Variation and distribution of virulence phenotypes of Bremia lactucae in natural populations of Lactuca serriola


*To whom correspondence should be addressed.


During 1997–2000 the virulence variation and distribution of virulence phenotypes of Bremia lactucae (lettuce downy mildew) in natural populations of Lactuca serriola (prickly lettuce) were studied. Altogether 139 isolates of B. lactucae originating from the Czech Republic, France and Germany were examined for the presence of 27 virulence factors (v-factors) and their combinations. In the Czech population, 37 different v-phenotypes (P1–P37) of B. lactucae were found to occur on L. serriola. Most v-phenotypes were characterized by v-factors that match resistance (Dm genes/R-factors) carried by L. serriola. A wide diversity of v-phenotypes was recorded every year, but most were rare and did not reappear in other populations of B. lactucae. The three v-phenotypes P21, P28 and P29 were most frequent and widely distributed. There was variation both between spatially isolated populations and within populations of the pathogen. Geographic differences in virulence were found for the southern parts of Moravia, where the phenotypic composition of pathogen populations was completely different from the remaining part of the area investigated. However, some populations had v-phenotypes similar to those of spatially distant populations in Moravia.


Downy mildew of lettuce, caused by Bremia lactucae, occurs worldwide and is a serious disease of glasshouse and field lettuce (Lactuca sativa) and many wild Asteraceae species (Crute & Dixon, 1981; Lebeda et al., 2002). In natural plant communities Lactuca serriola (prickly lettuce) is the most common wild host of B. lactucae (Lebeda et al., 2001a, 2002).

The relationships between L. sativa, L. serriola and B. lactucae conform to a gene-for-gene interaction (Crute, 1992). More than 40 genes or factors (Dm and R) for resistance of the host, and complementary virulence (v) genes (factors) in the parasite have been identified so far (Lebeda et al., 2001b, 2002; Michelmore et al., 2002; Lebeda & Zinkernagel, 2003a). Detailed analyses of the genetics of the host–parasite interaction made it possible to interpret the variability of virulence in B. lactucae in terms of virulence factors (v-factors) and virulence phenotypes (v-phenotypes) (Lebeda, 1982; Crute, 1987). Bremia lactucae is a highly variable pathogen with a number of forms differing in their virulence to genotypes of L. sativa, L. serriola and other wild Lactuca spp. (Lebeda et al., 2002; Lebeda & Zinkernagel, 2003b). In general the pathogen and its populations could be considered to have a high evolutionary potential sensuMcDonald & Linde (2002). During recent decades, studies of this phenomenon have focused on the population level, and only on lettuce as a crop (L. sativa) (Crute, 1987; Lebeda & Zinkernagel, 2003a). So far there has been no detailed research of B. lactucae in natural plant populations, and only very limited information about their virulence structure has been added (Lebeda, 2002). A study of the wild (natural) plant pathosystem L. serriolaB. lactucae which began in the early 1980s demonstrated the existence of various v-phenotypes among eight isolates of B. lactucae originating from naturally growing L. serriola (Lebeda, 1984). These v-phenotypes differed substantially from those isolated from L. sativa and made it possible to identify new race-specific resistance factors (R-factors) in L. serriola accessions (Bonnier et al., 1994; Lebeda & Zinkernagel, 2003b). Recently, more detailed research has focused on the natural distribution and virulence structure of B. lactucae populations on L. serriola (Lebeda, 2002; Petrželová & Lebeda, 2004).

The aim of the present study was to determine the variation of virulence for individual B. lactucae isolates and the distribution of v-phenotypes in natural populations of L. serriola in the Czech Republic and elsewhere in Europe.

Materials and methods

Pathogen collecting

Virulence variation was studied in a set of 139 isolates of B. lactucae. They were collected randomly in natural populations of L. serriola (prickly lettuce) during 1997–2000 (two isolates collected in 1997, 23 in 1998, 36 in 1999, and 78 in 2000). The isolates originated mostly from central and southern Moravia (districts Brno County, Břeclav, Nový Jičín, Olomouc, Prostějov, Přerov and Znojmo); one isolate was collected in central Bohemia (district Rakovník) (Petrželová & Lebeda, 2004). Figure 1 shows the distribution of localities in the Czech Republic where the isolates were collected. Five further isolates were collected elsewhere (three in France and two in Germany) (Lebeda et al., 2001a). The origin of B. lactucae isolates used for v-factor analysis is summarized in Table 1.

