Due to its physical properties and chemical composition, the soil is a suitable environment for the development of microorganisms, especially bacteria and fungi. Their biological activity affects, among other things, the fertility of the soil, as well as the availability to plants of nutrients derived from the decomposition of the biomass by microorganisms (Galus-Barchan and Paśmionka 2014; Pociejowska et al. 2014). Depending on environmental factors, the quantitative and qualitative composition of microorganisms may vary over time. Differences in bacterial counts may also result from their populations’ diversity, sometimes even within one site (Clark 1971; Steinauer et al. 2016).
The layer of soil most populated by microorganisms is the rhizosphere. This layer contains several to several hundred times more bacteria than the soil zone outside the rhizosphere. Foster (1988) states that in one gram of dry matter within the rhizosphere, there are 1010–1012 bacterial cells, and outside it, no more than 108 cells per gram of soil. In the rhizosphere, there are microorganisms with a different qualitative composition than in soil without plants. The relationships between microbes and plants are complex. Bacteria and fungi living in the rhizosphere are dependent on hosts (plant roots), nutrients, other microbial populations and climatic conditions that stimulate or inhibit their diversity (Bardgett 2011; Lau and Lennon 2011; Eisenhauer et al. 2017).
A diversified microbiome can be observed in the roots of plants. The plant’s condition is strongly influenced by the microbial community, which in turn is influenced by the host (Beckers et al. 2017; Proenca et al. 2017). This microbiome is made up of microorganisms that have a beneficial (plant growth-promoting rhizobacteria, PGPR), neutral (neutral rhizobacteria) or deleterious (deleterious rhizosphere microorganisms, DRMO) effect on plant development (Kurek and Kobus 1990). Symbiosis is one example of the beneficial influence of microorganisms on plant development.
It is well known that through symbiosis, organisms acquire capabilities and properties that they would not otherwise show (Frey-Klett et al. 2011). The Poole study (2001) shows that the rhizosphere layer is dominated by bacterial species belonging primarily to the genera
For many years, the direct influence of rhizospheric microorganisms on the stimulation of plant growth was associated with their ability to fix nitrogen. Rhizospheric bacteria have received particular attention because of their potential use in stimulating the colonization of mycorrhizal roots and, consequently, the industrial production of fungi, including truffles (Boersma et al. 2009, 2010; Antony-Babu et al. 2014). Free-living bacteria (PGPR) have a beneficial effect on plants by stimulating their growth directly and indirectly. Bacterial communities isolated from root secretions have shown an effect on stimulating the development of mycorrhizal mycelia (Ali and Jackson 1988). Some rhizospheric microorganisms can also induce changes in the quantity and composition of root secretions (Deveau et al. 2007). Direct support consists of stimulating mineral nutrition, for example, by facilitating the uptake of nitrogen, dissolving phosphorus compounds or binding iron by siderophores. This also includes the synthesis of phytohormones that affect plant development, that is, auxins, gibberellins and cytokinins, as well as lowering ethylene levels that affect plant rooting. Indirect support, on the other hand, is associated with bacterial-based biological control of phytopathogens. Bacteria can exhibit various symbiotic effects against fungi and also inhibit pathogen development through antagonism (Tsukamoto et al. 2002; Frey-Klett et al. 2007), improve spore distribution (Citterio et al. 1995; Splivallo et al. 2015), and provide vitamins and growth regulators (Rangel-Castro et al. 2002; Riedlinger et al. 2006).
Soil is inhabited not only by bacteria. It is also the natural habitat for fungi, including the genus
Research in many countries shows that the role of bacteria and fungi in the functioning of forest soils is not yet fully understood. This is especially true for the evaluation of bacterial diversity and their relationship with fungi (Baldrian et al. 2012), particularly regarding the summer truffle, in which communities of the coexisting bacteria were usually identified by the
This study aimed to elucidate the bacterial communities associated with the ascomata of the summer truffle
It has been assumed that the ‘truffle’ sites will exhibit a specific composition of the bacterial microbiome, which may help monitor soils conducive to truffle development and yield.
The research was conducted on three study sites within the Nida Basin, described as G, M and W (the abbreviations G, M, W come from area/location names). The research areas are located in forests (locations), which are at an altitude of 250 to 296 m above sea level on Rendsina soils. In each location, six research areas were designated: three in which summer truffle ascomata (variants T) were found (G1–3, M1–3, W1–3) and three control areas, where no summer truffle ascomata (variants C) were found (G4–6, M4–6 W4–6) (Siebyla and Hilszczańska 2020). For the analysis of bacterial communities, samples of fine roots with a diameter of ~1 mm with ectomycorrhiza visible to the naked eye were taken from a depth of 10–15 cm in spring and autumn of 2017 and 2018. A total of 72 samples were collected, 18 in spring and 18 in autumn each year. Root samples were taken from research areas where summer truffle ascomata had been recorded in the previous years (Rosa-Gruszecka et al. 2014; Hilszczańska 2016; Hilszczańska et al. 2019).
