Mycol. Res. 109 (11): 1276–1287 (November 2005). f The British Mycological Society 1276 doi:10.1017/S0953756205003928 Printed in the United Kingdom. Phylogenetic analysis and Real Time PCR detection of a presumbably undescribed Peronospora species on sweet basil and sage Lassaad BELBAHRI1*, Gautier CALMIN1, Jan PAWLOWSKI2 and Francois LEFORT1 1 Laboratory of Applied Genetics, School of Engineering of Lullier, University of Applied Sciences of Western Switzerland, 150 Route de Presinge, 1254 Jussy, Switzerland. 2 Department of Zoology and Animal Biology, University of Geneva, Sciences III ; 30, quai Ernest-Ansermet, 1211 Geneva, 4, Switzerland. E-mail : lassaad.belbahri@etat.ge.ch Received 20 February 2005; accepted 21 July 2005. Downy mildew of sweet basil (Ocimum basilicum) has become a serious disease issue for the producers of sweet basil in Switzerland since it was first recorded in 2001. Reported in Africa in Uganda as early as 1933, major outbreaks of this disease in Europe were first noted in Italy in 1999 and in the USA from 1993. Previous reports have named the pathogen as Peronospora lamii. Its preferential hosts belong to the Lamiaceae family including basils (Ocimum spp.), mints (Menta spp.), sages (Salvia spp.) and other aromatics. This study investigated the taxonomic status of the downy mildew pathogen, using both morphological characters and molecular analysis of the ITS region of the rDNA. The inherent variability of conidial dimensions made species differentiation difficult. Sequence homology and phylogenetic analysis of nine collections of the Peronospora on sweet basil showed unique ITS sequences distinct from those of P. lamii and any other sequenced Peronospora species. This paper describes and illustrates the morphology of this presumably undescribed species of Peronospora. Its taxonomic position and relationships with other related species in the same genus are presented and discussed. In addition to this work, PCR primers for real time PCR analysis have been developed for the specific detection of this downy mildew pathogen from infected tissues or seeds. It is shown that these primers can also be used in classic PCR. INTRODUCTION The common sweet basil (Ocimum basilicum) with its several cultivated varieties is an annual aromatic plant, widely grown because of its pleasant odour and taste. It is an economically important herb in several Mediterranean and European countries where it has been a culinary herb since before the written world. In Switzerland approximately 9 ha are grown annually for fresh and processed consumption and large quantities imported mainly from France, Morocco, Israel and Brazil. Downy mildew proved recently to be one of the most destructive diseases of sweet basil. Since the disease was first recorded in Switzerland in 2001 (Heller & Baroffio 2003, Lefort, Gigon & Amos 2003), it has become prevalent, affecting most plants each season. Typically, leaves of infected plants were initially slightly chlorotic, especially near the central vein. Within 2–3 d, a characteristic grey to brown, furry growth was evident * Corresponding author. on the lower epidermis of infected leaves. These symptoms sometimes occurred on the upper epidermis of leaves (Garibaldi et al. 2004). The pathogen can spread rapidly through the plants, under conditions of high humidity and cool temperatures, causing complete crop losses in some greenhouses. In the past years, based on epidemiological data and microscopic observations, the pathogen responsible for downy mildew disease on Lamiaceae was reported as the straminipile Peronospora lamii on basil plants in Benin (Gumedzoe et al. 1998), Italy (Martini et al. 2003), Switzerland (Heller & Baroffio 2003, Lefort et al. 2003), and as Peronospora sp. in Italy (Garibaldi et al. 2004). On Lamium, Salvia and other aromatic plants, it was reported as P. lamii in Austria (Plenk 2002), Israel (Gamliel & Yarden 1998), Italy (Minuto, Pensa & Garibaldi 1999, Voltolina 2001), New Zealand (Hill, Pearson & Gill 2004) and the USA (McMillan & Graves 1994, Byrne 2000, 2001, Holcomb 2000, Leahy 2001, Klassen 2002). It was also reported on Lamium maculatum and L. rubrum in the UK (Bentley 1980). L. Belbahri and others In recent times, four Peronospora species have been recorded as infecting species of the Lamiaceae family which includes basil (Ocimum spp.), mint (Menta spp.), sage (Salvia spp.) and other aromatics (Hansford 1933, Holcomb 2000, Voglmayr 2003, Garibaldi et al. 2004, Hill et al. 2004). In addition, it has to be mentioned that many more species have been described from Lamiacea, of which the status as distinct species is mostly unclear or doubtful (Constantinescu 1991). However, the status of these species is complicated. According to previous publications (Bentley 1980, Holcomb 2000, Voglmayr 2003), P. lamii infects Lamium, Salvia, and possibly a few other plants in the mint family. ITS sequences from this species are available from an isolate from L. purpureum (Voglmayr 2003). P. stigmaticola infects Mentha longifolia and ITS sequences are also available (Voglmayr 2003). The type host of P. swinglei also belongs to the genus Salvia (S. reflexa), and it has been recorded from several Salvia spp., such as S. officinalis, in both the USA and Europe (Müller 1999). It should be noted that the identity of the North American and European ‘P. swinglei ’ has not been demonstrated by molecular studies, and investigations from the North American type hosts are needed to clarify that question. Peronospora sp. has been recorded on sweet basil plants (Garibaldi et al. 2004) and on sage (this study). This last species was differentiated from P. lamii on the basis of the smaller average conidial dimensions of the latter on Salvia splendens and S. coccinea. Conidia measured were 21r18 mm (16–26r15–23) for P. lamii, and 28r22 mm (20–35r15–25) for the parasite of Salvia and Ocimum basilicum (Holcomb 2000, Garibaldi et al. 2004). Species differentiation on morphological characteristics alone is problematic as these can change with host genotype and environmental conditions (Hall 1996) or with the conservation state of the sample. A more accurate circumscription of species may be obtained by combining phylogenetic analysis of molecular