Recurrent outbreaks of root mat have occurred in the UK and France in cucumber and tomato. Root mat is caused by bacterial strains harbouring a Ri-plasmid (pRi). Fifteen root mat-associated (RMA) cucumopine pRi were analysed by PCR-restriction fragment length polymorphism (RFLP) and Southern blotting. These pRi were harboured by Agrobacterium biovar 1 strains isolated during a 1970s outbreak of root mat in UK soil grown cucumber, and also by Agrobacterium biovar 1, Ochrobactrum, Rhizobium and Sinorhizobium isolated during outbreaks of root mat in cucumber and tomato grown hydroponically in the UK and France since 1993. PCR-RFLP analysis of the T-DNA and virD2 regions showed sequence homology between all cucumopine pRi, indicating that these pRi are monomorphic, and thus this pRi remained in the UK without inducing symptoms for some 15 years between outbreaks in the 1970s and 1990s. Cucumopine pRi were also shown to possess the virE2 substitute GALLS gene by Southern blotting. Two other pRi, harboured by Agrobacterium isolated from a recent root mat outbreak in one tomato crop, were also shown to possess the GALLS gene but were shown not to be cucumopine pRi by PCR-RFLP.

Ochrobactrum CSL 2573, Rhizobium CSL 2411 and Sinorhizobium CSL 2611 strains harbouring the Agrobacterium cucumopine Ri plasmid (pRi), previously were shown to induce root‐mat symptoms in an in vitro cucumber cotyledon assay. In whole‐plant, rockwool‐grown cucumber host tests Rhizobium CSL 2411 was shown to be as efficient an inducer of root‐mat symptoms as the virulent Agrobacterium radiobacter strain NCPPB 4042, which also harbours a cucumopine pRi. Conjugal transfer of pRi to ingressing, avirulent Agrobacterium isolates was observed within root tissues with symptoms. Ochrobactrum CSL 2573 and Sinorhizobium CSL 2611 were not able to induce root‐mat symptoms on plants. Rhizobium CSL 2411 and Ochrobactrum CSL 2573 were reisolated from inoculated plants, but Sinorhizobium CSL 2611 was not detected or isolated from inoculated plants 68 days after inoculation. It was postulated that the differences in pathogenicity observed between the in vitro and in situ host tests were caused by a lack of proper attachment to inoculated root tissues by pRi‐harbouring Ochrobactrum and Sinorhizobium in the whole‐plant host tests.

ABSTRACT Thirty-eight bacterial strains isolated from hazelnut (Corylus avellana) cv. Tonda Gentile delle Langhe showing a twig dieback in Piedmont and Sardinia, Italy, were studied by a polyphasic approach. All strains were assessed by fatty acids analysis and repetitive sequence-based polymerase chain reaction (PCR) fingerprinting using BOX and ERIC primer sets. Representative strains also were assessed by sequencing the 16S rDNA and hrpL genes, determining the presence of the syrB gene, testing their biochemical and nutritional characteristics, and determining their pathogenicity to hazelnut and other plants species or plant organs. Moreover, they were compared with reference strains of other phytopathogenic pseudomonads. The strains from hazelnut belong to Pseudomonas syringae (sensu latu), LOPAT group Ia. Both fatty acids and repetitive-sequence-based PCR clearly discriminate such strains from other Pseudomonas spp., including P. avellanae and other P. syringae pathovars as well as P. syringae pv. syringae strains from hazelnut. Also, the sequencing of 16S rDNA and hrpL genes differentiated them from P. avellanae and from P. syringae pv. syringae. They did not possess the syrB gene. Some nutritional tests also differentiated them from related P. syringae pathovars. Upon artificial inoculation, these strains incited severe twig diebacks only on hazelnut. Our results justify the creation of a new pathovar because the strains from hazelnut constitute a homogeneous group and a discrete phenon. The name of P. syringae pv. coryli is proposed and criteria for routine identification are presented.