The occurrence of the dominant tree species in individual samples taken from all the areas, as well as information about the parent rock, are presented in Siebyla and Hilszczańska (2020). The root samples included:
Analyses of the diversity of bacterial communities were carried out using classical and molecular methods. Bacteria were cultured in pure cultures on agar medium imitating the natural environment, followed by the identification of strains according to the methods presented below.
Quantitative analysis (measuring the bacterial count) performed by culture methods began with sterilization of the root samples in 70% ethyl alcohol for 3 min, then in 2.5% sodium hypochlorite (NaOCl) for 5 min, then again in 70% ethyl alcohol for 30 seconds. After sterilization, the roots were rinsed 5 times with distilled water and dried (Sun et al. 2008; Kubiak et al. 2017). The roots were then cut in sterile conditions into 0.5 × 0.5 cm × 0.5 cm fragments. Seven fragments were laid out on each plate with a culture medium (with nutrient broth) (Gotkowska-Płachta et al. 2008). Two replicates were made for each soil sample. The control for the study was water kept from the last root rinsing. The cultures were incubated for 5 days at room temperature. The number of new colonies was recorded every day and after the determined number of days, the total number of colonies was recorded (Colony Forming Units – CFU).
In order to perform qualitative analyses, material was used from a single bacterial colony, obtained by means of streaking and multiplied in a liquid medium of nutrient broth. The DNA extraction was carried out according to the instructions provided with the bacterial isolation kit: Bacterial Genomic Miniprep Kit Sigma Aldrich (Merc Germany) and the isolated product was then amplified using the Polymerase Chain Reaction (PCR) method. The PCR reaction was performed according to the following protocol: pre-denaturation at 94°C for 4 min; amplification – 30 cycles, denaturation at 94°C for 60 s, annealing of primers at 62°C for 40 s, the extension of the primer at 72°C for 2 min and elongation at 72°C for 6 min. Amplifications were performed in 10 μl with 1 μl DNA, 0.2 U/μl Taq polymerase (Qiagen), 1 μl 10X buffer PCR (Qiagen), 1.5 mm Mg (25 mm) (Qiagen), 0.1 mm dNTP (5 mm) (Qiagen), 0.1 μl each primer (10 μm), 5 μl 25 × buffer Q (Qiagen). The universal bacterial domain primers were used for the PCR reaction: 530f (5’GTG CCA GCM GCC GCG G’3) and 1100R (5’GGGTTGCGCTCGTTG’3) (Lane 1991; Gryndler et al. 2013).
The products were cleaned using a Clean-up kit (from A&A Biotechnology). The PCR products were sequenced using the Sanger method (Tedersoo et al. 2010). The analysis was carried out by Genomed S.A., Warsaw, Poland. Finch TV software was used to analyse the sequencing products. The obtained sequences were compared with the NCBI Gene Bank database using Blast (
Temperature and hydrological data for the nearest measuring stations Kielce and Cracow were recorded on the basis of monthly Bulletins of the State Hydrological and Meteorological Service of the Institute of Meteorology and Water Management (IMGW-PIB). Using these values, Sielianinov’s hydrothermal coefficient
In 2018, DNA was isolated directly from the roots of the tree species tested, both in Tuber (T) and control (C) variants, and used for NGS sequencing and metagenome analysis based on the 16S RNA fragment (bacterial composition). Amplicon sequencing included the V3–V4 fragment of the 16S rRNA gene enabling the analysis of taxonomic groups of bacteria. Amplicons were prepared using samples of isolated DNA obtained by PCR and 16S libraries. Sequencing was performed using Illumina’s MiSeq in two reads of 300 base pairs. The expected mean number of reads per sample was 160,000–180,000. The bioinformatic analysis included filtering of the reads and analysis of the sample composition for each taxonomic category based on database homologues (Medinger et al. 2010; Staley et al. 2013).
Indicators of genetic and biological diversity were calculated on the basis of: a) alpha biodiversity of samples representing a given community (applied to individual communities within a specific delimited area, which was determined by a collective list of species occurring within a given geographical unit) and b) beta diversity determining species diversity when comparing communities (Kim et al. 2017).
