NC State
Oszust, K., and Frąc, M. (2020). "Apple pomace microbiome carrying fungal load against phytopathogens – Considerations regarding application in agriculture and horticulture," BioRes. 15(1), 945-966.


This experiment evaluated the taxonomic diversity of the fungal community in conventional (AP) and organic (OAP) apple pomace using high-throughput sequencing, applying fungal genetic barcodes to functional guilds. The most abundant taxonomic groups identified in both AP and OAP were the genera Aureobasidium, Cladosporium, and Alternaria, classified into the pathotroph-saprotroph-symbiotroph guild. The phenotype microarray provided insight into the role of the apple pomace fungal community in the ecosystem. It is theorized that adding apple pomace to the soil may improve the bioavailability of bioresource-based polyols. Evaluation of the antagonistic ability of the AP fungal community and Trichoderma atroviride G79/11 strain against pathogenic fungi was performed. Trichoderma G79/11 developed well on apple pomace and revealed the antagonistic mode against tested fungal plant pathogens. Therefore, it could be applied to soil as a formulation of AP with spores or AP with metaferm biopreparation.

Download PDF

Full Article

Apple Pomace Microbiome Carrying Fungal Load against Phytopathogens – Considerations Regarding Application in Agriculture and Horticulture

Karolina Oszust,* and Magdalena Frąc

This experiment evaluated the taxonomic diversity of the fungal community in conventional (AP) and organic (OAP) apple pomace using high-throughput sequencing, applying fungal genetic barcodes to functional guilds. The most abundant taxonomic groups identified in both AP and OAP were the genera Aureobasidium, Cladosporium, and Alternaria, classified into the pathotroph-saprotroph-symbiotroph guild. The phenotype microarray provided insight into the role of the apple pomace fungal community in the ecosystem. It is theorized that adding apple pomace to the soil may improve the bioavailability of bioresource-based polyols. Evaluation of the antagonistic ability of the AP fungal community and Trichoderma atroviride G79/11 strain against pathogenic fungi was performed. Trichoderma G79/11 developed well on apple pomace and revealed the antagonistic mode against tested fungal plant pathogens. Therefore, it could be applied to soil as a formulation of AP with spores or AP with metaferm biopreparation.

Keywords: Apple waste; Fungal diversity; Phytopathogens; Antagonism; Trichoderma atroviride; Agriculture

Contact information: Institute of Agrophysics, Polish Academy of Sciences, Doświadczalna 4, 20-290 Lublin, Poland; *Corresponding author:


A wide array of scientific activities and their resulting products, including biofertilizers, microbial pesticides, and bio-control agents, used to fight plant pathogens are present at the interface of applied microbiology and horticulture (Ray and Ward 2008). The use of advanced microbiological techniques and tools to characterize the genetic and functional diversity of the microbial community has also been recently proposed as a strategy in biowaste eco-toxicological evaluation (Oszust and Frąc 2018; Oszust et al. 2018a).

Of large-scale horticultural waste, the apple juice industry generates a large amount. The waste takes the form of apple pomace, the by-product that results from apple processing and consists of apple skin, seeds, and stems (Wang et al. 2019). Approximately 75% of apple fruit is processed for juice or cider, and the remaining 25% of the weight of the fresh fruit constitutes biowaste. Apples (Malus domestica) are the favoured fruit of millions of people, and are widely grown in the intemperate regions of the globe (Shalini and Gupta 2010). Poland is one of the major apple producers, with approximately 3170 Gg of apples produced every year (Lipiński et al. 2018).

Fruit pomace is a generally rich source of biological compounds and has become an important raw material from which to obtain various valuable by-products. Thus, it offers a logical basis for waste management and apple pomace is used for fuel and food purposes, biotransformation, a source of fibre and pectin, microcrystalline cellulose (Shalini and Gupta 2010), nanocellulose (Szymańska-Chargot et al. 2019), and other bioactive compounds, such as organic acids and flavonoids (Mourtzinos and Goula 2019). Apple pomace application for agricultural purposes could be especially important for local apple processing companies surrounded by orchards and crops, which often lack sources of exogenous organic matter. The EU Thematic Strategy on Soil Protection lists the decline in organic matter as one of the main threats to soil quality, and calls for cultivation and agricultural production systems that will lead to an increase in its content (Soil Quality and Policy 2018).