characters with morphological characters (Scott, Hay & Wilson 2004). Previous studies have shown the ITS regions of rDNA to be useful for the differentiation of downy mildew species and related organisms at the species level (Crawford et al. 1996, Cooke et al. 2000, Constantinescu & Fatehi 2002, Voglmayr 2003, De Cock & Lévesque 2004). Knowledge of host specificity of a pathogen has important implications, such as understanding the role of alternate hosts in the epidemiology of the disease or the potential for host range expansion (LindqvistKreuse, Koponen & Valkonen 2002). Host specificity of isolates of these three species has not been studied. However, the host ranges of the three Peronospora species in question exhibit a degree of overlap. Of the known host species present in Switzerland, Ocimum basilicum is an introduced aromatic plant, in contrast to Salvia officinalis and Mentha which have wild relatives growing all over Switzerland and often in close proximity to the fields of cultivated aromatic plants. 1277 The three main forces introducing genotypic variations into a pathogen population are mutation, recombination and migration (Burdon & Silk 1997). Neither the distribution nor the genetic variability of these species have been studied. However, since Peronospora on sweet basil is not endemic in Switzerland, the disease was not known and described before and no herbarium samples were found to be infected, it is likely that the pathogen was introduced. It should be noted that seeds of sweet basil and sage plants were imported by growers into Switzerland, suggesting that the origin could reside in contaminated seeds or the large quantities of basil cuttings imported from all over the world, mainly from France, Israel, Morocco, Germany, Brazil, and other parts of Europe. Basil downy mildew control will benefit from a rapid and reliable detection method. Since downy mildew species were not characterized at the genetic level and no proteins were described, diagnostic approaches based upon serology would not be available within a reasonable time. As a result, a variety of alternative diagnostic approaches have been used for detecting downy mildew pathogens, including microscopic examination (Holcomb 2000), bioassays (Garibaldi et al. 2004, Catherine Baroffio, pers. comm.), nucleic acid hybridization (De Cock et al. 1998) as well as PCR (Lindqvist-Kreuse et al. 2002, Scott et al. 2004). Despite the multiple advantages of PCR-based methods over other techniques, they are also subject to practical limitations for routine diagnostic purposes : in particular they tend to be labour intensive and as ‘open tube ’ methods, the risk of carry-over contamination and the possibility of false positive results remains. The use of homogeneous assay systems such as Real Time fluorescent PCR has been successfully applied to the detection of plant viruses (Mumford et al. 2000, Korimbocus et al. 2002, Boonham et al. 2002, 2004), the oomycete Phytophthora ramorum (Ivors, Wilkinson & Garbelotto 2003) and Peronospora parasitica (Brouwer et al. 2003). Typically, in a real-time PCR amplification, the accumulation of the amplicon is monitored for each cycle based on the emission of fluorescence (Mackay, Arden & Nitsche 2002). Amplicons are detected using different chemical methods, divided into either amplicon specific detection methods (Holland et al. 1991, Tyagi & Kramer 1996, Wittwer et al. 1997, Whitcombe et al. 1999) or amplicon nonspecific methods (Morrison, Weiss & Wittwer 1998). Though the advantage of amplicon specific detection methods is the high specificity of the generated fluorescent signal, different amplicon detection requires different probes. This disadvantage does not apply to amplicon non-specific detection chemistries that are based on fluorophores which intercalate inside double-stranded DNA. Such fluorophores can also associate with primer-dimers and non-specific amplification products, which may seriously disturb the interpretation of results. However, a melting curve analysis at the end of the PCR to identify the product An undescribed Peronospora sp. pathogen of basil and sage allows an accurate estimation of the PCR amplification. The main objectives of the present study were to determine the inter- and intra-specific relationships of downy mildew pathogens of the Lamiaceae based on DNA sequence analysis of the ITS1, 5.8S gene, and ITS2 region, and morphological dimensions of conidia. A secondary objective was to develop Real Time PCR primers for the specific detection of the sweet basil downy mildew pathogen, with the goal to assess seed borne downy mildew and for environmental monitoring of the pathogen. MATERIALS AND METHODS Sample collection Infected leaf samples were collected from commercial sweet basil and sage plants in greenhouses and open fields in the Lake Geneva region, and also obtained from some other areas with records of Peronospora lamii or Peronospora sp. on sage and sweet basil around the world (Table 1). When no conidiophores were present after observation under the binocular microscope, leaves were incubated in sealed trays containing damp tissue at 100 % humidity, and 12 xC in the dark for 12 h to induce sporulation. Conidia were washed from leaves into centrifuge tubes with distilled water. Conidial suspensions were concentrated by centrifugation at 3000 g for 20 min at room temperature, and the supernatant removal by pipetting. Conidia and conidiophores were stained with a standard Lugol’s iodine solution. Morphological observations The morphology of the available asexual structures of collected specimens was observed visually at a 400 fold magnification with an HBO50 AC microscope with CP-Achromat objectives (Carl Zeiss, Feldbach). Observations were compared with the published descriptions (Garibaldi et al. 2004) of Peronospora species known to infect members of the Lamiaceae family. Conidia were measured at a 400 fold magnification obtained with an AxioCam MRC camera (Carl Zeiss) and processed with the Axiovision 3.1.21 software (Carl Zeiss). The first 100 conidia observed from the different samples analysed were measured, with misshapen, or dehydrated, conidia ignored. Statistical analysis of conidial dimensions was carried out using analysis