Alpha biodiversity was determined on the basis of the diversity of identified operational taxonomic unit (OTU) sequences within a biological sample. The alpha biodiversity indices were determined by the biodiversity indices, that is,, Chao’s index, Faith’s phylogenetic diversity, observed OTUs, number of distinct features, Heip’s evenness measure, and Shannon’s and Simpson’s indices. The alpha biodiversity indices were calculated using Qiime2 software with the implemented alpha diversity, alpha-phylogenetic diversity, and alpha group significance diversity programs. All the samples, divided into individual groups as described above, were used for analysis. A detailed description of the indices can be found in Siebyła and Hilszczańska (2020).
Beta biodiversity determines the diversity of the identified OTU sequences within a given group of biological samples. Beta biodiversity was defined by biodiversity indicators, that is, unweighted and weighted UniFrac, Bray–Curtis dissimilarity (a measure of similarity between samples, referred to as semimetric), and Jaccard index. A detailed description of the indices can be found in Siebyła and Hilszczańska (2020).
All the tested groups of bacteria, depending on the biodiversity indicator used, were assessed with the Kruskal-Wallis non-parametric test, with the null hypothesis assuming an equal number of species in the group. The UniFrac measures used to determine beta biodiversity were calculated using the Qiime2 software with its implemented diversity core-metrics-phylogenetic program (Lozupone and Knight, 2005). As a result of the analysis, PcoA plots were prepared, illustrating the distances (differences in sequences) between individual groups of samples. The PcoA plots were prepared using R software (Team 2013). Analyses of the beta biodiversity results was performed using Qiime2 software with the implemented diversity beta-group-significance program. The PERMANOVA test, which determines the differences in distances between groups, was also used in the analysis.
The analysis of the correlation of alpha and beta biodiversity results was performed using Qiime2 software with the implemented diversity alpha-group-significance and diversity beta-group-significance programs. The plots for the alpha correlations were made using R software, preceded by the evaluation of the normality of the distribution using the Shapiro-Wilk test. Spear-man’s rank correlation coefficient (non-parametric test, for data without normal distribution) and Pearson correlation coefficient (parametric test for data with normal distribution) were used in the analyses. All the alpha and beta biodiversity indicators were calculated for 4 factors: spring 2017, spring 2018, autumn 2017 and autumn 2018.
The root colonization by bacteria assessed based on the CR coefficient expressed as a percentage was more varied in 2017 than in 2018. In 2018, regardless of the season, the root colonization coefficient value was ~1.0 for samples from all three areas, while in 2017, the root colonization coefficient ranged from 0.5 to 1.1. In spring, the lowest value of the root colonization coefficient was recorded on the surface of G6 (0.5%) and the highest on the surface of M2 (1.1%). In the case of the assessment in autumn 2017, the lowest value of the coefficient was characterized by the surface of W3 (0.5%) and the highest, equal to 1, by surfaces M2, M4, M5, M6 and W4.
In area M in autumn 2017, the degree (%) of root colonization (CR) by bacteria was similar to that of 2018. A similar situation was recorded in area W in autumn 2017, except for sample W3, where the colonization coefficient was ~ 0.5. An inverse situation was recorded in area G (autumn 2017), where the colonization coefficient was 0.65–1.0 in the truffle variant (T) and 0.65–0.8 in the control variant (C). The average value of the root colonization coefficient in spring 2017 was similar regardless of the location (Fig. 1). The root colonization coefficient (CR) in area G, regardless of the season, did not exceed 0.8% on average. The exceptions were the samples from G1, G2, where the coefficient was equal to ~ 1.0. The lowest value of the root colonization coefficient was recorded in spring in the truffle variant (T) for sample G3 with a value of ~ 0.2 (Fig. 1).
The uniform degree of root colonization in 2018 could have been affected by meteorological conditions, regardless of the season and location (Tab. 1). The value of hydrothermal coefficient
Average values of hydrothermal coefficient
Station name | K | Total precipitation by month | Years | |
---|---|---|---|---|
IV–X | I–XII | |||
Kielce | 1.44 | 378.6 | 619.5 | 2016 |
Cracow | 1.95 | 553.6 | 745.3 | |
Kielce | 2.27 | 548.9 | 731.8 | 2017 |
Cracow | 2.11 | 545.9 | 702.3 | |
Kielce | 0.95 | 340.1 | 486.7 | 2018 |
Cracow | 1.25 | 459.0 | 568.7 |
Of the 189 bacterial isolates from a single bacterial colony sampled between 2017 and 2018 in all the assessment variants, 14.3% of bacteria were no longer bred, while the remaining 85.7% were used for further molecular analyses. As a result of Sanger DNA sequencing, seven genera of bacteria were distinguished in the samples collected from areas G, M, W.