The agricultural use of fruit pomace as a natural fertilizer promoting plant growth in organic farming has been studied previously (Mercy et al. 2014). Pomegranate, orange, sweet lime, and banana pomaces have promoted the growth of plants and have helped achieve higher yields, due to the organic matter and nutrients introduced. The risks and opportunities of organic farming have recently been summarized (Röös et al. 2018). It should be emphasized that additional fungal species may follow in the wake of biofertilizer application and may be unincorporated into the soil with almost any kind of biofertilizer (Frac et al. 2014).

Various microbial inhabitants may be described as indigenous representatives of biowaste. They may undertake a pivotal ecological role in influencing plant health as symbionts or decomposers when they are introduced to fields in the form of biofertilizers (Oszust et al. 2018a). In contrast, they may be pathogenic to plants, produce toxins, or cause mycoses (Presterl et al. 2018). To date, the mycological compositions of biofertilizers have been poorly analysed and described compared with analysis of their physicochemical properties (Oszust et al. 2018b).

However, biofertilizers may influence soil fungal biodiversity after their agricultural application. Therefore, the aims of the present study are as follows: [i] to assess the potential threats and benefits of introducing fungal representatives into the soil with apple pomace as a fertilizer, attributing fungal genetic barcodes to functional guilds, and thus evaluating the taxonomic diversity using high-throughput sequencing; and [ii] to provide insight into the role of fungal community maintenance in apple pomace ecosystem functionality. Herein, the specific respiration rate demonstrates a metabolic effect, in addition to biomass presented as a ratio development in the phenotype microarray. The methods of Pinzari et al. (2016) were followed and applied to the fungal community for the first time.

The statement that apple pomace carries a fungal load may also be considered from another point of view: namely, that waste is also a carrier of microbial beneficial strains, which are intended to be incorporated into the soil to fulfil a positive function. In that case, the indigenous microbial fungal species of the organic waste may be introduced into plant cultivation. Accordingly, it was discovered that fruit pomace, which is a solid waste, may be used as a growth substrate for microorganisms that inhibit plant pathogen development (Kalidas 1999).

The Trichoderma genus, a well-known fungus, is able to parasitize a great number of other soil-borne fungi that are pathogenic to plants. Trichoderma sp. comprises numerous biopreparations. Among others, Trichoderma atroviride G79/11 was previously described as a strain with cellulolytic potential (Oszust et al. 2017b), and based on the strain culture metaferm, a multi-enzymatic biopreparation was developed (Oszust et al. 2017a). However, no analysis has been performed to determine its antagonistic ability.

It was assumed that Trichoderma atroviride G79/11, which revealed a relatively high cellulolytic activity, would also develop well on apple pomace and demonstrate an antagonistic activity against pathogens. Therefore, the third aim of this study is to [iii] evaluate the associated antagonism between apple pomace and the G79/11 strain against four fungal pathogens that have been known to devastate soft fruit plantations.

These pathogens were Colletotrichum sp., Botrytis sp., Verticillium sp., and Phytophthora sp. Colletotrichum sp. is a burdensome fungal pathogen in modern agriculture. C. acutatum is one of the most harmful species of this genus; it causes anthracnose in plants such as strawberries and raspberries (Dolan et al. 2018; Forcelini et al. 2018). Botrytis cinerea is a causal agent of grey mould, and its resistance to fungicide renders it one of the most harmful pathogens affecting raspberry and strawberry plants as it diminishes otherwise effective management strategies (Kozhar and Peever 2019; Weber and Hahn 2019). The Verticillium genus includes two major species, V. dahliae and V. alboatrum, that cause losses in agriculture. V. dahliae causes wilt on economically important crops including strawberries and raspberries (Fan et al. 2018). It is also theorized that the Phytophthora species is responsible for most strawberry and raspberry losses in all production areas of the world (Nellist 2018). Scientists distinguish between two varieties of this species: P. fragariae var. fragariae, which can only infect strawberries, and P. fragariae var. rubi, which is native to raspberries. The co-occurrence of those pathogens may appear, e.g.Phytophthora rubi and Verticillium dahliae in the form of late-summer disease symptoms in red raspberry fields (Weiland et al. 2017; Nellist et al. 2018). A universal natural enemy of those fungi is thus required in organic horticulture, especially because the worldwide area of organic cultivation has increased meaningfully over recent years (Kiełbasa 2015).



