Isolation and characterization of Streptomyces species from potato common scab lesions in Norway

Authors


E-mail: merete.wiken.dees@bioforsk.no

Abstract

The present survey was conducted to isolate and characterize Streptomyces species from common scab lesions of potato in Norway. Bacteria were isolated from scab lesions on tubers sampled in two consecutive years at different locations in Norway spanning ∼1400 km from south to north. In total, 957 independent isolations from individual tubers were performed, with 223 putative pathogenic isolates obtained from 29 different potato cultivars and 130 different fields. Streptomyces europaeiscabiei was the most abundant species isolated from common scab lesions (69%), while 31% of the isolates obtained were S. turgidiscabies. Streptomyces scabies was not found. Pathogenicity of selected Streptomyces isolates was tested on potato. The ability of the bacterial isolates to infect potato was consistent with the presence of the txtAB operon. The results revealed no pattern in geographical distribution of S. europaeiscabiei and S. turgidiscabies; both could be found in the same field and even the same lesion. Four different pathogenicity island (PAI) genotypes were detected amongst the txtAB positive isolates: nec1+/tomA+, nec1–/tomA+, nec1+/tomA− and nec1−/tomA−. The current findings demonstrate that there is genetic variability within species and that the species are not spread solely by clonal expansion. This is thought to be the most comprehensive survey of Streptomyces species that cause common scab of potato in a European country.

Introduction

Common scab (CS) constitutes a persistent disease and quality problem in the production of potato (Solanum tuberosum) throughout the world and it was first recorded in Norway in 1920 (Jørstad, 1922). In Norway, 14 332 ha of potatoes are grown, resulting in a total yield of 350 000 t potatoes (Statistics Norway, 2009). Plant health regulations prohibiting import of seed potatoes to Norway have been implemented to limit introduction of new potato pathogens and diseases. Most farmers therefore produce their own seed potatoes: approximately 20% of the seed used in the last 3 years were certified (Møllerhagen, 2011). The increase in sales of washed and prepacked potato products has increased the economic importance of skin blemish diseases as they affect the appearance and thereby the quality of the potato.

Management of CS is complicated and so far there have been no effective and reliable control strategies for the disease, with the exception of the use of more tolerant potato cultivars. However, none of the commercially available potato cultivars have proven to be completely resistant to CS (Hiltunen et al., 2005; Wanner, 2009).

Common scab is caused by soilborne, Gram-positive, filamentous bacteria in the genus Streptomyces. This saprophytic genus is well known for production of antibiotics (Chater, 2006; Chater et al., 2010) and relatively few of the several hundred species in the genus are pathogenic on plants. Amongst those that are pathogenic on plants, the species that cause CS on potato have been studied most extensively, because they reduce crop value and generate yield losses (Hiltunen et al., 2006; Loria et al., 2006; Wanner, 2009). Infection of potato cultivars by plant-pathogenic Streptomyces usually results in superficial, pitted or raised lesions on the tuber surface. In addition to potato, the host range of CS-causing streptomycetes includes tap root crops such as radish, beet, carrot and parsnip (Goyer & Beaulieu, 1997). Plant pathogenicity in the genus is based on production of the toxin thaxtomin which induces the characteristic symptoms of CS (King et al., 1989; Lawrence et al., 1990; Healy et al., 2000; Bignell et al., 2010). Thaxtomin induces plant cell hypertrophy in expanding plant tissues (Fry & Loria, 2002; Scheible et al., 2003).

The nec1 and tomA genes are present in a wide range of common scab-inducing Streptomyces strains, but not in all, and they are not required for pathogenicity (Seipke & Loria, 2008; Wanner, 2009). The nec1 gene encodes a protein that induces necrosis in plant tissue, and tomA encodes a virulence factor homologous to tomatinase, an enzyme that belongs to saponinases found in plant-pathogenic fungi (Bukhalid & Loria, 1997; Kers et al., 2005). In S. turgidiscabies, pathogenicity and virulence genes are clustered in a pathogenicity island (PAI), which can be transferred horizontally from pathogenic to saprophytic strains of Streptomyces, and thereby create new pathogenic strains (Healy et al., 1999; Bukhalid et al., 2002; Kers et al., 2005). The PAI of S. turgidiscabies consists of two non-overlapping modules, the first of which is similar to a genomic island in S. scabies, but the other module contains only a short region highly similar to the genome of S. scabies (Huguet-Tapia et al., 2011). The genes associated with thaxtomin production, such as txtA, txtB, txtC and txtR, are found in the first segment of the PAI, which is called the ‘toxicogenic region’, whereas the virulence-related genes, tomA and nec1, are located in the second segment, designated the ‘colonization region’ (Lerat et al., 2009).

