing zone II of mature transgenic V. hirsuta root nodules.

A full-length cDNA encoding the Vicia faba L. early nodulin VfEnod12 was isolated. The deduced protein sequence specified a 90 amino acid protein with a MW of 10206 and contained a putative signal peptide sequence followed by PPX(3) repeats characteristic of Enod12 proteins. The VfEnod12 gene was found to be expressed specifically in root nodules as early as 3 days post inoculation with Rhizobium leguminosarum bv. viciae. In mature nodules, VfEnod12 transcripts were confined to the prefixing zone II. A 3.3 kb genomic fragment carrying the complete VfEnod12 coding region was isolated. No intervening sequences were identified in the coding region. A promoter fragment carrying the -692/-41 region mediated reporter gene expression in root cortical cells, nodule primordia and the prefixing zone II of transgenic Vicia hirsuta root nodules. This fragment contained a putative binding site for the transcription factor ENBP1. In contrast to the highly conserved terminal AATAA motif of the ENBP1 binding site of known Enod12 promoters, the VfEnod12 promoter was characterized by an altered terminal AATAT sequence. This alteration did not interfere with VfEnod12 promoter activity in transgenic roots and nodules of V. hirsuta.

Genomic organization and expression properties of the MtSucS1 gene, which encodes a nodule-enhanced sucrose synthase in the model legume Medicago truncatula.

We have isolated and sequenced a sucrose synthase (SucS) cDNA from the model legume Medicago truncatula. This cDNA (MtSucS1) contains an ORF of 2418 bp, coding for a protein of 805 amino acids with a molecular mass of 92.29 kDa. The deduced amino acid sequence shows significant homology to other plant sucrose synthases, in particular to the nodule-enhanced sucrose synthases from pea and broad bean. Northern analysis revealed that the corresponding gene shows a ten-fold higher expression level in root nodules than in uninfected root, stem and leaf tissues. SucS protein was detected in root nodules from a variety of legumes, including M. truncatula. Whereas only one SucS isoform was detectable in root nodules, an additional sucrose synthase of slightly larger molecular weight was present in uninfected root, stem and flower tissues of M. truncatula. From our expression and sequence data we infer that the MtSucS1 gene encodes a nodule-enhanced sucrose synthase in M. truncatula. Southern hybridization data indicate that MtSucS1 is a single-copy gene. An analysis of a genomic MtSucS1 sequence revealed that the gene consists of 14 exons with the start codon being located on exon II. As is common for SucS genes, the MtSucS1 gene contains a large intron of 747 bp in the 5' untranslated region. The transcriptional start of MtSucS1 was mapped and putative regulatory elements in the MtSucS1 promoter were identified.

The temporal and spatial transcription pattern in root nodules of Vicia faba nodulin genes encoding glycine-rich proteins.

Four different transcript sequences encoding gene products with an unusually high glycine content were identified in Vicia faba root nodules. Northern blot analysis revealed a strong nodule specific expression of the corresponding genes. Time course experiments showed that two of these genes were transcribed before the onset of leghemoglobin expression and hence were designated VfENOD-GRP2 and VfENOD-GRP5, whereas the first detection of VfNOD-GRP1 and VfNOD-GRP4 transcripts coincided with the appearance of leghemoglobin transcripts in V. faba root nodules. A characteristic feature of all encoded nodulins was a hydrophobic N-terminus, which in the case of the nodulins ENOD-GRP2 and ENOD-GRP5 has the characteristics of a signal peptide. Such a structure is comparable to other plant glycine-rich proteins decribed as components of the plant cell wall. Based on tissue print hybridizations, we found that VfNOD-GRP1, VfENOD-GRP2 and VfNOD-GRP4 were expressed in the interzone II-III and in the whole nitrogen-fixing zone III. In contrast to VfENOD-GRP2 and VfNOD-GRP4, the signal intensity of hybridizing VfNOD-GRP1 transcripts was slightly reduced in the more proximal part of broad bean root nodules. Apart from the interzone II-III and the nitrogen fixing zone III, VfENOD-GRP5 RNA was also detected in large areas of the prefixing zone II.

The Vicia faba leghemoglobin gene VfLb29 is induced in root nodules and in roots colonized by the arbuscular mycorrhizal fungus Glomus fasciculatum.