The most numerous bacteria recorded were those of the genus
Percentage of bacteria of the dominant genera by season (S – spring; A – autumn) of the year (2017, 2018)
Genus | 2017 | 2018 | ||
---|---|---|---|---|
S | A | S | A | |
6.8 | 6.8 | 3.4 | 11.2 | |
9.6 | 0.0 | 3.4 | 3.4 | |
20.5 | 32.9 | 2.2 | 14.6 | |
1.4 | 1.4 | 30.3 | 0.0 | |
4.1 | 4.1 | 22.5 | 0.0 | |
Other | 6.8 | 5.5 | 6.7 | 2.2 |
The comparison of the variants: truffle –T and control – C (evaluated using the Sanger sequencing method) allowed us to distinguish three genera of bacteria. The genera
Percentage of bacteria of dominant genera, indicating variant (T – truffle; C– control) obtained by Sanger sequencing. The numbers in bold indicate the presence of a particular genus of bacteria in a particular sample
Genus | 2017 | 2018 | ||
---|---|---|---|---|
T | C | T | C | |
6.8 | 6.8 | 6.7 | 7.9 | |
32.9 | 20.5 | 11.2 | 3.4 | |
8.2 | 1.4 | 5.6 | – | |
– | 4.1 | 18.0 | 11.2 | |
6.8 | 1.4 | 9.0 | 5.6 | |
1.4 | – | – | – | |
1.4 | 1.4 | 1.1 | – | |
1.4 | 1.4 | – | – | |
– | 1.4 | – | – | |
– | 1.4 | – | – | |
– | – | 2.2 | – | |
– | – | 1.1 | – | |
– | – | – | 1.1 | |
– | – | – | 1.1 |
The Kruskal-Wallis statistical analysis for the indices of alpha biodiversity did not show significant differences in the values of Chao, Faith’s, OTUs, Shannon’s, Heip’s or Simpson’s indices for all the studied groups. Statistical analysis between different groups of samples for the above indices also showed no significant differences. The microbiomes for individual samples, including variants T and C, did not differ significantly within individual or between groups (Tab. 4).
Alpha biodiversity index values for Chao, Faith’s, OTUs, Shannon’s, Heip’s and Simpson’s indices for selected samples representing sites W, G and M
Sample | Chao | Faith’s | OTU | Shannon’s | Heip’s | Simpson’s |
---|---|---|---|---|---|---|
W1 | 188.00 | 13.77 | 188.00 | 7.12 | 0.74 | 0.99 |
W2 | 178.00 | 13.53 | 178.00 | 7.09 | 0.76 | 0.99 |
W4 | 362.00 | 14.56 | 362.00 | 7.78 | 0.61 | 0.99 |
W5 | 272.00 | 13.36 | 272.00 | 7.43 | 0.63 | 0.99 |
G1 | 246.00 | 13.44 | 246.00 | 7.49 | 0.73 | 0.99 |
G2 | 120.00 | 12.74 | 120.00 | 6.51 | 0.76 | 0.99 |
G3 | 233.00 | 15.23 | 233.00 | 7.25 | 0.65 | 0.99 |
G4 | 184.00 | 12.87 | 184.00 | 6.98 | 0.68 | 0.99 |
G5 | 239.00 | 13.73 | 239.00 | 7.46 | 0.74 | 0.99 |
G6 | 214.00 | 13.48 | 214.00 | 7.23 | 0.70 | 0.99 |
M1 | 162.00 | 12.44 | 162.00 | 6.35 | 0.50 | 0.98 |
M2 | 245.00 | 15.21 | 245.00 | 7.50 | 0.74 | 0.99 |
M3 | 252.00 | 14.80 | 252.00 | 7.53 | 0.73 | 0.99 |
M4 | 203.00 | 13.38 | 203.00 | 7.14 | 0.69 | 0.99 |
M5 | 200.00 | 13.54 | 200.00 | 7.07 | 0.67 | 0.99 |
M6 | 121.00 | 11.12 | 121.00 | 6.48 | 0.73 | 0.98 |
Both the weighted and unweighted measures showed an unequal distribution of beta biodiversity between sample groups, as shown in Fig. 2. The GC and WC sample groups are concentrated on the right side of the plots while the remaining samples are concentrated on the left side of the plots (Fig. 2). This means that the GC and WC samples had microbiomes, which were similar to each other, and different from the microbiome of the other samples that are concentrated on the left side of the plots.
