Assessing the efficacy of biological control agents in plate assays and field applications to control Cladosporium on raspberry

Raspberries are a profitable and delicate fruit that are susceptible to a range of fungal pathogens. In recent years, UK growers have reported problems with Cladosporium on raspberries, potentially due to the movement of production from open fields to polythene tunnels. As Cladosporium is an understudied pathogen of raspberries, there is insufficient research on its epidemiology and management. In the present research, five commercially available biological control organisms were initially screened in dual-culture plate assays to determine if they could significantly reduce Cladosporium cladosporioides mycelial growth. All tested biological control organisms significantly inhibited Cladosporium mycelial growth in plate assays. The tested Trichoderma sp. was the most effective, with the rest having similar effects. In addition to the Trichoderma sp., two other Bacillus -based biological control products were selected for field studies: Bacillus subtilis strain QST 713 and B. amyloliquefaciens strain FZB24. All three biological control organisms led to significant reductions in the proportion of fruit with skin lesions caused by Cladosporium in the field, irrespective of whether they were used before (preventive) or after (curative) Cladosporium inoculation. Of the three biological control organisms, the Trichoderma sp. was the most effective in reducing Cladosporium development on raspberry (28% preventatively, 27% curatively). Further large-scale experiments closer to industry standard applications are essential before these biological control products can be used in commercial raspberry production.

Raspberries (Rubus idaeus) are a valuable horticultural crop that are susceptible to several fungal pathogens, including Botrytis, Rhizopus, Mucor and Cladosporium (Williamson et al., 1991). Cladosporium has been increasingly reported as a problem on raspberries in the UK and the USA (Farwell et al., 2023; Swett et al., 2019), causing both skin lesions pre-harvest (Farwell et al., 2023; Swett et al., 2019) as well as post-harvest fruit decay (O’Neill et al., 2012). Raspberries can be heavily infected with Cladosporium in some years, with one farm in Kent, UK having 55% of fruit infected post-harvest (O’Neill et al., 2012). Hence, Cladosporium can cause mass fruit rejection; however, most recommendations for its control are generalised from measures used against other fungal pathogens, due to a lack of research on the epidemiology and management of Cladosporium in raspberry.

Until recently, the primary method of controlling fungal pathogens was chemical fungicides. The timing of chemical control applications against Cladosporium on raspberry is difficult due to their short ripening time and the resulting risk of pesticide residue on the fruit. Strains of Cladosporium isolated from raspberries were also resistant to multiple fungicides, including caspofungin, fluconazole and flucytosine (Farian and Wójcik-Fatla, 2022), indicating an urgent need for developing alternative management methods. Furthermore, due to the concern of plant and human pathogen resistance to pesticides (Bastos et al., 2021; Deising et al., 2008; Hahn, 2014), there are needs to develop and implement alternative control methods against plant diseases. One such alternative is the use of Biological Control Agents (BCAs). Biological control agents are microorganisms that inhibit pathogen development using various mechanisms, for example, outcompeting pathogens for resources, releasing noxious metabolites, and hyper parasitism of pathogen hyphae (Köhl et al., 2019). These BCA products are primarily applied to prevent pathogen infections (Moragrega et al., 2021).

Cladosporium cladosporioides, the most prevalent Cladosporium species on raspberries collected across the UK (Farwell et al., 2023), prefers colonising raspberries via wounds (Swett et al., 2019). Cladosporium cladosporioides was also shown to mainly colonise the outer surface of grapes (Briceño and Latorre, 2008). As such, there is potential for other microorganisms to outcompete Cladosporium on the fruit surface, reducing the development of fruit skin lesions. Several BCAs were demonstrated to be effective at reducing Cladosporium spp. growth in in vitro conditions. For example, Chaetomium globosum inhibited the growth of C. cladosporioides in plate assays, with the amount of radial reduction ranging from 4.28 to 33.33 mm (El-Dawy et al., 2021). However, to the author’s knowledge there is limited published evidence to suggest that BCAs can control Cladosporium on crops under field conditions (previous studies have investigated Cladosporium fulvum, however this species has since been reclassified to Passalora fulva; Braun et al., 2003).

The objective of the present study was to assess the efficacy of commercially available BCAs for controlling C. cladosporioides in vitro and in field applications. Only commercial products were screened due to their potential to be used in commercial raspberry production in the future if they were proven to be effective against C. cladosporioides. Commercial products were initially tested under in vitro conditions; then based on the in vitro screening results, some products were taken to field experiments where they were applied either before (preventatively) or after (curatively) inoculation of raspberry fruit with C. cladosporioides inoculum.

