Infection threads are absent from all of the pea mutants described above (Walker et al. 2000), and are greatly reduced in lines of pea homozygous for syminoculated with a strain lacking nodX. However normal infections are induced if R. I. viciae carrying nodX is inoculated onto sym/1 lines of pea (Geurts et al. 1997). This observation, together with analysis of infection of Medicago spp. by mutants of Sinorhizobium meliloti (Ardourel et al. 1994), demonstrates a role for specific Nod factor structures in initiation of infections. We have examined infection events induced by nod mutants of R. I. viciae and the effect of Cnb+ strains on infections in cv. Afghanistan, which is homozygous for sym
3.1. nodE or nodO required for infection thread growth but not for entry into root hairs. In the absence of nodE, the nodO gene is necessary for nodulation of pea and the vetch Vicia hirsuta (Economou et al. 1994). To assay where nodulation is blocked, nodO, nodE and nodO-nodE (double) mutants were marked with a constitutively-expressed lacZ gene and infection events were assayed by staining with X-gal. Normal infections were induced by the nodO mutant and the nodE mutant on both vetch and pea roots. These infection events could not be distinguished from those induced by a wild-type strain. However, with the nodO-nodE double mutant a strikingly different result was observed. Normally pea inoculated with a wild-type strain induces an average of 28 (± 6) infection threads per root section under the assay conditions used. These are found only very infrequently with the nodO-nodE double mutant (average 0.7 +/- 0.3 per root section). Instead, the mutant was found to have induced aberrant infection foci (average 24.3 +/- 3 per root section), that did not develop into infection threads. Such infection foci lacking infection threads were very rarely found on roots inoculated with the wild-type (average 0.6 +/- 0.3 per root section). This indicates that in the absence of nodO and nodE, entry can be made into root hairs. However infection thread growth can be stimulated by nodO or nodE, because if either gene is restored to the double mutant, normal infection threads are formed.
In vetch an analogous situation is seen (Walker, Downie 2000). Thus, normal infections are seen with nodE or nodO mutants but abnormal infection events are seen with the nodO-nodE double mutant. In vetch the number of infections induced by wild type averages about 130 and this is reduced by about two orders of magnitude in a nodO-nodE double mutant. Instead there are in excess of 700 aberrant infections in which bacteria are located within root hairs but normal infection threads are not seen in these infected root hairs (Walker, Downie 2000). Again, restoring either nodO or nodE restores normal infection. This implies that the pore-forming NodO protein may stimulate infection by forming an ion channel in the plant plasma membrane, where the plant cell wall has been degraded to enable the bacteria to make entry. Alternatively, if the host specific Nod factors (i.e. those carrying the «oiffi-determined C18:4 group) are made, infection threads can be established in the absence of NodO. This suggests that bacterial entry into root hairs requires minimal Nod factor structure, but that subsequent establishment of infection threads requires activation of some additional pathway.
3.2. Competitive nodulation blocking in cv. Afghanistan pea occurs at the infection stage. R. I. viciae strain TOM infects and nodulates cv. Afghanistan pea normally, but some strains of R. I .viciae (such as strain A34) competitively inhibit nodulation. Strain TOM was marked with a constitutively-expressed lacZ gene and infection events on cv. Afghanistan roots were assayed in the presence or absence of R. I. viciae A34 which is strongly Cnb+. Co-inoculation with A34 caused about a thirty-fold reduction in the number of infection threads formed by strain TOM. Previous work has demonstrated that the Cnb phenotype requires the host specific Nod factor structure, because nodE and nodL mutants of A34 are Cnb" (Dowling et al. 1989; Firmin et al. 1993).
One characteristic of the competitive nodulation blocking strain A34 is that it makes high levels of Nod factors, whereas strain TOM makes very low levels. These low levels are due in part to the presence of the nolR gene in strain TOM. NolR is predicted to be a repressor and it decreases expression of the nodABC operon and overall levels of Nod factors (Kiss et al. 1998). The cloned nolR gene was transferred to strain A34, and the resulting strain was greatly reduced for competitive nodulation blocking on cv. Afghanistan peas. This suggests that Cnb is due to high levels of Nod factor production.
Using a purified preparation of Nod factors from strain A34, it was possible to block nodulation of cv. Afghanistan by strain TOM, and this was correlated with a thirty-fold decrease in infection thread formation. Interestingly, a purified Nod factor preparation from a strain expressing the TOM nod genes was also found to inhibit nodulation of cv. Afghanistan by strain TOM. Therefore we conclude that if too much Nod factor is present, nodulation inhibition can occur with cv. Afghanistan, and it is the high levels of Nod factor, rather than the absence of the NodX-determined acetyl group, that determines nodulation blocking.
Nod factors with minimal structure (lacking host-specific decorations) can enable strains of R. I. viciae to enter root hairs, but host specific decorations (or alternatively the pore-forming protein NodO) are required for normal infection thread growth. Host-specific decorations are also required for competitive nodulation blocking, which occurs at the level of infection thread initiation. Taken together, these observations suggest that there are quantitative and qualitative differences in the recognition of undecorated or host-specifically decorated Nod factors. The mechanism by which the plant makes this discrimination has yet to be determined.
