An Inducible Activator Produced by a Serratia Proteamaculans Strain and Its Soybean Growth-Promoting Activity Under Greenhouse Conditions

An inducible activator produced by a Serratia proteamaculans strain and its soybean growth-promoting activity under greenhouse conditions

Yuming Bai, Alfred Souleimanov and Donald L. Smith1

Department of Plant Science, Macdonald Campus of McGill University, 21,111 Lakeshore Road, Ste Anne de Bellevue, Quebec, Canada H9X 3V9

Received 23 October 2001; Accepted 20 February 2002

Astract

Serratia proteamaculans 1-102 (1-102) promotes soybean-bradyrhizobia nodulation and growth, but the mechanism is unknown. After adding isoflavonoid inducers to 1-102 culture, an active peak with a retention time of about 105 min in the HPLC fractionation was isolated using a bioassay based on the stimulation of soybean seed germination. The plant growth-promoting activity of this material was compared with 1-102 culture (cells) and supernatant under greenhouse conditions. The activator was applied to roots in 83, 830 and 8300 HPLC microvolts (µV) per seedling when plants were inoculated with bradyrhizobia or sprayed onto the leaves in same concentrations at 20 d after inoculation. The root-applied activator, especially at 1 ml of 830 µV per seedling, enhanced soybean nodulation and growth at the same level as 1-102 culture under both optimal and sub-optimal root zone temperatures. Thus, this activator stimulating soybean seed germination is also responsible for the plant growth-promoting activity of 1-102 culture. However, when sprayed onto the leaves, the activator did not increase growth and in higher concentrations decreased average single leaf area. The results suggest that this inducible activator might be a lipo-chitooligosaccharide (LCO) analogue. LCOs act as rhizobia-to-legume signals stimulating root nodule formation. The activator could provide additional 'signal', increasing in the signal quality (the signal-to-noise ratio, SNR) of the plant-rhizobia signal exchange process.

Key words: Inducible activator, plant growth-promoting rhizobacteria, Serratia proteamaculans, soybean.

Introducion

Legume nodulation is a complex process involving interactions between the host plants and rhizobia. This process is also affected by many biotic and abiotic environmental factors (Hungria and Stacey, 1997 ; Valdssak and Vanderleyden, 1997 ). The first stage in the establishment of the symbiotic system is signal exchange between legume plants and rhizobia. The plant-to-bacteria signals are isoflavonoids which induce bradyrhizobial nod gene expression and in the case of soybean are mainly genistein and daidzein (Rao and Cooper, 1994 ). The rhizobia-to-plant return signals are lipo-chitooligosaccharides (LCOs), so-called Nod factors, which play pivotal roles in root nodule formation. LCOs are oligosaccharides of Я-1,4-linked N-acetyl-D-glucosamine and of some specifically modified side groups. LCOs are synthesized via sophisticated biochemical processes catalysed by a series of nod gene encoded enzymes (Perret et al., 2000 ). All individual rhizobial strains produce specific structurally diverse LCO mixtures (Spaink, 1996 ) and the major LCO molecule produced by Bradyrhizobium japonicum 532C is Nod Bj V (C18:1; MeFuc) (Prithiviraj et al., 2000 ).

During the signal exchange process, environmental factors affecting either signal production or signal perception can affect nodulation and subsequent nitrogen fixation. Sub-optimal (15-17.5 oC) root zone temperatures (RZTs), pH stress and mineral nitrogen inhibit the production of isoflavonoids by soybean roots as well as subsequent nodulation and nitrogen fixation (Streeter, 1988 ; Cho and Harper, 1990 ; Zhang and Smith, 1994 , 1996 a; Pan and Smith, 1998 ). High temperature (39 oC) increases the release of the isoflavonoid signals from soybean seeds during the first 24 h, but the compounds released have decreased nod gene-inducing activities (Hungria and Stacey, 1997 ). The addition of genistein to the inoculant or the rhizosphere could at least partially alleviate the deleterious effects of these environmental factors (Zhang and Smith, 1995 , 1996 b, 1997 ; Smith and Zhang, 1996 ; Hungria and Stacey, 1997 ; Pan et al., 1998 ). Besides inhibiting the synthesis and excretion of isoflavonoids by soybean roots, low RZTs also suppress bacterial nod gene expression, and this also could be partially overcome by genistein application (Zhang et al., 1996a ). In addition, LCOs are sensitive to chitinase and related hydrolases which cleave and inactivate Nod factors in the host rhizosphere (Perret et al., 2000 ). When Sinorhizobium fredii and S. meliloti were transconjugated with a chitinase gene from a Serratia marcescens strain, the enzyme was expressed and nodulation of soybean and alfalfa were impeded (Krishnan et al., 1999 ).

Many plant growth-promoting rhizobacteria (PGPR) have beneficial effects on legume growth, and at least some PGPR strains enhance legume nodulation and nitrogen fixation by affecting signal exchange between the plants and rhizobia. Co-inoculation of some Pseudomonas and Bacillus strains, along with effective Rhizobium spp., stimulates chickpea growth, nodulation and nitrogen fixation (Parmar and Dadarwal, 1999 ). Seed colonization by these PGPR or application of the ethyl acetate extract of the culture supernatant increase the concentration of flavonoid-like compounds in roots, and the rhizobacteria themselves are capable of producing fluorescent flavonoids similar to those produced by the plant (Parmar and Dadarwal, 1999 ). These lines of evidence indicate that PGPR may produce signal molecule analogues and/or stimulate the plant to produce more signal molecules. It may also be reasonable to postulate that some rhizobacteria produce LCO analogues or improve conditions for signal exchange.

Some Serratia strains, such as S. proteamaculans 1-102 and S. liquefaciens 2-68, have beneficial effects on legume plant growth (Chanway et al., 1989 ; Zhang et al., 1996b ). They are both partially able to overcome the effects of sub-optimal RZT on soybean nodulation and N2 fixation. Strain 1-102 generally performed better than 2-68 at sub-optimal RZTs (Zhang et al., 1996b ). Their culture supernatants had the same beneficial effects on soybean plants as the bacterial cultures, although not under sub-optimal conditions (Dashti, 1997 ). Combined application of these PGPR and genistein improved N2 fixation in soybean at suboptimal root zone temperatures (Dashti et al., 2000 ). Given their effects on soybean plants, it is hypothesized that the PGPR strains exert their influence via the production of specific compounds after they have been inoculated into plant rhizospheres. In testing this hypothesis, a series of experiments