Figure 1.

Collecting sites of Bremia lactucae isolates used for v-phenotype analysis in the Czech Republic.

Table 1.  Geographic origin of Bremia lactucae isolates from Lactuca serriola used for v-phenotype analysis (1997–2000)
Country/regionDistrictNumber of isolates by year
  • a


Czech Republic
JihomoravskýBrno County116 0 0
Břeclav0 0 0 6
Znojmo1 0 0 0
MoravskoslezskýNový Jičín0 0 0 6
OlomouckýOlomouc0 43040
Prostějov0 3 0 1
Přerov0 0 025
StředočeskýRakovník0 0 1 0
GermanyEichelborna0 0 2 0
FranceMégévea0 0 3 0
Total 2233678

One or two leaves with symptoms of B. lactucae infection (sporulation or well developed chlorotic spots) were taken from each L. serriola plant. Leaf samples were transported to the laboratory in plastic boxes with moistened filter paper. If there were areas of sporulation (surrounded by veins), spores were washed off with distilled water and inoculated onto seedlings of susceptible L. serriola genotypes (PI 273617, LSE/57/15 or LS-102). Leaves with chlorotic spots were incubated for 1–3 days under conditions described previously (Lebeda, 1982) to induce sporulation. Isolates were maintained and multiplied on L. serriola seedlings (PI 273617, LSE/57/15 or LS-102). For long-term storage the isolates were kept at −80°C on infected seedlings.

Differential set of host genotypes

The virulence of B. lactucae isolates was examined on a differential set of 49 genotypes of L. sativa and L. serriola (Lebeda & Zinkernagel, 2003b). Of these, 27 genotypes, each with one or two well characterized race-specific Dm genes and/or R-factors, were used for primary differentiation of isolates. The basic set was supplemented by several cultivars and accessions carrying a combination of race-specific Dm genes and R-factors or whose resistance had not yet been clearly defined. Accession LSE/57/15 of L. serriola served as a universally susceptible control. Seed of the differentials originated from the Lactuca spp. collection of the Department of Botany (Laboratory of Plant Pathology), Faculty of Science, Palacký University in Olomouc.

Inoculum preparation and inoculation

Seedlings of the L. sativa and L. serriola differentials were grown in plastic boxes (320 × 265 × 60 mm) lined with moistened filter paper and closed with a glass cover. They were inoculated with isolates of B. lactucae at the stage of fully expanded cotyledon leaves (7–9 days after sowing). Inoculum was prepared by washing 2–3-day-old spores off seedlings of L. serriola accessions (LSE/57/15, PI 273617 or LS-102) with distilled water and adjusting the concentration of inoculum to c. 105 spores mL−1. The spore concentration was the same in all tests. The conidial suspension was sprayed onto seedlings using a glass sprayer. Boxes of inoculated seedlings were incubated at 10–15°C, covered with black foil for 12–24 h after inoculation, then kept under a 12 h photoperiod.

Assessment of sporulation intensity

The degree of infection was assessed quantitatively using a 0–3 scale (Lebeda, 2002). The final evaluation was made 14 days after inoculation. Sporulation intensity was expressed as a percentage of maximum scores (Lebeda & Zinkernagel, 2003b). Each test contained 20–25 seedlings per differential. Two further replications were made if the sporulation intensity of an isolate on a differential was 50% or less.

The response of a differential to an isolate of B. lactucae was assessed as compatible (susceptible) if profuse sporulation developed on most inoculated seedlings, usually 8–10 days after inoculation. An incompatible (resistant) reaction was characterized either by a complete absence of sporulation, or by delayed and sparse sporulation with final sporulation intensity not exceeding 30% (incompletely compatible reaction). Sometimes an intermediate reaction was found.

Assessment of virulence variation

Isolates were tested for the presence of one or more of 27 v-factors. The variation of virulence was expressed by the known virulence phenotypes (v-phenotypes), which were differentiated on the basis of recorded virulence/avirulence. To quantify virulence, the complexity of v-phenotypes (Lebeda, 1982), expressed as the ratio between the number of v-factors in a given isolate and the total number of known v-factors, was used. Variation between geographically isolated populations of B. lactucae was expressed by the frequencies (%) of v-phenotypes found.


