Superior resistance to Phytophthora infestans in new pre-breeding potato clones under different nitrogen fertilisation regimes used in organic agriculture

ABSTRACT Late blight caused by Phytophthora infestans is the most devastating disease for cultivated potatoes, causing considerable financial loss annually due to loss of yield and costs of fungicide applications. For organic farming especially, the strict limitations on the use of fungicides makes research necessary to investigate factors that influence disease severity. Therefore, effects of nitrogen fertiliser rates on leaf and tuber blight in organically grown potato cultivars and pre-breeding clones were evaluated in the present study. It was found that the foliage of the pre-breeding clones was little, or not at all, affected by P. infestans, irrespectively of the nitrogen application rate, while the cultivars reacted differently with mild to strong late blight symptoms. Only some of the clones and cultivars showed increased infection of tubers at the higher nitrogen application rates, whilst most showed no significant differences between the N rates. Overall, the present study showed that the use of new resistant cultivars could be a suitable approach to reduce late blight in organic farming, where soil nitrogen levels may vary due to the use of different forms of nutrient inputs with varying mineralisation rates.


Introduction
Potato cultivation is affected by high yield losses due to late blight, caused by the oomycete Phytophthora infestans.This is a major problem, especially in organic farming, as the European Union organic agriculture standards only allow active substance copper compounds as control agents (EU Regulation 2018).Copper accumulates in the soil and can become harmful to soil organisms in high concentrations (Rajput et al. 2020).In addition to the use of copper, other agronomic measures can also be used to limit or prevent P. infestans infestation, including for example, crop rotations with a minimum three-year break in potato cultivation (Bødker et al. 2006;Hannukkala et al. 2007;Van Bruggen et al. 2016), the use of non-infected potato seed tubers (Ghorbani et al. 2004) or pre-sprouting so that the plants have a head start on P. infestans (Waschl and Hein 2010).Strip intercropping with cereals and a grass-clover mixture can also reduce infestation (Bouws and Finckh 2008).The use of microorganisms and plant extracts has so far proved to be less effective than copper (Ghorbani et al. 2005;Dorn et al. 2007), though this is an area that is still being actively researched and that may show potential in future.
The amount of fertiliser used can also influence the infestation.With regard to nitrogen, however, many studies have reported contrasting results.Juárez et al. (2000) studied two cultivars and three different nitrogen application rates and concluded that higher nitrogen rates led to a more severe course of the disease.Greenhouse trials with one cultivar and two nitrogen levels by Mittelstraß et al. (2006) also reported higher susceptibility in plants grown at the high nitrogen level.This was also the conclusion of Ros et al. (2008), who investigated four cultivars and two levels under greenhouse conditions, and also of Bangemann (2010) using two cultivars but four nitrogen levels.Jensen and Nielsen (2015) concluded that there was a tendency that higher nitrogen levels led to more severe infestations.Sunita et al. (2011), however, came to the opposite conclusion, and Cicore et al. (2012) found no difference between the nitrogen rates, although only two Argentine cultivars were used in their study.Agu (2004) came to the same conclusion, but with additional phosphorus application, a stronger infestation was found at the high nitrogen application rates, though the study only included one cultivar.In Jin et al. (2014) and Jha et al. (2019), the lowest infestation of P. infestans was detected at medium nitrogen levels (Table 1).
In the evaluation of data from several farms by Möller et al. (2006) and Möller and Reents (2007), it was unclear whether under organic farming conditions the nitrogen application rate had any influence on the P. infestans infestation.The aim of this study reported here was to clarify whether the nitrogen application rates can influence late blight infestation on both foliage and tubers.To the knowledge of the authors, no previous study has combined field and laboratory tests and most studies used four or less genotypes in their investigations, which makes it difficult to generalise results.Previous studies have focused mainly on foliage blight.Furthermore, with two exceptions (Möller et al. 2006;Möller and Reents 2007), only chemical nitrogen fertiliser was used in past publications.As organic farming is increasing strongly at present, the focus of this study was to use of a fertiliser approved for use in organic production systems, as well as to test a greater number of genotypes.Another special aspect of the present study was that the influence of the nitrogen fertiliser was not only investigated using released potato cultivars, but also resistant pre-breeding clones.