Apple pomace

Two types of apple waste were used for the experiments: apple pomace (AP), a waste consisting of a mixture of red-coloured apple varieties grown conventionally in the Grójec area of Poland, obtained from a local apple juice-processing factory; and organic apple pomace (OAP) of the Gala variety that was ecologically grown in Trentino-Alto Adige, Italy. The content of mineral ingredients (NPK) and the organic matter content were evaluated. The phosphorus (P) level was calorimetrically determined and the potassium (K) content was estimated by flame photometry according to the Spurway method (Spurway and Lawton 1949). The total nitrogen and organic matter contents were assessed according to the Kjeldahl and weight method, respectively. All analyses were determined at the District Chemical and Agricultural Station in Rzeszów, Poland. The results were obtained in the form of mean values. All chemical analyses of the tested waste were performed in triplicate.


Next generation sequencing – Meta-barcoding

An analysis of the fungal community structure of apple pomace was performed on the basis of the region of Internal Transcribed Spacers 1 (ITS1). The set of the following primers ITS1FI2: 5’-GAACCWGCGGARGGATCA-3’ (Schmidt et al. 2013) and 5.8S: 5’-CGCTGCGTTCTTCATCG-3’ (Vilgalys 1992) were used to amplify the selected region. The polymerase chain reaction (PCR) was completed in a Q5 Hot Start High-Fidelity 2X Master Mix (New England Biolabs, Ipswich, MA, USA) according to the conditions included in the manufacturer’s protocols. The DNA library was sequenced using an MiSeq platform (Illumina Inc., San Diego, CA, USA) with pair-end (PE) technology, 2 x 250 bp using the v2 Illumina kit following the manufacturers’ instructions (Genomed S.A., Warsaw, Poland). MiSeq Reporter (MSR) v2.6. software (Illumina Inc., San Diego, CA, USA) was used for a preliminary elaboration of the data and the Quantitative Insights into Microbial Ecology (QIIME) tool (Illumina Inc., San Diego, CA, USA) was used to process the raw sequencing data (Caporaso et al. 2010). The analysis included the following steps: reading quality evaluation, removing low quality sequences and chimeras, and generating operational taxonomic units (OTUs) that were defined by clustering at 97% similarity. The taxonomical classification of the OTUs was achieved using a Basic Local Alignment Search (BLAST) against the UNITE database.

A bioinformatics assay was based on the reference sequence database, Greengenes_13_05 (DeSantis, Jr. et al. 2006b), and was performed using an algorithm from Qiime software (Caporaso et al. 2010). The analysis included the following steps: demultiplexing of samples and adaptor cutting, conducting quality analysis, determining taxonomic composition, and performing diversity analysis. Sequences that were over 97% identical were grouped into one OTU using a distance-based OTU program. The application of MiSeq Reporter v2.3 (Illumina Inc., San Diego, CA, USA) allowed for species-level classifications. The taxonomy database for the metagenomics workflow was the Illumina version of the Greengenes database (DeSantis, Jr. et al. 2006a). FUNGuild was used to taxonomically parse fungal OTUs using the ecological guild (Nguyen et al. 2016).

Phenotype microarray

Phenotype fingerprinting of apple pomace (AP and OAP) fungal community was determined using the Biolog® System FF MicroPlates (Biolog®, Hayward, CA, USA), expressed as the ratio between the values of substrate use (respiration, OD 490 nm) and the growth pattern (biomass production, OD 750 nm); seven carbon source groups were evaluated (Pinzari et al. 2016, 2017). The protocol followed that of Jeszka-Skowron et al. (2018). The analyses were performed in tree technical replications (n=3). Results are presented as mean values, and standard errors are provided.

