Virulence of Spanish Phytophthora nicotianae isolates towards Capsicum annuum germplasm and pathogenicity towards Lycopersicum esculentum

The virulence of six northwestern Spanish Phytophthora nicotianae Breda de Haan isolates towards the pepper (Capsicum annuum) cultivars commonly used for the determination of P. capsici Leonian pathotype (SCM 334, PI201234 and Yolo Wonder), was similar to that of the latter oomycete but different to that of the German reference isolate P. nicotianae 411.87. When nine local C. annuum cultivars were inoculated with a P. nicotianae isolate of well known pathogenicity, the virulence recorded was significantly different to that recorded for P. capsici. Though the average level of resistance of the northwestern Spanish Capsicum germplasm to P. nicotianae was incomplete and weaker than that recorded for the SCM 334 and PI201234 cultivars, genotypes Co 12B and Co 3.25 showed high resistance to this oomycete. None of the six northwestern Spanish P. nicotianae isolates tested were pathogenic towards Lycopersicum esculentum cv. S. Pedro, unlike P. capsici, which produced clear collar and root rot in this host. Additional key words: collar rot, pepper, resistance, root rot.

Palabras clave adicionales: pimiento, podredumbre de cuello, podredumbre de raíz, resistencia.(Erwin and Ribeiro, 1996).Hosts susceptibility is variable, but isolates usually show the strongest pathogenicity towards the hosts from which they were isolated (Bonnet et al., 1978).Two different pathotypes of P. nicotianae have been identif ied among the isolates that affect tobacco (Apple, 1962;Van Jaarsveld et al., 2002), but no confirmed information is available regarding the pathotypes affecting other hosts.The existence of pathotypes within the P. nicotianae-Lycospersicum esculentum L. interaction has, however, been suggested (Boukema, 1982).The virulence of P. nicotianae towards tobacco (Nielsen, 1992;Van Jaarsveld et al., 2002) and tomato germplasm (Sharma et al., 1974;Kohli et al., 1996) has been examined, as well as towards that of pepper in Tunisia (Allagui, 1994) and India (Verma et al., 2001).As far as we know, no previous screening for sources of resistance to P. nicotianae in C. anuum germplasm has been performed in Europe.
The aims of the present work were: (i) to study the virulence of P. nicotianae isolates to well known reference C. annuum cultivars used for determining P. capsici pathotype, (ii) to evaluate the resistance of northwestern Spanish pepper germplasm to local isolates of P. nicotianae, and (iii) to study the pathogenic behaviour of these isolates towards L. sculentum, a well known worldwide host of this oomycete.

Pathogen material
The fungal material used comprised 12 Phytophthora isolates (Table 1), 10 of which belonged to P. nico-tianae and two of which belonged to P. capsici (employed as reference controls).Nine Phytopthora spp.isolates -eight of which were P. nicotianae and one of which was P. capsici-were obtained from diseased pepper plants in four locations in northwestern Spain.The pathogen was isolated from the infected plants on potato dextrose agar (PDA) (Rapilly, 1968) and on a selective medium (Ponchet et al., 1972).The taxonomic criteria used for the identification of Phytophthora species were those described by Stamps et al. (1990).Isolates 370.72 and 411.87 provided by the Centraalbureau voor Schimmelcultures (The Netherlands) were used as reference controls.Isolate 00/307, collected in Burgos (northern Spain), was provided by NEIKER (the Research Centre of the Basque Regional Government).

Host plants
The pepper plants used represented 12 genotypes (Table 2), nine of which were C. annuum lines from northwestern Spain and conserved at the CIAM (Centro de Investigaciones Agrarias de Mabegondo).The remaining Yolo Wonder and SCM334 and PI201234 lines were supplied by the University of La Coruña (Spain).The commercial San Pedro tomato (Lycopersicum esculentum L.) cultivar was used in the virulence tests.

