Skip to main content

Acquisition of genes from indigenous bacteria by inoculant strains at long- term release sites

Objective

Genetic exchange between inoculant and native bacteria could generate undesirable novel genetic combinations: the possibility limits agricultural and environmental exploitation of genetically modified (GM) bacteria. The capacity of GM bacteria to act as genetic donors can be minimized, but it is difficult to ensure that they cannot act as recipients. Natural populations show indirect evidence of genetic interactions. Information about the frequency of such events is essential for risk assessement prior to release of GM bacteria. Our objective was to gain information on the types of genetic elements involved (these can serve as markers of gene transfer, and may themselves confer significant new properties), and to gather quantitative data on gene acquisition by bacterial inoculants in the field, using rhizobia, the most important plant inoculants in Europe and worldwide. We isolated natural genetic elements (bacteriophages, plasmids, transposable IS elements), developed detection methods for them, determined their distribution in indigenous populations; screened established introductions for their acquisition; developed and released strains designed as potential recipients; and analyzed those with potential genetic acquisitions.
The plasmid replication origin oriC, cloned and sequenced from R. leguminosarum (York) and Sinorhizobium meliloti (Granada) enabled construction of PCR primers for repC amplification from a majority of field isolates tested: four repC families were identified in R. leguminosarum from Dijon, Bielefeld, Erlangen and Rothamsted, providing information on plasmid distribution. Analysis of S. meliloti repC regions indicated two groups; one widespread, with a homolog on an R. tropici symbiotic plasmid (pSym).
Insertion elements were isolated, sequenced and primers for PCR designed. ISRl2 from Dijon R. leguminosarum was identical to independantly-isolated ISRlF7-2 from Bielefeld, where nine S. meliloti IS were investigated: ISRM220-12-3, had 76% sequence homology to ISRlF7-2. IS distribution in Bielefeld, Dijon, Erlangen, Granada and Rothamsted field populations and in inoculant strains was assessed. Primers designed in Granada for two S. meliloti IS elements amplified related elements in R. leguminosarum (strain B10 had 90% homology to ISRm3) and S. fredii. Five new S. meliloti IS were identified; one homologous to a Methylobacterium IS.
Bacteriophages were isolated from soil in Rothamsted, field isolates in Erlangen, infection profiles were compared, DNA isolated and analysed, PCR primers designed and used to detect infected bacteria, and phage particles in soil. Most R. leguminosarum field isolates were lysogenic. A virulent Rothamsted phage isolate could transduce genes in inoculant strains RSM2004.
Screening for acquisition of plasmids from native rhizobia by R. leguminosarum CT0370 two years after release at Rothamsted (pSym-cured, chromosomal gusA insertion, constructed Bielefeld/Rothamsted) gave no evidence for pSym transfer (20,000 root nodules screened for GUS activity). Other inoculant strains reisolated in Dijon, Erlangen, Granada and Rothamsted were screened with primers for IS and repC present only in native strains. In Granada, 27000 nodules formed by S. fredii inoculant HH102 were screened for ISRm3 and ISRm6; none were detected. At Rothamsted, 500 RSM2004 colonies reisolated from soil after 8 years were screened for ISRm3, 1000 CT0370 reisolates (after 2 years) were screened for repCI/II acquistion but none was detected. However, in Dijon, 130 soil samples containing an estimated 125,000 cells of R. leguminosarum strain T2 (released 7 years before) were screened for IS and repC following enrichment. Some gave positive results; one T2 derivative was reisolated from a repCII-positive sample and found to have acquired a small plasmid which proved unstable upon subculture.
MAJOR SCIENTIFIC BREAKTHROUGHS:
Isolation and characterization of repC groups and IS elements, together with design of PCR primers has provided important data on their distribution in rhizobia at different european sites, and valuable tools for future investigations, a significant step in bacterial population genetics. Screening reisolated strains for gene acquisition has provided quanitiative data: fewer than 4E-5 S. fredii acquired IS elements from native strains in Spain; pSym transfer frequency from native R. leguminosarum to CT0370 at Rothamsted was less than 5E-5 per donor, but Dijon results indicate a plasmid transfer frequency of 8E-6 per T2 recipient.

Funding Scheme

CSC - Cost-sharing contracts

Coordinator

ROTHAMSTED RESEARCH LIMITED
Address
West Common
AL5 2JQ Harpenden, Herts
United Kingdom

Participants (5)

CONSEJO SUPERIOR DE INVESTIGACIONES CIENTIFICAS
Spain
Address
Calle Serrano 117
Madrid
INSTITUT NATIONAL DE LA RECHERCHE AGRONOMIQUE
France
Address
Rue De Sully 17
21065 Dijon
UNIVERSITY OF BIELEFELD
Germany
Address
Universitätsstraße 25
33501 Bielefeld
UNIVERSITY OF ERLANGEN-NUREMBERG
Germany
Address
Egerlandstrasse 1
91058 Erlangen
University of York
United Kingdom
Address
Heslington
York