Introduction

Vegetables and fruits are a crucial part of the planetary health diet EAT-Lancet Commission [1]. This diet aims to simultaneously provide health to the population and the planet as required by the Food and Agriculture Organization (FAO) and World Health Organization [2]. In addition to vegetables and fruits, this diet is based on the predominant consumption of greens and whole grains, and reduced consumption of meat, fish, eggs, refined cereals, and tubers [3]. Vegetables and fruits contain various ingredients and bioactive plant-derived secondary metabolites which are considered to have beneficial health effects. Moreover, they harbor millions of microorganisms [4], which potentially serve as one of the main direct sources of environmental microbiota. The human gut microbiome is regarded as an internal environmental factor, while the impact of the external environment, including the food microbiota, is less well understood in the exposome concept. The exposome concept was first suggested by Wild [5] to encompass the totality of human environmental exposures from conception onwards, complementing the genome. The concept differentiates three categories of non-genetic exposures: internal, specific external, and general external [6]. Recently, intervention trials to demonstrate the importance of external, nature-based microbe exposures on the human microbiota and immune functions were reported [7, 8].

as well as factors that shape them are still scarce. During the last years, an accumulating amount of evidence has shown that fruits are colonized by distinct microbial communities. Studies have examined the impact of host genetics [9] and environmental influences, e.g., soil and climate [10, 11] on the composition of the fruit microbiota. Moreover, post-harvest treatments, such as washing, waxing, storage, and thermal treatment, were shown to strongly influence the composition of fruit microbiota [12,13,14]. Recently, the beneficial effects of fruit consumption on the gut microbiota and human health have been increasingly recognized [15, 16]. However, our knowledge of the microbial composition of fruits produced in different growing systems and geographic regions is still very limited.

Despite the importance of the fruit and vegetable associated microbiota, studies on fresh fruits and vegetables. Our objective was to understand the fruit microbiome in the context of the exposome concept. We expected that the fruit microbiome is an important external factor that influences the gut microbiome especially during the early life. First, however, it is important to understand the variability of the fruit microbiome between different growing systems. Therefore, we have selected apples and blueberries, which are among the most commonly consumed raw fruits in the world and, more importantly, are commonly eaten in early childhood. Apples can be grown in home gardens as well as in commercial orchards. Blueberries are grown in commercial farms and can be also found in the wild. These different growing systems make both fruits ideal models to further study the variability of the fruit microbiota that are commonly consumed. The growing systems are characterized by different management practices. While no chemicals, nor fertilization were used in fruits that were grown in the wild and within home gardens, typical horticultural systems are intensively treated. In this study, we attempted to address the following questions: (i) do fruits of natural origin have a different microbial diversity compared to horticulturally grown fruits; (ii) are there differences in the microbial composition between these two groups; (iii) are there differences between fruits from distinct geographical locations; and (iv) which taxa explain the differences between the two groups? Overall, this study provides important insights into the impacts of growing systems on the apple and blueberry microbiota.

Materials and Methods

Sampling Procedures and DNA Extraction

Apple (Malus domestica) and blueberry samples were collected between July and August 2020 at 29 locations (Austria — 20 locations; Finland — 9 locations, Supplementary Table S1) in Austria and Finland. We chose these countries to test if geographic distance had an effect of fruit microbiota compositions. Here, we have defined naturally grown fruits as those grown in the wild or in private gardens, away from commercial orchards, and have not undergone any post-harvest treatment. A total of 15 and 6 naturally grown apple and blueberry samples, respectively, were collected in Austria while a total of 5 and 4 naturally grown apple and blueberry samples, respectively, were collected in Finland. The naturally grown apple and blueberry samples were collected from ecologically isolated individuals. Ripe fruits were collected using sterile gloves and instruments. We randomly selected at least two apple fruits per sampling tree. For blueberry samples, we collected four composite samples (containing at least 10 berries) from four adjacent bushes/shrubs in one location. All fruits that represented horticultural production, were obtained from local supermarkets in Austria and Finland. We decided to obtain fruits from the local supermarket because it is the point that fruits are purchased and consumed. It should be noted that, in this study, most of the naturally grown blueberries belong to Vaccinium myrtillus whereas horticultural blueberries mostly belong to Vaccinium corymbosum (Supplementary Table S1). Sampling was carried out by using hand gloves and changing the hand gloves between handling various samples. All samples were put in sterile bags, kept in a cooling box during transportation, and stored at 4 °C before processing. Upon arrival in the laboratory, all samples were processed under sterile conditions. A total of 108 apple samples and 100 blueberry samples were analyzed. Details related to the samples and the associated metadata are presented in Supplementary Table S1.