To investigate similarities between symbiotic interactions of broad bean (Vicia faba) with rhizobia and mycorrhizal fungi, plant gene expression induced by both microsymbionts was compared. We demonstrated the exclusive expression of 19 broad bean genes, including VfENOD2, VfENOD5, VfENOD12 and three different leghemoglobin genes, in root nodules. In contrast, the leghemoglobin gene VfLb29 was found to be induced not only in root nodules, but also in broad bean roots colonized by the mycorrhizal fungus Glomus fasciculatum. In uninfected roots, none of the 20 nodulin transcripts investigated was detectable. VfLb29 has an unusually low sequence homology with all other broad bean leghemoglobins as well as with leghemoglobins from other legumes. It can be regarded as a novel kind of leghemoglobin gene not described until now and the induction of which is common to symbiotic interactions of broad bean with both Rhizobium and a mycorrhizal fungus.

The Vicia faba lipoxygenase gene VfLOX1 is expressed in the root nodule parenchyma.

A full-length cDNA encoding the broad bean lipoxygenase VfLOX1 was isolated from a nodule cDNA library. The VfLOX1 gene was strongly expressed in nodules, and only weakly in roots. VfLOX1 transcripts were localized in the nodule parenchyma and in the cells surrounding the root stele.

The modular nodulins Nvf-28/32 of broad bean (Vicia faba L.): alternative exon combinations account for different modular structures.

The broad bean late nodulins, Nvf-28/32, are composed of two types of repetitively occurring sequence modules flanked by unique N- and C-terminal modules. Six isoforms of these nodulins were characterized by a specific modular structure resulting from a different individual order of repetitive sequence modules. A detailed analysis of genomic PCR fragments revealed that the repetitive modules and the N-terminal unique module exactly corresponded to exons, whereas the C-terminal module was specified by two exons. Since those exons encoding the repetitive modules missing in specific Nvf-28/32 isoforms were consistently present within genomic sequences, a post-transcriptional generation of VfNOD28/32 transcripts specifying six Nvf-28/32 nodulins was concluded. Using tissue-print hybridizations, these transcripts were localized in the interzone II-III and the nitrogen-fixing zone III of root nodules. From this and from cDNA-cDNA hybridizations demonstrating a comparable timing of expression of VfNOD28/32 and of leghemoglobin transcripts in root nodules, a function of the modular nodulins Nvf-28/32 in late developmental stages of broad bean nodules was inferred.

Male sterility in transgenic tobacco plants induced by tapetum-specific deacetylation of the externally applied non-toxic compound N-acetyl-L-phosphinothricin.

A system for the inducible destruction of plant tissues based on the deacetylation of the non-toxic compound N-acetyl-L-phosphinothricin (N-ac-Pt) has been developed. The argE gene product of Escherichia coli, representing a N-acetyl-L-ornithine deacetylase was identified to remove the acetyl-group from N-ac-Pt giving the cytotoxic compound L-phosphinothricin (Pt, glufosinate). Transgenic Nicotiana tabacum plants constitutively expressing the argE gene were constructed. No effect of the bacterial N-acetyl-L-ornithine deacetylase on plant growth and reproduction could be traced. However, application of N-ac-Pt on leaves of the transgenic plants led to the formation of necrotic areas due to the release of Pt. Additionally, due to the uptake of the N-ac-Pt by roots, transgenic shoots grown on medium containing N-ac-Pt bleached within 6-7 days and finally died. Untransformed controls showed no reaction to high amounts of N-ac-Pt applied, either under sterile or under unsterile conditions. In order to construct inducible male-sterile plants, the argE coding region was fused to a DNA fragment carrying sequences homologous to the tobacco TA29 promoter, known to function exclusively in the tapetum. Owing to the tapetum-specific expression of the chimeric gene the application of N-ac-Pt led to empty anthers resulting in male-sterile plants. The sanity of the female reproductive part of the male-sterile flowers could be demonstrated by cross-pollination. Without N-ac-Pt treatment the plants turned out to be completely fertile making fertility restoration in the F1 generation superfluous. The system presented is easy to handle and might be applicable to a wide range of crop plants.

The broad bean gene VfNOD32 encodes a nodulin with sequence similarities to chitinases that is homologous to (alpha/beta)8-barrel-type seed proteins.

