Detection of Acidovorax avenae ssp. avenae in Rice Seeds Using BIO-PCR

The bacterium, Acidovorax avenae ssp. avenae causes several important plant diseases including bacterial stripe of rice, bacterial stalk rot of corn, bacterial leaf blight of oats, and red stripe of sugarcane and millet. Although the organism is seed-borne in rice, no reliable seed assay is available. A semiselective liquid medium based on d-sorbitol and l-pyroglutamic acid (SP medium), and two sets of polymerase chain reaction (PCR) primers were designed for use in a BIOPCR assay for detection of A. avenae ssp. avenae in rice seeds. External primer set, Aaaf3 and Aaar2, and internal primer set, Aaaf5 and Aaar2, designed from a 619 bp fragment of the internal transcribed spacer region of the 16S–23S rDNA of A. avenae ssp. avenae strain CAa4 were specific at the subspecies level. A nested-PCR assay produced the expected DNA product from 58 rice strains tested but not with DNA from 27 strains of A. avenae ssp. avenae from corn and other hosts. Furthermore, the primers failed to amplify a PCR product from two strains of A. avenae ssp. cattleyae, 10 strains of A. avenae ssp. citrulli, or 64 other bacteria. When A. avenae ssp. avenae was enriched prior to PCR by incubating washings of naturally contaminated rice seeds for 12 h in SP liquid medium, populations of the pathogen increased over 1000-fold, whereas populations of saprophytic bacteria remained stable. Populations of the target bacterium routinely reached PCR-detectable levels after seed samples containing as few as 1–2 target cells and up to 10 cfu of other bacteria were soaked overnight in phosphate buffer and then enriched for 12 h in SP liquid medium. Assays of 64 naturally infected rice seed samples resulted in 53 being positive by BIO-PCR and 34 by classical PCR. The BIO-PCR assay provides a sensitive, reliable tool for the specific detection of A. avenae ssp. avenae in rice seeds. The use of trade, firm or corporation names in this publication (or page) is for the information and convenience of the reader. Such use does not constitute an official endorsement or approval by the United States Department of Agriculture or the Agricultural Research Service of any product or service to the exclusion of others that may be suitable. Introduction Acidovorax avenae ssp. avenae (1⁄4Pseudomonas avenae) is the causal agent of several important plant diseases including bacterial stripe of rice (Goto, 1964; Shakya et al., 1985; Kadota, 1996), bacterial stalk rot of corn (Rosen, 1926; Summer and Schaad, 1977), bacterial leaf blight of oats (Manns, 1909), red stripe of sugarcane and millet (Martin and Wismer, 1989), and brown stripe of foxtail (Rosen, 1922). The organism was originally described by Manns (1909) as the cause of blade blight of oats in Ohio in 1909 and named P. avenae. A similar non-fluorescent pseudomonad described several years later on foxtail (Rosen, 1922) and corn (Rosen, 1926) as P. alboprecipitans was shown to be synonymous with P. avenae (Schaad et al., 1975). Furthermore, the same organism has been referred to as P. setariae and P. panici on rice (Goto, 1964) and as P. rubrilineans on sugarcane (Martin and Wismer, 1989). The latter three non-fluorescent pseudomonads were later reclassified as P. avenae based on numerical and genomic analysis (Hu et al., 1991). A recent proposal places all the above organisms in the new acidovorans DNA–rRNA homology group as A. avenae ssp. avenae (Willems et al., 1992). This group contains many pathogens previously classified in the non-fluorescent pseudomonad group, including P. avenae (Hu et al., 1991), P. cattleyae (Savulescu, 1947), and P. pseudoalcaligenes ssp. citrulli (Schaad et al., 1978), which have been reclassified as A. avenae ssp. avenae, cattleyae, and citrulli, respectively, based upon the results of DNA–DNA hybridization, DNA– rRNA hybridization, polyacrylamide gel electrophoresis of whole-cell proteins and a numerical analysis of carbon assimilation tests (Willems et al., 1992). The natural host range of A. avenae ssp. avenae includes U. S. Copyright Clearance Centre Code Statement: 0931–1785/2004/15212–0667 $ 15.00/0 www.blackwell-synergy.com J. Phytopathology 152, 667–676 (2004) 2004 Blackwell Verlag, Berlin ISSN 0931-1785 several monocotyledonous plants but strains from rice and millet apparently infect only rice and millet, respectively (Nishiyama et al., 1979; Kadota, 1996). Acidovorax avenae ssp. avenae is a widely distributed (Shakya et al., 1985) seed-borne pathogen of rice (Goto, 1964; Shakya and Chung, 1983) and can be recovered from seeds of some lots using general agar media (W. Y. Song, unpublished data). However, because A. avenae ssp. avenae is frequently overgrown by other seed-borne saprophytic bacteria, its recovery and identification is often difficult. Since the pathogen does not always produce distinct symptoms under field conditions, the disease is very difficult to diagnose. In such cases, disease diagnosis must rely on isolation of the pathogen on semiselective agar media and pathogenicity tests. However, isolation is difficult because many