What Are Artificial Soil Mixes Made of

Abstract

Artificial soils were used in this study to analyse the importance of different mineral compositions for the diversity of soil microorganisms. Variants containing montmorillonite (MT), illite (IL) and illite + ferrihydrite (IL+FH) were compared to each other. Bulk material and their particle size fractions, as obtained by ultracentrifugation and wet-sieving, were characterised for abundance and diversity of Bacteria, Archaea and Fungi. Samples were analysed 6 and 18 months after inoculation with sterilised manure and a soil-extracted microbial community. Generally, IL, and even more pronouncedly IL+FH, supported the growth of more Bacteria, Archaea and Fungi, than MT. This trend was most pronounced in the finest fraction (< 20 μm). The structural diversity of Fungi responded more strongly to the different mineral compositions than the Bacteria, for which particle size fractions were more important. Archaea established a specific community in the finest fraction and showed no response to the different mineral compositions. Overall, this study demonstrates that the mineral composition and the particle size fractions have specific and different selective effects on the three domains and, thus, suggests that these factors strongly contribute to niche separation and the high diversity of microbial communities in natural soils with complex mineral compositions.

Artificial soils composed of different minerals reveal soil particle-specific colonisation by Bacteria, Archaea and Fungi. These domain-specific responses to minerals may contribute to the high microbial diversity in natural soils.

Artificial soils composed of different minerals reveal soil particle-specific colonisation by Bacteria, Archaea and Fungi. These domain-specific responses to minerals may contribute to the high microbial diversity in natural soils.

Introduction

Soil provides a huge diversity of microhabitats, which exhibit different physicochemical conditions for the selection of specifically adapted microbial communities. An approach to unravel this diversity of niches is to fractionate soils according to their different particle size classes, which include sand- (63–2000 μm), silt- (2–63 μm) and clay-size (< 2 μm). Typically for soils, such classes differ in their mineral composition (e.g. Acosta et al., 2011) and in the quantity and quality of soil organic carbon (Christensen, 1992). Carbon nitrogen ratios (C/N) decline with decreasing particle sizes, probably due to a replacement of plant material by microbial material (Ahmed & Oades, 1984), and no plant debris can be found in aggregates of < 20 μm (Oades & Waters, 1991). Activities of single enzymes (e.g. Stemmer et al., 1998; Kandeler et al., 2000; Marx et al., 2005) as well as metabolic potentials including denitrification (Lensi et al., 1995) and methanogenesis (Zhang et al., 2007; Zheng et al., 2007) can differ between particle size classes. Furthermore, microbial biomass typically increases with decreasing particle size fractions (e.g. Jocteur Monrozier et al., 1991; Lensi et al., 1995; Stemmer et al., 1998), excluding, however, the particulate organic material (POM; Kanazawa & Filip, 1986) prevalent in the coarse fractions. This negative correlation between particle size fractions and microbial abundance can be related to the overall increasing surface area accessible for microbial colonisation (Neumann et al., 2013). Microbial diversity is also affected by particle size fractions: microbial community analyses based on fingerprinting techniques of PCR-amplified small subunit (SSU) rRNA genes or phospholipid fatty acid (PLFA) analysis revealed different microbial communities in distinct particle size classes (e.g. Sessitsch et al., 2001; Poll et al., 2003; Zhang et al., 2007; Neumann et al., 2013), thus underlining the fact that particle size fractions represent different organo-mineral surface properties and select for specifically structured microbial communities. Results from previous studies suggested a selective effect on microbial community structure by the quality and quantity of soil organic carbon (Neumann et al., 2013, 2014).

A number of studies already demonstrated that various minerals may support different amounts of microbial biomass (e.g. Rogers et al., 1998; Roberts, 2004; Mauck & Roberts, 2007) and that minerals of different compositions and structures are linked to distinct microbial communities (e.g. Certini et al., 2004; Gleeson et al., 2005; Nishiyama et al., 2012). This effect of minerals on microbial cells in soil may be caused by taxon-specific sorption patterns of the microbial cells (Ams et al., 2004), a requirement of taxa for specific nutrients (Bennett et al., 2001), or the inability of some microorganisms to colonise certain minerals (Gleeson et al., 2005, 2006). Considering the diversity of minerals in soil, one can assume a complex composition of different selective niches just based on these inorganic compounds.

The objective of this study was to analyse whether soil minerals have a specific effect on the selection of surface-attached microbial communities. To experimentally approach this problem, we decided to work with artificial soils. These are mixtures of defined organic and mineral compounds, which mimic natural soils and thereby allow the importance of specific components to be analysed (Madhok, 1937), but, to date, they have rarely been used to study microbial colonisation patterns. In a joint project, quartz, clay minerals (montmorillonite = MT or illite = IL), iron oxides (ferrihydrite = FH) and sterilised farmyard manure were mixed, moistened with CaCl₂ solution and inoculated with a microbial suspension from a soil extract. These soil-like mixtures were incubated for 6 and 18 months to study the development of biogeochemical interfaces. Previous studies already demonstrated that microbial communities of different composition developed in the selected artificial soil variants (Babin et al., 2013; Ding et al., 2013), but the importance of the different surface properties as provided by the different particle size fractions still remained unclear. In the present work, we therefore studied the microbial communities attached to different particle size fractions, which were obtained by mild sonication and wet-sieving (Neumann et al., 2013). Artificial soil variants MT, IL and IL+FH were selected to characterise the importance of the clay material and the presence of ferrihydrite, respectively. Cultivation independent analyses of SSU rRNA gene or ITS sequences based on directly extracted DNA allowed the quantification of the abundance and characterisation of the diversity of Bacteria, Archaea, and Fungi, respectively. To our knowledge, this is the first study integrating mineral composition, particle size fractions and succession for all three microbial domains in a soil-like environment.

Materials and methods

Artificial soil preparation

The artificial soils were prepared by mixing quartz of sand- and silt-size with the clay minerals montmorillonite (MT) or illite (IL). Montmorillonite (Ceratosil^® WG) was obtained from Süd-Chemie AG (Moosburg, Germany) and used without further preparation, while illite (Inter-ILI, Mérnöki Iroda, Hungary) was ground and dry-sieved to a grain size < 63 μm. A third artificial soil variant consisted of illite plus 6-line ferrihydrite (synthesised according to Schwertmann & Cornell (1991)) as an iron oxide (IL+FH) (Table 1). In these mixtures, the fraction < 6.3 μm represented 5.6% of the total mineral mass. Soil microcosms were prepared in PVC flower pots with a diameter of about 15 cm and a depth of about 7–10 cm. All three variants received dried and sterilised manure from a long-term fertilisation experiment in Bad Lauchstädt (Germany) as a source for organic carbon (final concentration 4.3% w/w). The completely dry artificial soil variants were initially wetted with 138 mL 0.01 M CaCl₂ for MT and 126 mL for illite containing variants and homogenised by gentle stirring. Three days later, the soils were inoculated with 60 mL of a microbial inoculum reaching a final water content of 60% of the water holding capacity and a total weight of 1 kg of artificial soil. The inoculum was obtained by water-extraction of an arable Cambisol of < 2 mm particle size from Ultuna (Sweden) under shaking with gravel for 2 h. The suspension was centrifuged at 1000 g for 12 min, and subsequently, the supernatant was centrifuged at 3470 g for 30 min (Ding et al., 2013). After resuspension of the pellet, the ratio of soil mass extracted (g) to the final volume of inoculum (mL) was 1 : 15. The soil variants were incubated at 20 ± 5 °C in the dark. Water content was kept constant at 60% of water holding capacity by weekly addition of water according to loss of weight and gentle stirring to guarantee homogenous distribution. Six and 18 months after inoculation, batches were sacrificed and stored at −20 °C. Each variant was represented by three independent replicates. Further details on the origin and treatment of the materials, the preparation of the artificial soils and their incubation conditions have been described elsewhere (Pronk et al., 2012).