Streptomyces scabies, S. turgidiscabies and S. europaeiscabiei are amongst the most well-known CS-causing species worldwide. Streptomyces scabies was the first described species causing CS on potatoes (Lambert & Loria, 1989) and is the dominant CS-causing species in the USA, although it seems to be infrequent in Europe (Bouchek-Mechiche et al., 2000a; Flores-Gonzalez et al., 2008; Dees et al., 2012). More recently, S. turgidiscabies has been found to cause CS on potato. It was first described in Japan (Miyajima et al., 1998) and Finland (Kreuze et al., 1999), and later found to be present in other parts of the world including China, Korea, Norway, North America, Sweden and the UK (Kim et al., 1999; Lehtonen et al., 2004; Zhao et al., 2008; Wanner, 2009; Thwaites et al., 2010; Dees et al., 2012). Thus far, S. europaeiscabiei has been found to be the main CS-causing species in Europe. It was first described in France in 2000 and has frequently been found in Europe, as well as Korea and North America (Bouchek-Mechiche et al., 2000a; Song et al., 2004; Flores-Gonzalez et al., 2008; Wanner, 2009; Dees et al., 2012).

An investigation of scab diseases in Norway was carried out during 1967–1968, in which CS turned out to be the most frequent scab disease followed by powdery scab (Værdal, 1973). Since then, no further surveys of scab diseases have been performed in Norway.

Before the advent of molecular techniques it was assumed that S. scabies was the causal agent of CS in Norway. However, in addition to S. scabies, S. turgidiscabies has been shown to be a causal agent of CS in Sweden and Finland, the two countries neighbouring Norway (Kreuze et al., 1999; Lehtonen et al., 2004). The aim of the present study was to isolate and characterize Streptomyces species from CS lesions on potato in Norway and compare the results to the Streptomyces species causing CS in the other Nordic countries.

Materials and methods

Source of potato tubers

Potato tubers showing symptoms of CS were obtained from potato growers, farmer advisers and the potato industry. The tubers sampled originated from 256 fields in 15 counties in Norway spanning ∼1400 km from south to north and were collected after the growing seasons in 2008 and 2009.

Bacterial isolates

Symptoms on the scabby potato tubers were recorded before the tubers were surface-disinfected in 70% ethanol for 20 s and then rinsed several times in sterile distilled water (SDW). Thereafter, from each tuber, a small piece of potato tissue was cut from under the surface of a single lesion, at the border between healthy and infected tissue, and then homogenized in 200 μL SDW and incubated for 30 min at room temperature. A 100 μL aliquot of the homogenate was plated out on water agar and incubated at 28°C in the dark. One isolation was performed from each of a total of 957 independent tubers. From each plate, up to three single colonies, phenotypically characteristic of Streptomyces, were transferred to yeast malt extract agar (YME) (Loria, 2001) after approximately 6 days. Subsequent transfer to fresh medium was done to obtain pure cultures. Streptomyces isolates were grown on YME agar plates at 28°C and stored on YME agar plugs at –80°C.

Isolation of genomic DNA

Streptomyces isolates were grown on YME at 28°C for 6–8 days and thereafter cells were scraped from the plate into a mortar and ground with a pestle in liquid nitrogen. DNA was extracted using the DNeasy Plant Mini Kit (QIAGEN) according to the manufacturer’s instructions. Lysis buffer and RNase were added and the samples were incubated at 65°C for 1 h. DNA was eluted in 40 μL AE buffer. The quality and amount of DNA were determined by agarose gel electrophoresis.

Identification of putative pathogenic Streptomyces by PCR

Polymerase chain reaction using primers txtAB 1 and txtAB 2 (Table 1) was performed as previously described (Wanner, 2006) to detect the txtAB operon encoding thaxtomin synthetase. The PCR mix (15 μL) contained primers at 0·5 μm, 0·2 mm dNTPs, 0·02 U μL−1 Phusion High-Fidelity DNA polymerase (Finnzymes) 5 ×  Phusion HF buffer, 3% dimethyl sulphoxide (DMSO) and 2·5 μL template DNA. The PCR cycle included initial DNA denaturation at 94°C for 3 min followed by 30 cycles of denaturation at 94°C for 30 s, primer annealing at 46°C for 30 s, extension at 72°C for 35 s, and final extension at 72°C for 7 min. The experiment was repeated once.