Microbial Strains and metaferm Biopreparation

The following fungal pathogens were used in the study: two strains of Colletotrichum sp. (G166/18 and G168/18) isolated from infected strawberry fruit, two strains of Botrytis sp. (G277/18 and G276/18), two strains of Verticillium sp. (G296/18 and G297/18) isolated from infected strawberry roots, and two strains of Phytophthora sp. (G373/18 and G368/18) from the collection of the Research Institute of Horticulture in Skierniewice (Sierniewice, Poland). Concerning the beneficial and possibly antagonistic strain of Trichoderma atroviride G79/11, its origin and characteristics were described by Oszust et al. (2017b). Metaferm is a multi-enzymatic biopreparation previously developed and described by Oszust et al. (2017a). It reveals hydrolytic activities such as those involving: xylanase, β-glucosidase, carboxymethyl cellulase, poligalactouronase, pectinesterase, amylase, lactase, and protease (Frąc et al. 2014; Oszust et al. 2017a). The metaferm used for the purposes of the experiment was stored at -20 °C for 4 years.

Antagonism Experiment

To evaluate the interactions between the antagonistic variants and the pathogens in the in vitro culture, the following antagonistic variants were used: (1) Trichoderma atroviride (G79/11) (5-mm fragments of potato dextrose agar (PDA) cultured for 10 days), (2) AP (0.2 g), (3) AP inoculated with Trichoderma atroviride (AP-G79/11) (0.2 g AP inoculated via an inoculation needle with spores of G79/11), (4) metaferm (MET) (100 µL into the 5-mm diameter hole in PDA), and (5) AP with metaferm (AP-MET) (0.2 g AP and 100 µL MET). They were tested against four selected pathogenic species (Colletotrichum sp., Botrytis sp., Verticillium sp., and Phytophthora sp.) represented by two isolates each.

The discs of 5-mm diameter PDA (with antibiotics addition) containing 10-day-old cultures of pathogenic fungi were placed on a Petri dish with 20 cm3 of PDA. The scheme of the antagonism experiment is shown in Fig. 1. Controls of all of the pathogenic strains without any treatment were provided. After incubating the pathogenic fungus and antagonism variant at 22 °C for approximately 5 days, the horizontal culture and vertical growth diameter (mm) were measured. The experiment was set up in triplicates (n=3). Results were presented as mean values with standard error provided.

Fig. 1. Antagonism experiment diagram (explanations: G79/11 – Trichoderma atroviride G79/11; AP – apple pomace; AP-G79/11 – apple pomace inoculated with Trichoderma atroviride G79/11; MET – metaferm multi-enzymatic biopreparation based on post-culture, concentrated liquid of Trichoderma atroviride G79/11 in soy flour-cellulose-lactose medium; AP-MET – apple pomace with the addition of metaferm; n = 3)


Apple pomace and OAP exhibited differences with regards to their chemical properties, as shown in Table 1. The latter contained twice as much total nitrogen and phosphorus (1.18% and 0.14%, respectively) as AP. Organic matter and potassium content were relatively equal in AP and OAP (0.8% and 98%, respectively).

Table 1. Chemical Properties of Apple Pomace

The study evaluated the share of fungal OTUs in apple pomace organized into groups that referred to trophic modes (Fig. 2). Among all of the obtained OTUs for AP, 20% were assigned and organized into trophic modes. For OAP, the total assignment was 26%. Therefore, 74% to 80% of the OTUs found were unassigned. In detail, the most numerous group of AP (constituting 18% of all OTUs) and of OAP (constituting 23% of all OTUs) was assigned to the pathotroph-saprotroph-symbiotroph group. The pathotroph-saprotroph group was the second organized entry detected, but much less numerous (1% in AP and 3% in OAP).

Fig. 2. The share of fungal OTUs entries organized into groups that referred to trophic modes in apple pomace

The relative abundance (%) of fungal OTUs ID entries have been qualified directly into taxa and indicated with trophic modes and guilds, as shown in Table 2. Both yeasts and moulds were found, including soft rot, endophyte pathogens, fungal parasites, likely opportunistic human pathogens, biotrophic, necrotrophic, or saprobic, on various plant tissues or/and necrotrophic on stems or leaves.