Biological control agent products and storage

All BCA products were bought in their product formulations and stored as recommended on their label (Table 1), which involved either storing them in a fridge at 4 °C, or in ambient room conditions.

Dual-culture plate screening

The effects of five commercially available BCAs in the UK (Table 1) were tested in vitro for their ability to inhibit Cladosporium cladosporioides mycelial growth. In total, there were six treatments (five BCAs – Table 1, and an untreated control) with a minimum of seven replicates (plates) per treatment. Plates were first inoculated with C. cladosporioides, incubated for three days, and then commercial formulations of the BCAs were applied to plates (described further below). Once the plates were inoculated, ten blocks of plates were created by placing one plate of each treatment into a stack with the position of each treatment (plate) being randomly assigned. In the in vitro test, BCAs were applied in their product formulation at the maximum dose recommended on the label.

One isolate of C. cladosporioides (Isolate #65 as in Farwell et al., 2023) was grown in an incubator set to 25 °C in darkness on Potato Dextrose Agar (PDA) plates (Oxoid, Basingstoke, UK, CM0139) for one week. The plates were then flooded with 20 mL of a 0.5% KCl and 0.1% Tween-20 solution and rubbed with a sterile spreader to release conidia. The inoculum was then poured through sterile muslin into a sterile Universal bottle, to remove large mycelial clumps. Spore concentration was counted with a haemocytometer and adjusted to 1 x 10 7 spores/mL. This was subsequently used to inoculate dual culture plates.

For the dual-culture screening (Raymaekers et al., 2020), PDA media was used for bacterial BCAs and 3% raspberry agar was used for fungal BCAs. To make 100 mL of the 3% raspberry agar, 3 g of freeze-dried raspberry powder was added to 47 mL of water in one Duran bottle, then 2 g of technical agar (Millipore, Livingston, UK, 05039) was added in a separate Duran bottle containing 48 mL of water. Duran bottles were autoclaved separately (121 °C for 15 mins), then the liquids were combined into one Duran bottle, and finally poured into 90 mm Petri plates.

For BCA treatments, 20 µL of the 1 x 10 7 spores/mL C. cladosporioides conidial suspension was placed onto one of the two pre-marked points on the plate (two points with a gap of 30 mm in the centre of the plate), or onto the centre of the plate for the control treatment. The plates were then left inside a laminar flow hood for approximately 20 mins until the liquid had evaporated. The plates were then placed into an incubator in darkness at 25 °C for three days to allow germination and growth of the C. cladosporioides. Then, a 20 µl droplet of the designated commercial formulation of the BCA was added to the other point (30 mm away from the C. cladosporioides colony). The plates were left for 20 mins in a laminar flow cabinet to dry as before, then placed back into the incubator in their pre-designated position within each stack.

The diameters of C. cladosporioides and BCA colonies in each plate were recorded nine days post-BCA application along the inoculation axis (the line where the BCA and Cladosporium were inoculated; this line can be seen in Supplementary Figures S3, S4). The raw values of C. cladosporioides diameter growth were then used in statistical analyses.

Field applications of biological control agents

Two field inoculations were conducted in 2022 and 2023 to examine the effects of preventative and curative treatments of three BCAs (selected on the basis of in vitro screening) under field conditions: Trichoderma sp., B. subtilis strain QST 713 and B. amyloliquefaciens strain FZB24, with two control treatments (an uninoculated and untreated as a negative control; and a Cladosporium inoculated but untreated as a positive control). Experiments were performed on an experimental plot under polythene at Niab, East Malling (Latitude: 51.288350, Longitude: 0.454766). The raspberry plants used were propagated by a commercial nursery and supplied as established floricane plants with 3-canes per pot and approximately one year of age at the time of experimenting, and were of a popular UK proprietary variety. New plants were used across the two years of experiments to standardise their ages, and from the same commercial supplier across both years. The pest Drosophila suzukii was managed using industry-standard pesticide sprays, but no fungicides were applied, and plants were subjected to industry standard fertigation.

In 2022, a total of 60 plants (in a single row) were arranged into six blocks of ten plants, with five plants randomly assigned to the preventative and five plants to curative BCA applications (Supplementary Figure 1). In 2023, a total of 80 plants (in a single row) were arranged into eight blocks of ten plants, with five plants in each block randomly assigned to the preventative and five plants to curative BCA applications. To prevent spray drift, each preventative and curative plot had a 0.5 m section of canopy gap left as a buffer zone. Then, between pairs of plants receiving different BCA treatments, a 1 m gap was present as a buffer. This also included the use of spray curtains during inoculations to limit spray drift. Inoculation and BCA treatments were only performed on one side of the plants (the side with heaviest fruit set) to limit the number of fruit used in experiments.