ABSTRACT Root mat of cucumbers and tomatoes has previously been shown to be caused by Agrobacterium radiobacter strains harboring a root-inducing Ri plasmid (pRi). Nine other pRi-harboring α- Proteobacteria have subsequently been isolated from root mat-infected crops. Fatty acid profiling and partial 16S rRNA sequence analysis identified three of these strains as being in the genus Ochrobactrum , five as being in the genus Rhizobium , and one as being in the genus Sinorhizobium . An in vitro pathogenicity test involving inoculation of cucumber cotyledons was developed. All pRi-harboring α- Proteobacteria induced typical root mat symptoms from the cotyledons. Average transformation rates for rhizogenic Ochrobactrum (46%) and Rhizobium (44%) strains were lower than those observed for rhizogenic A. radiobacter strains (64%). However, individual strains from these three genera all had transformation rates comparable to those observed from cotyledons inoculated with a rhizogenic Sinorhizobium strain (75%).

During a serious epidemic of crown and cane gall on the blackberry-raspberry (Rubus occidentalis-Rubus idaeus) hybrid Lockness in a specialized crop in the province of Treviso (northern Italy) Gram-negative bacteria were found associated with tumours. Following experimental inoculation these bacteria caused tumours on tomato stems and on pot-grown hybrid canes in the greenhouse. These bacteria were found to possess a Ti plasmid, common to Agrobacterium. The resulting fatty acid profile did not correspond to any known Agrobacterium species, but did indicate an affinity with the genus Agrobacterium. Partial 16S rRNA sequencing revealed that the bacteria were closely related to three strains of Agrobacterium rhizogenes, in particular with a strain isolated from a peach tumour.

The genus Burkholderia comprises over 28 species and species-specific, recA-based polymerase chain reaction (PCR) tests are available for several species, but not for some soil-inhabiting species including B. fungorum. Previous analysis of several novel rhizospheric, environmental isolates belonging to the B. cepacia complex suggested they may be closely related to B. fungorum. To discover any relationship between these isolates and B. fungorum we set out to clone and sequence a portion of the B. fungorum recA gene in order to design species-specific primer pairs for use in a recA-based PCR assay. Using a similar procedure we extended the recA-based PCR assay to identify B. sacchari and B. caledonica, two additional soil-inhabiting Burkholderia spp.

AIMS:The development of a fluorogenic, 5' nuclease, TaqMan PCR assay for the detection of Ri-plasmids from root mat inducing Agrobacterium biovar 1 strains.METHODS AND RESULTS:A TaqMan probe and primer set were designed within the T-DNA sequence of a known root mat inducing Agrobacterium strain. One hundred and ten Agrobacterium and closely related bacteria were tested using this novel PCR and compared with results from a conventional PCR which detects Ti and Ri-plasmids. The Agrobacterium selective media, Medium 1A was modified into broth form for use as an enrichment of the pathogen from samples prior to the TaqMan PCR.CONCLUSIONS:The root mat pathogen was detected successfully from a range of sample types using the enriched fluorogenic PCR assay, negating the need for complex DNA extraction procedures and post-PCR processing techniques such as gel electrophoresis. The technique is therefore a rapid and cost-effective detection method.SIGNIFICANCE AND IMPACT OF THE STUDY:This is the first known report of a fluorogenic, 5' nuclease, TaqMan assay designed to detect an Agrobacterium plant pathogen. The method can be used as a model system for the detection of other Agrobacterium pathogens.

Aims: Determination of genetic diversity among UK Burkholderia cepacia isolates from various environmental niches, principally woodland tree rhizospheres and onions.Methods and Results: Genus determination was made using polymerase chain reaction (PCR) amplification and fatty acid methyl ester profiling. Genetic diversity was investigated by repetitive sequence genetic PCR fingerprinting. Several onion isolates were similar to clinical isolates but others were diverse. Some environmental isolates were possibly synonymous with B. cepacia and B. gladioli but most from woodland rhizospheres were distinct and clustered together. The 16S rRNA genes of representatives from these clusters were PCR amplified, sequenced and phylogenetically compared with all known Burkholderia and related species. This revealed that the rhizospheric isolates had closest affinity with Burkholderia spp. with known bioremediative and biocontrol capabilities and were unrelated to taxa comprising plant or human pathogenic strains.Conclusions: All of the analyses investigated revealed that environmental and onion isolates of B. cepacia complex bacteria are genetically diverse but that woodland rhizospheric isolates are related to each other and unrelated to plant or human pathogenic strains.Significance and Impact of the Study: Woodland rhizospheric isolates of B. cepacia are potentially good candidates for use in bioremediation and biocontrol, as they appear distinct from plant or human pathogenic strains.

