Rhizosphere colonization of hexaploid wheat by Pseudomonas fluorescens strains Q8r1-96 and Q2-87 is cultivar-variable and associated with changes in gross root morphology Patricia A. Okubara, a, * Jeremy P. Kornoely, a and Blanca B. Landa b a US Department of Agriculture, Agricultural Research Service, Washington State University, Pullman, WA, USA b Departamento de Agronom  ıa, Escuela T ecnica Superior de Ingenieros, Agr onomos y de Montes (ETSIAM), Universidad de C ordoba, Apartado 3048, C ordoba 14080, Spain Received 5 August 2003; accepted 12 November 2003 Abstract In the Pacific Northwest, natural suppression of take-all disease of wheat (take-all decline) has been attributed to certain strains of the gram-negative bacterium Pseudomonas fluorescens. To examine naturally occurring variation in the ability of Triticum aes- tivum L. (hexaploid wheat) to support these bacteria, we have surveyed 27 Pacific Northwest (PNW) cultivars for the ability to undergo root colonization with the aggressive colonizer P. fluorescens strain Q8r1-96, and P. fluorescens strain Q2-87, a less effective colonizer. In seed inoculation experiments, Q8r1-86 colonized roots of all of the cultivars equally or more effectively than did Q2-87 in a non-pasteurized, non-agricultural soil. Seven cultivars supported significantly (P < 0:05) higher rhizosphere populations of Q8r1-96 than Q2-87 within 14 days post-inoculation (dpi), two cultivars supported relatively high population densities of each bacterial strain, and three cultivars supported low population densities of the strains. Population densities normalized to root weight reached maximum steady-state levels within 4 dpi, and differential colonization was seen as early as 7 dpi. In pairwise comparisons, the bacterial treatments differentially affected the root morphology of some of the cultivars at 14 dpi. However, principal com- ponents (factor) and correlation analysis showed that preferential colonization by Q8r1-96 was independent of root fresh weight, total length, surface area, volume, and average diameter, and that differential colonization was not correlated with changes in any specific root morphometric variable. Variation in root colonization of specific cultivars suggests useful genetic stocks for mapping and identifying host genes involved in wheat–rhizosphere interactions. Ó 2003 Elsevier Inc. All rights reserved. Keywords: Biological control; Triticum aestivum 1. Introduction The practice of direct seeding, or sowing of seed into the stubble of the previous crop, has increased in the US since the mid-1990s. The benefits of direct seeding, specifically topsoil conservation and reduction of input costs, can be offset by the increased incidence and se- verity of fungal diseases. Increased disease severity has been attributed in part to the build-up of pathogen populations on wintering stubble (Cook et al., 2000; Windels, 2000). Triticum aestivum L. (hexaploid wheat) and related cereal species have limited genetic resistance to necrotrophs such as Gaeumannomyces tritici var. graminis, Rhizoctonia solani, and Pythium spp. that are among the major soilborne pathogens of small grain cereals in the Pacific Northwest. Furthermore, trans- ferable sources of resistance are rare (Smith et al., 2003a,b). Plant growth-promoting rhizobacteria (PGPR) can be highly effective in the suppression of these diseases (Becker and Cook, 1988; Kim et al., 1997; Weller and Cook, 1983, 1986), and offer an alternative supportive of sustainable agriculture. In the Pacific Northwest, naturally occurring sup- pression of G. graminis var. tritici (Ggt), which causes take-all disease of wheat, is attributed to certain strains of Pseudomonas fluorescens that produce the anti- fungal metabolite 2,4-diacetylphloroglucinol (2,4-DAPG) * Corresponding author. Fax: 1-509-335-7674. E-mail address: pokubara@wsu.edu (P.A. Okubara). 1049-9644/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.biocontrol.2003.11.003 Biological Control 30 (2004) 392–403 www.elsevier.com/locate/ybcon