Characterization of individual v-phenotypes

There were clear differences in virulence between the isolates of B. lactucae and a broad diversity of v-phenotypes. Altogether 37 different v-phenotypes (designated as P1–P37) were recognized (Tables 2 and 3). The two isolates from 1997 belonged to the same v-phenotype. In the 23 isolates from 1998, 12 different v-phenotypes were identified. In 1999 the 31 Czech isolates belonged to 13 v-phenotypes, while three v-phenotypes were recognized for the five isolates from France and Germany. The 78 isolates collected in 2000 were characterized as having 19 different v-phenotypes.

Table 2.  Number of isolates and v-phenotypes of Bremia lactucae from Lactuca serriola
Year determinedCountry of originNumber of isolates testedNumber of v-phenotypes
  • a

    A complete list of v-phenotypes determined is given in Table 3; six v-phenotypes were detected in several successive years.

1997Czech Republic  2 1
1998Czech Republic 2312
1999Czech Republic 3113
France  3 1
Germany  2 2
2000Czech Republic 7819
Total 13937a
Table 3.  List of v-phenotypes found in Bremia lactucae from Lactuca serriola
DesignationVirulence phenotypeaNumber of virulencesbNumber of isolates
  • a

    v1, v2, v3, v4, v10, v12, v13, v14 = v-factors matching Dm genes or R-factors in cvs of Lactuca sativa.

  • v7, v15, v16, v17, v23, v24, v25, v26, v27, v28, v29, v30 = v-factors matching Dm genes or R-factors in Lactuca serriola.

  • v5/8, v6, v11, v18, v38 = v-factors matching Dm genes or R-factors in cvs of Lactuca sativa derived from L. serriola.

  • v36, v37 = v-factors matching Dm genes or R-factors in cvs of L. sativa derived from Lactuca saligna.

  • vSP = virulence to cv. Spartan Lakes, corresponding resistance gene for the reaction not known (Dm1 + ?).

  • vTI = virulence to cv. Titan, corresponding resistance gene for the reaction not known (Dm6 + R36).

  • vCR = virulence to L. serriola ×L. sativa (CS-RL), corresponding resistance gene for reaction not known (R18 + ?).

  • vPI = virulence to L. serriola (PI 273617), corresponding resistance gene for the reaction not known (R?).

  • b

    27 v-factors and four additional virulences are included in Table 3.

  • c

    Virulence to some additional lettuce cultivars (Cobham Green, Cristallo, Mildura, Reskia, Saffier, Blondine, Amplus, Bourguignonne, Valverde, Sucrine, Vanguard, Kinemontepas, Iceberg, Regina di Maggio and Santa Anna).

  • d

    v-phenotype P33 is identical with previously described and internationally agreed race NL16 (BL-1b).