of variance (ANOVA) in the software program Minitab v.13.31 (Minitab, State College). DNA extraction DNA was purified from conidial suspensions, infected basil tissue and seeds with the use of the DNA-Easy Plant Mini kit (Qiagen, Basel), according to manufacturer’s specifications. Quality was checked by visualization under UV light following electrophoretic 1278 separation with a molecular mass standard (HindIII/ EcoRI DNA Marker, Biofinex, Praroman) in 1 % agarose (Biofinex) gel in 1r TBE, subjected to 100 V for 1 h and stained with ethidium bromide (0.5 mg mlx1). Concentrations were assayed in a S2100 Diode Array spectrophotometer (WPA Biowave, Cambridge, UK). DNA amplification ITS amplifications of downy mildew samples (Table 1) were carried out using previously described universal primers ITS4 and ITS6 that target conserved regions in the 18S and 28S rRNA genes (Table 2 ; White et al. 1990, Cooke et al. 2000). The reaction mixture contained 1r PCR buffer (75 mM Tris-HCl (pH 9.0), 50 mM KCl, 20 mM (NH4)2 SO4), 0.1 mM dNTPs, 0.25 mM of each primer, 1.5 mM MgCl2, 1 U of Taq Polymerase (Biotools, Madrid) and 1 ml of conidial DNA in a total volume of 50 ml. Amplifications were carried out in a Master Gradient thermocycler (Eppendorf, Schönenbuch) according to the following amplification programme : an initial denaturation step of 95 x for 2 min followed by 30 cycles including denaturation for 20 s at 95 x, annealing for 25 s at 55 x and extension for 50 s at 72 x. Amplification was terminated by a final extension step of 10 min at 72 x (Cooke et al. 2000). PCR products were separated in 1% agarose (Biofinex) gels in 1r TBE, subjected to 100 V for 1 h, stained with ethidium bromide (0.5 mg lx1) and visualised under UV light. Conidial suspensions known to contain contaminants from non-oomycete species or infected plant tissues were submitted to a preliminary step in a seminested protocol using the primer DC6 (Table 2). DC6, used in combination with the primer ITS4, specifically amplified members of the orders Pythiales and Peronosporales (Cooke et al. 2000). The first step of the nested protocol was carried out using the primers DC6 and ITS4 in a 25 ml reaction mixture. Reaction mixture contents and thermocycler settings were as described above. Following amplification, a 1 ml aliquot of the reaction product was substituted to the conidial DNA as template in reaction described above. DNA sequencing and phylogenetic analysis Amplicons were purified using a Minelute PCR Purification Kit (Qiagen, Holubrechtikon), according to manufacturer’s specifications. Quantity and quality were checked as described above for DNA extraction. Amplicons were sequenced directly in both sense and antisense directions. All pathogen samples were sequenced twice and a consensus sequence was created from the duplicates. DNA sequences have been deposited in GenBank (Table 1). Sequences were aligned manually using Seaview (Galtier, Gouy & Gautier 1996). The maximum likelihood (ML) trees were obtained using the PhyML program (Guindon & Gascuel 2003) with the HKY L. Belbahri and others 1279 Table 1. List of species and specimens included in the study. Species Collectiona Host Origin Peronospora sp. Peronospora sp. Peronospora sp. Peronospora spb. Peronospora sp. Peronospora sp. Peronospora sp. Peronospora sp. Peronospora sp.d P. conglomeratac P. conglomerata P. sordidac P. sordida P. radii P. lamii c P. lamii P. stigmaticola P. arborescens P. chenopodii-polyspermi P. kochiae-scopariae P. tabacina P. obovata P. rumicis P. boni-henrici P. chenopodii P. holostei P. destructor P. polygoni P. claytoniae P. arenariae P. alsinearum P. trivialis P. paula P. parva P. arthurii P. esulae P. chrysosplenii P. dicentrae P. bulbocapni P. alpicola P. pulveracea P. aff. Ranunculi P. illyrica P. ranunculi P. alta P. flava P. arvensis P. agrestis P. grisea (V. beccabunga) P. grisea (V. serpyllifolia) P. violae UASWS0007 UASWS0008 UASWS0009 UASWS0010 UASWS0033 UASWS0034 UASWS0035 UASWS0036 UASWS0037 WU22894 Ocimum basilicum O. basilicum O. basilicum Salvia officinalis Ocimum basilicum O. basilicum O. basilicum O. basilicum O. basilicum Geranium molle G. molle Scrophularia nodosa S. nodosa Anthemis austriaca Lamium purpureum L. purpureum Mentha longifolia Papaver rhoeas Chenopodium polyspermum Bassia scoparia Nicotiana alata Spergula arvensis Rumex acetosa Chenopodium bonus-henricus C. album Holosteum umbellatum Geneva, CH Geneva, CH Geneva, CH New Zealand Geneva, CH Geneva, CH Geneva, CH Geneva, CH Italy Austria GenBank Austria GenBank GenBank Austria Genbank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank a b c d WU22894 WU22907 Polygonum aviculare Claytonia virginica Moehringia trinervia Stellaria media Cerastium holosteoides C. holosteoides Stellaria holostea Oenothera biennis agg. Euphorbia esula Chrysosplenium alternifolium Dicentra canadensis Corydalis cava Ranunculus aconitifolius Helleborus niger Ranunculus recurvatus R. illyricus R. acris Plantago major Linaria vulgaris Veronica triloba V. chamaedrys V. beccabunga V. serpyllifolia Viola arvensis Source year GenBank accession no. 2003 2003 2004 2003 2004 2004 2004 2004 2005 1999 AY831719 AY831720 AY831721 AY831722 AY884603 AY884604 AY884605 AY884606 AY919301 AY919304 AY198246 AY919302 AY198247 AY198296 AY919303 AY198261 AY198295 AY198292 AY198291 AY198290 AY198289 AY198288 AY198287 AY198286 AY198285 AY198283 AB021712 AY198282 AY198281 AY198280 AY198279 AY198278 AY198277 AY198276 AY198284 AY198275 AY198274 AY198273 AY198272 AY198271 AY198270 AY198269 AY198268 AY198267 AY198248 AY198245 AY198244 AY198243 AY198242 AY198241 AY198240 1999 1999 Collection numbers for specimens sequenced in this study. Sample provided by Eric McKenzie. Sample provided by Hermann Voglmayr. Sample provided by Patrizia Martini. (Hasegawa, Kishino & Yano 1985) model allowing transitions and transversions to have potentially different rates and General Time Reversible (GTR) model allowing all rates to be different (Lanave et al. 1984, Rodriguez et al. 1990). In order to correct the amongsite rate variations, the proportion of invariable sites (I) and the alpha parameter of gamma distribution (G), with eight rates categories, were estimated by the program and taken into account in all