Ardourel M et al. (1994) Plant Cell 6,1357-1374
Davis EO etal. (1988) Mol. Gen. Genet. 212, 531-535
Dowling DN etal. (1987) J. Bacteriol. 169,1345-1348
Dowling DN et al. (1989) Mol. Gen. Genet. 216, 170-174
Economou A etal. (1994) Microbiol. 140, 2341-2347
Ehrhardt DW et al. (1996) Cell 85, 673-681
Engvild KC (1987) Theor. Appl. Genet. 74, 711-713
Firmin JL (1993) Mol. Microbiol. 10, 351-360
Geurts R et al. (1997) Plant Physiol. 115, 351-359
Kiss E et al. (1998) Mol. Plant-Microbe Interact. 11, 1186-1195
Markwei CM, LaRue TA (1992) Can. J. Microbiol. 38, 548-554
Ovtsyna AO et al. (1998) Mol. Plant-Microbe Interact. 11, 418-422
Schneider A et al. (1999) Mol. Gen. Genet. 262,1-11
Spaink HP (1991) Nature 354,125-130
Sutton JM et al. (1994) Proc. Natl. Acad. Sci. USA 91, 9990-9994 Tsyganov VE et al. (1998) Mol. Gen. Genet. 259, 491-503 Walker SA etal. (2000) Proc. Natl. Acad. Sci. USA 97, 13413-13418 Walker SA, Downie JA (2000) Mol. Plant-Microbe Interact. 13, 754-762
We thank Karen Wilson for preparing Nod factors, and H. Spaink and E. Kondorosi for kindly giving us plasmids carrying nodX and nolR. This work was supported principally by the Biotechnology and Biological Sciences Research Council, with additional support from the European Union (Contract FMRX-CT98-0243).
Section 5: Developmental Biology
LEGUME PLANT NITROGEN FIXATION SYMBIOTIC MUTANTS - A PROMISING TOOL FOR UNDERSTANDING FUNDAMENTAL PROCESSES AND MOLECULAR BASES OF PLANT DEVELOPMENT
All-Russia Research Institute for Agricultural Microbiology, St Petersburg, Pushkin 8, Podbelsky chaussee, 3,196608, Russia
Plant-microbe interactions provide a unique model based on the signaling process during nodule formation. The process of induction of nodule formation on the roots of legumes requires reorganization of a plant developmental program. Details of such reorganization remain unclear. The most exciting phenomenon is the action of a Nod factor in plant development. Among the problems to be solved in the connection with Nod factor action, the problem of receptor molecules should be treated as the most urgent. Also the intriguing data about the possible role of Nod factor during late stages of nodule development have been recently obtained (Timmers et al. 1998; V.A. Voroshilova et al. unpublished). We still do not understand what is the way of the regulation of the nodule formation genetic system of higher plants under the bacterial influences.
There are two sets of plant genes involved in symbiosis: nodulin-genes and sym-genes. Sym genes are thought to be good candidates for plant genes regulating the process of nodule development. To date the collections of legume mutants have been obtained and the terminus of sym-genes actions have been determined. Such data allow to describe the process of nodule formation as sequential functioning of the blocks of genes from both partners.
The highest number of legume plant symbiotic genes has been identified in pea (Pisum sativum L.) as a result of all efforts in the field of isolation of pea symbiotic mutants with abnormalities of nodule formation and function throughout the world. More than two hundred independently obtained symbiotic mutant lines are known to date. More than one hundred isolated symbiotic mutants have been isolated in various labs, and more than forty pea symbiotic genes have been identified to date (reviewed in Borisov et al. 2000). In parallel a majority of genetically characterized mutants have been characterized to identify nodule developmental stages (reviewed in Borisov et al. 2000). This characterization allowed subdivision of nodule morphogenesis into eight discrete developmental stages. These results have made it necessary to modify the previously used system of phenotypic codes describing the process of symbiotic nodule development (Vincent 1980; Caetano-Anolles, Gresshoff, Tsyganov et al. 1998). To date, the sequence of nodule developmental stages is defined as follows: (i) root hair curling (Hac), (ii) infection thread growth initiation (Iti), (iii) infection thread growth inside root hair (Ith), (iv) infection thread growth inside root tissue (Itr), (v) infection thread growth inside nodule tissue (Itn), (vi) infection droplet differentiation (Idd), (vii) bacteroid differentiation (Bad) and nodule persistence (Nop) (reviewed in Borisov et al. 2000).
Among the identified plant symbiotic genes there should be genes involved in the recognition and processing of a Nod factor. The molecular products of those genes and other sym genes, which are regulatory proteins, have been revealed with the use of the two-hybrid system (reviewed in Brent, Finley 1997) in proteomic analysis. The combination of this method with the mutational analysis of nodule formation provides a powerful tool for understanding the molecular bases of plant development as well as the process involving prokaryotes and eukaryotes into this highly integrated system, the nitrogen-fixing symbiotic nodule.
Borisov AY, Barmicheva EM, Jacobi LM, Tsyganov VE, Voroshilova VA, Tikhonovich IA (2000)
Czech J. of Gen. Plant Breeding 36,106-110 Brent R, Finley Jr. RL (1997) Annu. Rev. Genet. 31, 663-704 Caetano-Anolles G, Gresshoff PM (1991) Annu. Rev. Microbiol. 45, 345-382 Timmers ACJ, Auriac M-C, de Billy, Truchet G (1998) Develop. 125, 339-349 Tsyganov VE, Morzhina EV, Stefanov SY, Borisov AY, Lebsky VK, Tikhonovich IA (1998)
Mol. Gen. Genet. 256,491-503 Vincent JM (1980) In Newton WE, Orme-Johnson WH (ed) Nitrogen Fixation, Vol. 2, pp. 103-129, University Park Press, Baltimore, MD
This work was financially supported by the Russian Foundation for Basic Research (01-04-48580 and 01-04-49643) and the Netherlands Organisation for Scientific Research (NWO 047-007.017).
Was this article helpful?