Table 1. Primers used for detection of genes from Streptomyces isolates
Primer name Primer sequence (5′–3′)TargetReferenceb
  1. aF, forward primer; R, reverse primer.

  2. bReferences: 1 (Wanner, 2006); 2 (Bukhalid et al., 1998); 3 (Songet al., 2004; Flores-Gonzalez et al., 2008); 4 (Edwards et al., 1989); 5 (Kreuze et al., 1999; Lehtonen et al., 2004); 6 (Clark et al., 1998).

txtAB 1FaCCACCAGGACCTGCTCTTC txtAB operon1
txtAB 2RaTCGAGTGGACCTCACAGATG txtAB operon
NfFATGAGCGCGAACGGAAGCCCCGGA nec1 gene2
NrRGCAGGTCGTCACGAAGGATCG nec1 gene
Tom3FGAGGCGTTGGTGGAGTTCTA tomA gene1
Tom4RTTGGGGTTGTACTCCTCGTC tomA gene
ITS-LFGTCAAGTCATCATGCCCCTT16S intergenic region3
ITS-RRAAACTTGGCCACAGATGCTC16S intergenic region
pAFAGAGTTTGATCCTGGCTCAGUniversal Streptomyces (16S rRNA gene)4
pH′RAAGGAGGTGATCCAGCCGCAUniversal Streptomyces (16S rRNA gene)
ScabIFCAACACTCTCGGGCATCCGA S. scabies (16S rRNA gene)5
ScabIIRTTCGACAGCTCCCTCCTTAC S. scabies (16S rRNA gene)
TurgIFCCTCGCATGGGGGTGGGTTA S. turgidiscabies (16S rRNA gene)5
TurgIIRCGACAGCTCCCTCCCCGTAA S. turgidiscabies (16S rRNA gene)
BOXA1R CTACGGCAAGGCGACGCTGACGRepetitive DNA sequences6

Species identification by PCR

The primers developed for the 16S rRNA gene sequences by Lehtonen et al. (2004) were used to detect DNA of S. scabies and S. turgidiscabies by PCR as described by the authors. However, S. scabies and S. europaeiscabiei cannot be distinguished by investigating the 16S rRNA gene sequences because those sequences are almost identical. Therefore, the intergenic transcribed spacer (ITS) region of the 16S operon was amplified using the primer pair ITS-R/ITS-L (Table 1) and the amplicons were subjected to digestion with the restriction enzyme Hpy99I (New England Biolabs), as previously described (Song et al., 2004; Flores-Gonzalez et al., 2008). The S. scabies type isolate ATCC49173 (Hpy99I+, i.e. amplicon-digestible with Hpy99I) and the three isolates ME01-11h (S. scabies; Hpy99I+), ID02-12 (S. europaeiscabiei; Hpy99I–) and PE07-1C (Hpy99I–) were used as controls (courtesy of Dr L. Wanner, USDA, Beltsville, MD, USA).

PAI marker genes

The genes nec1 and tomA were detected using the primer pairs Nf/Nr and Tom3/Tom4, respectively (Table 1), as previously described (Bukhalid et al., 1998; Wanner, 2006). The PCR cycle was the same as above, except that the annealing temperatures were 57 and 54°C for nec1 and tomA, respectively.

Pathogenicity assay on radish

A selection of 46 putative pathogenic Streptomyces isolates (txtAB-positive) was subsequently tested for pathogenicity on radish seedlings as previously described (Flores-Gonzalez et al., 2008). In short, radish cv. Cherry Belle seeds were disinfected in 0·5% sodium hypochlorite solution for 1 min and then rinsed several times in SDW. Thereafter, the seeds were placed on an 8-day old culture of a Streptomyces isolate growing on Difco™ Oatmeal Agar (OMA) and incubated at room temperature for 8 days. Growth of seeds on OMA plates with four txtAB-negative Streptomyces isolates and OMA plates without bacteria were used as controls. The appearance of the seedlings after growth with the bacteria was recorded. A bacterial isolate was considered pathogenic if the seedlings showed abnormal growth and hypertrophy, or if the seeds did not even germinate. The experiment was performed twice in duplicate.