analyses. Nonparametric ML bootstraps (with 100 replicates) were calculated using PhyML. Bayesian inferences (BI) were obtained with MrBayes v.3.0 (Huelsenbeck & Ronquist 2001), using the same models of DNA evolution as for the ML analyses. The program was run for 1 M An undescribed Peronospora sp. pathogen of basil and sage 1280 Table 2. PCR primers used in this study: oligonucleotide sequences and location within the genomic ribosomal RNA gene. Primer a DC6 ITS6b ITS4b Bas-F Bas-R LAM-F LAM-R a b Sense Sequence (5k p 3k) Location Forward Reverse Reverse Forward Reverse Forward Reverse GAGGGACTTTTGGGTAATCA GAAGGTGAAGTCGTAA TCCTCCGCTTATTGATATGC CCGTATCAACCCAATAATTTGGGGGTTAAT TTACAATCGTAGCTACTTGTTCAGACAAAG CACGTGAACCGTATCAAACACTTT CGCCATGATAGGGCTTTCTCAA 18S gene 18S gene 28S gene ITS1 ITS1 ITS1 ITS1 Bonants et al. (1997). White et al. (1990). generations, and sampled every 100 generations, with four simultaneous chains. The trees, sampled before the chains reached stationarity, were discarded. NJplot and Treeview were used to view ML and Bayesian trees, respectively. Primer design and Real Time PCR analysis Primers specific to the different samples of downy mildew pathogens were developed through visual comparison of the alignment of all sequences used for the phylogenetic analysis (Table 2). Primers were specifically designed to discriminate between Lamiales downy mildew pathogens, particularly those infecting the Lamiaceae family (Peronospora lamii, P. stigmaticola and Peronospora sp.). The specificity of potential primers was further tested by BLAST (Altschul et al. 1997) searching GenBank for compatible sequences. Following the identification of suitable target regions in the sweet basil and sage Peronospora sequence, primers were tested in real time PCR reactions against all Peronospora spp. samples from Lamiaceae. Amplification mixtures (20 ml final volume), were made using the Light Cycler Fast Start DNA Master SYBR Green I kit, to which 1 ng of template DNA and primers at a final concentration of 5 mM were added. Amplifications were carried out in capped capillary tubes in a Lightcycler 2.0 Real time PCR systems (Roche Diagnostics, Rotkreuz) using an initial denaturation at 95 x for 10 min, followed by 45 cycles of 95 x for 10 s, 70 x for 5 s and 72 x for 15 s. No post-PCR manipulations, such as gel electrophoresis analysis, were required, hence removing many of the contamination problems associated with conventional PCR. RESULTS Morphological observations All downy mildew samples obtained from all basil and sage plants displayed sexual structures with general morphological characteristics consistent with Peronospora spp. (Martini et al. 2003, Garibaldi et al. 2004) known to infect members of the mint family and several Lamium spp. (Lamiaceae). Microscopic examination revealed conidiophores branching 2–7 times. Conidiophores terminate with conidiogenous cells bearing single conidia (data not shown). Conidia were elliptical and greyish as described by Garibaldi et al. (2004). The pathogen was identified as a Peronospora sp. based on its morphological characteristics (Spencer 1981). Mean conidial lengths of downy mildew samples obtained from Switzerland were between 23–36 mm (av. 29 mm) ; the mean conidial breadths varied between 18–29 mm (av. 23 mm). Previously published mean conidial lengths for P. lamii varied between 16 and 26 mm (av. 21 mm) and between 20 and 35 mm (av. 28 mm) for Peronospora sp. Published mean conidial widths varied between 15 and 23 mm (av. 18 mm) for Peronospora lamii and between 15 and 25 mm (av. 22 mm) for Peronospora sp. (Table 3). Sequence analysis The nine samples of downy mildew pathogens obtained from Ocimum basilicum and Salvia officinalis in this study were differentiated into five ITS sequence types. These sequences differed at five nucleotide positions in the ITS1 and ITS2 regions. Group 1 consisted of three samples (UASWS0007, UASWS0008 and UASWS0009). Group 2 included also three samples (UASWS0033, UASWS0034 and UASWS0036). However, Groups 3, 4 and 5 consisted of single samples (UASWS0035, UASWS0037, and UASWS0010). Due to the small number of samples analysed for conidial dimensions (two samples), no correlation between ITS grouping and conidial length could be established. The sequence homology for the ITS region between samples collected from sweet basil plants in this study and P. lamii (GenBank accession No. AY198261 ; Voglmayr 2003) is about 88 %. However, downy mildew samples, of the present study, had sequence homologies of 91–92 % with the relevant sequences obtained from GenBank for the species Peronospora conglomerata, P. sordida, P. radii and P. stigmaticola (GenBank accession nos AY198246, AY198247, AY198296 and AY198295, respectively; Voglmayr 2003). Phylogenetic analysis The position of representatives of each of the five groups (UASWS0007 : AY831719, 10 : AY831722, 33: AY884603, 35 : AY884605, and 37: AY919301) L. Belbahri and others 1281 Table 3. Conidia dimensions of downy mildew species recorded on Ocimum and Salvia spp. Length Breadth Collection min. mean max. min. mean max. Host Reference UASWS0007 UASWS0009 Peronospora sp. Peronospora lamii Peronospora lamii 25.5 23.5 20 23 16 29.54 28.81 28 25.3 21 37 33.5 35 27.5 26 18.5 20 15 15 15 23.76 23.38 22 21 18 29 27.5 25 27 23 This study This study Garibaldi et al. (2004) Martini et al. (2003) Holcomb (2000) Peronospora lamii 20 21 22 16 17 18 Ocimum basilicum O. basilicum O. basilicum O. basilicum Salvia splendens S. coccinea S. lanceolata sequences and 37 additional sequences of Peronosporales, including representatives on Lamiales and Ranunculates, was illustrated in Fig. 1. The tree was rooted with Peronospora violae following Voglmayr (2003). In all analyses, the sequences AY831719, AY831722, AY884603, AY884605 and AY919301 form a monophyletic group supported by high values of bootstrap (BS) and posterior probabilities (PP) (100 % BS and 100 % PP). In the ML tree with GTR+G+I model (Fig. 1), the new sequences form a sister group to the clade of P. arvensis+P. agrestis+P. grisea. However, the bootstrap support for this topology is very weak (20 % BS). In the ML tree calculated with the HKY+I model (results not shown), the new sequences form a sister group to the