Isolates from Czech Republic
P1v7, v15, v23, v24, v26, v27, v29, v30, vTI 9 – – 1
P2v5/8, v7, v15, v24, v25, v26, v27, v29, v30, vTI10 1 – –
P3v7, v15, v17, v23, v24, v25, v26, v28, v29, v30, vCR11 – – 1
P4v5/8, v7, v15, v16, v17, v24, v25, v26, v27, v28, v29, v3012 – – 1
P5v5/8, v7, v15, v17, v23, v24, v25, v26, v28, v29, v30, vCR12 – 1 1
P6v5/8, v7, v15, v23, v24, v25, v26, v27, v29, v30, vTI, vPI12 2 – –
P7v7, v11, v15, v17, v23, v24, v25, v26, v28, v29, v30, vCR12 – – 4
P8v7, v15, v17, v23, v24, v25, v26, v28, v29, v30, vTI, vCR12 – – 3
P9v7, v11, v15, v17, v24, v25, v26, v28, v29, v30, vTI, vCR12 – – 1
P10v7, v11, v15, v23, v24, v25, v26, v27, v29, v30, vTI, vPI12 – 1 –
P11v5/8, v7, v11, v15, v23, v24, v25, v26, v27, v29, v30, vTI, vPI13 – 2 –
P12v5/8, v7, v15, v16, v17, v24, v25, v26, v27, v28, v29, v30, vCR13 1 – –
P13v5/8, v7, v11, v15, v17, v23, v24, v25, v26, v28, v29, v30, vCR13 1 – –
P14v5/8, v7, v15, v17, v23, v24, v25, v26, v28, v29, v30, vTI, vCR13 – 1 –
P15v7, v11, v14, v15, v16, v17, v24, v25, v27, v28, v29, v30, vCR13 – – 1
P16v7, v11, v15, v16, v17, v24, v25, v26, v27, v28, v29, v30, vCR13 – – 2
P17v7, v11, v15, v17, v23, v24, v25, v26, v28, v29, v30, vTI, vCR13 – – 9
P18v1, v2, v4, v5/8, v6, v7, v10, v12, v13, v14, v16, v23, vSP, vPIc14 2 – –
P19v5/8, v7, v15, v16, v17, v23, v24, v25, v26, v27, v28, v29, v30, vPI14 1 – –
P20v5/8, v7, v11, v15, v16, v17, v23, v24, v25, v26, v27, v28, v29, v3014 – 1 –
P21v5/8, v7, v11, v15, v16, v17, v24, v25, v26, v27, v28, v29, v30, vCR14 – 117
P22v5/8, v7, v11, v15, v17, v23, v24, v25, v26, v28, v29, v30, vTI, vCR14 1 4 1
P23v7, v11, v14, v15, v16, v17, v24, v25, v26, v27, v28, v29, v30, vCR14 – – 7
P24v7, v15, v16, v17, v23, v24, v25, v26, v27, v28, v29, v30, vSP, vPI (+ avirulence to cv. Iceberg)14 – – 2
P25v7, v15, v16, v17, v23, v24, v25, v26, v27, v28, v29, v30, vSP, vPI (+ virulence to cv. Iceberg)14 (+1) 4 – –
P26v1, v2, v3, v4, v5/8, v6, v7, v10, v11, v12, v13, v16, vSP, vPIc142 – – –
P27v1, v2, v4, v5/8, v6, v7, v10, v11, v12, v13, v14, v16, v23, vSP, vPIc15 1 – –
P28v5/8, v7, v11, v14, v15, v16, v17, v24, v25, v26, v27, v28, v29, v30, vCR15 – 220
P29v5/8, v7, v15, v16, v17, v23, v24, v25, v26, v27, v28, v29, v30, vSP, vPI15 7 7 4
P30v5/8, v7, v11, v15, v16, v17, v23, v24, v25, v26, v28, v29, v30, vTI, vCR15 – 1 –
P31v5/8, v7, v11, v15, v17, v23, v24, v25, v26, v27, v28, v29, v30, vTI, vCR15 – 2 –
P32v7, v11, v14, v15, v16, v17, v23, v24, v25, v26, v27, v28, v29, v30, vCR15 – – 1
P33dv1, v2, v3, v4, v5/8, v6, v7, v10, v11, v12, v13, v16, v23, vSP, vPI15 1 – –
P34v5/8, v7, v11, v14, v15, v16, v17, v24, v25, v26, v27, v28, v29, v30, vSP, vCR16 – – 1
P35v5/8, v7, v15, v16, v17, v23, v24, v25, v26, v27, v28, v29, v30, vSP, vCR, vPI16 – 7 1
P36v5/8, v7, v11, v15, v16, v17, v23, v24, v25, v26, v27, v28, v29, v30, vSP, vCR, vPI17 – 1 –
P37v1, v2, v4, v5/8, v7, v10, v12, v13, v14, v15, v16, v23, v24, v25, v26, v27, v28, v29, v30, vSP, vPIc21 1 – –
Total number of tested isolates2233178
Isolates from France (FRA) and Germany (GER)
P24 (FRA)v7, v15, v16, v17, v23, v24, v25, v26, v27, v28, v29, v30, vSP, vPI14 – 3 –
P35 (GER)v5/8, v7, v15, v16, v17, v23, v24, v25, v26, v27, v28, v29, v30, vSP, vCR, vPI16 – 1 –
P36 (GER)v5/8, v7, v11, v15, v16, v17, v23, v24, v25, v26, v27, v28, v29, v30, vSP, vCR, vPI17 – 1 –
Total number of tested isolates – 5 –

The number of v-factors per isolate (v-phenotype) ranged between 9 and 19, i.e. the complexity of the identified v-phenotypes was 0·33–0·70 (Table 3). Only one isolate (P37) contained 19 previously known v-factors. A comparison of the number of L. sativa and L. serriola differentials on which each isolate was virulent is presented in Fig. 2. In most isolates the number of virulences ranged between 12 and 17. Isolates with lower or higher numbers of virulences were rare.