ubiquitous bacteria such as P. fulva, P. corrugata and P. fluorescens are often present in/on rice seeds (Cottyn et al., 1996; W. Y. Song, personal observations). Pathogenicity tests are very time-consuming and expensive because rice seedlings are difficult to grow and symptoms on seedlings are often nondiscriminating. Several methods to detect A. avenae ssp. avenae have been proposed including growing-on tests (Shakya and Chang, 1983; Kihupi et al., 1996), isolation on semiselective agar media (Zeigler and Alvarez, 1989; Kadota, 1996), and serology (Shakya and Chung, 1983; Kadota, 1996), but none have become widely used. Plant disease diagnosis has been improved greatly by polymerase chain reaction (PCR)-based detection methods (Henson and French, 1993). Primers have been described for many plant pathogenic bacteria (Schaad et al., 2001) including A. avenae at the species level (Song et al., 1997). However, the A. avenaespecific primers react with all subspecies of A. avenae and are not specific to strains pathogenic only to rice. Based on preliminary characterization of the 16S–23S ribosomal DNA intergenic transcribed spacer region (ITS) of A. avenae (Kim and Song, 1996), PCR primers were designed for the specific identification of all rice strains of A. avenae ssp. avenae (Song et al., 2001). Although PCR can be a very useful method for detecting bacteria in pure culture, the technique has limitations with natural seed or soil samples because of inhibition of PCR by numerous compounds and cells of other bacteria present in plant and soil extracts. Inhibitors can be reduced by extraction of DNA, however, this can reduce assay sensitivity and it requires the use of toxic chemicals. Also, classical PCR can result in false positives from detection of dead cells and/or free DNA. One way to avoid PCR inhibitors while increasing assay sensitivity is to enrich the target bacteria on agar or in liquid media prior to PCR, a technique termed BIO-PCR (Schaad et al., 1995). By using a short, 15–24 h enrichment, as few as 10 cfu/ml of target bacteria can be detected. We describe a BIO-PCR procedure using a new semiselective liquid medium and PCR primers for a nested-PCR technique for the specific detection of A. avenae ssp. avenae in rice seeds. Materials and Methods Bacterial strains and identification The sources of bacterial strains used in this study are shown in Table 1. Bacteria were maintained on yeast extract-dextrose-calcium carbonate (YDC) (Schaad, 1988) or King et al’s medium B (KB) (King et al., 1954) and stored at )80 C. For short-term storage, cultures were stored on YDC slants at 22–24 C. To confirm the identification of A. avenae, strains were checked for colony morphology on YDC agar, polyb-hydroxybutyrate accumulation on nile blue medium (Pierce and Schroth, 1994) and PCR amplification using A. avenae-specific primers, Oaf1 and Oar1 (Song et al., 1997). Final identity of all rice strains of A. avenae ssp. avenae was confirmed by inoculating rice seedlings at the 3or 4-leaf stage, as described (Schaad et al., 1975), with slight modification. Briefly, rice strains grown overnight in nutrient broth-yeast extract (NBY) (Vidaver, 1967) liquid medium were diluted 100-fold in phosphate-buffered saline (PBS) (20 mm sodium phosphate, 0.85% NaCl, pH 7.6) to a concentration of approximately 10 cfu/ml. The whorl of 5–10 seedlings per strain was then inoculated using a 1 ml syringe and a 26-gauge needle. After 24 h in a dew chamber at 30 C and 100% relative humidity (RH), all plants were moved to a growth chamber (12 h day/12 h night) at 30 C and 60% RH. Control plants were inoculated with PBS in the same manner. Symptoms were recorded 5–7 days after inoculation. Development of a semiselective liquid medium for BIO-PCR The key to developing a reliable BIO-PCR enrichment technique is to use a medium that allows specific and rapid growth of the target bacterium and suppresses the growth of non-target bacteria. This can be difficult with rice because most rice seeds contain large numbers of bacteria, which may inhibit growth of A. avenae ssp. avenae (W. Y. Song, personal observations). Since no reliable agar or liquid media were available, we developed a new semiselective liquid medium. Presumptive utilization of carbon and nitrogen compounds for the specific growth of A. avenae ssp. avenae was determine by using Biolog GN Microplates (Biolog, Hayward, CA, USA) with eight rice strains (CAa4–6, MAFF 301502–4, and ATCC 19882) and one corn (ATCC 19860) strain. The most promising carbon and nitrogen compounds were then selected and tested by comparing the specificity and recovery efficiency by dilution-plating techniques. From these quantitative results a basal liquid medium containing 2.0 g d-sorbitol, 0.2 g l-pyroglutamic acid, 0.5 g of K2HPO4, and 3.0 g of Na2HPO4 per litre was selected and used to evaluate several inhibitors for reducing growth of rice-associated bacteria, without reducing growth of A. avenae ssp. avenae. The growth of the above-mentioned strains of A. avenae ssp. avenae along with 24 other known bacterial pathogens of rice, 668 Song et al.