Foliar Chlorosis in Symbiotic Host and Nonhost Plants Induced by Rhizobium tropici Type B Strains

Abstract

Rhizobium tropici CIAT899 induced chlorosis in the leaves of its symbiotic hosts, common bean (Phaseolus vulgaris L.), siratro (Macroptilium atropurpureum Urb.), and Leucaena leucocephala (Lam.) de Wit. Chlorosis induction by strains CIAT899 and CT9005, an exopolysaccharide-deficient mutant of CIAT899, required carbon substrate. When the bacteria were added at planting in a solution of mannitol (50 g/liter), as few as 1o cells of CIAT899 were sufficient to induce chlorosis in bean plants. All carbon sources tested, including organic acids and mono-and disaccharides, supported chlorosis induction. The addition of a carbon source did not affect the growth rate or the population density of CT9005 in the bean plant rhizosphere. Cell-free filtrates of cultures of CT9005 did not induce detectable chlorosis. All type B strains of R. tropici tested also induced chlorosis in common bean. Type A strains of R. tropici and all other species of bacteria tested did not induce chlorosis. Several lines of evidence indicated that nodulation was not required for chlorosis induction. Strain RSP900, a pSym-cured derivative of CIAT899, induced chlorosis in wild-type P. vulgaris. In addition, NOD125, a nodulation-defective line of common bean, developed chlorosis when inoculated with CIAT899, but did not develop nodules. CL4T899 consistently induced severe chlorosis in the leaves of the nonhost legumes alfalfa (Medicago sativa L.) and Berseem clover (Trifolium alexandrinum L.), and induced chlorosis in 29 to 58% of the plants tested of sunflower, cucumber, and tomato seedlings, but it did not induce chlorosis in the leaves of corn or wheat. Chlorosis induction in nonhost plants also required carbon substrate. The data are consistent with the hypothesis that R. tropici type B strains produce a chlorosis-inducing factor that affects a wide range of plant species. * Corresponding author. roots of bean (15, 18) and siratro (15) and retain resistance to aluminum (11), but many EPS-deficient mutants are reduced in competitiveness relative to that of CIAT899 (15). We showed previously that EPS-deficient mutants of strain CIAT899 induce severe interveinal chlorosis in common bean and Leucaena leucocephala (18). In this report we show that CIAT899, the EPS-producing, wild-type strain

Mutants of Rhizobium tropici strain CIAT899 that do not induce chlorosis in plants

“…Since mutants CT4812 and ME245 (containing mutations in the genomic glnD gene) do not induce chlorosis, the ability of plasmids carrying mutations in glnD to restore chlorosis induction suggests that a wild-type copy of glnD may have been reconstructed by recombination, since the strains are Rec'. Only a small minority of individuals in the rhizosphere population need to have experienced a marker rescue to see chlorosis in the host plant, because as few as 4 x lo3 chlorosis-inducing bacteria cause symptoms in the presence of 9 x 10' non- The finding that a mutation in a homologue of glnD, a sensor of nitrogen status in enteric bacteria, prevented R. tropici strain CIAT899 from inducing chlorosis in bean plants, coupled with our earlier finding that chlorosis induction by type B strains of R. tropici requires a carbon source added to the rhizosphere (O'Connell & Handelsman, 1993) led us to determine whether the ratio of carbon and nitrogen sources in the rhizosphere affects chlorosis induction by CIAT899.…”

“…The apparent lack of expression of glnD provides, in addition to the nutritional phenotypes of glnD mutants described above, another contrast with the glnD gene of E. coli, which is expressed constitutively (van Heeswijk et al, 1993). If the glnD gene of R. tropici requires some condition or compound, or the lack thereof, in the plant rhizosphere for its expression, the inducing compound or condition must be widespread among dicotyledonous plants, since R. tropici can induce chlorosis in the leaves of sunflower, cucumber, tomato, alfalfa and berseem clover as well as common bean, but not wheat, corn, or bluegrass (O'Connell & Handelsman, 1993). The nitrogen-sensing role of GlnD proteins in enteric bacteria suggests that the product of the glnD homologue in R. tropici may function to regulate chlorosis induction in response to environmental nitrogen levels.…”

“…(1988) Figurski & Helinski (1979) Simon et at. 1983Selveraj & Iyer 1983Quandt & Hynes (1993) Promega Stachel et al (1985) Bonas et al (1989) al ., 1990;O'Connell & Handelsman, 1993). CIAT899 also induces chlorosis in other host legumes a n d nonhost dicotyledonous plants (O'Connell & Handelsman, 1993).…”

“…1983Selveraj & Iyer 1983Quandt & Hynes (1993) Promega Stachel et al (1985) Bonas et al (1989) al ., 1990;O'Connell & Handelsman, 1993). CIAT899 also induces chlorosis in other host legumes a n d nonhost dicotyledonous plants (O'Connell & Handelsman, 1993). T h e induction of chlorosis depends o n the availability of a carbon source in the rhizosphere, the presence of which does n o t affect bacterial growth (O'Connell & Handelsman, 1993).…”

“…CIAT899 also induces chlorosis in other host legumes a n d nonhost dicotyledonous plants (O'Connell & Handelsman, 1993). T h e induction of chlorosis depends o n the availability of a carbon source in the rhizosphere, the presence of which does n o t affect bacterial growth (O'Connell & Handelsman, 1993). Symptoms t h a t accompany foliar chlorosis often include stunting a n d loss of apical dominance (O'Connell & Handelsman, 1993) ; superficially, the chlorosis induced by CIAT899 resembles iron deficiency in Phaseolus beans (Schwartz, 1991).…”