Nodulin gene transcripts isolated from a broad bean (Vicia faba L.) root nodule cDNA library and designated VfNOD32 are detectable in the nitrogen-fixing zone III of nodules and in much smaller amounts in flowers. In nodules, these transcripts are detectable for the first time 7 d after inoculation, at least 1 d before leghemoglobin gene transcription starts. Two putative full-length cDNAs representing different transcript sequences of 92.5% identity were sequenced. The corresponding broad bean genes were termed VfNOD32-A1 and VfNOD32-A2, and the encoded proteins were termed Nvf32-A1 and Nvf32-A2. The derived amino acid sequences of the Nvf32 proteins are highly homologous to the Vicia narbonensis (alpha/beta)8-barrel seed protein narbonin. Considering this homology, Nvf32 is assumed to have a similar structure consisting of beta-sheets forming a central barrel surrounded by alpha-helices. The two Nvf32 sequences also contain two conserved amino acid motifs that are characteristic of class-III chitinases. Several amino acids demonstrated to be essential for chitinase activity are conserved in both regions, whereas one essential glutamic acid was changed to glycine in the Nvf32-A1 isoform but not in the Nvf32-A2 isoform.

The promoter of the Vicia faba L. VfENOD-GRP3 gene encoding a glycine-rich early nodulin mediates a predominant gene expression in the interzone II-III region of transgenic Vicia hirsuta root nodules.

We recently reported on the broad bean gene VfENOD-GRP3 encoding a glycine-rich early nodulin. This gene was predominantly expressed in the interzone II-III region of Vicia faba root nodules. The VfENOD-GRP3 promoter contained several sequence motifs potentially involved in the regulation of gene expression. To investigate the molecular basis for the specific VfENOD-GRP3 expression, defined VfENOD-GRP3 promoter fragments were fused to an intron-containing gusAint gene. Agrobacterium rhizogenes ARqual strains carrying these fusions integrated into the TL DNA were used to generate hairy roots on Vicia hirsuta, which subsequently were nodulated. Histochemical analysis of transgenic nodules indicated that a strong gusAint expression in the interzone II-III region was mediated by the -1252/+10 VfENOD-GRP3 promoter region. This reporter gene expression in V. hirsuta was comparable to the location of VfENOD-GRP3 transcripts in V. faba nodules. An analysis of defined promoter fragments revealed that a strong gusAint expression in the interzone II-III region was also mediated by the -737/+10 promoter, whereas the -239/+10 promoter only mediated a weak gusAint expression in the interzone II-III region. Since the -239/+10 promoter fragment did not resemble published nodulin gene promoters, we propose that it contains new sequence motifs involved in mediating gene expression in the interzone II-III region of Vicia nodules.

The nodule-specific VfENOD-GRP3 gene encoding a glycine-rich early nodulin is located on chromosome I of Vicia faba L. and is predominantly expressed in the interzone II-III of root nodules.

A nodule-specific cDNA was isolated from a Vicia faba L. nodule cDNA library. Since time course experiments revealed an early expression of this transcript in the nodule, this cDNA coded for an early nodulin and was designated VfENOD-GRP3. Based on tissue print hybridizations, we found a predominant expression of VfENOD-GRP3 transcripts in the interzone II-III region of broad bean root nodules. The encoded early nodulin ENOD-GRP3 was characterized by an N-terminal signal peptide and a C-terminal domain displaying a glycine content of 31%. Sequence analysis of a genomic VfENOD-GRP3 clone revealed that the signal peptide and the glycine-rich domain were specified by two separate exons. Primer extension experiments identified two adjacent transcription start sites for VfENOD-GRP3 transcripts. The common nodulin sequences 'AAAGAT' and 'CTCTT' were present five and three times on both DNA strands of the putative VfENOD-GRP3 promoter, respectively. Additionally, three sequence motifs resembling organ-specific elements of the soybean lbc3 gene promoter and a sequence similar to the binding site 1 for the nodule trans-acting factor Nat2 were identified. From Southern blot data and from sequence analysis of genomic PCR fragments, the presence of a VfENOD-GRP3 gene family was inferred. By PCR experiments using sequence-specific primers and DNA of microisolated chromosomes as a template, this family was located on the long arm of chromosome I.