TaqMan™ PCR exploits the 5‘ nuclease activity of Taq DNA polymerase (4) in conjunction with fluoroenic DNA probes (6). Each probe, designed to hybridise specifically to the target PCR product, is labelled with a fluorescent reporter dye and a quencher dye. During PCR amplification the probe is digested by Taq DNA polymerase, separating the dyes, resulting in an increase in reporter fluorescence. Repeated PCR cycles result in exponential amplification of the PCR product and corresponding increase in fluorescence intensity.

PCR-amplification has been used to screen 75 isolates of the Burkholderia cepacia complex for the cblA pilin gene. PCR-amplified products of the correct size (664 bp) were cloned and sequenced and the sequences compared. Apart from in the control, epidemic cystic fibrosis (CF)-associated B. cepacia lineage we also identified, for the first time, cblA genes in a unique, non-CF clinical isolate from France and a plant (onion) pathogenic isolate from Italy. The sequence of the cblA gene amplified from the clinical isolate was more diverged from the epidemic lineage than that amplified from the onion pathogenic isolate.

Modern methods suitable for characterising, identifying and detecting bacteria associated with plants (pathogens, contaminants, epiphytes, endophytes) are reviewed. Preferred methods for characterisation and identification include fatty acid profiles and repetitive sequence - PCR profiles. Preferred methods for detection include well established methods such as isolation and immunofluorescence and novel PCR assays based on fluorigenic probes.

Since 1993 many hydroponic cucumber (Cucumis sativus) crops in the UK have been affected by a root disorder known as root mat, a condition first described in soil and straw bed crops in the 1970s. The predominant symptom is extensive root proliferation within the rockwool propagation cube and across the rockwool slab surface. In severe cases a loss in marketable yield has been reported. Recently, the causal agent of this disease was shown to be wild-type Agrobactrium biovar 1 strains harbouring an Ri-plasmid (Weller et al. 2000). Since 1997, sporadic cases of a similar disorder have also been observed in commercial tomato (Lycopersicon esculentum) crops. Tomato plants affected by root mat were collected from crops across the UK. All crops were grown hydroponically in rockwool and the incidence of affected plants ranged from < 1% to 50%. Rhizogenic strains of Agrobacterium biovar 1 were isolated from affected but not from healthy roots, of symptomatic crops. Suspensions of three isolates of the bacterium (NCPPB 4060, 4061, 4062–National Collection of Plant Pathogenic Bacteria. Central Science Laboratory, Sand Hutton, York YO41 1LZ.) were inoculated, as in previous experiments on cucumber (Weller et al. 2000), onto six rockwool slabs each containing four young plants of one of two tomato varieties (cv. Pronto – a round variety; cv. Favorita – a cherry variety). Symptoms were initially seen five weeks after inoculation and at three months 16 plants from five of the six slabs were symptomatic. Symptoms were observed as increased root production across the propagation cube surface and swelling of the cube and/or root proliferation across the slab surface. Although root growth was excessive compared with normal tomato plants, it was less than that seen in root mat affected cucumbers. Rhizogenic Agrobacterium bv. 1 strains were re-isolated from four of the five slabs with symptoms and from the one slab without symptoms. No symptoms developed and no rhizogenic Agrobacterium strains were isolated from uninoculated control plants. Investigations on a heavily infected cherry tomato crop (cv. Favorita) at a commercial nursery indicated that fruit diameter was reduced by 2–3 mm in infected compared with healthy crops, leading to a higher proportion of unmarketable fruit (< 28 mm). Fruit numbers did not appear to be affected, although overall fruit yield was reduced by 15%. This is the first UK record of rhizogenic Agrobacterium causing root mat symptoms in tomato. Although tomato is susceptible to crown gall producing Agrobacterium strains, crown gall is a rare disease of glasshouse tomatoes. Non-pathogenic strains of Agrobacterium (without the plasmid) are commonly found on hydroponic tomato crops. This work was funded by the Horticultural Development Council (HDC) Project PC 149.