Table 1

Original composition of the artificial soils in mass fractions

	MT			IL			IL+FH
63–2000 μm		2–63 μm			< 2 μm			63–2000 μm			2–63 μm		< 2 μm		63–2000 μm		2–63 μm		< 2 μm
Quartz	42.9		46.8			0			41.2			46.8		0		41.2		46.8		0
Montmorillonite	0.5		1.5			4.0			0			0		0		0		0		0
Illite	0		0			0			0			3.8		3.8		0		3.3		3.3
Ferrihydrite	0		0			0			0			0		0		0		– –1.0– –
Manure	3.7		– –0.6– –		3.7				– –0.6– –	3.7			– –0.6– –

	MT			IL			IL+FH
63–2000 μm		2–63 μm			< 2 μm			63–2000 μm			2–63 μm		< 2 μm		63–2000 μm		2–63 μm		< 2 μm
Quartz	42.9		46.8			0			41.2			46.8		0		41.2		46.8		0
Montmorillonite	0.5		1.5			4.0			0			0		0		0		0		0
Illite	0		0			0			0			3.8		3.8		0		3.3		3.3
Ferrihydrite	0		0			0			0			0		0		0		– –1.0– –
Manure	3.7		– –0.6– –		3.7				– –0.6– –	3.7			– –0.6– –

Sums of mineral components and manure equal 100%. For values indicated by dashes, distribution between particle size classes is not known. Data based on Pronk et al. (2012).

Table 1

Original composition of the artificial soils in mass fractions

	MT			IL			IL+FH
63–2000 μm		2–63 μm			< 2 μm			63–2000 μm			2–63 μm		< 2 μm		63–2000 μm		2–63 μm		< 2 μm
Quartz	42.9		46.8			0			41.2			46.8		0		41.2		46.8		0
Montmorillonite	0.5		1.5			4.0			0			0		0		0		0		0
Illite	0		0			0			0			3.8		3.8		0		3.3		3.3
Ferrihydrite	0		0			0			0			0		0		0		– –1.0– –
Manure	3.7		– –0.6– –		3.7				– –0.6– –	3.7			– –0.6– –

	MT			IL			IL+FH
63–2000 μm		2–63 μm			< 2 μm			63–2000 μm			2–63 μm		< 2 μm		63–2000 μm		2–63 μm		< 2 μm
Quartz	42.9		46.8			0			41.2			46.8		0		41.2		46.8		0
Montmorillonite	0.5		1.5			4.0			0			0		0		0		0		0
Illite	0		0			0			0			3.8		3.8		0		3.3		3.3
Ferrihydrite	0		0			0			0			0		0		0		– –1.0– –
Manure	3.7		– –0.6– –		3.7				– –0.6– –	3.7			– –0.6– –

Sums of mineral components and manure equal 100%. For values indicated by dashes, distribution between particle size classes is not known. Data based on Pronk et al. (2012).

Particle size fractionation

For particle size fractionation, the protocol of Neumann et al. (2013), which was based on a previously published protocol (Amelung et al., 1998), was further modified as follows: a total of 20 g dry weight of artificial soil material was suspended in 100 mL distilled water. To disrupt the aggregates while maintaining as many microorganisms attached as possible, a low sonication energy of 30 J mL⁻¹ was applied using a Sonopuls HD 2200 homogeniser (Bandelin electronic, Berlin, Germany) at a total output of 70 W. The tip of the sonotrode VS 70T was submersed by 2 cm into the suspension. During sonication, the suspension was kept water-cooled. The fractions were separated by two wet-sieving steps using mesh sizes of 63 and 20 μm, respectively. The suspension of the fraction < 20 μm was supplied with 1 M MgCl₂ solution to a final concentration of 3.3 mM and stored at 4 °C over night. It should be noted that the illite containing < 20 μm fractions and among them, especially those supplied with ferri-hydrite, did not fully precipitate after addition of MgCl₂. Therefore, these suspensions of all three soil variants were additionally centrifuged at 2450 g for 10 min at room temperature and supernatant was decanted.

DNA extraction

DNA was extracted from 0.5 g fresh weight artificial soil material. The starting material originated either from the nonfractionated samples or from the above mentioned size fractions. Material from the coarse fraction (63–2000 μm) was taken directly from the sieves, the pellet of the 20–63 μm-fraction was suspended with distilled water at a ratio of 1 : 5, and the finest fraction (< 20 μm) at a ratio of 1 : 12. A total of 1.8 mL of the suspensions was transferred to 2 mL Eppendorf-tubes. After centrifugation at 12 700 g for 5 min, the supernatants were discarded. The soil material used for DNA extraction corresponded to 0.4 g dry weight of the coarse and the medium-sized fraction, respectively, and to 0.15 g of the finest fraction.

The DNA extraction was conducted with the FastDNA^™ SPIN Kit for soil using the FastPrep^®-24 Instrument (both MP Biomedicals, Santa Ana) according to the manufacturer's instructions, using following modifications: Volumes of sodium phosphate buffer and supplied 'MT-buffer' were adjusted to 950 and 120 μL, respectively. Bead-beating was run twice for 45 s at a speed of 6.5 m s⁻¹. The samples were then centrifuged for 5 min at 14 000 g and room temperature. The DNA bound to the binding matrix of the FastDNA^™ SPIN Kit was washed twice with 1 mL 5.5 M guanidinthiocyanate to remove coextracted contaminants. After elution of DNA with 100 μL distilled water, this step was repeated using the eluate.

The concentration of the DNA solutions was determined in a Mithras LB 940 Fluorometer (Berthold Technologies, Bad Wildbad, Germany) using the Quant-iT^™ PicoGreen^® dsDNA Assay Kit (Life Technologies/Thermo Fisher Scientific, Carlsbad).

Quantitative PCR (qPCR) of microbial domains

The abundance of SSU genes in the soil-extracted DNA solutions was determined by qPCR using the StepOnePlus^™ Real-Time PCR System apparatus (Life Technologies/Thermo Fisher Scientific). Partial bacterial SSU rRNA genes were amplified using the primers BAC338F and BAC805R and quantified with the BAC516F probe (Yu et al., 2005). Accordingly, Archaea were analysed with the primers ARC787F and ARC1059R and the probe ARC915F (Yu et al., 2005). Both prokaryotic domains were amplified with the Maxima Probe qPCR Master Mix (2x) including ROX solution (Thermo Fisher Scientific, Waltham). To quantify Fungi (Dikarya), the ITS1 region was amplified with NSI1 and 58A2R (Martin & Rygiewicz, 2005) and the PCR products were quantified by SYBR Green using Maxima SYBR Green/ROX qPCR Master Mix (2x) (Thermo Fisher Scientific). A total volume of 20 μL contained 500 nM of single primers and, in the case of the prokaryotes, 200 nM of the single probes. For all samples, 2 μL of a 10-fold dilution was used as a template. For measurement of the bacterial SSU rRNA genes from the particle size fractions 63–2000 μm and < 20 μm, 2 μL of a 100-fold dilution was used as a template. The PCR of the prokaryotes started with an initial denaturing step of 95 °C for 10 min, followed by 45 cycles at 95 °C for 15 s and 60 °C for 1 min. Thermal cycling conditions for Fungi were initial denaturation at 95 °C for 10 min, 40 cycles at 95 °C for 15 s, 52 °C for 30 s, 72 °C for 30 s and fluorescent data collection at 79 °C for 15 s. For melting curve analyses, there was a denaturing at 95 °C for 15 s, annealing at 60 °C for 1 min and melting in 0.3 °C steps up to 95 °C for 15 s. DNA from pure cultures of Bacillus subtilis, Methanobacterium oryzae and Fusarium culmorum served as positive controls for the three domains.