The experiment consisted of: (1) on day one, curative treatments were inoculated with a mixed C. cladosporioides inoculum; (2) on day two, applications of the three BCAs to all plants allocated to both curative (inoculated the day before) and preventative treatments (inoculated the day after); (3) on day three, plants allocated to preventative treatments (those treated with BCAs the day before) were inoculated with a mixed C. cladosporioides inoculum, and finally (4) on day four, all experimental fruits (from preventative and curative experiments) were picked for incubation and assessment. BCA inoculation of fruits for preventative and curative treatments was carried out on 07/09/2022 and 04/08/2023. Experiments were timed to target inoculation and BCA applications to under-ripe fruit, so that when all fruit were picked, they were considered ripe to an industry standard. To prevent fruit being over-ripe when collecting, fully ripe fruit were removed at both day one and day two during inoculations and BCA applications. To ensure enough ripe fruit were available, experiments were targeted at the first initial peak of raspberry production.

A 1 x 10 7 spores/mL mixed C. cladosporioides conidial suspension was prepared as described previously, but with three isolates of C. cladosporioides (Isolate #73, #78 and #80 as in Farwell et al., 2023) adjusted to 1 x 10 7 spores/mL separately, then combined in equal parts. Inoculum was applied to each individual ripening fruit (just turned pink) with a fine misting hand sprayer. Biological control agents were applied using handheld pressurised sprayers (one sprayer per treatment) at the maximum dosage as described on the product label (Table 1). Each plot consisted of a pair of plants, one preventatively and one curatively inoculated with the same BCA treatment. Each individual plant received 75 mL of BCA spray on the inoculated side, equivalent to a spray volume of 1000 L/ha. Spray screens were placed between neighbouring plants to prevent drift; the curtains were thoroughly rinsed with sterile water and 70% ethanol between applications of different BCA products.

Fruits from all treatments were removed by cutting the petiole with a pair of sterile scissors and placed into trays (one tray for each plot) with at least 2 cm between each fruit. Individual trays were then enclosed and sealed in a polythene bag to increase humidity, before being incubated at ambient room temperature (average temperature 23 °C) and 24 hours of light for four days. After four days of incubation, fruit were assessed for the presence or absence of Cladosporium skin lesions. Four days was previously found to be the optimal time to characterise fungal infections of raspberries, as symptoms of disease were well developed, but fruit had not deteriorated completely (Farwell et al., 2023). Visual identification of raspberry fungal diseases was conducted using a range of written descriptions and images from various sources (Martin et al., 2017; Swett et al., 2019; Bensch et al., 2012). In the present study, no re-isolation or molecular identification of fungal diseases were conducted, hence some fungi may have been falsely identified. However, previous studies investigating Cladosporium skin lesions did confirm positive visual identification with multi-locus phylogenetics (Farwell et al., 2023), and the morphological characteristics of Cladosporium skin lesions from this study were used to help identify Cladosporium in the present study. Cladosporium skin lesions were identified as appearing as dark olive-green and velvety mycelial growth upon the surface of the raspberries (example images of Cladosporium fruit rot are provided in Supplementary Figure 2 and in Swett et al., 2019). Other fungal rots were also noted when assessing the fruit, including Botrytis which was identified by “cottony and whitish to grey” mycelia (as described by Martin et al., 2017), Penicillium which appears sooty but is typically pale-blue with a white edge and Rhizopus/Mucor which have long sporangiophores where dark sporangiospores are seen at the tips.

Statistical analyses

Analyses were performed with R (v. 4.1.2). The raw diameter of C. cladosporioides from the dual-culture plate assays were analysed with a one-way ANOVA, separately for each medium type. A qq-plot and Shapiro test were used to confirm if errors of the model were normally distributed, and a Levene test to confirm homogeneity of variance. Post-hoc t-tests were performed using the R package emmeans v. 1.8.8 (Lenth, 2023), with p-values adjusted using the Tukey method. For field experiments, the preventative and curative applications of BCAs were analysed separately. To calculate the proportion of infected fruit, the total number of fruit infected with Cladosporium was divided by the total number of fruit for each plot, which was then used in the analysis. The proportion of infected fruit was analysed with two separate models for preventative and curative BCA applications using the R package glmmTMB (Brooks et al., 2017) with a binomial mixed model. In the binomial mixed models, BCA treatment was a fixed effect factor, while the date of BCA application and block were set as random effect factors. Due to overdispersion in the curative model, an additional observation-level random effect was added to the model. To determine if there was a significant effect of the BCAs, a Chi-square analysis of deviance was implemented. To test differences between treatments, emmeans (v. 1.8.8.) was used, which were adjusted using Sidak (Šidák, 1967).