Host preparation
The inoculum assays were conducted under greenhouse and growth chamber conditions.Seeds of the different genotypes were first sown on plastic trays

Inoculum assays
The Phytophthora isolates were grown on V8 juice agar (Erwin and Ribeiro, 1996) at room temperature for 8 days.Each inoculum was prepared by seeding isolate samples in sterile 0.01 M potassium nitrate solution in Petri dishes.Sporangia were formed by growing this culture under UV light at 24ºC for 7 days.When abundant, the potassium nitrate solution was replaced by sterile distilled water, under aseptic conditions.The Petri dishes were then kept at 5ºC for 30 min and then at room temperature (20-24ºC) for 3 h to release the zoospores.The zoospore solution was f iltered using sterile Whatman paper, vibrated for 1 min and adjusted to 2 × 10 4 zoospores per ml (Bartual et al., 1991).At the six leaf stage, the collar of each plant was inoculated with 5 ml of the zoospore solution (Gil Ortega et al., 1995).

Experimental design
In the virulence tests, seven P. nicotianae isolates -six from Spain and one reference isolate of German origin-were inoculated into the well known reference C. annuum cultivars PI201234, SCM 334 and Yolo Wonder, commonly used for P. capsici pathotype identification.The experiments for each isolate had a complete random block design with three replications and 20 plants per replication.Assays were performed in a growth chamber inoculating with one isolate at a time.All experiments included non-inoculated controls.
The assessment of resistance to P. nicotianae in the local pepper germplasm was also based on a complete random block design for each isolate, with three replications and 20 plants per replication.In these tests, 10 pepper lines -nine local germplasm types and the susceptible Yolo Wonder-were inoculated with two Phytopthora isolates -P.nicotianae and P. capsicifrom northwestern Spain.These experiments were conducted under greenhouse conditions.
The pathogenic behaviour of P. nicotianae towards tomato was studied in a growth chamber.Five isolates of the oomycete -four from northwestern Spain and one of German origin-plus a reference P. capsici strain, were inoculated into tomato cv.San Pedro plants.This experiment also had a complete randomised block design (three replications per each cultivar-isolate combination and 20 plants per replication).

Disease severity ratings
Disease severity was rated 14 days after inoculation in the virulence tests, at 30 days post-inoculation for the evaluation of the resistance of C. annuum to P. nicotianae, and 14 and 21 days post-inoculation in the study of the pathogenicity of P. nicotianae towards tomato.Each plant in each experiment was rated on the following scale: 0, no symptoms (0% disease); 1, small rots on the base of the stem, to up to 30% of the plant affected; 2, 31-50% of the plant affected; 3, 51-70% of the plant affected; 4, 71-90% of the plant affected; 5, dead plant, 100% of the plant affected (Kim and Hwang, 1992).

Statistical analysis
Mean disease severity ratings were compared after transforming the data for each plant using the following formula: Y = arcsin √  X/100, where X is the disease rating as a percentage.The virulence tests were performed separately for each individual isolate;

Results
The P. nicotianae isolates from northwestern Spain produced clear disease symptoms on cv.Yolo Wonder but were not pathogenic towards cvs.PI201234 and SCM334 (Table 3).The virulence of these isolates was not different to that of the two P. capsici isolatesincluding isolate 370.72 from the USA [which is classified as pathotype 0 according to Gil Ortega et al. (1995)], although the P. capsici strains were clearly more pathogenic than P. nicotianae towards cv.Yolo Wonder.The reference P. nicotianae isolate 411.87 showed significantly different virulence to all other isolates, in fact it seemed not to be pathogenic towards any of the three inoculated cultivars.
No complete resistance to P. nicotianae was observed among the northwestern Spanish pepper lines, whereas the reference Capsicum cultivars PI201234 and SCM334 were completely resistant (Table 4).The average level of resistance to this pathogen was higher than that recorded for P. capsici.Although quite susceptible to P. capsici, lines Co 12 B and Co 3.25 showed a high level of resistance to P. nicotianae.
None of the P. nicotianae strains inoculated into tomato San Pedro produced any kind of symptoms, while the P. capsici isolate used as a control caused clear collar and root rots 21 days post-inoculation (although it caused no plant deaths) (Table 5).The pathogenicity of the P. capsici isolate was confirmed

