Bias Caused by Using Different Isolation Media for Assessing the Genetic Diversity of a Natural Microbial Population S. Tabacchioni, L. Chiarini, A. Bevivino, C. Cantale, C. Dalmastri Dipartimento Innovazione, Divisione Biotecnologie e Agricoltura, ENEA (Ente Nazionale per le Nuove Tecnologie, l’Energia e l’Ambiente) C.R. Casaccia,, 00060 Rome, Italy Received: 29 April 1999; Accepted: 27 January 2000; Online Publication: 28 August 2000 A B S T R A C T The influence of isolation medium on the biodiversity of Burkholderia cepacia strains recovered from the rhizosphere of Zea mays was evaluated by comparing the genetic diversity of isolates obtained by plating serial dilutions of root macerates on the two selective media TB-T and PCAT. From each medium, 50 randomly chosen colonies were isolated. On the basis of the restriction patterns of DNA coding for 16S rRNA (16S rDNA) amplified by means of PCR (ARDRA), all strains isolated from TB-T medium were assigned to the B. cepacia species, whereas among PCAT isolates only 74% were assigned to the B. cepacia species. Genetic diversity among the PCAT and TB-T isolates was evaluated by the random amplified polymorphic DNA (RAPD) technique. The analysis of molecular variance (AMOVA) method was applied to determine the variance component for RAPD patterns. Most of the genetic diversity (90.59%) was found within the two groups of isolates, but an appreciable amount (9.41%) still separated the two groups (P < 0.001). Mean genetic distances among PCAT isolates (10.39) and TB-T isolates (9.36) were significantly different (P < 0.0001). The results indicate that the two different isolation media select for B. cepacia populations with a different degree of genetic diversity. Moreover, a higher degree of genetic diversity was observed among strains isolated from PCAT medium than among those isolated from TB-T me- dium. Introduction Understanding the genetic diversity of a bacterial species in nature is shaped by the need to isolate pure cultures of microorganisms. Since cultivable bacteria are necessary for genetic studies and these represent only 0.1–20% of the bac- teria present in microbial communities, it is likely that only a small portion of genetic diversity of a microbial species has been documented [1]. Even within the cultivable fraction of bacteria, obtaining microorganisms that are representative of the diversity existing in nature can be difficult as a result of the inherent bias due to culture media. Numerous selective and nonselective media have been used for enumerating and isolating soil and rhizosphere mi- croorganisms. Previous work suggests that the choice of the Correspondence to: S. Tabacchioni; Fax: +39 06 30484808; E-mail: microbi@ casaccia.enea.it MICROBIAL ECOLOGY Microb Ecol (2000) 40:169–176 DOI: 10.1007/s002480000015 © 2000 Springer-Verlag New York Inc.