Biological control agent in vitro dual-culture plate screens against Cladosporium cladosporioides

There were significant differences in the growth of C. cladosporioides between BCA treatments on raspberry media (F (2) = 2361.2, p < 0.001) and PDA (F (3) = 507.1, p < 0.001). For tests on raspberry media, A. pullalans (t (18) = 25.6, p < 0.001) and the Trichoderma sp. (t (18) = 68, p < 0.001) significantly reduced C. cladosporioides mycelial growth (Figure 1; Supplementary Figure 3), and the Trichoderma sp. led to a greater reduction in the Cladosporium growth than A. pullalans (t (18) = 42.4, p < 0.001; Figure 1). On the PDA media, B. amyloliquefaciens strain D747 (t (23) = 32.9, p < 0.001), B. amyloliquefaciens strain FZB24 (t (23) = 30.7, p < 0.001) and B. subtilis (t (23) = 30.7, p < 0.001) all significantly reduced C. cladosporioides mycelial growth compared to the control (Figure 1; Supplementary Figure 4).

Biological control agent field applications against Cladosporium cladosporioides

The total number of fruits inoculated each year for preventative treatments are shown in Supplementary Table 1, and for curative treatments in Supplementary Table 2. In both years, BCAs were applied when temperatures were around 25 °C on overcast days (Figures 2, 3) with low wind speeds.

Preventative applications (i.e., 24 h before Cladosporium inoculation) of BCAs led to significant reductions in the proportion of infected fruit (χ2(4) = 159.86, p < 0.001). For the positive control treatment, inoculation with C. cladosporioides led to a 26% increase in the incidence of infected fruits (z = 9.95, p < 0.001; Figure 4). Bacillus amyloliquefaciens strain FZB24 reduced the incidence of infected fruit by 19% (z = -7.24, p < 0.001), B. subtilis by 15% (z = -5.31, p < 0.001) and the Trichoderma sp. by 28% (z = -10.82, p < 0.001; Figure 4) compared with the positive control.

Curative applications (i.e., 24 h after Cladosporium inoculation) of BCAs reduced the proportion of fruit infected with Cladosporium (χ2(4) = 119.33, p < 0.001). For the positive control, inoculations led to a 24% increase in the incidence of infected fruit (z = 8.44, p < 0.001; Figure 5). Bacillus amyloliquefaciens strain FZB24 reduced the incidence of Cladosporium by 13% (z = -4.28, p < 0.001), B. subtilis by 14% (z = -4.74, p < 0.001), and the Trichoderma sp. by 27% (z = -9.70, p < 0.001; Figure 5) compared with the positive control.

In plate assays, all BCAs tested significantly reduced C. cladosporioides mycelial growth, with the Trichoderma sp. being the most effective. Biological control agents applied in the field were also effective at reducing the incidence of fruit with Cladosporium skin lesions; the Trichoderma sp. once again was the most effective in both preventative and curative applications. Bacillus amyloliquefaciens strain FZB24 and B. subtilis had similar efficacy in both curative and preventative treatments in the field.

In vitro testing showed that all five BCAs tested were effective at reducing mycelial growth of a single C. cladosporioides strain. Further testing on more strains of C. cladosporioides may be needed to determine if there is intra-species variation of resistance to BCAs. While no literature currently exists discussing variation in Cladosporium species susceptibility to BCAs (to the authors knowledge), Botrytis cinerea was documented to show variation across strains to pyrrolnitrin, an anti-biotic produced by some BCA species (Ajouz et al., 2011). Furthermore, the efficacy of the BCA Pseudomonas helmanticensis CT22 varied depending on isolates of B. cinerea in detached-stem tomato assays (Bouaoud et al., 2018). Therefore, the importance of screening a representative population instead of single isolates can be argued for when searching for new BCAs with efficacy against pathogens of interest, and will be needed for further evidence of the efficacy of the BCAs tested in this study on C. cladosporioides isolates affecting raspberries. It is further important to note that in dual-culture assays the commercial formulation of the BCA products were used, instead of isolating and culturing the microbial agent from the products. This may have affected the accuracy of inhibition due to the strain identity and purity not being confirmed and the potential for other compounds present in the product to inhibit Cladosporium growth, besides the BCA microbe. Therefore, future efforts into screening BCAs against Cladosporium would benefit from independent verification of the BCAs before screening.