Discussion
All of the northwestern Spanish P. nicotianae isolates were similar in their virulence towards the reference varieties proposed by Gil Ortega et al. (1995) for determining the P. capsici pathotype.Certain variation in pathogenicity towards the pepper cv.Yolo Wonder was seen among the P. nicotianae isolates, but these were not pathogenic towards SCM334 and PI201234; similar results were obtained with P. capsici.These results differ from those of Allagui and Lepoivre (1996) who reported the virulence of the P. nicotianae strains isolated from pepper in Tunisia when inoculated into these same varieties: the Tunisian strains were pathogenic to PI201234 but not to SCM334.Whether this means that the Tunisian and Spanish isolates belong to different pathotypes needs to be determined in future work.
These results provide information valuable in breeding programs.Most of the pepper genotypes were moderately resistant to the oomycete, and some showed strong (although incomplete) resistance.The resistant pepper germplasm from northwestern Spain (Co-12B and Co 3.25) might be useful in breeding programs after the source of its resistance is determined.The lack of complete resistance to the pathogen among the Spanish germplasm tested, in comparison with that observed for the American lines SCM334 and PI201234, implies they might have different mechanisms of resistance.This has been reported for the host-pathogen interaction of P. nicotianae and carnation: one mechanism appears to be related to aggressiveness and another to compatibility (Maia and Venard, 1978).A similar situation was also confirmed recently for the interaction between P. capsici and C. annuum in Spain (Andrés et al., 2005b).
The lack of pathogenicity of the P. nicotianae isolates obtained from pepper when inoculated into tomato was also observed in Tunisia by Allagui and Lepoivre (2000): these authors proposed the possible existence of a well-defined forma specialis within this species.Elicitins produced by P. nicotianae have been reported as microbial signalling molecules involved in the host-specificity of this pathogen (Kamoun et al., 1993) and specif ically in the P. nicotianae-tomato interaction (Colas et al., 1998;Capasso et al., 1999).This possible specialisation could be important for horticultural growers in northwestern Spain where crop rotation involving tomato and pepper (both potential hosts of P. nicotianae) is common.
Overall, these f indings conf irm the existence of C. annuum germplasm in northwestern Spain with relatively high levels of resistance to P. nicotianae.This material could be useful in breeding programs with the aim of producing more resistant plants that require the use of less fungicide.also like to thank Dr. Santiago Larregla of NEIKER (Vizcaya) and Prof. Fuencisla Merino of the University of La Coruña for providing the Phytophthora isolates and reference pepper genotypes.

Table 1 .
Isolates of Phytophthora capsici and P. nicotianae used (32 × 30 × 20 cm) filled with a mixture of sterilized peat (50% in volume) and sand.For greenhouse experiments, pepper and tomato plants were grown in plastic trays at 18-22ºC.For growth chamber experiments, these trays were kept at 22-28ºC with a photoperiod of 16 h at 30,000 lux.

Table 2 .
Origin of the Capsicum annuum genotypes used means were compared using Duncan's multiple range test.The means of cultivar tolerance to P. nicotianae and P. capsici were compared separately for each oomycete using the Waller-Duncan's multiple range test.The mean pathogenicity of P. nicotianae isolates towards tomato San Pedro were also compared using Duncan's multiple range test.Significance was set at P < 0.05 for all tests.All calculations were performed using SAS System v. 8 software(SAS Institute, 1999).

Table 4 .
Mean disease severity ratings of northwestern Spanish pepper germplasm inoculated with one isolate of P. capsici (Pa-1) and another of P. nicotianae (Ar-3) collected in northwestern Spain, at 30 days post-