cluster comprising P. conglomerata and the clade containing P. arvensis+P. agrestis+P. grisea, but without any significant bootstrap support (data not shown). There is also weak support for the position of the new sequences in the BI trees (results not shown), where they branch as a sister group to P. conglomerata with rather weak posterior probabilities (51x66% PP). In general, the relations within the Lamiales clade are not resolved, which is in agreement with previously published analyses of Peronospora sequences (Voglmayr 2003). In contrast to that study, however, we did not find a strong support for a Lamiales clade, possibly due to the limited number of outgroup sequences in our data. On the other hand, both our and Voglmayr’s analyses provide a congruent evidence for the monophyly of Ranunculales clade, supported in our analyses by 100 % BS and 99 % PP. The relationships among this clade are also very similar in both our and Voglmayr’s study. Primer design and Real Time PCR analysis The primer pair Bas-F/Bas-R produced a single band approximately 134 base pairs in size for all downy mildew (Peronospora sp.) specimens collected in this study. No bands were produced when tested against specimens of Peronospora lamii, P. conglomerata, P. sordida or P. radii (Fig. 2). Cross amplification was sometimes noticed with P. sordida only at high DNA template concentration. Cross-reacting amplification Hansford (1933) peaks were delayed and under gel electrophoresis analysis, related bands were fainter than the observed bands for the sweet basil Peronospora sp. This is similar to results of a study of Phytophthora ramorum and its close relative Phytophthora lateralis (Ivors et al. 2003, L. B., unpub. data). For Peronospora lamii, the same specificity was observed with the primer pair Lam-F/ Lam-R which produced also a single band approximately 106 base pairs in size. No cross reactions were noticed when this primer pair was tested against Peronospora sp., P. conglomerata, P. sordida or P. radii (Fig. 3). The primer pair Bas-F/Bas-R proved to be very efficient for detecting Peronospora sp. in infected seed batches (Fig. 4). Further screening of 11 commercial sweet basil batches with the protocol developed in this study detected the presence of the Peronospora sp. in nine of them (results not shown). DISCUSSION Mean conidial dimensions of specimens of downy mildew pathogens collected from basil plants in this study were generally larger than those in previously published accounts for Peronospora spp. known to infect sweet basil species. Based on published records, conidial dimensions of P. lamii ranged from 16–26r15–23 mm, whilst those of Peronospora sp. were 20–35r15–25 mm. In the present study, conidial dimensions were 23–36r18–29 mm. This overlap highlights the difficulty of differentiating Peronospora spp. only on spore dimensions, and emphasizes the need for alternative means of identification. It is possible that part of the overlap in conidial dimensions between P. lamii and Peronospora sp. could be attributed to misidentification of species in the past. In order to unambiguously define the morphological dimensions of these two species, it would be necessary to first correctly identify samples, using methods such as the molecular techniques developed here. Based on discrete variations at a few nucleotide sites over the ITS regions, sweet basil Peronospora samples of the present study were divided into two groups of three samples, and three single distinct samples. This level of variation is comparable to variability recorded for P. sparsa samples on arctic bramble An undescribed Peronospora sp. pathogen of basil and sage 1282 sp Fig. 1. Phylogenetic position of UASWS0007, UASWS0010, UASWS0033, UASWS0035 and UASWS0037 inferred from ITS sequences by using ML method with GTR+G+I model. The numbers at nodes are non-parametric bootstrap values higher than 80 %. The length of branches is proportional to the number of substitutions per site as indicated in the scale. L. Belbahri and others 1283 1 2 3 4 5 6 7 8 9 10 11 (A) Peronospora sp. P. lamii P. sordida P. conglomerata P. radii (B) Amplification Curves 3.5 3.2 2.9 Fluorescence (530) 2.6 2.3 2 1.7 1.4 1.1 0.8 0.5 0.2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Cycles Fig. 2. Reaction of downy mildew samples to the PCR primer pair BAS-F and BAS-R. (A) Specificity of the primer pair BAS-F and BAS-R at different temperatures with DNA from infected plant material with Peronospora sp., P. sordida, P. conglomerata and P. radii. 1–11: annealing temperatures 53–73 xC. (B) Specificity of the primer pair BAS-F and BAS-R in real time experiments. Rectangle and circle refer respectively to100 and 10 ng DNA of Peronospora sp. Triangle, inverted triangle, cross and diagcross refer to 10 ng DNA from P. radii, P. lamii, P. sordida and P. conglomerata. Star refers to negative control. (Rubus arcticus) in a similar study in Finland, where six variable nucleotide sites was recorded for the ITS1 and ITS2 regions (Lindqvist-Kreuse et al. 1998). Also in a study of P. cristata on oilseed poppy in Tasmania (Scott et al. 2004), six variable nucleotide sites were recorded. In our study, no geographical basis was evident for the groupings based on ITS sequences of the Swiss and Italian samples ; this might be indicative of a more recent introduction of the pathogen, primarily spread by seed, and (or) long distance dispersal by wind from a single geographic region each season. The New Zealand specimen on Salvia officinalis has a slightly different sequence from the specimens found on Ocimum spp. Since Salvia officinalis is not closely related to Ocimum basilicum, the Peronospora from Salvia and Ocimum may represent separate lineages. However, to date, the main source of the primary inoculum of Swiss epidemics has not been identified. Sequence analysis of the ITS region of sweet basil Peronospora samples indicated that they have unique ITS sequences different enough from those of any described Peronospora species and mainly Peronospora lamii (AY198261 ; Voglmayr 2003). This unknown Peronospora sp. shared 88% sequence homology with P. lamii, which was comparable to the sequence homology observed between P. stigmaticola and P. lamii (90 %), P. sordida and