Results and Discussion

Genetic mapping of the non-nodulation trait was started with the identification of RAPD markers using the bulked segregant analysis method described by Michelmore et al. (1991). Using isolated DNA from five Nod" and five Nod+ plants, respectively, four RAPD markers were identified and mapped in an extended tetraploid population in proportion to the nodulation phenotype. These Nod-linked RAPD markers were then mapped in the diploid alfalfa mapping population on linkage group (LG) 5. From this genetic region linked RFLP markers were selected and re-mapped in the tetraploid population. These RFLP markers mapped to the exact same region in the tetraploid map as the RAPD markers and the Nod" phenotype. The summary of this mapping work is shown in Figure 1.

Nod" Bulk

Nod* Bulk mi mi mi mi mi

RAPD OPW8a

RAPD OPE8c

RAPD OPB13b

RAPD OPA6a mi mi mi mi mi

RAPD OPW8a

OPW8a

OPE8c

Nod" OPB13b

OPA6a

Tetraploid alfalfa

OPW8a

OPE8c U071

U0224

CG13, Tm, EOK U0492

OPB13b OPA6a

OPE8c U071

U0224

CG13

U0492

Diploid alfalfa

Tetraploid alfalfa

Figure 1. Genetic mapping of the Nod" trait in tetraploid alfalfa (see text for explanation).

Two RFLP markers tightly linked to the non-nodulation trait were used to isolate primary BAC clones from the Medicago truncatula BAC library (Nam et al. 1999). The end sequences of these primary BACs were used to identify additional BAC clones. Restriction endonuclease mapping (BAC fingerprinting) was used to orient the isolated 10 BAC clones and construct the so-called Nod-contig, which overlaps more than 600 kb DNA sequence. Genetic markers generated from these BAC clones were used to map the Nod" mutation more precisely within the contig.

The Nod-region (-200 kb) covering the Nod" mutation (nn {) was sequenced and the gene content of the region was determined. Twenty-six genes located in this region were used to search for similar gene content in the Arabidopsis thaliana genome. This analysis revealed partial synteny between the two species: one particular gene (see ORF8 in Figure 2) in this region was present as a one-copy gene in the genome of the Medicago species (in M.t. and M.s. as confirmed by DNA-DNA hybridization and RFLP mapping of this gene), while it was a two-copy gene in the Arabidopsis genome (see gene A and B in Figure 2). None of the genes in the flanking region of gene A was present in the Nod-region of M. truncatula. On the other hand, 9 out of the 26 genes were present in the vicinity of gene B. Some genes of the Nod-region were present elsewhere in the Arabidopsis genome with a configuration of same order and orientation. Interestingly, one gene in the Nod-contig (see ORF7 in Figure 2) was not present in the Arabidopsis genome according to any BLAST search available in the NCBI web site. From this finding it is concluded, that Medicago and Arabidopsis share limited microsysteny in certain locations of their genome (see Figure 2).

9 10 1112 13 14 15 16 17 1 19 20 212223

M.t. Nod contig

ORF5 ORF8

ORF7 ORF9

XZ3033Q: B

13'i 41'516'' 17-181920 2122 2

Figure 2. Comparison of the gene content of the Nod-region of Medicago truncatula with specific regions of the Arabidopsis genome (see text for explanation).

The non-nodulation phenotype of MN-1008 could be explained by potential mutations in the genes coded by ORF2, ORF5, ORF7 and ORF9 (see Figure 2). To search for possible genetic alterations in these ORFs, these genes were sequenced from the MN-1008 mutant plant material. Genomic DNA as well as RT-PCR-produced root cDNA template DNA was used in amplifications with exon-specific primer pairs designed according to the available sequence information. Amplified fragments were cloned into pUC19 vector and the inserts were sequenced in ABI377 automatic sequencer using the BigDye labeling procedure. Similarly, DNA templates originating from M. truncatula R38 mutant, a Nod" but Myc+ dmi2 allele (K. VandenBosch, personal communication), and Pisum sativum wild type (Frisson), and P4 and P55 mutants (Nod", Myc" syml9 alleles, Schneider et al. 1999) were similarly sequenced. According to the obtained sequence information, mutation in ORF7 of each mutant plant could be identified. The nature of the mutations either generated a stop codon (nni) evoking premature translation termination or in two cases (R38 and P4) changed a highly conserved amino acid into a dissimilar residue.