Genetic profiling of microbial communities

The diversity of the dominant members of the three microbial domains was characterised by genetic profiling using terminal restriction fragment length polymorphism (T-RFLP). The 16S rRNA genes of Bacteria were amplified using primers F27* (Lane, 1991) and 926r (Liu et al., 1997), those of Archaea with A364aF (Burggraf et al., 1997) and A934bR* (Grosskopf et al., 1998), and for Fungi, the region between the SSU and the large subunit rRNA gene was amplified using ITS1f* (Gardes & Bruns, 1993) and ITS4r (White et al., 1990). Asterisks indicate fluorescence-labelled primers. PCR was conducted in a total volume of 25 μL which contained 0.025 U μL⁻¹ HotStar Taq DNA Polymerase, 1× buffer (both Qiagen, Hilden, Germany), 0.2 μM of each dNTP (Carl Roth, Karlsruhe, Germany), and 0.5 μM of each primer. The concentration of MgCl₂ was 1.75 mM for amplifications targeting Bacteria, 2.75 mM for Fungi and 3 mM for Archaea. For the Fungi-specific PCR, the reaction solution also contained 0.4 mg mL⁻¹ bovine serum albumin (New England Biolabs, Ipswich). Templates for PCR consisted of 1 μL of a 100-fold dilution for Bacteria (10-fold for the coarsest fraction) or 10-fold for Archaea (undiluted for the coarsest fraction) and Fungi. The PCR started at 95 °C for 15 min, then followed 30 cycles at 94 °C for 30 s, 72 °C for 90 s for Bacteria or 60 s for Archaea and Fungi and a final extension at 72 °C for 5 min (8 min in the case of Bacteria). The annealing conditions were 40 s at 50 °C for Bacteria, 50 s at 60 °C for Archaea and 40 s at 55 °C for Fungi, respectively. To minimise variability, all PCRs were carried out in triplicates, which were afterwards unified. The size and purity of the amplicons were analysed by agarose gel electrophoresis. A total of 70 μL of the PCR products were purified with the Hi Yield^® Gel/PCR DNA Fragments Extraction Kit (Süd-Laborbedarf GmbH, Gauting, Germany) according to manufacturer's instructions and eluted with 30 μL elution buffer. The DNA concentration was determined with NanoDrop 2000c (Thermo Fischer Scientific), and then 100 ng of DNA was used for digestion with restriction endonuclease (final concentration of 0.3 U μL⁻¹) and the corresponding buffer in a total volume of 30 μL. Three independent fingerprints were generated for Bacteria and Fungi with HaeIII (BsuRI), FastDigest HhaI (both Fermentas/Thermo Fisher Scientific, Waltham), and MspI (New England Biolabs), respectively, while archaeal PCR products were digested only with BstUI (Fermentas/Thermo Fisher Scientific). The incubations were run for 4 h at 37 °C. The reactions were stopped by freezing at -20 °C. The DNA was then precipitated with 5 volumes of 95% (v/v) ethanol and 0.1 volumes of 3 M sodium acetate (pH 4.6) at −20 °C for 30 min. After centrifugation with 14 000 g for 30 min at 4 °C, the DNA pellets were washed with 100 μL ice-cold 70% (v/v) ethanol and centrifuged again for 10 min. The pellets were then air-dried, dissolved in 30 μL sample loading solution and supplied with internal standard 600 (both Beckman Coulter, Brea). The DNA fragments were size-separated by capillary electrophoresis (CEQ^™ 8800; Beckman Coulter). The conditions of the runs corresponded to the Frag-4 programme described by the manufacturer, but the duration of the runs was extended to 80 min. Profiles were analysed by setting the maximum bin width to 2 bp. Total peak heights were normalised to 100%. Fragments representing < 1.5% of the total peak heights were considered to be noise, and bins, which appeared in only one sample, were outliers. Independent profiles of the tree restriction enzymes for Bacteria and Fungi were additively combined to single profiles for each domain.

Measurement of carbon and nitrogen content

Artificial soil samples, including particle size separates, were oven-dried at 105 °C and ground. Contents of total carbon and total nitrogen were determined by combustion at 1350 °C with TrueMac^® CN (Leco Corp., St. Joseph).

Statistical analysis

The data of the soil variants were Box–Cox-transformed and analysed by two-way analysis of variance (anova) and Tukey's honestly significant difference (HSD) using sigmaplot (Version 11.0, Systat Software Inc., San Jose). Community profiles (T-RFLPs) were compared by nonmetric multidimensional scaling (nmds) using r (Version 2.15.1; R Core Team, 2012) together with the package vegan (Version 2.0-7; Oksanen et al., 2013) and by two-way analysis of similarity (anosim) using past (Version 2.17; Hammer et al., 2001). In the case of the particle size separates, two-way anosim was calculated separately for every single level of one of the factors mineral composition, particle size fraction and time point, respectively. For these multiple comparisons, the significance level was halved to 2.5%.

Results

Carbon and nitrogen contents and microbial colonisation of the artificial bulk soils

No differences were detected for the amount of total carbon or total nitrogen, when the three soil variants (MT, IL, IL+FH) were compared with each other (Fig. 1, Table 2, for individual values see Supporting Information, Table S1). There was a significant decline of the C/N ratio from 10.5 to 9.5 caused by a loss of carbon during 12 months of incubation.