P. lamii (90 %), and less than that observed for P. stigmaticola and P. radii (97 %). Phylogenetic analysis of the ITS region consistently indicated that sweet basil downy mildew samples in this An undescribed Peronospora sp. pathogen of basil and sage (A) 1 2 3 1284 4 5 6 7 8 9 10 11 P. lamii Peronospora sp. P. sordida P. conglomerata P. radii (B) Amplification Curves 4.1 3.7 3.3 Fluorescence (530) 2.9 2.5 2.1 1.7 1.3 0.9 0.5 0.1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Cycles Fig. 3. Reaction of downy mildew samples to the PCR primer pair LAM-F and LAM-R. (A) Specificity of the primer pair LAM-F and LAM-R at different temperatures with DNA from infected plant material with Peronospora sp., P. sordida, P. conglomerata and P. radii.1–11 : annealing temperatures 53–73 xC. (B) Specificity of the primer pair LAM-F and LAM-R in real time experiments. Rectangle and circle refer respectively to100 and 10 ng DNA of P. lamii. Triangle, inverted triangle, cross and diagcross refer to 10 ng DNA from P. radii, P. sordida, P. conglomerata and Peronospora sp. Star refers to the negative control. study shared a common ancestor with all Peronospora sequences we used. In all analyses, the sequences of sweet basil downy mildew samples formed a monophyletic group supported by high bootstrap and posterior probability values confirming the species status of this pathogen. Analysis of the ITS region of members of the Peronosporomycetes indicated that all species parasitic on leaves of the Lamiales (sensu Angiosperm Phylogeny Group 1998) were contained within a highly supported clade in the Bayesian analysis, but not in the MP bootstrap tree generated by Voglmayr (2003). In addition, P. conglomerata, a species infecting the unrelated Geraniaceae, is a member of this clade, supported by the close phylogenetic position and the high sequence homology we found between the Peronospora sp. and P. conglomerata. However, in contrast to Voglmayr’s study, we did not find strong support for the Lamiales clade, possibly due to the limited number of outgroup sequences in our data. Interestingly, the floricolous Peronospora group of the Lamiales is not closely related to this group but has its closest relatives amongst the other floricolous Peronospora species (Voglmayr 2003). In the study of Voglmayr (2003), the floricolous Peronospora group L. Belbahri and others 1285 Amplification Curves 4.4 4 3.6 Fluorescence (530) 3.2 2.8 2.4 2 1.6 1.2 0.8 0.4 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Cycles Fig. 4. Amplification plots showing the specific detection of Peronospora sp. in infected plant material and contaminated seed batch using primers BAS-F and BAS-R. Rectangle, circle, Triangle, inverted triangle and cross refer to 100, 10, 1, 0.1 and 0.1 ng DNA from infected Ocimum basilicum leaves with Peronospora sp. Diamond refers to 100 ng DNA from infected seeds. Star refers to negative control. formed a highly supported monophyletic clade, embedded within the species parasitic on leaves of Rubiaceae and Dispacaceae. Except for P. radii, which was also found on leaves, the floricolous Peronospora’s were confined to flowers or flower parts (e.g. petals, stigmata) of their hosts. In our study, the Peronospora sp. displayed a higher sequence homology with P. radii than with the mint pathogen P. stigmaticola. The apparent relationship between pathogen and host systematics was fully supported by the results of the present study. Sweet basil downy mildew samples of the present study grouped within the Lamiales downy mildews clade and were more closely related to P. conglomerata, P. arvensis, P. agrestis, and P. grisea, than to P. lamii. The Peronospora sp. isolated from Salvia officinalis in the New Zealand sample seems to be different from the other basil samples with a high value of support (BS 99 %). Due to the limited sampling and the lack of cross-infection studies, it is of course not possible to yet confidently claim the presence of two different species, but this should at least be considered as a possibility that there may be host specific races which could evolve into or already represent different species, depending on the species concept applied. Low ITS sequence divergence does not necessarily mean conspecificity, as there are examples of low ITS variation between sexually incompatible species. Thus, molecular evidence indicated that the Peronospora sp. found on Ocimum basilicum is a Peronospora species not yet characterised by ITS sequence analysis and may represent a new species. Since Peronospora on sweet basil is not endemic in Switzerland and the disease has not been known before, it is likely that the pathogen has been introduced, via either contaminated seeds or cuttings or wind dispersal. Hall (1996) argued that while the morphology of asexual structures could be used to reliably differentiate species of downy mildew pathogens at the genus level, molecular techniques such as those incorporating rDNA sequencing may need to be used to differentiate them at the species level. To date, examples for the use of ITS sequences to redefine the taxonomy of Peronospora species include P. rubi and P. sparsa, which have been shown to be conspecific (Kokko et al. 1999), and Peronospora spp. parasitic on Brassicaceae, which have been transferred into two new genera (Constantinescu & Fatehi 2002). In addition to the identification of the causal organism of downy mildew disease in sweet basil this study also led to the development of real time PCR primers for the specific detection of the pathogens P. lamii and Peronospora sp. DNA sequence comparison indicated that PCR primers would differentiate between P. sp. and P. lamii. BLAST searches of the GenBank database, An undescribed Peronospora sp. pathogen of basil and sage which is known to include sequences from the common pathogens of Lamiaceae (i.e. Fusarium oxysporum f. sp. basilici, Rhizoctonia solani, Sclerotinia spp., Botrytis cinerea, Fusarium tabacinum, Pythium sp., and Colletotrichum gloeosporioides) and testing against the most common Peronospora species in Switzerland did not indicate any cross reactions. These results indicate that these primers specifically discriminated between the Peronospora sp. and P. lamii. Previous studies aiming