The value of the Bray-Curtis dissimilarity is in the range 0–1, where 0 means that both groups have the same composition and 1 means that the groups do not have a single species in common.
The PcoA plot, which includes Bray-Curtis dissimilarity, shows the breakdown between sample groups. As with the UniFrac measures, the results of a group of WC and GC samples on the right-hand side of the plot indicate that these samples have similar microbiomes. The remaining samples concentrated on the left side of the plot differed in microbiome composition from the samples concentrated on the right side of the plot. In addition, these calculations show a division between the sample groups GT and MC (upper left corner) and WT and MT (lower left corner) (Fig. 3).
The Jaccard index values, which determine the common and different elements in the sets (sample groups) in the PcoA plot, confirmed the differences between the sample groups. Similar to the UniFrac and Bray-Curtis measures, the groups of WC and GC samples were concentrated on the right side of the plot, while the remaining samples were on the left side of the plot. The PcoA distance plot confirms the results obtained using the Bray-Curtis dissimilarity index. In addition, both plots show the division between sample groups GT and MC (upper left corner) and WT and MT (lower left corner) (Fig. 4).
The PERMANOVA test for all the tested groups showed significant differences in the probability level
Prepared boxplot charts with Bray-Curtis, Jaccard, Unweighted and Weighted indices show differences in the distances between the analysed groups. The value
Analysis of the alpha correlation with CR data at different time intervals showed a positive correlation relevant to Simpson’s and CR coefficient in spring 2017 (
The involvement, abundance, metabolic activity, and role of bacteria in shaping the forest soil environment are not yet fully understood. Bacteria in forest soils can serve many functions, including helping plants adapt to adverse conditions, that is, lack of water or excessive salinity, assisting in the formation of mycorrhizal systems, increasing the resistance of trees to stressors, and inhibiting the development of pathogens (Tsukamoto et al. 2002; Frey-Klett et al. 2007; Tarkka and Frey-Klett 2008). Soil bacteria show various endophytic and symbiotic effects on fungi, including enhancing the distribution of their spores (Citterio et al. 1995; Splivallo et al. 2015) or providing vitamins and growth regulators (Rangel-Castro et al. 2002; Riedlinger et al. 2006). Bacteria are found not only in soil and green plant tissue, but also in wood and roots (Kubiak et al., 2018). The abundance and composition of soil bacterial communities depends on the type of host plant, on the activity and nutrients of plant roots and their secretions, on fungi coexisting in the substrate, but also on climatic conditions (Bardgett 2011, Lau and Lennon 2011; Eisenhauer et al. 2017). Ambient temperature influences the number of bacteria, the ratio of bacteria to fungi, but also favourable interactions between species.
Studies by Deveau et al. (2016) in relation to
The differences in root colonization by bacteria in the years 2017–2018 may have been due to meteorological conditions. The total precipitation in the vegetative season was ~ 1.5 times higher in 2017 compared to 2018. The difference between the averages of the two measuring stations was higher by ~ 1.00 in 2017 than in 2016 and by ~ 2.18 in 2017 than in 2018. From the results obtained by Siebyła and Hilszczańska (2020) on the identification of soil bacterial communities, it can be concluded that the weather conditions (here: in the autumn season of 2017) resulted in an increase in the amount of bacterial population. Similar relationships are described by Przemieniecki et al. (2021), who point to the significant influence of weather conditions at different times of the year on the activity of the bacterial microbiome in
The Sielianinov hydrothermal coefficient in presented research was above 2 at that time, which indicates high water saturation of the soil after the dry summer months in 2015 and 2016. In 2017, the Sielianinov hydrothermal coefficient (average of results in both stations) was ~2.19, twice as high as in 2018 when it was ~1.1.
Research by Deveau et al. (2016) concerning the analysis of ectomycorrhizal communities associated with
Studies conducted by López-Mondéjar et al. (2015) and Deveau et al. (2016) indicate that the bacterial community in soil changes over time. Changes in the composition of the microbiome depending on the season may therefore affect the growth of
The NGS sequencing of root samples showed differences in the proportion of
Significant differences were found between the communities determined by Sanger sequencing and NGS. Bacteria of the genus
Comparison of the results with previous soil studies on the same sites shows significant differences in the ratio of bacteria of a given type in the roots compared to their presence in the soil.
A negative correlation between the presence of bacteria isolated from roots of the genus