Pathogenicity assay on the potato cultivar Blue Congo

Twenty-one txtAB-positive and three txtAB-negative Streptomyces isolates were tested for pathogenicity on potato (Table 3). Disease free minitubers of cv. Blue Congo (Overhalla Klonavlsenter), highly susceptible to CS, were used in the experiment.

Streptomyces isolates stored at −80°C were grown on YME agar for 2 weeks. P-Soil (mixture of peat and clay; Tjerbo Torvfabrikk) and agra-perlite (Pull Rhenen) were autoclaved three times and mixed 50:50 (v/v) in 5-L pots. Two plates of each Streptomyces isolate were mixed thoroughly in the upper layer of the soil/perlite mixture before planting one potato tuber in each pot. Pots containing soil mixed with YME from uninoculated plates and pots containing soil without any treatment were used as controls. The plants were grown indoors from July to August at 18°C under natural light conditions. An automated drip-watering system provided water containing fertilizer separately to each pot. Each treatment was run in three replicates and the tubers were harvested after 9 weeks. The lesion types and the surface area covered with symptoms were recorded for all tubers from each pot.

Results

Symptoms of common scab on tubers sampled in the survey

The tubers sampled in this survey exhibited diverse symptoms, ranging from superficial to deep-pitted lesions. A few tubers had raised, warty lesions. Furthermore, scab severity varied from the occurrence of a few discrete lesions to deep-pitted, coalescent lesions that covered nearly the entire surface of a tuber.

Putative pathogenic Streptomyces isolates

Streptomycetes were isolated from potato tubers displaying symptoms of CS. After extracting DNA from pure cultures, the primer pair txtAB 1/txtAB 2 was used to distinguish putative pathogenic isolates and probable non-pathogenic isolates. The 223 txtAB-positive isolates originated from 190 independent tubers and 130 different fields in 15 counties in Norway (Table 2). The various isolates came from 29 different potato cultivars and 40% of them were obtained from cvs Saturna, Mandel and Folva, which are amongst the most popular cultivars grown in Norway.

Table 2. Putative pathogenic Streptomyces isolates characterized in this study
Geographic originFields S. europaeiscabiei S. turgidiscabies
PAI genotypeaPAI genotype
IsolatesTxN1ToTxTxToTxN1IsolatesTxN1ToTxTxToTxN1
  1. aPAI genotype: Tx = presence of txtAB operon; N1 = presence of nec1 gene; To = presence of tomA gene.

Akershus34400020020
Aust-Agder45320020020
Buskerud34400021010
Finnmark14211000000
Hedmark2727617401532100
Møre og Romsdal23021030120
Nord-Trøndelag16345252230120
Oppland1414390241030
Rogaland21001033000
Sør-Trøndelag34040000000
Telemark11100000000
Troms15113431105320
Vest-Agder21010042020
Vestfold5121110040130
Østfold33111040040
Unknown292414280156090
Total130152576921571218420

Species identification by PCR and distribution of the species

All of the 223 putative pathogenic isolates were positive by PCR using the universal primers for the 16S rRNA gene of Streptomyces spp. Species determination was conducted using the primers developed by Lehtonen et al. (2004) for the 16S rRNA gene sequences. All the isolates were first tested with the ScabI/ScabII-primer pair (Table 1). Of the 223 isolates, 152 (69%) produced amplicons with the ScabI/ScabII primer pair: none of those 152 isolates could be restricted with Hpy99I and thus they were all assigned to S. europaeiscabiei (Table 2).

All 223 of the putative pathogenic isolates were also tested with the primer pair TurgI/TurgII, because S. turgidiscabies is found in some of the Nordic countries (Kreuze et al., 1999; Lehtonen et al., 2004). Of a total of 223 isolates, DNA of 71 of the isolates could be amplified using the primer pair TurgI/TurgII and hence these were assigned to S. turgidiscabies (Table 2). Distribution of the species was the same both years; 69% of the isolates were assigned to S. europaeiscabiei and 31% to S. turgidiscabies.