Table 2

Differences between mineral compositions and time points for bulk soil and every fraction given by F- and P-values derived from two-way anova

	Bulk soil		63–2000 μm	20–63 μm		< 20 μm
F		P		F			P		F		P		F		P
Total carbon
Composition	1.620		0.238		0.193			0.827		0.255		0.779		2.328		0.140
Date	7.068		0.021		16.334			0.002		0.021		0.887		27.079		< 0.001
Comp. × Date	1.050		0.380		0.591			0.569		3.077		0.083		2.890		0.094
Total nitrogen
Composition	3.589		0.060		1.087			0.368		1.220		0.329		4.734		0.031
Date	0.296		0.596		8.859			0.012		0.040		0.845		20.285		< 0.001
Comp. × Date	2.304		0.142		0.693			0.519		0.086		0.918		0.400		0.679
C/N ratio
Composition	2.664		0.110		7.875			0.007		2.884		0.095		2.722		0.106
Date	38.778		< 0.001		57.286			< 0.001		0.403		0.537		6.469		0.026
Comp. × Date	2.391		0.134		1.492			0.264		8.294		0.005		7.572		0.007
Bacterial abundance
Composition	67.309		< 0.001		1.422			0.279		13.356		< 0.001		35.197		< 0.001
Date	0.163		0.693		53.635			< 0.001		4.675		0.052		2.546		0.137
Comp. × Date	0.298		0.748		0.982			0.403		1.170		0.343		0.465		0.639
Archaeal abundance
Composition	32.096		< 0.001		12.038			0.001		13.563		< 0.001		64.887		< 0.001
Date	66.810		< 0.001		24.220			< 0.001		0.520		0.485		509.179		< 0.001
Comp. × Date	1.568		0.248		1.433			0.277		0.215		0.809		11.152		0.002
Fungal abundance
Composition	6.673		0.011		1.708			0.222		4.205		0.041		5.470		0.020
Date	0.088		0.771		12.909			0.004		0.066		0.801		0.093		0.766
Comp. × Date	1.721		0.220		3.325			0.071		0.949		0.415		0.769		0.485

	Bulk soil		63–2000 μm	20–63 μm		< 20 μm
F		P		F			P		F		P		F		P
Total carbon
Composition	1.620		0.238		0.193			0.827		0.255		0.779		2.328		0.140
Date	7.068		0.021		16.334			0.002		0.021		0.887		27.079		< 0.001
Comp. × Date	1.050		0.380		0.591			0.569		3.077		0.083		2.890		0.094
Total nitrogen
Composition	3.589		0.060		1.087			0.368		1.220		0.329		4.734		0.031
Date	0.296		0.596		8.859			0.012		0.040		0.845		20.285		< 0.001
Comp. × Date	2.304		0.142		0.693			0.519		0.086		0.918		0.400		0.679
C/N ratio
Composition	2.664		0.110		7.875			0.007		2.884		0.095		2.722		0.106
Date	38.778		< 0.001		57.286			< 0.001		0.403		0.537		6.469		0.026
Comp. × Date	2.391		0.134		1.492			0.264		8.294		0.005		7.572		0.007
Bacterial abundance
Composition	67.309		< 0.001		1.422			0.279		13.356		< 0.001		35.197		< 0.001
Date	0.163		0.693		53.635			< 0.001		4.675		0.052		2.546		0.137
Comp. × Date	0.298		0.748		0.982			0.403		1.170		0.343		0.465		0.639
Archaeal abundance
Composition	32.096		< 0.001		12.038			0.001		13.563		< 0.001		64.887		< 0.001
Date	66.810		< 0.001		24.220			< 0.001		0.520		0.485		509.179		< 0.001
Comp. × Date	1.568		0.248		1.433			0.277		0.215		0.809		11.152		0.002
Fungal abundance
Composition	6.673		0.011		1.708			0.222		4.205		0.041		5.470		0.020
Date	0.088		0.771		12.909			0.004		0.066		0.801		0.093		0.766
Comp. × Date	1.721		0.220		3.325			0.071		0.949		0.415		0.769		0.485

Significant values are shown in bold.

Table 2

Differences between mineral compositions and time points for bulk soil and every fraction given by F- and P-values derived from two-way anova

	Bulk soil		63–2000 μm	20–63 μm		< 20 μm
F		P		F			P		F		P		F		P
Total carbon
Composition	1.620		0.238		0.193			0.827		0.255		0.779		2.328		0.140
Date	7.068		0.021		16.334			0.002		0.021		0.887		27.079		< 0.001
Comp. × Date	1.050		0.380		0.591			0.569		3.077		0.083		2.890		0.094
Total nitrogen
Composition	3.589		0.060		1.087			0.368		1.220		0.329		4.734		0.031
Date	0.296		0.596		8.859			0.012		0.040		0.845		20.285		< 0.001
Comp. × Date	2.304		0.142		0.693			0.519		0.086		0.918		0.400		0.679
C/N ratio
Composition	2.664		0.110		7.875			0.007		2.884		0.095		2.722		0.106
Date	38.778		< 0.001		57.286			< 0.001		0.403		0.537		6.469		0.026
Comp. × Date	2.391		0.134		1.492			0.264		8.294		0.005		7.572		0.007
Bacterial abundance
Composition	67.309		< 0.001		1.422			0.279		13.356		< 0.001		35.197		< 0.001
Date	0.163		0.693		53.635			< 0.001		4.675		0.052		2.546		0.137
Comp. × Date	0.298		0.748		0.982			0.403		1.170		0.343		0.465		0.639
Archaeal abundance
Composition	32.096		< 0.001		12.038			0.001		13.563		< 0.001		64.887		< 0.001
Date	66.810		< 0.001		24.220			< 0.001		0.520		0.485		509.179		< 0.001
Comp. × Date	1.568		0.248		1.433			0.277		0.215		0.809		11.152		0.002
Fungal abundance
Composition	6.673		0.011		1.708			0.222		4.205		0.041		5.470		0.020
Date	0.088		0.771		12.909			0.004		0.066		0.801		0.093		0.766
Comp. × Date	1.721		0.220		3.325			0.071		0.949		0.415		0.769		0.485

	Bulk soil		63–2000 μm	20–63 μm		< 20 μm
F		P		F			P		F		P		F		P
Total carbon
Composition	1.620		0.238		0.193			0.827		0.255		0.779		2.328		0.140
Date	7.068		0.021		16.334			0.002		0.021		0.887		27.079		< 0.001
Comp. × Date	1.050		0.380		0.591			0.569		3.077		0.083		2.890		0.094
Total nitrogen
Composition	3.589		0.060		1.087			0.368		1.220		0.329		4.734		0.031
Date	0.296		0.596		8.859			0.012		0.040		0.845		20.285		< 0.001
Comp. × Date	2.304		0.142		0.693			0.519		0.086		0.918		0.400		0.679
C/N ratio
Composition	2.664		0.110		7.875			0.007		2.884		0.095		2.722		0.106
Date	38.778		< 0.001		57.286			< 0.001		0.403		0.537		6.469		0.026
Comp. × Date	2.391		0.134		1.492			0.264		8.294		0.005		7.572		0.007
Bacterial abundance
Composition	67.309		< 0.001		1.422			0.279		13.356		< 0.001		35.197		< 0.001
Date	0.163		0.693		53.635			< 0.001		4.675		0.052		2.546		0.137
Comp. × Date	0.298		0.748		0.982			0.403		1.170		0.343		0.465		0.639
Archaeal abundance
Composition	32.096		< 0.001		12.038			0.001		13.563		< 0.001		64.887		< 0.001
Date	66.810		< 0.001		24.220			< 0.001		0.520		0.485		509.179		< 0.001
Comp. × Date	1.568		0.248		1.433			0.277		0.215		0.809		11.152		0.002
Fungal abundance
Composition	6.673		0.011		1.708			0.222		4.205		0.041		5.470		0.020
Date	0.088		0.771		12.909			0.004		0.066		0.801		0.093		0.766
Comp. × Date	1.721		0.220		3.325			0.071		0.949		0.415		0.769		0.485

Significant values are shown in bold.

Fig. 1

Carbon and nitrogen contents of the artificial soils in this study: total carbon (a), total nitrogen (b), C/N ratio (c). Means are displayed in accordance to two-way anova results (Table 2). Error bars represent standard deviation. In the case of mineral composition and interactions, letters indicate significant differences.