To extract microorganisms from the fruits, approx. 10 g of each fruit sample was homogenized in a BagMixer laboratory blender (Interscience, Saint-Nom-la-Bretèche, France) with 10 ml sterile NaCl (0.85%) solution for 3 min. A total of 2 ml of homogenized suspensions was then centrifuged for 20 min at 16,000 g and pellets were used for DNA extraction. DNA extraction was carried out using the FastDNA SPIN Kit for soil and the FastPrep Instrument (MP Biomedicals, Santa Ana, CA, USA) according to the manufacturer’s protocol. DNA quality and yield were determined using the Nanodrop 2000 UV–Vis spectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA, USA) and then stored at − 20 °C for further PCR reactions.

Bacterial and Fungal Quantification Using Quantitative Real-Time PCR (qPCR)

By implementing a qPCR-based analysis, we first calculated microbial abundance in the fruit samples (copies maker genes/gram). The qPCR analysis was based on SYBR Green fluorescence using KAPA SYBR FAST qPCR Kit (Kapa Biosystems, Woburn, USA) using the primer pair 515f–806r [17] and ITS1f-ITS2r [18] for bacterial and fungal quantification, respectively. The qPCR reactions and standard preparations were conducted as described previously [19]. Fluorescence quantification was performed using the Rotor-Gene 6000 real-time rotary analyzer (Corbett Research, Sydney, Australia) with initial denaturation at 95 °C for 10 min, followed by 40 cycles of denaturation at 95 °C for 30 s, annealing at 54 °C for 30 s, and extension at 72 °C for 30 s and a final melting curve. The calculated PCR efficiencies were in a range of 94–98% (R2 = 0.955–0.965) for 515f–806r primers and 80–88% (R2 = 0.993–0.995) for ITS1f-ITS2r primers.

16S rRNA Gene Fragment and Internal Transcript Spacer (ITS) PCR Amplification and Illumina Sequencing

A one-step PCR approach using primers 515F/806R [17] and ITS1f-ITS2r [18] was employed for targeted amplification of the bacterial 16S rRNA V4 region and fungal ITS1 region. The primers contained Illumina indexes (barcode sequences) for multiplexing. Peptide nucleic acid (PNA) clamps were included in the PCR mix to block amplification of the plant’s plastid and mitochondrial DNA. To verify successful amplification, PCR products were visualized on 1% agarose gels and subsequently combined and purified using the Wizard® SV Gel and PCR Clean-Up kit (Promega). All barcoded amplicons were pooled in equimolar concentrations. The pooled samples were sequenced by the commercial sequencing provider Eurofins (Ebersberg, Germany) using the Illumina MiSeq platform (2 × 300 bp paired end reads). Amplicon sequences were deposited at the European Nucleotide Archive (ENA) under the project number PRJEB51939.

Bioinformatic and Statistical Analysis

Cutadapt was used to remove low quality reads, primer sequences and demultiplex the reads according to the assigned barcode [20]. The DADA2 algorithm [21] was executed in QIIME2 [22] to quality filter, denoise, and remove chimeric sequences. The resulting representative sequences, known as amplicon sequences variants (ASVs), were further classified using the vsearch algorithm against the SILVA v132 and UNITE v7.1 database [23,24,25].

Statistical analysis and graph rendering were conducted in R studio version 2021.09.0 [26] unless stated otherwise. Prior to statistical analysis, a normality test was performed using the Shapiro test. The data were not normally distributed; therefore, the non-parametric Kruskal–Wallis test was carried out to determine significant differences (P < 0.05) of bacterial gene copy numbers per gram of fruits countries and growing systems (commercial versus wild/home-grown). Groups were compared using Dunn’s test of multiple comparisons and the P values were adjusted using Benjamini–Hochberg procedure. ASV tables and taxonomic classifications that were generated with the DADA2 algorithm were used as an input for bacterial community analysis. The bacterial community analysis was performed using the software packages Phyloseq and MicrobiomeAnalyst [68,69]. To answer this question, targeted studies would be required.

Conclusion

In conclusion, growing systems were shown to substantially affect the variability of the fruit microbiome. Horticultural production results in a more homogenous fruit microbiome in comparison to naturally grown fruits (wild or home gardens). Moreover, specific changes in the composition of the microbiomes were observed that could have implications for human health. The microbiota associated with fruits and other fresh produce is considered a potential source and a key exposome for the gut microbiota. Hence, consuming naturally grown fruits could potentially expose our gut to diverse microbiota. Moreover, for future research, it is also important to consider the impact of management practices on the indigenous fruit microbiota, an element that is mostly overlooked.