to detect downy mildew inoculum in seed showed that microscopic examination of soaked seeds (results not shown) or of seeds germinated on Petri dishes with the potential pathogen (Catherine Baroffio, pers. comm.) could not be used reliably to identify the downy mildew pathogen. Our assay demonstrated that the downy mildew pathogen could easily be detected in seed batches. This real time PCR assay could be adequate for large-scale detection of P. lamii and Peronospora sp. providing the perennial problem of large-scale nucleic acid extraction can be solved. It would offer a number of advantages over other systems currently available to diagnosticians, including improvements in sensitivity, a format suitable for automation using laboratory liquid handling robotics and easily interpreted results. Evidence we presented here suggests that the downy mildew pathogen on basil plants may represent an undescribed species. To our knowledge, this would constitute the first molecular recognition and characterisation of Peronospora sp. infecting sweet basil plants. Further work is required in order to determine whether P. lamii could also infect sweet basil plants. If this were not the case, previous studies attributing downy mildew of basil plants to P. lamii should be reappraised. It would also be useful to compare the morphology of P. swinglei to the Peronospora sp. described here and to determine the ITS sequences of North American and European isolates of P. swinglei. A wider sampling of Peronospora spp. infecting Lamiaceae, cross-infection studies with type hosts of these pathogens, and sexual incompatibility tests should also help to determine the full number of Peronospora species involved on this family, their host range, and maybe the occurrence of host-specific races. Such studies will conclusively determine if the species described in this paper is a new species, or if it is conspecific with a previously described ones. ACKNOWLEDGEMENTS We thank Bruno Amos and Vincent Gigon (Office Technique de Culture Maraichère de Genève, Jussy) for providing us with sweet basil leaves and potted plants infected with Peronospora sp.; Guy Auderset (School of Engineering of Lullier) for conidial measurements; Eric McKenzie (Landcare Research, Auckland) for providing Salvia leaves infected with Peronospora sp.; Patrizia Martini (Instituto Regionale Per La Floricoltura, Servizio Di Patologia, San Remo) for providing potted sweet basil plantlets from the San Remo region infected with Peronospora sp.; Hermann Voglmayr (Institute of Botany and Botanical Garden, University of Vienna, Vienna) for 1286 providing leaf samples of Geranium molle infected with P. conglomerata, Anthemis austriaca with P. radii, Lamium purpureum with P. lamii, and Scrophularia nodosa with P. sordida; and Catherine Baroffio (Agroscope Wädenswil) for fruitful exchanges. REFERENCES Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W. & Lipman, D. J. (1997) Gapped BLAST and PSIBLAST: a new generation of protein database search programs. Nucleic Acids Research 25: 3389–3402. Angiosperm Phylogeny Group (1998) An ordinal classification for the families of flowering plants. Annals of the Missouri Botanical Garden 85: 531–553. Bentley, J. I. (1980) Downy mildew (Peronospora lamii) on Lamium maculatum. Plant Pathology 29: 201–202. Bonants, P., de Weerdt, M. H., van Gent Pelzer, M., Lacourt, I., Cooke, D. & Duncan, J. (1997) Detection and identification of Phytophthora fragariae Hickman by the polymerase chain reaction. European Journal of Plant Pathology 103: 345–355. Boonham, N., Walsh, K., Preston, S., North, J., Smith, P. & Barker, I. (2002) The detection of tuber necrotic isolates of Potato virus Y, and the accurate discrimination of PVY(O), PVY(N) and PVY(C) strains using RT-PCR. Journal of Virological Methods 102: 103–112. Boonham, N., Perez, L. G., Mendez, M. S., Peralta, E. L., Blockley, A., Walsh, K., Barker, I. & Mumford, R. A. (2004) Development of a real-time RT-PCR assay for the detection of potato spindle tuber viroid. Journal of Virological Methods 116: 139–146. Brouwer, M., Lievens, B., van Hemelrijck, W., van den Ackerveken, G., Cammue, B. P. & Thomma, B. P. (2003) Quantification of disease progression of several microbial pathogens on Arabidopsis thaliana using real-time fluorescence PCR. FEMS Microbiology Letters 228 : 241–248. Burdon, J. J. & Silk, J. (1997) Sources and patterns of diversity in plant pathogenic fungi. Phytopathology 87: 664–669. Byrne, J. (2000) Downy mildew on Lamium. Michigan State University’s Landscape Crop Advisory Team Alert 15: 1. Byrne, J. (2001) Downy mildew on Salvia. Michigan State University’s Greehouse Alert 9: 1. Constantinescu, O. & Fatehi, J. (2002) Peronospora-like fungi (Chromista, Peronosporales) parasitic on Brassicaceae and related hosts. Nova Hedwigia 74: 291–338. Cooke, D. E. L., Drenth, A., Duncan, J. M., Wagels, G. & Brasier, C. M. (2000) A molecular phylogeny of Phytophthora and related oomycetes. Fungal Genetics and Biology 30: 17–32. Crawford, A. R., Bassam, B. J., Drenth, A., Maclean, D. J. & Irwin, J. A. G. (1996) Evolutionary relationships among Phytophthora species deduced from rDNA sequence analysis. Mycological Research 100: 437–443. De Cock, A. W. A. M. & Lévesque, A. (2004) New species of Pythium and Phytophthora. Studies in Mycology 50: 481–487. De Cock, A. W. A. M., Quail, A., Descalzo, R., Mayes, C. & Lévesque, C. A. (1998). Identification and detection of Pythium species by reverse dot blot hybridization. Proceedings of the International Congress of Plant Pathology, Edinburgh, UK. Aug. 1998: 3.3.26. Gamliel, A. & Yarden, O. (1998) Diversification of diseases affecting herb crops in Israel accompanies the increase in herb crop production. Phytoparasitica 26: 1–6. Galtier, N., Gouy, M. & Gautier, C. (1996) SEAVIEW and PHYLO_WIN, two graphic tools for sequence alignment and molecular phylogeny. Computer Applications in Bioscience 12: 543–548. Garibaldi, A., Minuto, A., Minuto, G. & Gullino, M. L. (2004) First report of downy mildew on basil (Ocimum basilicum) in Italy. Plant Disease 84: 1154. L. Belbahri and others Guindon, S. & Gascuel, O. (2003) A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology 52: 696–704. Gumedzoe, M. Y. D., Hemou, P., Lepage, A. & Lecat, N. (1998) Inventaire et identification de Peronospora lamii (Al. Br.) de By, agent causal du mildiou du basilic (Ocimum basilicum) dans les exploitations de Gyma Cultures. Les VIIIe Journées Scientifiques de l’Universite´ du Bénin, 11–15 mai 1998, Lome´, Bénin. Hall, G. (1996) Modern approaches to species concepts in downy mildews. Plant Pathology 45 : 1009–1026. Hansford, C. G. (1933) Annual report of the mycologist. Review of Applied Mycology 12: 421–422. Hasegawa, M., Kishino, H. & Yano, T.