Plant materials and experimental design
The trials were set up in 2020 and 2021 on the experimental field of the Julius Kuehn Institute in Groß Lüsewitz, Germany in a randomised block design with three nitrogen fertiliser rates, 50 (N1), 100 (N2) and 200 (N3) kg N ha −1 total available nitrogen (including N min in the soil), and three replicates.Each plot consisted of 10 plants in a row and planted at a distance of 28 cm.The distance between the rows was 75 cm.The ware cultivars Adretta, Jelly, Krone (susceptible controls) and Otolia (moderately resistant control) were included in the trial, as well as 20 pre-breeding clones with medium-early to medium-late maturity that had proven resistant in previous studies and were developed together with the Bavarian State Research Centre for Agriculture, Germany (LfL) in a previous project.The N-fertiliser 'Diaglutin N pellet' (Biofa, Münsingen, Germany), consisting mainly of heated feather meal and which is approved for use in organic agriculture, was applied two weeks after planting at the beginning of May before the final ridge shaping.Although only approved organic fertiliser inputs were used on the land, the area in this trial was not managed to full organic standards, as in both years the Colorado potato beetles (Leptinotarsa decemlineata) had to be controlled to not endanger the trial.The phytosanitary measures used that deviate from the practice of organic farming are listed in Supplemental Table S1.

Assessment of late blight infestation
Foliage blight infestation was determined both directly in the field and with a detached leaf assay.Tuber blight infestation was determined with a tuber slice test.At the end of flowering of the cultivar Adretta, one plant per experimental plot was inoculated with 5 ml of a P. infestans suspension of common field isolates, a composition which was supplemented each year with newly collected pathogen material (1.2 × 10 4 sporangia ml −1 , Darsow 2008).Additionally, the pathogen spectrum was analysed by the James Hutton Institute using FTA cards, showing that in 2020, primarily the pathotypes EU_41_A2 and EU_13_A2 were detected and in 2021, primarily the pathotype EU_37_A2 was identified.The infestation was assessed as the percentage of the infested foliage area of each plot, excluding the inoculated plant, every three to four days until an infestation of 100 % was reached, or until the respective clone had matured.Based on these assessments, the relative Area Under Disease Progress Curve (rAUDPC) values were calculated (Fry 1978).Since late blight resistance is often associated with late maturity, the rAUDPC values were additionally maturity-corrected according to Truberg et al. (2009) and presented as ΔrAUDPC values.
For the detached leaf assay, five leaves of different plants per plot were taken shortly before inoculation in the field trial and 20 μl drops of the P. infestans suspension were applied onto the underside of each leaf in the laboratory.After 24 h, the leaves were turned and incubated for another five days at 16°C, 95 % relative humidity (RH) and 150 Lux.Scoring was based on the necrotic leaf area and mycelium formation on the underside of the leaf, with scores from 1 (no infestation) to 9 (full infestation).The tuber slice test was performed on four tubers per plot.One slice from each tuber was inoculated with 20 μl of a low (1900 sporangia ml −1 ) and a high (15,000 sporangia ml −1 ) concentrated P. infestans suspension.After 24 h, the slices were rotated and then incubated for another five days at 16°C, 95 % RH.Scoring was based on mycelium formation, rotting and browning with scores from 1 (no infestation) to 9 (full infestation) for both concentrations.The mean value was used for evaluation.

Statistical analysis
The statistical evaluation was carried out by an analysis of variance (ANOVA) using the program R version 3.6.3.Fixed effects were the genotype and the N rate and random effects were the year and the replicates.The differences between the fertiliser levels were then determined with the Tukey test after verifying the normal distribution of residuals (Shapiro-Wilk test).Means were considered to be statistically significant at p < 0.05.

Results
The effect of the N fertiliser was confirmed by the significantly different yields of the three levels (N1 1.003 kg plant −1 , N2 1.210 kg plant −1 , N3 1.320 kg plant −1 , Supplemental Table S2).ANOVA revealed for all traits that the N rate, the year, the genotype and their interactions (fertiliser x genotype, fertiliser x year and genotype x year) influenced the infestation (Table 2).Therefore, years and genotypes were analysed separately.The results of the various tests are summarised below and are presented in detail in Supplemental Table S3.The figures show four of the cultivars and two clones as examples.

