Contrasted responses of Botrytis cinerea isolates developing on tomato plants grown under different nitrogen nutrition regimes

The nutritional status of a plant is known to influence its susceptibility to pathogens. In the case of Botrytis cinerea the role of nitrogen fertilization of various host plants on disease development appears to be variable. This study was carried out to characterize possible variability associated with strains and inoculum density of B. cinerea in its ability to infect leaf-pruning wounds and to develop stem lesions on tomato plants as affected by the nitrogen input. Six strains differing in their aggressiveness to tomato were compared. They all had similar reaction patterns in vitro in response to differential nitrogen levels. In tests on plants grown with contrasted regimes of nitrate fertilization, overall disease severity was lower for all strains on plants with higher nitrogen inputs, regardless of inoculum concentration. However, differences among strains were observed in the effect of plant nitrogen nutrition on infection and on lesion expansion. Disease onset was delayed on all plants with higher nitrogen inputs, but the response was greater for strains with lower aggressiveness on tomato. The highest contrast among strains was observed with the colonization of stems. The daily rate of stem lesion expansion decreased with increasing nitrogen fertilization levels for the more aggressive strains, while it increased for the less aggressive strain. Hypotheses to explain these results are discussed in light of the possible physiological effects of nitrogen fertilization on nutrient availability for the pathogen in the host tissue and of possible production of defence metabolites by the plant. Additional key words Solanum lycopersicum, Lycopersicon esculentum, gray mould, host resistance Version définitive du manuscrit publié dans / Final version of the manuscript published in : Plant Pathology, 2010, 59, 891-899 DOI: 10.1111/j.13653059.2010.02320.x. The original publication is available at http://onlinelibrary.wiley.com/doi/10.1111/ppa.2010.59.issue-5/issuetoc. M a n u sc ri t d ’a u te u r / A u th o r m a n u sc ri p t M a n u sc ri t d ’a u te u r / A u th o r m a n u sc ri p t M a n u sc ri t d ’a u te u r / A u th o r m a n u sc ri p t Introduction It has long been recognized that the nutritional status of a plant can play a role in its susceptibility to pathogenic fungi. Nitrogen, in particular, is deemed to strongly influence the host-pathogen interactions (Huber & Watson, 1974, Huber & Thompson, 2007). However, the nitrogen (referred to as "N" in the rest of this paper) status of a plant can be either favourable or unfavourable to the infection process, depending on the pathosystem (Huber & Thompson, 2007). This variable effect can be explained by the fact that plant N nutrition may have an influence on many factors involved in the epidemiological cycle. Such factors include the molecules involved either in host defence or in the virulence and aggressiveness of the pathogen, the quantity and nature of host N-based substrates acquired by the pathogen and the microclimate around the plant (through an effect on plant vigour and architecture). The N level of plant tissues has often been correlated to host susceptibility, one explanation being that at high plant N content more substrate is available for the development of the pathogen (Jensen & Munk, 1997, Neumann et al., 2004, Walters & Bingham, 2007). In contrast, it has been shown in Arabidopsis thaliana that the constitutive and induced levels of some proteins involved in plant resistance to infection are higher at high N nutrition (Dietrich et al., 2004). But other defence compounds and molecules acting as structural barriers against pathogens can be lowered at low C/N ratio, which are the consequence of a high N nutrition (Talukder et al., 2005). It has also been suggested that plant soluble carbohydrate content – which is negatively correlated with N nutrition – has a positive influence on plant susceptibility (Hoffland et al., 1999). Although these contradictory effects highlight the need for pathosystem-specific studies, a general rule has been suggested (Solomon et al., 2003): the development of biotrophic fungi would be enhanced by nitrate and inhibited by