The current results showed no pattern of geographical distribution of S. europaeiscabiei and S. turgidiscabies. Both species could occur in the same field and even in the same lesion. Streptomyces europaeiscabiei was found in all 15 counties included in the survey, whereas S. turgidiscabies was detected in 12 counties (i.e. not in Finnmark, Sør-Trøndelag or Telemark; Table 2).

anova was performed to detect correlations between the following: Streptomyces species and geographical regions; PAI genotypes and geographical regions; species and percentage of the tuber surface covered with CS; and PAI genotypes and percentage of the tuber surface covered with CS. According to the results, there was no correlation between Streptomyces species and PAI genotype or the other factors. Furthermore, no specific pattern was observed in the geographical distribution of the species or the PAI genotypes (Table 2). Streptomyces europaeiscabiei and S. turgidiscabies could be isolated from tubers with symptoms ranging from only a few superficial lesions to almost complete coverage with deep pitted lesions. No correlation was found between the percentage of tuber surface covered with CS and either Streptomyces species or PAI genotypes.

PAI marker genes

All the putative pathogenic isolates and about 10 of the non-pathogenic isolates were tested for presence of the genes nec1 and tomA, which are characteristic of the Streptomyces PAI. Amongst the isolates that were txtAB-positive, four different PAI genotypes were detected; nec1+/tomA+, nec1−tomA+, nec1+/tomA− and nec1−/tomA− (Table 2). The nec1 gene was missing in 60% of all the Streptomyces isolates and 37% of the isolates lacked tomA. The nec1−/tomA+ PAI genotype predominated in S. turgidiscabies, whereas nec1−/tomA− and nec1+/tomA+ were detected most frequently in the S. europaeiscabiei isolates. The combination nec1+/tomA+ was found in 41% of the S. europaeiscabiei isolates and 26% of the S. turgidiscabies isolates and nec1−/tomA+ was observed in up to 63% of the S. turgidiscabies isolates. The combination nec1+/tomA− was detected in 2% of the S. europaeiscabiei isolates, but not in S. turgidiscabies. The combination nec1−/tomA− was found in almost half of the S. europaeiscabiei isolates (Table 2). None of the non-pathogenic isolates tested harboured the nec1 and tomA genes. Isolates with different combinations of the PAI marker genes nec1 and tomA were derived from the same field.

Pathogenicity assay on radish

Forty-six of the putative pathogenic Streptomyces isolates were tested for pathogenicity on radish seedlings. The ability of the bacterial isolates to inhibit radish seed germination and early seedling growth was consistent with the presence of the txtAB operon, the pathogenicity determinant (Table 3). The radish seeds that were grown with the txtAB-positive isolates showed hypertrophy or did not even germinate. By comparison, the seeds grown with txtAB-negative isolates displayed normal germination and growth, similar to the seedlings grown on OMA plates without bacteria (Fig. 1).

Table 3. Streptomyces isolates selected for pathogenicity tests on potato cv. Blue Congo and radish cv. Cherry Belle
 IsolateLesion typeaPotatoRadish
Percentage of tuber surface area covered with symptomsbPathogenic on radish seedlings
  1. aLesion type: 0 =  none; 1 = deep pitted lesions; 2 = superficial lesions; 3 = both deep pitted lesions and superficial lesions; nt = not tested on potato.

  2. bMean percentage of tuber surface covered with symptoms, results from three independent replicates per isolate.

  3. cYeast malt extract agar.

S. europaeiscabiei 08-05-02-1310+
08-05-04-1ntnt+
08-06-03-4ntnt+
08-08-01-1ntnt+
08-08-02-1ntnt+
08-12-01-1310+
08-15-01-1ntnt+
08-20-01-1ntnt+
08-30-4b-135+
08-74-04-1313+
08-88-5-135+
09-63-2-1318+
09-185-2-1313+
09-192-3-122·5+
09-196-4-236+
09-204-2-134+
09-210-2-134+
S. turgidiscabies 1B213+
14ntnt+
08-02-05-1a118+
08-04-02-1antnt+
08-04-04-1ntnt+
08-13-01-135+
08-18-02-2ntnt+
08-35-01-1123+
08-45-02-3123+
08-54-05-1111+
09-22-1-3147+
09-176-3-3137+
09-213-1-1130+
10-129-3-1140+
Control08-06-03-1b00
08-23-01-100
YMEc00
YME00
No treatment00
No treatment00
No treatment00
Figure 1.