Field test
Overall, in both years, the susceptible cultivars were clearly more susceptible to P. infestans than the clones, with Otolia showing the lowest infestation in the cultivar comparison (Figure 1).Adretta was the most susceptible cultivar in the trials and the only one that showed a significant increase of infection levels when N-fertilisation was increased (Supplemental Table S3, Mean rAUDPC 2020).However, the difference of infection between N-levels for Adretta was only significant in 2020, when the disease severity across the trial was higher due to seasonal differences.The opposite behaviour was observed for the cultivar Krone, which showed a tendency of less (though non-significant at p < 0.05) infection when N-fertilisation was higher (Supplemental Table S3, Mean rAUDPC).
In 2020, no significant effects of the nitrogen level on the infestation (∆rAUDPC values) were found in 18 out of the 20 pre-breeding clones and in only two of the clones the infestation was significantly increased at the higher nitrogen application rates (Supplemental Table S3).Of the cultivars, Krone, showed a significantly lower incidence with the higher N application rates (∆rAUDPC).The rAUDPC values were significantly higher with higher N application rates for Adretta and tended to be lower for Krone (Figure 2a).The rAUDPC values of all pre-breeding clones remained unaffected by the nitrogen rate (Supplemental Table S3; Figure 2b).In 2021, one clone and Adretta had significantly higher ∆rAUDPC values with higher N application rates (Supplemental Table S3).

Detached leaf assay
In the detached leaf assay in 2020, Adretta was significantly more affected at the highest N level.The opposite was the case for one of the clones and Krone.In 2021, Otolia showed significantly less infestation with higher N rates.The rest of the clones and cultivars showed no significant differences between the N rates (Supplemental Table S3).

Tuber slice test
In the tuber slice test in 2020, no effects of the N level on the infestation in the tuber slices were determined (Figure 3a, Supplemental Table S3).In 2021, the infestation of two clones and Krone were significantly affected by N dosage (Figure 3b), whereby, the infestation increased at higher nitrogen rates (Supplemental Table S3).

Discussion
The results reported in the literature show different effects of N fertilisation levels on P. infestans (Table 1).Older theories of plant pathogen interactions describe a promoting effect of high N fertilisation on the development of late blight (Herms and Mattson 1992), possibly due to higher foliar density providing a more suitable environment for infection (Runno-Paurson et al. 2020), or lower plant defence (Snoeijers et al. 2000).However, this theory has often been contested and many studies have found no significant effects of N-fertilisation on severity of late blight infestation (Cicore et al. 2012;Jensen and Nielsen 2015).In the present study complex interactions of genotype and N-fertilisation affected the severity of the late blight infection.For the cultivars, a significant increase of disease severity with higher N-fertilisation was only observed in the susceptible control cultivar Adretta in one of the two years of the experiment.The resistant pre-breeding clones were hardly, or not at all, infested with late blight regardless of the nitrogen supply.In 2020 and 2021, respectively, 11 and 19 of the 20 clones tested were only infected by up to 10 % of leaf surface area with late blight at complete maturity.In contrast, the cultivars showed infestation at every nitrogen level, but an increase or decrease in infestation with the change in the fertiliser application rates depended on both the respective cultivar and, to a considerable extent, on the year of the trial.In general, the influence of the application rates of nitrogen was not decisive for the level of infestation with P. infestans and the sporadically significant results were likely influenced by additional factors not accounted by the experimental design.Further increases in the nitrogen application rates may have produced significant results, but higher N-fertilisation outside of the recommended rates for organic farming would lack agronomic significance.Since only four cultivars were tested together with a larger panel of pre-breeding clones the results were less representative for cultivars currently in use in organic farming, which are generally more susceptible to blight.In addition, it should be noted that although the study presented here investigated the effect of a nitrogen fertiliser approved for use in organic farming, it was recognised that in organic farming systems soil nitrogen levels may vary due to the use of different forms of nutrient inputs with varying mineralisation rates.It was also acknowledged that the trial plots in this study were not managed to full organic agriculture standards, due to the high loads of Colorado potato beetles.
Besides N fertilisation, other nutrients can also influence the infestation of P. infestans.It was shown that higher potassium applications led to a lower infestation (Kowalska and Drożdżyński 2018;Bista and Bhandari 2019).Dey and Chakraborty (2016) obtained similar results, but in combination with phosphorus application.These interactions could be investigated in further experiments together with the application of nitrogen.
Overall, it was concluded that N fertilisation rates only had a minor influence on the occurrence of late blight, and that it is much more relevant to grow resistant or moderately resistant cultivars such as Otolia.This is especially important in organic farming, because of the limited options for controlling the disease.For this purpose, the pre-breeding clones investigated here should be included in breeding programmes in the future.

Figure 1 .
Figure 1.Average development of P. infestans infestation of all nitrogen application rates of four potato cultivars and two prebreeding clones in 2020 (a) and 2021 (b).

Table 1 .
Current literature on the influence of nitrogen application rate on infestation with P. infestans in potatoes.Field experiments, b Glasshouse experiments, c Field and glasshouse experiments.

Table 2 .
p-values of ANOVA for all traits.