ABSTRACT A fluorogenic (TaqMan) PCR assay was developed to detect Ralstonia solanacearum strains. Two fluorogenic probes were utilized in a multiplex reaction; one broad-range probe (RS) detected all biovars of R. solanacearum , and a second more specific probe (B2) detected only biovar 2A. Amplification of the target was measured by the 5′ nuclease activity of Taq DNA polymerase on each probe, resulting in emission of fluorescence. TaqMan PCR was performed with DNA extracted from 42 R. solanacearum and genetically or serologically related strains to demonstrate the specificity of the assay. In pure cultures, detection of R. solanacearum to ≥10 2 cells ml −1 was achieved. Sensitivity decreased when TaqMan PCR was performed with inoculated potato tissue extracts, prepared by currently recommended extraction procedures. A third fluorogenic probe (COX), designed with the potato cytochrome oxidase gene sequence, was also developed for use as an internal PCR control and was shown to detect potato DNA in an RS-COX multiplex TaqMan PCR with infected potato tissue. The specificity and sensitivity of the assay, combined with high speed, robustness, reliability, and the possibility of automating the technique, offer potential advantages in routine indexing of potato tubers and other plant material for the presence of R. solanacearum .

An unusual root disease of cucumbers has affected an increasing number of hydroponic crops in the UK since 1993. Symptoms include upward growth of roots, root proliferation sufficient to distort the shape of rockwool cubes and slabs, thickened roots and occasionally reduced root production and an increase in bent fruit. Rhizogenic Agrobacterium biovar 1 has previously been associated with the problem. In a survey of 37 nurseries in 1997, rhizogenic Agrobacterium bv. 1 isolates were obtained from 17 of 22 nurseries with symptoms and only occasionally from nurseries without symptoms. Young cucumber plants inoculated with rhizogenic Agrobacterium bv. 1 isolates, previously obtained from cucumber roots with symptoms, developed typical symptoms. The survey data and pathogenicity test support rhizogenic Agrobacterium bv. 1 as the primary causal agent of root mat in the UK. Although Agrobacterium bv. 1 isolates that do not harbour an Ri-plasmid are considered avirulent, their abundance in rockwool slabs as potential vectors of the Ri-plasmid may be a major factor in the epidemiology of the disease.

Potato brown rot disease, caused by the bacterium Pseudomonas solanacearum (Smith) Smith biovar 2A, has occurred in Southern Europe (1940-1970) and in Sweden (1972). More recently, an isolated outbreak recorded in England (1992) was followed by reports of other sporadic outbreaks in northern Europe. In the Netherlands, more widespread outbreaks were reported in 1995 and infected seed potatoes were implicated as sources of infection. Evidence from several countries suggests that infected Solanum dulcamara growing in watercourses is an important secondary host in which the pathogen can successfully overwinter and from which it can spread to potato crops when associated water is used for irrigation. Current control measures, governed by national and European Union plant health legislation, rely on (i) accurate detection and reporting of brown rot outbreaks and distribution of the pathogen, (ii) prevention of importation of potatoes from known infected areas and (iii) restrictions on use of infested land and irrigation water for potato production.

Declining Turkey oaks (Quercus cerris) in Central Italy were studied. Bacteria were isolated from woody tissues between necrotic areas and healthy tissues beneath the bark, from apparently healthy tissues and from exudates oozing from the trunk in spring. A total of 98 bacterial isolates were collected. The determination of the bacteria was by fatty acid profiling and biochemical tests. Erwinia herbicola was frequently found both in necrotic and healthy tissues. Ice-nucleation active species were present. None of the species isolated was considered responsible for the decline.