Fig. 1

In contrast to carbon and nitrogen contents, the microbial population sizes, as indicated by the SSU rRNA gene and ITS1 copy numbers, responded to the different soil variants (Fig. 2). Compared to MT, bacterial 16S rRNA gene copy numbers were about 1.7 times higher with IL, and 2.5 times higher with IL+FH (Fig. 2a, for individual values see Table S2). Differences between variants were significant, but there were no differences between samples taken after 6 and 18 months (Table 2). Archaeal rRNA gene copy numbers were 2.5–3 times higher with IL and IL+FH compared to MT, but the differences between IL and IL+FH were not significant (Fig. 2b, Table 2; Table S2). In contrast to Bacteria, there was a significant threefold increase of the archaeal 16S rRNA gene copy numbers over time. For fungal ITS1 sequences, IL+FH showed 8 times higher abundances than MT, while no significant differences were observed over time (Fig. 2c, Table 2; Table S2).

Fig. 2

Abundances of Bacteria (a), Archaea (b) and Fungi (c) represented by gene copy numbers obtained by quantitative PCR. Means are displayed in accordance to two-way anova results (Table 2). Error bars represent standard deviation. In the case of mineral composition and interactions, letters indicate significant differences.

Abundances of Bacteria (a), Archaea (b) and Fungi (c) represented by gene copy numbers obtained by quantitative PCR. Means are displayed in accordance to two-way anova results (Table 2). Error bars represent standard deviation. In the case of mineral composition and interactions, letters indicate significant differences.

Fig. 2

Carbon, nitrogen and microorganisms associated with the particle size fractions

The particle size fractionation procedure indicated that the artificial soils were composed of 45–47% of 63–2000 μm, 27–30% of 20–63 μm and 25–27% of < 20 μm, respectively. The sum of the particle size fractions recovered after the separation accounted for 98–102% (w/w) of the amount with which the fractionation procedure was started (data not shown). The recovery rates of carbon and nitrogen found with the three particle size fractions were comparable to the amounts detected in bulk soil, with 96–107% and 80–93%, respectively.

The loss of carbon found in the bulk soil, when months 6 and 18 were compared with each other, was only seen in the coarsest fraction, while there was no effect in the medium-sized fraction and even an increase in total carbon over time in the finest fraction (Fig. 1a, Table 2, individual values in Table S1). This finest fraction was also characterised by an increase in total nitrogen, which was not detected in the two coarser fractions, and it showed higher amounts in IL+FH than in MT (Fig. 1b). The significant decrease in the C/N ratio coincided with changes in the coarsest fraction, while there was no main effect over time for the medium and even the opposite trend of an increased C/N ratio in the finest fraction (Fig. 1c).

For total DNA, the overall recovery rates of the particle size fractions were low (Fig. 3): Compared to bulk (nonfractionated) soil, the recovery of total DNA from all particle size fractions ranged from 35% to 97%. Independent of the incubation period, the amount of total DNA extracted from bulk soil was lowest in the MT variant, higher in IL and highest in IL+FH (Fig. 3). The amount of bulk soil DNA found after 18 months was higher than after 6 months, which was mainly caused by an increase of DNA in the finest particle size fraction. Only a minor fraction of the DNA was associated with the medium particle size fraction (4%). No differences between variants were found for the DNA belonging to the coarsest fraction, which accounted for c. 50% of total soil DNA and decreased two- to threefold over time. The relatively low recovery rates of SSU rRNA gene copies and ITS1 sequences from particle size fractions compared to bulk soil for Bacteria (38–62%), Archaea (18–35%) and Fungi (17–32%) indicated that significant proportions of the microbial cells were detached from the particles during the soil fractionation procedure, which confirmed the results found for the quantification of the total DNA. In fact, for Bacteria and Archaea, but not for Fungi, there was a significant correlation between gene copy numbers and total DNA (R ² = 0.60, 0.74, and 0.30, respectively).

Fig. 3

$Comparisons of the DNA contents of the artificial bulk soils and the sums extracted from their respective particle size fractions. Means of triplicates are stacked.$

Comparisons of the DNA contents of the artificial bulk soils and the sums extracted from their respective particle size fractions. Means of triplicates are stacked.

Fig. 3

$Comparisons of the DNA contents of the artificial bulk soils and the sums extracted from their respective particle size fractions. Means of triplicates are stacked.$

Comparisons of the DNA contents of the artificial bulk soils and the sums extracted from their respective particle size fractions. Means of triplicates are stacked.

For the particle-attached bacterial and archaeal 16S rRNA genes, only the coarsest fraction indicated a decline over time (by 45%), while in the finest fraction the archaeal rRNA gene copy numbers increased by c. two orders of magnitude. Independent of the incubation time, for both Bacteria and Archaea, differences in the soil variants were seen in the medium-sized fraction, where IL+FH was significantly higher in gene copy numbers than MT, and in the finest size fraction, where both IL and IL+FH carried more gene copy numbers than MT (Fig. 2a and b, Table 2). Soil fractionation also indicated that the higher abundance of fungal ITS1 sequences seen in bulk soil with IL+FH was caused by the medium and finest, but not by the coarsest fraction (Fig. 2c).

Responses of the microbial community structure to the soil variants

The T-RFLP profiles for Bacteria for each sample were composed of an average of 51 terminal restriction fragments (TRFs). The complexity of the profiles which were used to characterise the diversity of Archaea and Fungi was much lower with an average of 6 TRFs and 19 TRFs, respectively (for profiles see, Figs S1–S7).

A strong change in community composition over time was observed for Bacteria and Archaea (Fig. 4a and b). Independent of the incubation period, nmds analyses of the T-RFLP profiles indicated that MT selected for differently structured communities compared to IL and IL+FH for both Bacteria and Fungi (Fig. 4). The effect of ferri-hydrite was not so pronounced, except for Fungi after 18 months. The effect of the soil variants on the diversity of the Archaea was not clear. anosim confirmed differences of bacterial and fungal, but not of archaeal communities, in response to the mineral compositions (Table 3). The R statistic of anosim also indicated that the community structure of Bacteria and Archaea, but not of the Fungi, changed between 6 and 18 months.

Table 3

Relationships between microbial communities revealed by two-way anosim. The R statistic indicates the degree of separation, while P indicates the significance of differences

	Bulk soil	Fractions
Composition		Date	Fraction
MT		IL			IL+FH		6 months		18 months		63–2000 μm		20–63 μm		< 20 μm
Bacteria
Composition
R	0.440	–		–		–		0.476		0.221		0.160		0.313		0.572
P	0.002	–		–		–		< 0.001		0.016		0.065		0.010		< 0.001
Date
R	0.901	0.654		0.815		0.938		–		–		0.679		0.778		0.951
P	< 0.001	0.004		0.001		0.001		–		–		0.005		0.001		0.001
Fraction
R	–	0.708		0.823		0.926		0.830		0.808		–		–		–
P	–	0.001		< 0.001		< 0.001		< 0.001		< 0.001		–		–		–
Archaea
Composition
R	0.202	–		–		–		–0.064		0.287		0.074		0.165		0.095
P	0.021	–		–		–		0.787		0.003		0.242		0.037		0.199
Date
R	1.000	0.741		0.753		0.444		–		–		0.321		0.617		1.000
P	< 0.001	0.002		0.003		0.009		–		–		0.021		0.003		0.001
Fraction
R	–	0.173		0.481		0.502		0.004		0.767		–		–		–
P	–	0.138		0.004		0.002		0.457		< 0.001		–		–		–
Fungi
Composition
R	0.523	–		–		–		0.517		0.443		0.609		0.317		0.514
P	0.001	–		–		–		< 0.001		< 0.001		0.001		0.004		< 0.001
Date
R	0.160	–0.025		0.099		0.370		–		–		0.309		0.000		0.136
P	0.105	0.519		0.251		0.005		–		–		0.037		0.381		0.160
Fraction
R	–	0.045		0.687		0.671		0.599		0.336		–		–		–
P	–	0.351		< 0.001		< 0.001		< 0.001		< 0.001		–		–		–