The incubation temperature for BCA dual plate assays was maintained at an optimum for C. cladosporioides growth (Briceño and Latorre, 2008; Solairaj et al., 2022), whereas field conditions can vary greatly both diurnally and throughout the growing season. The inoculated control fruit in the curative treatments in 2022 had a higher incidence of Cladosporium lesions than in 2023, perhaps due to extreme weather conditions (temperatures in excess of 35 °C, Figure 2) earlier in the 2022 season. These temperatures may have negatively impacted plant health, making plants more susceptible to infection. There may have also been a higher disease pressure in the 2022 season due to decreased competition from the death of microbes from the extreme temperatures. Future screening of BCAs should be conducted across the entire fruiting season, and therefore, under a range of environmental conditions. The present study only investigates existing commercial BCA products; future research into microbes that exist on the raspberry phylloplane and endophytes may yield BCAs adapted to surviving on raspberries, with further improved efficacy against raspberry pathogens. Epiphytes of raspberry fruits may be of particular interest, if they show improved survival and efficacy on the delicate surface of the fruit.

Applications of BCAs were effective in the present study when applied both preventatively and curatively. Out of all the BCAs tested, the most effective product was the Trichoderma sp., which reduced the Cladosporium incidence to a similar level as the un-inoculated control. This corroborates with previous findings where multiple Trichoderma spp. were effective at controlling a range of Cladosporium spp. in plate assays (Barbosa et al., 2001; Rolim et al., 2019). While Bacillus and Trichoderma spp. are known to have several biocontrol mechanisms such as volatiles and directly competing for resources with pathogens, Trichoderma spp. are also capable of mycoparasitism, whereby the hyphae release digesting enzymes as they coil around their host (Asad, 2022). This mode of action may have explained why the Trichoderma sp. was more effective at reducing the incidence of fruit skin lesions than the Bacillus spp. Trichoderma spp. can also prime plant defences for resistance to pathogens and abiotic stressors (Pacheco-Trejo et al., 2022). The Bacillus spp. tested have similar modes of action, being able to produce antifungal metabolites (Siahmoshteh et al., 2018) and promote expression of plant defence genes (Liu et al., 2019). However, the specific mechanisms underlying the inhibition of Cladosporium growth by each BCA were beyond the scope of the present study. This gap in research is needed to optimise the selection of BCAs within a wider disease management strategy; for example by informing BCA application timing in conjunction with chemical fungicides to maximise BCA efficacy.

It should be noted that BCAs were applied at the maximum dose in the present study, which may not be cost-effective for growers; further work on the doses of BCAs that can retain their efficacy is therefore needed. Furthermore, the extent of pathogen-BCA contact achieved in the present artificial control study is unlikely to be achieved in commercial spray applications. Therefore, while the current study shows some promising results, applications of BCAs need to be repeated under commercial conditions to show efficacy can be achieved reliably. A better understanding of the pathogen itself will also be required to control the pathogen in commercial settings.

Building upon these findings, it is important to note that Cladosporium is ubiquitous (Harvey, 1967; Farwell et al., 2024), and using an inoculum threshold to predict disease risk and to inform when to apply BCAs may not be practical. Further research into the conditions conducive for Cladosporium development is required to develop disease risk models. This is particularly important for curative applications where models are vital to determine the risk of an infection and make applications cost-effective for growers. It may also be warranted to investigate the efficacy of BCAs applied curatively for > 24 hours after pathogen inoculation. In commercial fruit production, BCAs are likely to be incorporated into a larger spray programme and work is needed to see how best this should be done with consideration of timing, product selection, and climatic conditions. In addition, we need to equally understand their survival on plant tissue surfaces following their application under natural conditions (Tut et al., 2023). This may determine their application interval, which is particularly important for a preventive control strategy. While this study has demonstrated the clear potential BCAs have to manage Cladosporium on raspberries, more work is vital to make a holistic management strategy for this pathogen.

One such strategy that may compliment BCAs well is post-harvest cool-chain management; this technique maximises the time fruit remains cooled (often to 0 – 2 °C for soft fruit) from the initial picking to being bought by the consumer, to decrease a pathogen’s ability to germinate and grow during transport. This method effectively delayed fruit Botrytis rot development in raspberry (O’Neill et al., 2012) and strawberry (Berrie et al., 2022). Thus, it may be worthwhile investigating the effectiveness of combining pre-harvest applications of BCAs with post-harvest cool-chain management in managing Cladosporium skin lesions post-harvest to provide an alternative approach to Cladosporium infections on raspberries.