-A. (1985) Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. Journal of Molecular Evolution 22: 160–174. Heller, W. & Baroffio, C. (2003) Le mildiou (Peronospora lamii) du basilic progresse! Der Gemüsebau/Le Maraıˆcher 8: 12–13. Hill, C. F., Pearson, H. G. & Gill, G. S. C. (2004) Peronospora dianthi and Peronospora lamii, two downy mildews recently detected in New Zealand. New Zealand Plant Protection 57 : 348. Holcomb, G. E. (2000) First report of downy mildew caused by Peronospora lamii on Salvia Splendens and Salvia coccinea. Plant Disease 84: 1154. Holland, P. M., Abramson, R. D., Watson, R. & Gelfand, D. H. (1991) Detection of specific polymerase chain reaction product by utilizing the 5k-3k exonuclease activity of Thermus aquaticus DNA polymerase. Proceedings of the National Academy of Sciences, USA 88: 357–362. Huelsenbeck, J. P. & Ronquist, F. (2001) MRBAYES: Bayesian inference of phylogeny. Bioinformatics 17: 754–755. Ivors, K., Wilkinson, C. & Garbelotto, M. (2003) Nested TaqMan PCR for detection of Phytophthora ramorum in environmental plant samples. Mycological Society of America/British Mycological Society Joint Meeting, 26–31 July, Asilomar Conference Center, California, US. Klassen, W. (2002) Some exotic pathogens of plants which threaten horticultural production and trade in southern Florida. In Major Production Problems affecting Miami-Dade Agriculture and Emerging Technological Developments (R. L. Degner, T. J. Stevens & K. L. Morgan, eds): 29–31. Florida Agricultural Market Research Center, Miami. Kokko, H., Virtaharju, O., Karenlempi, S., Cooke, D. E. L., Williams, N. A. & Williamson, B. (1999) Downy mildew (Peronospora rubi) in Rubus arcticus: a threat to commercial berry production in Finland. Acta Horticultural 505: 137–141. Korimbocus, J., Coates, D., Barker, I. & Boonham, N. (2002) Improved detection of Sugarcane yellow leaf virus using a real-time fluorescent (TaqMan) RT-PCR assay. Journal of Virological Methods 103: 109–120. Lanave, C., Preparata, G., Saccone, C. & Serio, G. (1984) A new method for calculating evolutionary substitution rates. Journal of Molecular Evolution 20: 86–93. Leahy, R. (2001) Peronospora lamii A. Braun, downy mildew collected at a nursery in Tampa. Triology (Plant Pathology section) : 40: 1. Lefort, F., Gigon, V. & Amos, B. (2003) Le mildiou s’étend. Déjà détecté dans de nombreux pays européens, Peronospora lamii responsable du mildiou du basilic, a été observé en Suisse dans la région lémanique. Réussir Fruits et Légumes 223 : 66. Lindqvist, H., Koponen, H. & Valkonen, J. P. T. (1998) Peronospora sparsa on cultivated Rubus arcticus and its detection by PCR based on ITS sequences. Plant Disease 82: 1304–1311. 1287 Lindqvist-Kreuze, H., Koponen, H. & Valkonen, J. P. T. (2002) Variability of Peronospora sparsa (syn.P. rubi) in Finland as measured by amplified fragment length polymorphism. European Journal of Plant Pathology 108 : 327–335. Mackay, I. M., Arden, K. E. & Nitsche, A. (2002) Real-time PCR in virology. Nucleic Acids Research 30: 1292–1305. Martini, P., Rapetti, S., Bozzano, G. & Bassetti, G. (2003) Segnalazione in Italia di Peronospora lamii su basilico (Ocimum basilicum L.). Atti del Convegno ‘Problemi fitopatologici emergenti e implicazioni per la difesa delle colture’, Sanremo IM 27–29 novembre 2003: 79–82. McMillan, R. T. jr & Graves, W. R. (1994) First report of downy mildew of Salvia in Florida. Plant Disease 78: 317. Minuto, A., Pensa, P. & Garibaldi, A. (1999) Peronospora lamii, nuovo parassita fogliara della salvia, Colture-Protette 28: 63–64. Morrison, T. B., Weis, J. J. & Wittwer, C. T. (1998) Quantification of low-copy transcripts by continuous SYBR Green I monitoring during amplification. Biotechniques 24 : 954–962. Müller, J. (1999) Peronospora swinglei – ein neuer Falscher Mehltaupilz für die Tschechische Republik. Czech Mycology 51: 185–191. Mumford, R. A., Walsh, K., Barker, I. K. & Boonham, N. (2000) Detection of Potato mop top virus and Tobacco rattle virus using a multiplex real-time fluorescent reverse transcription polymerase chain reaction assay. Phytopathology 90 : 448–453. Plenk, A. (2002) Peronospora lamii A. Braun, eine noch in Österreich seltene Krankheit an Salvia officinalis. Fachgruppe Phytomedizin, Austrian Association of Agricultural Research Institutions, ALVA Tagung 2002: 1. Rodriguez, F., Oliver, J. F., Martin, A. & Medina, J. R. (1990) The general stochastic model of nucleotide substitution. Journal of Theoretical Biology 142 : 485–501. Scott, J. B., Hay, F. S. & Wilson, C. R. (2004) Phylogenetic analysis of the downy mildew pathogen of oilseed poppy in Tasmania, and its detection by PCR. Mycological Research 108: 198–205. Spencer, D. M. (1981) The Downy Mildews. Academic Press, London. Tyagi, S. & Kramer, F. R. (1996) Molecular beacons: probes that fluoresce upon hybridization. Nature Biotechnology 14: 303–308. Voglmayr, H. (2003) Phylogenetic relationships of Peronospora and related genera based on nuclear ribosomal ITS sequences. Mycological Research 107: 1–11. Voltolina, G. (2001) Sclarea sclarea L. Veneto Agricoltura Piante Officinale 2: 1–12. Whitcombe, D., Theaker, J., Guy, S. P., Brown, T. & Little, S. (1999) Detection of PCR products using self-probing amplicons and flourescence. Nature Biotechnology 17: 804–807. White, T. J., Burns, T., Lee, S. & Taylor, J. (1990) Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In PCR Protocols: a guide to methods and applications (M. A. Innis, D. H. Gelfand, J. J. Sinsky & T. J. White, eds): 315–322. Academic Press, San Diego. Wittwer, C. T., Herrmann, M. G., Moss, A. A. & Rasmussen, R. P. (1997) Continuous fluorescence monitoring of rapid cycle DNA amplification. BioTechniques 22: 130–138. Corresponding Editor: D. E. L. Cooke