	Bulk soil	Fractions
Composition		Date	Fraction
MT		IL			IL+FH		6 months		18 months		63–2000 μm		20–63 μm		< 20 μm
Bacteria
Composition
R	0.440	–		–		–		0.476		0.221		0.160		0.313		0.572
P	0.002	–		–		–		< 0.001		0.016		0.065		0.010		< 0.001
Date
R	0.901	0.654		0.815		0.938		–		–		0.679		0.778		0.951
P	< 0.001	0.004		0.001		0.001		–		–		0.005		0.001		0.001
Fraction
R	–	0.708		0.823		0.926		0.830		0.808		–		–		–
P	–	0.001		< 0.001		< 0.001		< 0.001		< 0.001		–		–		–
Archaea
Composition
R	0.202	–		–		–		–0.064		0.287		0.074		0.165		0.095
P	0.021	–		–		–		0.787		0.003		0.242		0.037		0.199
Date
R	1.000	0.741		0.753		0.444		–		–		0.321		0.617		1.000
P	< 0.001	0.002		0.003		0.009		–		–		0.021		0.003		0.001
Fraction
R	–	0.173		0.481		0.502		0.004		0.767		–		–		–
P	–	0.138		0.004		0.002		0.457		< 0.001		–		–		–
Fungi
Composition
R	0.523	–		–		–		0.517		0.443		0.609		0.317		0.514
P	0.001	–		–		–		< 0.001		< 0.001		0.001		0.004		< 0.001
Date
R	0.160	–0.025		0.099		0.370		–		–		0.309		0.000		0.136
P	0.105	0.519		0.251		0.005		–		–		0.037		0.381		0.160
Fraction
R	–	0.045		0.687		0.671		0.599		0.336		–		–		–
P	–	0.351		< 0.001		< 0.001		< 0.001		< 0.001		–		–		–

In the case of significance at 5% level for bulk soil and at 2.5% for fractions (bold P-values), well separated communities (R > 0.75) are indicated in bold, while separated, but overlapping communities (R > 0.5) are shown in italics. R < 0.25 is considered to express barely separated communities.

Table 3

Relationships between microbial communities revealed by two-way anosim. The R statistic indicates the degree of separation, while P indicates the significance of differences

	Bulk soil	Fractions
Composition		Date	Fraction
MT		IL			IL+FH		6 months		18 months		63–2000 μm		20–63 μm		< 20 μm
Bacteria
Composition
R	0.440	–		–		–		0.476		0.221		0.160		0.313		0.572
P	0.002	–		–		–		< 0.001		0.016		0.065		0.010		< 0.001
Date
R	0.901	0.654		0.815		0.938		–		–		0.679		0.778		0.951
P	< 0.001	0.004		0.001		0.001		–		–		0.005		0.001		0.001
Fraction
R	–	0.708		0.823		0.926		0.830		0.808		–		–		–
P	–	0.001		< 0.001		< 0.001		< 0.001		< 0.001		–		–		–
Archaea
Composition
R	0.202	–		–		–		–0.064		0.287		0.074		0.165		0.095
P	0.021	–		–		–		0.787		0.003		0.242		0.037		0.199
Date
R	1.000	0.741		0.753		0.444		–		–		0.321		0.617		1.000
P	< 0.001	0.002		0.003		0.009		–		–		0.021		0.003		0.001
Fraction
R	–	0.173		0.481		0.502		0.004		0.767		–		–		–
P	–	0.138		0.004		0.002		0.457		< 0.001		–		–		–
Fungi
Composition
R	0.523	–		–		–		0.517		0.443		0.609		0.317		0.514
P	0.001	–		–		–		< 0.001		< 0.001		0.001		0.004		< 0.001
Date
R	0.160	–0.025		0.099		0.370		–		–		0.309		0.000		0.136
P	0.105	0.519		0.251		0.005		–		–		0.037		0.381		0.160
Fraction
R	–	0.045		0.687		0.671		0.599		0.336		–		–		–
P	–	0.351		< 0.001		< 0.001		< 0.001		< 0.001		–		–		–

	Bulk soil	Fractions
Composition		Date	Fraction
MT		IL			IL+FH		6 months		18 months		63–2000 μm		20–63 μm		< 20 μm
Bacteria
Composition
R	0.440	–		–		–		0.476		0.221		0.160		0.313		0.572
P	0.002	–		–		–		< 0.001		0.016		0.065		0.010		< 0.001
Date
R	0.901	0.654		0.815		0.938		–		–		0.679		0.778		0.951
P	< 0.001	0.004		0.001		0.001		–		–		0.005		0.001		0.001
Fraction
R	–	0.708		0.823		0.926		0.830		0.808		–		–		–
P	–	0.001		< 0.001		< 0.001		< 0.001		< 0.001		–		–		–
Archaea
Composition
R	0.202	–		–		–		–0.064		0.287		0.074		0.165		0.095
P	0.021	–		–		–		0.787		0.003		0.242		0.037		0.199
Date
R	1.000	0.741		0.753		0.444		–		–		0.321		0.617		1.000
P	< 0.001	0.002		0.003		0.009		–		–		0.021		0.003		0.001
Fraction
R	–	0.173		0.481		0.502		0.004		0.767		–		–		–
P	–	0.138		0.004		0.002		0.457		< 0.001		–		–		–
Fungi
Composition
R	0.523	–		–		–		0.517		0.443		0.609		0.317		0.514
P	0.001	–		–		–		< 0.001		< 0.001		0.001		0.004		< 0.001
Date
R	0.160	–0.025		0.099		0.370		–		–		0.309		0.000		0.136
P	0.105	0.519		0.251		0.005		–		–		0.037		0.381		0.160
Fraction
R	–	0.045		0.687		0.671		0.599		0.336		–		–		–
P	–	0.351		< 0.001		< 0.001		< 0.001		< 0.001		–		–		–

Fig. 4

nmds of genetic fingerprints of bulk soil derived from T-RFLP for Bacteria (a), Archaea (b) and Fungi (c). Please note the single 18-month outlier in (b).

nmds of genetic fingerprints of bulk soil derived from T-RFLP for Bacteria (a), Archaea (b) and Fungi (c). Please note the single 18-month outlier in (b).

Fig. 4

nmds of genetic fingerprints of bulk soil derived from T-RFLP for Bacteria (a), Archaea (b) and Fungi (c). Please note the single 18-month outlier in (b).

nmds of genetic fingerprints of bulk soil derived from T-RFLP for Bacteria (a), Archaea (b) and Fungi (c). Please note the single 18-month outlier in (b).