Figure 2.

Comparison of Bremia lactucae isolates from Lactuca serriola according to the number of virulences present.

The B. lactucae isolates also displayed great variation in the reaction of several supplementary differentials [e.g. cv. Spartan Lakes, cv. Titan, L. serriola × L. sativa (CS-RL), L. serriola (PI 273617)]. Therefore data on the virulence of these accessions was used to provide a more detailed description of differences between v-phenotypes. In v-phenotypes of the pathosystem L. serriolaB. lactucae the most frequent v-factors were v5/8, v7, v15, v16, v17 and v24–v30. Except for P37 the v-phenotypes differed mostly by the presence or absence of only a few (1–5) v-factors (Table 3). The most variable components of v-phenotypes were the v-factors v5/8, v11, v14, v16, v23 and v27; and virulences to cvs Spartan Lakes (Dm1) and Titan (Dm6 + R36), interspecific hybrid L. serriola ×L. sativa (CS–RL) (R18 + ?) and L. serriola (PI 273617) (R?: the genetic background of resistance of this accession has not yet been specified).

Shifts in the structure of v-phenotypes

A comparison of individual v-phenotypes over several years in the Czech Republic revealed that six (P5, P21, P22, P28, P29, P35) occurred in successive years. Virulence phenotypes P22 and P29 occurred in 1998–2000 (Table 3). The other v-phenotypes were recorded only in one season. In particular years the population consisted of a limited number of v-phenotypes, and the dominant ones varied from year to year. In 1998 v-phenotype P29 (v5/8, v7, v15, v16, v17, v23–v30, vSP, vPI) was the most frequent. In 1999 there were three prevailing v-phenotypes: P22 (v5/8, v7, v11, v15, v17, v23–v26, v28–v30, vTI, vCR), P29 and P35 (v5/8, v7, v15, v16, v17, v23–v30, vSP, vCR, vPI). In 2000 four v-phenotypes were among the most frequent: P17 (v7, v11, v15, v17, v23–v26, v28–v30, vTI, vCR), P21 (v5/8, v7, v11, v15, v16, v17, v24–v30, vCR), P23 (v7, v11, v14, v15, v16, v17, v24–v30, vCR) and P28 (v5/8, v7, v11, v14, v15, v16, v17, v24–v30, vCR). Possible differences in the incidence of individual v-phenotypes during each growing season (April–September) were evaluated. It was found that isolates with the same v-phenotype may occur throughout the year.

Geographic and spatial distribution of v-phenotypes

The geographic distribution of v-phenotypes in the Czech Republic was assessed for Moravia (regions 1–3), as only one isolate came from Bohemia (region 4; Fig. 1). At the level of larger geographic areas (regions and districts) there were mostly no substantial differences in the structure of virulence. In most areas a composite of similar v-phenotypes occurred, differing by the presence of only a few v-factors. Some v-phenotypes (P8, P21, P23, P28, P29) were distributed in a ‘mosaic structure’ (randomly) more-or-less throughout the whole area surveyed. Larger differences were recorded in the districts of Brno County and Znojmo (localities Hajany I, Hajany II, Ořechov, Lechovice; years 1997 and 1998). There, several isolates were found with v-phenotypes (P18, P26, P27, P33, P37) that were not typical of B. lactucae originating from L. serriola.

Differences were recorded in the occurrence of v-phenotypes for spatially separated populations of B. lactucae. Only 15 v-phenotypes occurred at two or more locations, and three of these were present in 10 localities. To characterize quantitative differences in the distribution of v-phenotypes between spatially separated populations, the frequencies of occurrence at the localities surveyed were determined for all v-phenotypes (Fig. 3). Frequencies were expressed as a percentage of localities at which certain v-phenotypes were recorded between 1997 and 2000; this evaluation was done for one Bohemian and 63 Moravian localities. The most widely distributed v-phenotype was P28, found at 21·1% of localities surveyed mainly in year 2000. The next most frequent were v-phenotypes P21 (20·3% of localities) and P29 (18·8%); P21 predominated in 2000, while P29 predominated during the period 1998–2000. Quite frequent was v-phenotype P35 (10·9%), recorded mainly in 1999; further P17 (14·1%) and P23 (9·4%) were found in 2000, and P22 (7·8%) was recorded from 1998 to 2000. A number of other v-phenotypes were recorded at only one locality during the whole observation period (Fig. 3).