The sequence of ORF7 is similar to receptor kinase genes identified in plants and therefore it was designated as NORK (NOdulation specific Receptor Kinase). The NORK gene of the above plant species contain 925 amino acid residues (as deduced from the cDNA sequence) with an N-terminal signal sequence. There is a transmembrane domain-like region in the middle of the molecule, therefore it is supposed that the N-terminal part of the protein is extracellular containing leucine rich repeats, while the C-terminal domain is intracellular coding for serine-threonine kinase sequences with conserved ATP binding sites and kinase active site.

Using the DNA region coding for the postulated extracellular part as the hybridization probe, distinct specific hybridizing bands could be detected in each legume species tested (Desmodium, Glycine, Macroptilium, Medicago, Melilotus, Phaseolus, Trifolium, Sesbani, Vicia, Vigna) even in Cassia emerginata, a legume in which symbiotic nodule formation capability had not evolved during evolution. On the other hand, representative members of non-legume plants (maize, tobacco, wheat, rice) did not dispose specific hybridization signals.

4. Concluding Remarks

The similar nodulation phenotypes of the alfalfa and pea mutants (Nod-, Mac-, no Ca2+ spiking, etc.), the similar map position of the mutations conditioning non-nodulation phenotype, and the mutation found in each NORK gene originating from the non-nodulating plants make NORK a promising candidate gene which must play a key role in the initiation of the signal transducing cascade leading to symbiotic nodule development by binding directly or indirectly the specific the Nod factor of Sinorhizobium meliloti. Transformation experiments are in progress to complement the mutations and to demonstrate the direct correlation between the non-nodulating phenotype and the mutation in the NORK gene.

5. References

Caetano-Annoles G et al. (1993) In Palacios R, Mora J, Newton WE (eds), New Horizons in

Nitrogen Fixation, pp. 297-302, Kluwer Academic Publisher, Dordrecht, The Netherlands Endre et al. (1996) Theor. Appl. Genet 93,1061-1065 Ehrhardt et al. (1996) Cell 85, 673-681 Felleetal. (1996) Plant J.10, 295-301 Dudley ME, Long SR (1989) Plant Cell 1, 65-72 Hirsch AM et al. (1997) Plant and Soil 194, 171-184 Kalo et al. (2000) Theor. Appl. Genet. 100, 641-657 Kiss GB et al. (1998) Acta Biol. Hung. 49, 125-142 Michelmore RW et al. (1991) Proc. Natl. Acad. Sci. USA 88, 9828-9832 Nam YW et al. (1999) Theor. Appl. Genet. 98, 638-646 Peterson, Barnes (1981) Crop Sci. 21, 611-616 Schneider et al. (1999) Mol. Gen. Genet. 262, 1-11 Truchet G et al. (1989) Mol. Gen. Genet. 219, 65-68

6. Acknowledgements

We are particularly indebted to K. VandenBosch and G. Due for providing non-nodulating lines of M. truncatula R38 and P. sativum P4/P55, respectively, and to D. Cook for providing the M. truncatula BAC library. This work was supported by the BRC, Szeged, Hungary and by grants from the MTA (AKP 96 -360/62, and AKP 00-246/35), from the OTKA (F030408, T025467), from the OMFB (EU-97-D8-063), from the NKFP (Grant No. Medicago Genomics 4/023/2001, and from the European Union (EuDicotMap Grant No. BI04 CT97 2170, and Medicago Grant No. QLTR-2000-30676).

Was this article helpful?

0 0

Post a comment