The nmds analyses of the particle-attached microbial communities revealed, with incubation periods of 6 and 18 months for Bacteria, clear differences between the finest fraction and the two other fractions (Fig. 5). For Archaea, this effect was only pronounced after 18 months, while a comparably strong effect was not seen for Fungi. In contrast, only the fungal communities detected at the particle size fractions responded to the mineral composition, with differences between MT and IL plus IL+FH after 6 months. In accordance with bulk soil, nmds indicated an additional differentiation between IL and IL+FH after 18 months, which, however, was not supported by anosim (Table 3).

Fig. 5

$nmds of T-RFLP profiles as obtained from DNA extracted from the respective particle size fractions for Bacteria (a), Archaea (b) and Fungi (c) after 6 months (left) and 18 months (right) of incubation.$

nmds of T-RFLP profiles as obtained from DNA extracted from the respective particle size fractions for Bacteria (a), Archaea (b) and Fungi (c) after 6 months (left) and 18 months (right) of incubation.

Fig. 5

anosim of the same data set revealed that the effect of mineral composition on Bacteria was restricted to the finest fraction. Similarly, Fungi responded to the minerals in this fraction, but, additionally, also in the coarsest fraction. In contrast, there was no fraction with responsive Archaea. For Bacteria, the differences caused by the minerals diminished over time, from 6 to 18 months, which was not equally reflected for Fungi or Archaea. A strong change of the archaeal community over time was detected in the finest but not in the other two particle size fractions analysed. anosim revealed an effect of particle size fractions on the fungal diversity, but this was restricted to the two illite containing soil variants.

Discussion

Microbial colonisation of artificial soils was followed in this study over a period of 18 months. The artificial soils clearly showed characteristics of natural soils (Heister et al., 2012; Pronk et al., 2012, 2013). The population sizes of the three microbial domains, as revealed by qPCR of rRNA genes or ITS1 sequences and the diversity of their dominant community members seen with T-RFLP profiling, which developed during this period of ageing from inoculation of the same microbial consortium, was comparable to studies with natural soils (Neumann et al., 2013). The selected three soil variants allowed a distinction of effects caused by the minerals montmorillonite (MT) and illite (IL) and, in addition, for illite the influence of ferrihydrite (IL+FH). It was found that the community changed in quantity and composition during the monitoring period, and these changes were affected by the respective mineral compositions. In contrast, previous analyses of important physicochemical parameters, including pH (7.4–7.7), macro-aggregation, and soil organic matter composition and distribution showed no differences between these soil samples when they were collected at the same dates (Pronk et al., 2012, 2013). This suggests that, in fact, the particular mineral compositions directly affected the abundance and diversity of the three microbial domains.

For samples from the same experiment, analysed by DGGE-profiling of rRNA gene fragments after only 3 months of incubation, Ding et al. (2013) had already identified the clay mineral as main driver for the bacterial community structure. However, after 12 months, the bacterial community structure appeared to be temporarily affected by the presence of the iron oxide (Babin et al., 2013). In this study, based on T-RFLP profiling of the same genes, the bacterial community structure responded to the clay minerals after 6 months, but, after 18 months, these effects diminished, suggesting a decreasing relevance of the mineral composition in favour of differences in soil organic matter quality (see below). For Fungi and samples analysed after 12 months of incubation, DGGE profiles revealed differences between clay minerals (Babin et al., 2013).

The T-RFLP analyses of the fungal communities in this study demonstrate that the differences between MT and IL were already pronounced after 6 months, and were maintained later on, as seen with the samples aged 18 months. These mineral-specific effects on the bacterial and fungal community structures were most likely triggered by differences in elemental compositions and physicochemical characteristics of MT and IL (Heister et al., 2012; Pronk et al., 2012, 2013), which suggest that these materials provide quite different surface properties and thus contrasting selective conditions for microbial colonisation. The fact that Bacteria and Fungi responded differently to the mineral compositions is in line with results from other studies, which demonstrated shifts in community composition of Bacteria by addition of mica, while Fungi responded to addition of basalt and rock phosphate (Carson et al., 2007).

No selective effect of the different clay minerals was found for Archaea, which may suggest that members of this domain have no strong interactions with these clay minerals. However, clay minerals in general supported the establishment of archaeal populations as revealed by the increase of rRNA genes associated with the smallest particle size fraction between 6 and 18 months. Furthermore, it should be noted that the diversity and, therefore the resolution obtained with only one restriction endonuclease (see Materials and Methods), was very low and may thus have overlooked more subtle changes. The method was, however, good enough to detect a change over time in a comparison of T-RFLP of the 6 and 18 months-aged samples. Interestingly, this differentiation only took place at the finest particle size fraction, where it was accompanied by an increase in the archaeal rRNA gene copy numbers, indicating their growth.

There was no time-dependent effect on the fungal community structure, which may have been masked by the high variability between replicates from the same soil variants. In contrast, the bacterial community changed, and these alterations occurred in all three particle size fractions analysed. These time-dependent responses correlate with a loss of soil organic carbon, which can mainly be attributed to respiration, and a change of soil organic matter quality (Pronk et al., 2012, 2013).

Generally, the abundances of the microbial domains, as revealed by qPCR of their respective rRNA genes or ITS1 sequences, increased in the order MT < IL < IL+FH. In contrast, the specific surface area (SSA) of bulk soil was shown to increase in the order IL < MT < IL+FH (Pronk et al., 2012). Assuming a correlation between microbial abundances and SSA, as demonstrated by natural soils (Neumann et al., 2013), it appears that IL was better colonised by all three domains than MT. One explanation could be the presence of iron as a macronutrient which is lacking in MT but present in IL, and, even more, in IL+FH. The higher amount of iron in IL+FH in comparison to IL, however, only correlated with a higher abundance of Bacteria and Fungi, and not of Archaea.

A further approach in this study to characterise the importance of the minerals for the selection of soil microbial communities was the particle size fractionation. The recovery of the DNA from the fractions was 70% or less, compared to amounts extracted from bulk soil and, thus, lower than previous findings from natural soil (Neumann et al., 2013). As the microbial inoculum represented an easily extractable fraction of the original community (Ding et al., 2013), this may have excluded some microorganisms strongly attached to particle surfaces. Furthermore, the incubation of 18 months may not have enriched microbial communities as strongly attached to particles as in a naturally aged soil. Considering that microbial community members could differ in their stickiness to the particle surfaces, the fractionation procedure applied in this study may have introduced a bias, for example, by overlooking species which might have been detached from one mineral, but not from another. However, the finest fraction, independent of the mineral composition, generally reflected the results of bulk soil.

In contrast to the finest fraction, the two coarser fractions consisted almost entirely of quartz (see Table 1), and thus, differences in the finest fraction should be more indicative of effects by the mineral composition than those in both coarser fractions. Indeed, the fractionation procedure yielded different microbial communities attached to the respective particle size fraction. In accordance to the bulk soil data, the fungal communities responded to the different minerals, and this response was seen across all particle size fractions. As fungal mycelium participates in aggregation of soil particles (Stotzky, 1986), it is obvious that Fungi interact simultaneously with particles of several sizes. Thus, even if filamentous Fungi were enriched in a coarser fraction, they might also have experienced a direct interaction with clay particles before. As clay particles can bind to mycelia (Stotzky, 1986), this might explain why the effect of the mineral composition on Fungi was seen across all fractions.