Approximately 500 fatty acid profiles were prepared for 340 strains of plant-pathogenic and other bacteria currently or recently classified in the genus Pseudomonas Migula 1984. Strains representing some infraspecific taxa were included. The fatty acid profiles were stable and reproducible provided that cultural and chemical techniques were standardized. The 2- and 3-hydroxy fatty acids were found to be useful in grouping strains into six major groups, several of which were further differentiated into subgroups. Group 1 contained strains of the following species and subspecies: Pseudomonas aeruginosa, P. agarici, P. asplenii, P. aureofaciens, P. caricapapayae, P. chlororaphis, P. cichorii, P. ficuserectae, P. fluorescens, P. fuscovaginae, "P. gingeri," P. marginalis, P. meliae, P. putida, "P. reactans," P. syringae, P. tolaasii, and P. viridiflava (subgroup 1a); P. corrugata (subgroup 1b); P. rubrisubalbicans (subgroup 1c); P. alcaligenes, P. pseudoalcaligenes subsp. Pseudoalcaligenes, and P. stutzeri (subgroup 1d); P. amygdali (subgroup le); and P. cattleyae NCPPB 1874 (subgroup 1f). All group 1 strains contained 10:0 3-OH and 12:0 3-OH, and most group 1 strains also contained 12:0 2-OH. Group 2 contained strains belonging to the following taxa: P. andropogonis, P. caryophylli, P. cepacia, P. gladioli, P. plantarii, and P. glumae (in part) (subgroup 2a); P. glumae (in part) (subgroup 2b); and P. solanacearum, P. syzygii, and the banana blood disease bacterium (subgroup 2c). All of the group 2 strains contained 14:0 3-OH, 16:0 3-OH, and 18:1 2-OH; most also contained 16:1 2-OH and 16:0 2-OH. Group 3 contained strains belonging to the following taxa: Comamonas acidovorans, P. avenae, P. cattleyae NCPPB 961, P. pseudoalcaligenes subsp. citrulli, P. pseudoalcaligenes subsp. konjaci, P. rubrilineans (subgroup 3a); and Comamonas testosteroni (subgroup 3b). All of the group 3 strains contained 10:0 3-OH. The group 4 strains were members of Sphingomonas paucimobilis, and all contained only 14:0 2-OH. The group 5 strains were members of P. ftectens and contained 12:0 2-OH, 14:0 2-OH, and 14:0 3-OH. The group 6 strains were P. betle, P. cissicola, P. hibiscicola, Xanthomonas maltophilia, and Xanthomonas campestris pv. campestris strains, and all contained 12:0 3-OH, 11:0 iso 3-OH, and 13:0 iso 3-OH. Within each group or subgroup, qualitative and quantitative differences in profiles occurred for most species. Differences were also found at the infraspecific level for some taxa. My results support genomic and other data which show that the plant-pathogenic and other pseudomonads tested should be placed in at least six genera.

D.E. STEAD, J.E. SELLWOOD, J. WILSON AND I. VINEY, 1992. Fatty acid profiles of 773 strains representing 25 taxa of plant pathogenic and related saprophytic bacteria were compared with two commercially available broad‐spectrum libraries and one self‐generated library based primarily on cultures from the National Collection of Plant Pathogenic Bacteria. The accuracy of identification at specific level was often 100%, although for some closely related species and infraspecific taxa accuracy was sometimes significantly less than this. The accuracy of identification of Xanthomonas campestris pathovars was much better than for Pseudomonas syringae pathovars. Almost all identifications were made within24–48 h. Standardization of cultural conditions was essential. Hydroxy fatty acids were of great taxonomic value in classification of Gram‐negative bacteria. Improved library development and standardization of cultural and analytical techniques will further increase the accuracy of identification. Fatty acid profiling offers a valuable rapid, accurate method for identification of many bacteria.