ammonium, while the contrary would be true for necrotrophic fungi. Regarding the influence of N nutrition on the necrotroph Botrytis cinerea, the situation is not as clear (Dik & Wubben, 2004). Higher plant susceptibility to B. cinerea was reported at high N fertilization rates in legumes (Davidson et al., 2004), grape (R'Houma et al., 1998), sweet basil (Yermiyahu et al., 2006) and begonia (Pitchay, 2007). In three of these studies (R'Houma et al., 1998, Yermiyahu et al., 2006, Pitchay, 2007), the N source was nitrate or a mixture of nitrate and ammonium. In contrast, a high N nutrition seems to lower the level of disease in tomato (Verhoeff, 1968, Hoffland et al., 1999). It has been reported that the susceptibility of bean to B. cinerea was 2.5 fold higher with an ammonium nutrition compared to a nitrate-based source of N (Huber & Watson, 1974). The explanation for this was that ammonium enhanced cell permeability and increased leaf exudates, both factors being favourable to infection (Huber & Watson, 1974). For grape, pruning of leaves and fruits lead to a reduction in infection, even at high N, suggesting that the effect of N was mainly to increase leaf surface, thus rendering the microclimate around infection sites more conducive to infection (R'Houma et al., 1998). For sweet basil, the percentage of stems carrying sporulating lesions was higher at higher nitrate concentration in the fertigation solution, but the percentage of infected plants, the lesion size and the rate of disease progression were not affected (Yermiyahu et al., 2006). For begonia, disease incidence was higher at 42 mmol.L N (brought as ammonium nitrate), a concentration hardly found in agricultural situations. However, between 1.7 mmol.L and 28 mmol.L N, the relationship between N nutrition and susceptibility was quadratic, with a maximum at 7 mmol.L N (Pitchay, 2007). Thus the effect of plant internal N content on B. cinerea infection and lesion growth appears to be highly dependent on the host species. This suggests that there could be a trade-off between a “trophic component” and a “defence component” of the host-pathogen interaction. For example, a high plant N status could raise the level of nutrients accessible to the pathogen, while at the same time enhance host defences. An unexplored hypothesis to explain the reported variability could be that the host-pathogen interaction could be differentially affected by N fertilisation depending on the strain of B. cinerea. The Version définitive du manuscrit publié dans / Final version of the manuscript published in : Plant Pathology, 2010, 59, 891-899 DOI: 10.1111/j.13653059.2010.02320.x. The original publication is available at http://onlinelibrary.wiley.com/doi/10.1111/ppa.2010.59.issue-5/issuetoc. M a n u sc ri t d ’a u te u r / A u th o r m a n u sc ri p t M a n u sc ri t d ’a u te u r / A u th o r m a n u sc ri p t M a n u sc ri t d ’a u te u r / A u th o r m a n u sc ri p t response to N availability, in vitro and in planta, of B. cinerea strains of contrasting aggressiveness, as well as the effect of inoculum density, should help to better understand the nature of this interaction. In the present study, our objective was to confirm that the N content of nitrate-fed tomato plants has an influence on their susceptibility to B. cinerea, and to test the influence of initial inoculum density and fungal strain on the growth response of the fungus, both in vitro and in planta. Materials and methods Three experiments were conducted with tomato plants produced under differing N nutrition regimes between May and July 2008 (experiment E1), between June and August 2008 (experiment E2) and between March and May 2009 (experiment E3). Although comparable in its design, experiment E3 differed from E1 and E2 by the number of N treatments and strains of B. cinerea tested. Additionally, two in vitro tests with B. cinerea on contrasting nutrient medium were conducted in summer 2009. Production of plant material and fertigation treatments Tomato (Solanum lycopersicum var. esculentum) seeds (cv Swanson, De Ruiter Seeds, Saint Andiol, France) were sown in 1 cm rock wool cubes in a greenhouse. Ten days after sowing, the cubes, each containing one plantlet, were transferred onto rock wool blocks 7.5 x 7.5 x 6cm (Grodan®, Roermonds, the Netherlands). During