Figure 3.

Frequency of v-phenotypes in Bremia lactucae from Lactuca serriola at survey sites (1997–2000).

Variability of v-phenotypes within populations

Variability was also found within individual populations (subpopulations) of B. lactucae. There were 28 sites and/or populations where more than one isolate was collected from individual host plants in a given year. Sometimes isolates differed in virulence even though they were collected from one host population on plants that often grew close (1–3 m) to each other. Table 4 shows that isolates collected at the same site and in the same year frequently differed in v-phenotype structure.

Table 4.  Virulence variation (expressed as number of different v-phenotypes) in selected populations of Bremia lactucae from Lactuca serriola
DistrictSiteYear of collectionNumber of isolates collected at a siteaNumber of recognized v-phenotypes
  • a

    At the remaining sites only one sample (isolate) of Bremia lactucae per site was collected in a given year.

Brno CountyHajany I199832
Hajany II199822
Road from Silůvky to199832
Moravské Bránice   
Nový JičínOdry200053
 Grygov I199822
Grygov II199942
Hlušovice I200022
Horní Sukolom199932
Nový Dvùr199822
Olomouc – Holice II199922
Olomouc – Holice III199922
Surroundings of Blatec & Tážaly199922
Velká Bystřice II200022
Velký Týnec –čechovice200022
PřerovLipník nad Bečvou – Loučka200021
Přerov –čekynì I200042
Přerov –čekynì II200022
Přerov – Vinary200022
Road from Hrabůvka to Velká200021
Surroundings of Pavlovice200022
Teplice nad Bečvou200022


The interaction between Lactuca spp. and B. lactucae displays diversity at different levels (Lebeda et al., 2001b, 2002; Lebeda & Zinkernagel, 2003b). The present paper deals with the structure and diversity of virulence in individual isolates of a pathogen population in natural plant communities. In general, such studies are rather limited (Burdon, 1993, 1997; Thrall & Burdon, 2003). So far they have mainly considered the crop pathosystems (McDonald & Linde, 2002), with some considering L. sativaB. lactucae (Crute, 1987; Lebeda & Zinkernagel, 2003a).

It is evident that the composition of v-phenotypes of B. lactucae originating from L. serriola is highly specific and completely different from that of isolates originating from L. sativa (Lebeda, 1984, 2002). Experiments have shown that the virulence structure of natural populations of B. lactucae is generally uniform, because the majority of v-phenotypes were composed of v-factors that matched resistance genes and factors found in either L. serriola, or L. sativa with resistance derived from L. serriola (Table 3; Lebeda, 2002). Also, B. lactucae isolates originating from France and Germany have comparable v-phenotypes (P24, P35, P36) with isolates collected in the Czech Republic (Table 3). It appears that this phenomenon could be based not only on the background of specific resistance, but also on the ‘general’ genetic background of L. serriola, but further detailed research must be carried out to verify this. However, variation in the presence of ‘L. serriola’-related v-factors was frequently recorded among isolates of the same pathogen population.

Detailed information about the distribution of virulence in natural populations of B. lactucae in other European countries where L. serriola occurs frequently (Doležalováet al., 2001; Lebeda et al., 2001a, 2004) is still lacking. It will be interesting to know whether there is uniformity on a larger geographic scale, or whether there are locally adapted populations in each geographic region similar to those in the crop pathosystem L. sativaB. lactucae (Crute, 1987; Lebeda & Zinkernagel, 2003a).

Nevertheless, considerable variation not associated with the composition of v-phenotypes described above was detected in a region of southern Moravia. There was a group of a few isolates, whose virulence patterns were completely different (P18, P26, P27, P33; Table 3). They expressed highly complex virulence to L. sativa differentials, and their v-phenotypes differed significantly by the presence of various v-factors matching resistance genes localized in L. sativa cultivars. One isolate even had a v-phenotype P33 (Table 3), the same as that of the previously described and internationally accepted race NL16 (BL-16) (Lebeda, 2002). A comparison with available literature (Bonnier et al., 1994; Lebeda & Zinkernagel, 2003a, 2003b) and the authors’ unpublished results has revealed that the other isolates are not similar in virulence to any recently found isolate originating from L. sativa. However, these isolates were collected in a region (southern Moravia) where production of field lettuce is quite frequent, but where naturally growing L. serriola is also common (Lebeda et al., 2001a). It was suggested (Lebeda, 2002) that when plants of L. serriola grew close to a field of lettuce, pathogens from the crop pathosystem can sometimes be transferred to the wild plant pathosystem. However, recent data showed that migration of B. lactucae spores among crop and wild pathosystem is probably rare (Petrželová & Lebeda, 2004). There are various possible reasons for this, e.g. low density of host and pathogen populations, completely different virulence structure in both pathosystems, low probability of cross-infections, short-distance movement of spores. Some aspects related to this topic are discussed by Lebeda & Zinkernagel (2003a).