 Appearance of radish seedlings 8 days after seeds were sown in a Petri dish containing an 8-day-old culture of Streptomyces. (a) Radish seedlings grown with a txtAB-negative isolate of Streptomyces and showing normal growth. (b) Ungerminated radish seeds grown with the txtAB-positive S. europaeiscabiei isolate 08-12-01-1. (c) Two radish seedlings grown with the txtAB-positive S. turgidiscabies isolate 08-04-04-1 showing abnormal growth and hypertrophy (left) compared with a radish seedling grown with a txtAB-negative isolate which showed normal growth (right).

Pathogenicity test on potato

All 21 txtAB-positive isolates tested for pathogenicity on potato induced symptoms characteristic of CS on the tubers (Table 3; Fig. 2). All of the tubers harvested from pots inoculated with txtAB-positive isolates showed symptoms ranging from discrete superficial lesions to coalescing deep-pitted lesions. The scab lesions varied in appearance and severity and this symptom variation was observed between and within species. Lesion severity was reproducible for each isolate tested in three independent replicates. All the tubers inoculated with txtAB-negative isolates were completely free of symptoms (Fig. 2). The mean surface area covered by symptoms was 8·5% (range 2·3–17·7%) in tubers infected with S. europaeiscabiei, but was 24·8% (5·0–46·7%) in tubers infected with S. turgidiscabies.

Figure 2.

 Potato cv. Blue Congo tubers (a) grown with a txtAB-negative Streptomyces isolate and completely free of symptoms of common scab; (b–d) grown with txtAB-positive Streptomyces turgidiscabies isolates 1B, 10-129-3-1 and 08-54-05-1, respectively, showing a range of symptoms characteristic of common scab.

Discussion

The present study would appear to be the most comprehensive survey of CS-causing Streptomyces species conducted in a European country thus far. The aim was to isolate and characterize plant-pathogenic Streptomyces species from CS lesions on potatoes grown in Norway. Streptomyces europaeiscabiei was found to be the most abundant species (69%) isolated from CS lesions in Norway, while 31% of the isolates obtained in the study were S. turgidiscabies. Surprisingly, S. scabies was not found in this study, nor were other streptomycetes that are pathogenic on potato, such as S. acidiscabies and S. stelliscabiei.

The results, in combination with the findings of other European studies of CS (Bouchek-Mechiche et al., 2000a; Flores-Gonzalez et al., 2008), suggest that S. scabies is not the predominant species causing CS in Europe, in contrast to North America, where S. scabies is the most common (Flores-Gonzalez et al., 2008; Wanner, 2009). Streptomyces europaeiscabiei can be mistaken for S. scabies if restriction analysis of the ribosomal DNA spacer region is not performed, and thus the real global distribution of S. scabies might differ from the picture presented in the literature.

Considering the findings of the present study in Norway, there was no pattern in the geographical distribution of S. europaeiscabiei and S. turgidiscabies. The latter species was not found in three of 15 counties studied, but that might be explained by the low number of samples, rather than the absence of S. turgidiscabies from these areas. The lack of a pattern in geographical distribution of S. europaeiscabiei and S. turgidiscabies or in the PAI genotypes may have been caused by the use of certified seed potatoes and the subsequent dispersal of infected seed tubers throughout the country. This is probably because Norwegian seed certification standards allow 5% (by weight) surface blemishes which can suffice to spread strains around the country. Another possible explanation is that both species may be natural inhabitants of soil in Norway. Streptomyces europaeiscabiei was the most abundant species isolated from CS-lesions in Norway, and it is also found to be the prevalent species in western Europe (Flores-Gonzalez et al., 2008). The present study found that S. turgidiscabies is widespread in Norway, although it is less abundant than S. europaeiscabiei. Streptomyces turgidiscabies was also reported to be one of the main causal agents of CS in Finland (Lehtonen et al., 2004; Hiltunen et al., 2009), whereas it appears that this species is absent or less common in other parts of the world, such as western Europe and North America (Bouchek-Mechiche et al., 2000b; Flores-Gonzalez et al., 2008; Wanner, 2009). However, in the cited studies, 23 of all the isolates from western Europe came from France and 84 isolates originated from several different countries in northern Europe; hence, a larger survey is needed to discover the true distribution of the species in western Europe. In Wanner’s extensive collection of 1074 txtAB-positive isolates from North America, only two isolates could be assigned to S. turgidiscabies (Wanner, 2009). Besides the 71 isolates of S. turgidiscabies described in Norway, the largest collections of this species were gathered in Japan and Finland, with 22 and 38 isolates, respectively (Miyajima et al., 1998; Kreuze et al., 1999; Lehtonen et al., 2004). The prevalence of S. turgidiscabies in Norway and Finland may be partly explained by the Nordic climatic conditions. Hiltunen et al. (2009) suggested that S. turgidiscabies competes with S. scabies for an ecological niche and is a potential major cause of CS in northern Europe.