Members of a broadbean nodulin family with partial homologies to the alfalfa nodulin 25 are composed of two types of amino acid repeats flanked by unique amino acid sequence termini.

Five cross-hybridizing cDNAs from clone group 'VfNDS-L' of a broadbean nodule-specific cDNA library differed by five specific in-frame deletions within the coding region. Northern blot analysis revealed that the transcripts represented by these clones were expressed in a nodule-specific way and therefore encoded a new family of broadbean nodulins. These nodulins have been designated Nvf-28/32. The Nvf-28/32 proteins were between 269 and 299 amino acids long with a high proportion of charged amino acids and contained a putative signal peptide of 20 amino acids. Sequence analysis indicated that the central part of the Nvf-28/32 proteins was composed of two different types of repeating amino acid stretches designated 'repeat 1' and 'repeat 2', whereas the N- and C-termini were unique. In contrast to 'repeat 1' stretches, which contained both positively and negatively charged amino acid residues, 'repeat 2' sequences did not contain any positively, but a total of 13 negatively charged amino acid residues. We could demonstrate that the five deletions identified exactly corresponded to complete repeats. The unique sequence termini of the Nvf-28/32 proteins displayed strong homologies to the late nodulin 25 from alfalfa. In addition, the repeating units identified were significantly homologous to several, but not all exon parts of this protein. We speculate that the 'VfNDS-L' transcripts were derived from a differential splicing mechanism and that the Nvf-28/32 proteins fulfil a structural rather than an enzymatic function within the broadbean root nodule.

A survey of transcripts expressed specifically in root nodules of broadbean (Vicia faba L.).

More than 600 potentially nodule-specific clones have been detected by differential hybridization of a broadbean cDNA library constructed from root nodule poly(A)+ RNA. These isolated cDNAs belong to at least 28 different clone groups containing cross-hybridizing sequences. The number of clones within a clone group varies from about 200 to only one single clone. Northern hybridization experiments revealed nodule-specific transcripts for 14 clone groups and markedly nodule-enhanced transcripts for another 7 clone groups. Sequence homologies indicate that three transcript sequences code for different leghemoglobins. Two other transcripts encode a nodule-specific sucrose synthase and a nodule-enhanced asparagine synthetase, respectively. Four deduced gene products are proline-rich, two of them being the homologues of PsENOD2 and PsENOD12. The third proline-rich protein (PRP) is composed of similar amino acid repeats as the nodule-specific PsENOD12 but is expressed in nodules and roots in comparable amounts. The fourth PRP is a nodule-enhanced extensin-type protein built up by Ser-Pro4 repeats. Two further nodule-specific transcripts encode gene products showing some similarity to structural glycine-rich proteins. Additionally, transcripts could be identified for broadbean homologues of the nodulins MsNOD25, PsENOD3 and PsENOD5 and transcripts specifying a nodule-enhanced lipoxygenase and a translation elongation factor EF-1 alpha, which is expressed in all broadbean tissues tested.

The sucrose synthase gene is predominantly expressed in the root nodule tissue of Vicia faba.

To analyze nodule-specific gene expression in broadbean, we have isolated and sequenced sucrose synthase (SUCS) cDNAs from a broadbean nodule-specific cDNA library. The most 5' sequences identified from these partial cDNAs were used as a molecular probe to isolate a full-length sucrose synthase transcript sequence from a cDNA library derived from broadbean nodule mRNA. This cDNA (VfSUCS) contained a reading frame of 2,418 bp, coding for a protein of 806 amino acids with a deduced molecular weight of 92.5 kDa. The DNA as well as the deduced amino acid sequence displayed substantial homologies (68-95%) to other plant SUCS sequences. Northern and RNA dot blot experiments demonstrated that this gene is strongly expressed in the broadbean nodule tissue. An at least 10-fold lower VfSUCS expression could be detected in the uninfected root, hypocotyl, stem, and flower tissues of broadbean, whereas only traces of VfSUCS transcripts were recognizable in the broadbean leaf tissues. VfSUCS transcripts could not be detected in mature seeds of broadbean. Because of this significantly nodule-amplified type of expression, we refer to VfSUCS as a nodulin gene and propose to designate it VfNOD93 (Nuf-93) for the sucrose synthase enzyme).