Four isolates from the site at Pravlov had a v-phenotype P25 (Table 3) which was, for the most part, similar to the v-phenotype typical of isolates from L. serriola (very similar to P24). However, its clear virulence to cv. Iceberg (R?) is similar to isolates originating from L. sativa. The virulence pattern of an isolate from the site Ořechov was unusual. It possessed nearly all known v-factors (Table 3) and is considered to be the most virulent isolate, with a highly complex v-phenotype (P37) found in the wild pathosystem L. serriolaB. lactucae. It is interesting that this isolate possessed a high proportion of v-factors that match resistance genes and/or R-factors carried by L. serriola and various L. sativa cultivars. This does not agree with the findings mentioned above. Lannou & Mundt (1997) pointed out that complex races of pathogens have a disadvantage in fitness proportional to the number of virulences they possess. However, the pathogen on the lettuce crop is characterized by very high virulence complexity and homogeneity of v-phenotypes, with the presence of many ‘unnecessary’ v-factors (Crute, 1987; Lebeda & Zinkernagel, 2003a). This indicates that the conclusion of Lannou & Mundt (1997) is not wholly correct as far as B. lactucae on lettuce is concerned.

Most of the v-phenotypes described in this study (Table 3) differed from each other, mostly by the presence or absence of only a few v-factors or virulences. These differences, and thus the genetic distances between isolates, would be expected to be far larger if sexual reproduction was the main factor in generating variation (Drenth & Goodwin, 1999). However, recent analyses have shown that in Czech populations of B. lactucae from L. serriola the B2 mating type prevails, while B1-type isolates were rarely found (Petrželová & Lebeda, 2003). Also, the spontaneous occurrence of oospores in natural populations of B. lactucae is extremely rare (Petrželová & Lebeda, 2003). This further supports the view that mechanisms other than sexual reproduction should be considered (Drenth & Goodwin, 1999) as the cause of the variability for virulence in natural populations of B. lactucae. Nevertheless, the uniqueness of some v-phenotypes (Table 3) is unlikely to occur in pathogen populations that are restricted by asexual reproduction. The occurrence of sexual reproduction, i.e. hybridization and/or recombination of B. lactucae strains originating from the crop (L. sativa) and those from the wild lettuce (L. serriola) pathosystem, appears to be a likely explanation for the existence of these isolates (Lebeda & Blok, 1990).

This study has shown the existence of virulence variation both between individual spatially isolated pathogen populations and within subpopulations. From the results obtained, it can be concluded that individual plants growing within a host population may be infected with different v-phenotypes of B. lactucae. The existence of inter- and intrapopulation variation was demonstrated previously for populations of B. lactucae on cultivated lettuce (Lebeda, 1982; Gustafsson et al., 1985; Crute, 1987; Lebeda & Zinkernagel, 2003a). The distribution of virulence within and among populations is influenced significantly by the process of selection (McDermott & McDonald, 1993). Differences in the virulence structure found among various populations and subpopulations of B. lactucae are probably the product of different selection pressures exerted by a specific genetic background of resistance in local L. serriola populations. The existence of this phenomenon was demonstrated by Crute (1990) and, more recently, by Sicard et al. (1999). The diversity of pathotypes in one B. lactucae population could be caused by heterogeneity and temporal coevolution with a host population (Lebeda & Zinkernagel, 2003a; Petrželová & Lebeda, 2004; Thrall & Burdon, 2003).


The authors gratefully thank Dr B. McDonald (FIT, Zurich, Switzerland) and Dr J. Nielsen (Winnipeg, Canada) for valuable comments on the manuscript. The research was supported by grant MSM 153100010 (Ministry of Education, Praha).