The most abundant taxonomic groups at the genus level that were identified in both AP and OAP were Aureobasidium sp. (2.75% and 12.7%, respectively), Cladosporium sp. (4.16% and 7.31%, respectively), and Alternaria sp. (0.93% in AP and 2.56% in OAP) were classified into pathotroph-saprotroph-symbiotroph group. Malassezia sp. representatives (0.1% in AP and 0.05% in OAP) and Leptosphaeria sp. were also found in both AP and OAP, though in a smaller amount (0.5% and 0.07%, respectively). The following genera: Phoma, Acremonium, Mycosphaerellaceae, and Exobasidium primarily occurred in AP. The genera Leptosphaeriaceae, Didymella, Sporobolomyces, and Rhodotorula were only found in OAP.

Among all the taxa described, only Malassezia sp. (< 0.1% in AP and OAP) and Sporobolomyces sp. (0.2% in OAP) were not thought to behave as plant pathogens. The animal pathogens in both AP and OAP were Cladosporium sp., Alternaria sp., and Malassezia sp.; in AP, Acremonium sp.; and, in OAP, Didymella sp. and Rhodotorula sp. Most of entries were classified as probable saprotrophs. Only two representatives were assigned to perform fungal parasitism, Acremonium sp. (0.15% AP) and Sporobolomyces sp. (0.2% OAP).

An evaluation of in vitro antagonism treatments on PDA against Colletotrichum sp. isolates is shown with photographic documentation in Fig. 3 (a to d), Botrytis sp. in Fig. 4 (a to d), Verticillium sp. in Fig. 5 (a to d), and Phytophthora sp. in Fig. 6 (a to d). Botrytis sp. and Colletotrichum sp. were the most expansive in this experiment. Botrytis sp. isolates outgrew the petri plate (diameter 80 mm), which meant that it was the most expansive sample among the pathogens tested. Colletotrichum sp. growth was slightly lower (60 mm). The slightest growth was noted for the Verticillium sp. (< 40 mm) and Phytophthora sp. isolates (< 12 mm).

Almost all proposed antagonism treatments were effective for pathogenic fungi growth inhibition, which was demonstrated by the decrease in horizontal and vertical growth diameter. Colletotrichum sp. isolates had an approximately 70% diameter decrease in G79/11 and AP-G79/11 strains, and 30% for AP, MET, and AP-MET. For Botrytis sp. G279/18, a 50% decrease was noted for AP, G79/11, and AP-G79/11, and 25% for MET and AP-MET. For Botrytis sp. G277/18, a 50% decrease was observed for G79/11, AP-G79/11, AP-MET, and a 25 to 30% decrease for AP and MET. In contrast, for Verticillium sp. G269/18 and G297/18, a 50% decrease for AP, and 70% decrease was noted for the rest of tested treatments. The only exception was Phytophthora sp. G373/18, which was not influenced by antagonists. The isolate Phytophthora sp. G368/18 was revealed to be immune.

The inhibition effect differed with respect to the particular treatment. Generally, AP treatment revealed the lowest inhibition activity. However, the direct addition of Trichoderma atroviride G79/11 to AP in the form of spores (AP-G79/11), or indirectly with metaferm (AP-MET), predominantly improved the efficacy of its antagonistic activity. This improvement was manifested by the overgrowth of G79/11. For Colletotrichum sp. G166/18 (Fig. 3c) and Verticillium sp. G269/18 (Fig. 5c), no G79/11 development was noted in the AP-MET treatment, and other fungi exceeded its growth. As shown in various photos (Figs. 3c, 3d, 4c, 4d, 5c, 5d, 6c, and 6d), for AP treatment, macroscopic evaluation showed that AP treatment favoured development of different species, demonstrating a probable antagonistic treatment.

Table 2. Relative Abundance (%) of Fungal OTUs with Trophic Modes and Guilds Identified

Fig. 3. Evaluation of in vitro antagonism treatments on the plate with potato dextrose lab (PDA) agar against Colletotrichum sp. (a) strain G166/18 and (b) strain G168/18, with photographic documentation (c) and (d), respectively. Explanations: C is pathogen growth control; antagonism treatments are as follows: AP, apple pomace; G79/11, Trichoderma atroviride G79/11; AP-G79/11, apple pomace inoculated with Trichoderma atroviride G79/11; MET, metaferm biopreparation; AP-MET, apple pomace with addition of metaferm; n = 3, the error bars represent standard error.