the first month, the plants were fertigated twice a day with a standard commercial nutrient solution (Duclos international, Lunel, France). After that period, the plants (bearing 3-4 leaves) were placed on the top of 2-liter pots filled with a mixture (1:1 V/V) of vermiculite and pozzolana (inert crushed volcanic rock) to start the nutrition treatments. Thirty plants were used in each treatment. In experiments E1 and E2, three levels of NO3 concentrations were tested: 0.5 mmol.L NO3 , 5 mmol.L NO3 and 15 mmol.L NO3 ; in E3, five levels were tested: 0.5 mmol.L NO3 , 2 mmol.L NO3 , 5 mmol.L NO3 , 10 mmol.L NO3 and 20 mmol.L NO3 The equilibrium in electric charges was maintained by replacing nitrates by sulphates in the solutions with less nitrate. The concentration of other major nutrient elements was kept constant, at the following levels: 11 mmol.L K, 3.5 mmol.L Mg, 3.5 mmol.L Ca and 1 mmol.L P. Oligo-elements were also added at the following concentrations (in μmol. L ): 20.6 B, 0.5 Cu, 10.7 Fe, 11.6 Mn, 0.28 Mo, and 3.2 Zn. The plants were fertigated with a drip irrigation system (one dripper per pot) up to 6 times a day depending on the climatic demand, with one minute pulses. Three pots chosen at random were weighted continuously to evaluate their loss of water, and thus the climatic demand in the greenhouse. The pH was adjusted to 6 in each treatment by addition of H2SO4. Plants were grown with those solutions for four (in E1) or three (in E2 and E3) weeks and were then inoculated. Evaluation of plant susceptibility to Botrytis cinerea Two monoconidial strains of B. cinerea (BC1 and BC21), previously collected in commercial greenhouses and used routinely in the laboratory, were used in E1 and E2. Four additional strains (BC43, BC44, BC84 and NHPm4) were added in experiment E3. From previous work of the laboratory, strains BC1, BC43 and BC44 were known to have a high level of aggressiveness on tomato, while strains BC21, BC84 and NHPm4 had a medium to low level of aggressiveness (Ajouz, 2009, Ajouz et al., 2010). For each strain, inoculum was produced on potato dextrose agar medium (39 g L, Difco, Detroit, USA) in a growth chamber (18°C night, 22°C day, and 14h daylight). Spores were collected in sterile distilled water from the surface of 14-day old cultures. Each suspension was filtered through a 30 μm mesh sterile filter to remove mycelium fragments and adjusted to the desired concentration with the help of a hemacytometer. Three infection concentrations where tested in E1 and E2: 10 spores per mL (e5), 10 spores per mL (e6) and 10 Version définitive du manuscrit publié dans / Final version of the manuscript published in : Plant Pathology, 2010, 59, 891-899 DOI: 10.1111/j.13653059.2010.02320.x. The original publication is available at http://onlinelibrary.wiley.com/doi/10.1111/ppa.2010.59.issue-5/issuetoc. M a n u sc ri t d ’a u te u r / A u th o r m a n u sc ri p t M a n u sc ri t d ’a u te u r / A u th o r m a n u sc ri p t M a n u sc ri t d ’a u te u r / A u th o r m a n u sc ri p t spores per mL (e7). Only concentrations e6 and e7 were retained in experiment E3. The third, fourth, fifth and sixth leaves of 5 randomly selected plants were excised, leaving 5-10 mm petiole stubs on the stems. The wounds were inoculated with 10 μL aliquots of a spore suspension. In E1 and E2, strain BC1 was inoculated on petioles 3 and 5 and strain BC21 on petioles 4 and 6, with different groups of plants for the different inoculum concentrations. A total of 15 plants were inoculated per N treatment. In E3, each plant was inoculated with one strain only, at concentration e6 on petioles 4 and 6 and at concentration e7 on petioles 3 and 5. A total of 30 plants were inoculated per N treatment. After inoculation, the plants were placed in a growth chamber for 7 days. The chamber was set at 21°C, 90% RH and 14h daylight. During this period, the plants were irrigated manually, twice a day, using the same fertilization solutions as those used before inoculation. Symptoms were assessed between the 3 and 7 day after inoculation. The incidence of stem lesions and the length of developing lesions (in mm) were recorded daily. Area under the disease progress curves (AUDPC) were computed as described by Aissat et al. (2008) and Decognet et al. (2009), as AUDPC = [Y1/2 + 1