In the present study, S. turgidiscabies was detected in samples from 22 different cultivars, which confirms that this species can infect a broad range of the potato cultivars grown in Norway. Many fields harboured both S. europaeiscabiei and S. turgidiscabies, and these two species could also be found in the same lesion. This is consistent with previous studies of CS showing that a single field could be infested with multiple pathogenic Streptomyces species, and even a single lesion could contain more than one species (Lehtonen et al., 2004; Wanner, 2009).

The Norwegian Streptomyces isolates showed variation in PAI marker genes and the dominant PAI genotypes differed between S. europaeiscabiei and S. turgidiscabies. Although nec1 and tomA are both present in S. scabies, they are not located in the same chromosomal regions as the thaxtomin biosynthesis gene cluster and the separate presence or absence of these regions suggests that they are independently transferable (Aittamaa et al., 2010). Aittamaa et al. (2010) concluded from one of their studies that pathogenicity- and virulence-related gene clusters in S. turgidiscabies have multiple origins, and that a PAI consists of a mosaic of regions that may undergo independent evolution.

The results of the present study demonstrate that there is genetic variation within species and thus they do not spread simply through clonal expansion. This confirms the novel findings concerning the genetic variability within S. europaeiscabiei that were obtained in a previous investigation (Dees et al., 2012).

The pathogenicity tests on potato revealed that, compared with S. europaeiscabiei, S. turgidiscabies induced more severe damage of the skin in general and caused symptoms over a larger proportion of the surface of tubers. The results of pathogenicity tests on potato also showed that species had a greater effect than PAI genotype on the proportion of the tuber surface displaying symptoms. No correlation was found between the surface area of the tuber covered with symptoms and the Streptomyces species isolated from it. Indeed, both S. europaeiscabiei and S. turgidiscabies could be isolated from tubers showing a wide range of symptoms ranging from the presence of a few discrete lesions to coalescent lesions covering nearly the entire surface. The development of symptoms depends on a number of factors including potato cultivar, environmental conditions and the pathogenicity and virulence of the bacterial species (Bouchek-Mechiche et al., 2000b; Wanner, 2009).

Much knowledge has been gained in recent years on various aspects of CS, including pathogenic species and their distribution, detection methods, mechanisms of pathogenicity and interactions between the bacteria and the plant. Nonetheless, there are still no reliable methods for controlling CS. Disease-resistant potato cultivars would be the best and most desirable control method, but no commercially available cultivar has yet been shown to be completely resistant. In as much as CS leads to diminished market value and impaired appearance of infected tubers, this disease continues to be an important quality problem in production of potatoes worldwide. Most of the work done with the aim of controlling CS is based on S. scabies, but it is possible that the other pathogenic species may respond differently to agricultural practices than S. scabies. Therefore, when developing methods to manage CS, it is important to begin with a survey of the pathogenic species that are present in the country of interest.

In summary, a total of 223 putative pathogenic Streptomyces isolates were obtained from CS lesions on potato tubers originating from all over Norway. PCR, using species-specific primers, and restriction analysis of parts of the ITS region, identified the isolates as S. europaeiscabiei or S. turgidiscabies. The distribution of PAI genotypes amongst the isolates was high.

It would be valuable to include the Norwegian collection of plant-pathogenic streptomycetes in population genetic studies of isolates obtained from CS lesions. Furthermore, it might be possible to apply the new knowledge about the Norwegian Streptomyces population to develop strategies for managing CS and national breeding programmes aimed at acquiring CS-tolerant varieties.

Acknowledgements

We are grateful to Dr L. Wanner for critical reading of the manuscript; Dr R. Nærstad for help with statistical analysis; E. Gauslå and T. Slørstad for assistance in isolating Streptomyces from scabby tubers; potato farmers, farmer advisers and the potato industry for providing scabby potato tubers; and Dr Carl Spetz for comments on the manuscript. This work was supported by grants from the Research Council of Norway, the Foundation for Research Levy on Agricultural Products, the Agricultural Agreement Research Fund, and Norwegian food industries.

Ancillary