In contrast, Bacteria and Archaea responded to the clay minerals, as expected, only in the finest fraction. In a similar experiment, where, however, the coarse fraction contained the discriminating minerals and the fine fraction consisted only of quartz, responses for Bacteria were found only in the coarse fraction (Carson et al., 2009). Interestingly, in our study, the bacterial response was more pronounced after 6 than after 18 months, while for archaeal growth, it was the opposite. These time-dependent responses may reflect the different growth rates of Bacteria and Archaea in soil, considering that Bacteria, in contrast to Archaea, include typical r-strategists (Fierer et al., 2007).

In summary, this study demonstrates, with the use of artificial soils, the importance of clay minerals for the establishment of distinct microbial communities. Changes in the community structure were not necessarily accompanied by changes in abundance and vice versa. Bacteria, Archaea and Fungi responded differently to the three compositions. The presence of ferrihydrite generally fostered microbial abundances. While Fungi responded across particle size separates to the different clay minerals, the impact of clay minerals on Bacteria and Archaea was found in the finest fraction, suggesting the importance of particle size separates. Thus, the specific selective effects of mineral composition and particle size fractions indicate that these factors strongly contribute to niche separation. The interactions between specific minerals may contribute to a high microbial diversity in natural soils with complex mineral compositions and different textures.

Acknowledgements

We thank Britta Müller, Karin Trescher, Jana Usarek, Claudia Wiese and especially Frederike Imbusch for their excellent assistance, and Anja Dohrmann, Doreen Gabriel, Sebastian Klimek and Astrid Näther for discussion. We also thank Joanna Hanzel and Kai Totsche for their support of the DFG Priority Programme SPP1315 of the German Research Foundation, which financially supported this study (Grant number TE 383/3-2).

References

(

2011

)

Accumulations of major and trace elements in particle size fractions of soils on eight different parent materials

Geoderma

161

–

(

1984

)

Distribution of organic matter and adenosine triphosphate after frationation of soil by physical procedures

Soil Biol Biochem

465

–

470

(

1998

)

Carbon, nitrogen, and sulfur pools in particle-size fractions as influenced by climate

Soil Sci Soc Am J

172

–

181

(

2004

)

Experimental measurements of the adsorption of Bacillus subtilis and Pseudomonas mendocina onto Fe-oxyhydroxide-coated and uncoated quartz grains

Geomicrobiol J

511

–

519

(

2013

)

Metal oxides, clay minerals and charcoal determine the composition of microbial communities in matured artificial soils and their response to phenanthrene

FEMS Microbiol Ecol

–

(

2001

)

Silicates, silicate weathering, and microbial ecology

Geomicrobiol J

–

(

1997

)

Reclassification of the crenarchaeal orders and families in accordance with 16S rRNA sequence data

Int J Syst Bacteriol

657

–

660

(

2007

)

Altering the mineral composition of soil causes a shift in microbial community structure

FEMS Microbiol Ecol

414

–

423

(

2009

)

Minerals in soil select distinct bacterial communities in their microhabitats

FEMS Microbiol Ecol

381

–

388

(

2004

)

Rock fragments in soil support a different microbial community from the fine earth

Soil Biol Biochem

1119

–

1128

(

1992

)

Physical fractionation of soil and organic matter in primary particle size and density separates

Adv Soil Sci

–

(

2013

)

Mineral composition and charcoal determine the bacterial community structure in artificial soils

FEMS Microbiol Ecol

–

(

2007

)

Toward an ecological classification of soil Bacteria

Ecology

1354

–

1364

(

1993

)

ITS primers with enhanced specificity for Basidiomycetes – Application to the identification of mycorrhizae and rusts

Mol Ecol

113

–

118

(

2005

)

Characterization of fungal community structure on a weathered pegmatitic granite

Microb Ecol

360

–

368

(

2006

)

Characterization of bacterial community structure on a weathered pegmatitic granite

Microb Ecol

526

–

534

(

1998

)

Diversity and structure of the methanogenic community in anoxic rice paddy soil microcosms as examined by cultivation and direct 16S rRNA gene sequence retrieval

Appl Environ Microbiol

960

–

969

(

2001

)

past: Paleontological software package for education and data analysis

Palaeontol Electronica

–

(

2012

)

NanoSIMS as a tool for characterizing soil model compounds and organomineral associations in artificial soils

J Soils Sediments

–

(

1991

)

Components and microbial biomass content of size fractions in soils of contrasting aggregation

Geoderma

–

(

1986

)

Distribution of microorganisms, total biomass, and enzyme activities in different particles of brown soil

Microb Ecol

205

–

215

(

2000

)

Structure and function of the soil microbial community in microhabitats of a heavy metal polluted soil

Biol Fertil Soils

390

–

400

(

1991

)

16S/23S rRNA sequencing

Nucleic Acid Techniques in Bacterial Systematics

( , eds), pp.

115

–

175

John Wiley & Sons Ltd

New York, NY

(

1995

)

Denitrifiers and denitrifying activity in size fractions of a Mollisol under permanent pasture and continuous cultivation

Soil Biol Biochem

–

(

1997

)

Characterization of microbial diversity by determining terminal restriction fragment length polymorphisms of genes encoding 16S rRNA

Appl Environ Microbiol

4516

–

4522

(

1937

)

Synthetic soil as medium for the study of certain microbiological processes

Soil Sci

319

–

322

(

2005

)

Fungal-specific PCR primers developed for analysis of the ITS region of environmental DNA extracts

BMC Microbiol

(

2005

)

Exploring the enzymatic landscape: distribution and kinetics of hydrolytic enzymes in soil particle-size fractions

Soil Biol Biochem

–

(

2007

)

Mineralogic control on abundance and diversity of surface-adherent microbial communities

Geomicrobiol J

167

–

177

(

2013

)

Response of microbial communities to long-term fertilization depends on their microhabitat

FEMS Microbiol Ecol

–

(

2014

)

Importance of soil organic matter for the diversity of microorganisms involved in the degradation of organic pollutants

ISME J

1289

–

1300

(

2012

)

Community structure of Bacteria on different types of mineral particles in a sandy soil

Soil Sci Plant Nutr

562

–

567

(

1991

)

Aggregate hierarchy in soils

Aust J Soil Res

815

–

828

et al. . (

2013

)

vegan: Community Ecology Package

. R package version 2.0-7.

(

2003

)

Micro-scale distribution of microorganisms and microbial enzyme activities in a soil with long-term organic amendment

Eur J Soil Sci

715

–

724

(

2012

)

Development of biogeochemical interfaces in an artificial soil incubation experiment; aggregation and formation of organo-mineral associations

Geoderma

189

585

–

594

(

2013

)

Is turnover and development of organic matter controlled by mineral composition?

Soil Biol Biochem

235

–

244

R Core Team

(

2012

)

R: A Language and Environment for Statistical Computing

R Foundation for Statistical Computing

Vienna

(

2004

)

Inhibition and enhancement of microbial surface colonization: the role of silicate composition

Chem Geol

212

313

–

327

(

1998

)

Feldspars as a source of nutrients for microorganisms

Am Mineral

1532

–

1540

(

1991

)

Iron Oxides in the Laboratory. Preparation and Characterisation

VCH Verlag GmbH

Weinheim

(

2001

)

Microbial population structures in soil particle size fractions of a long-term fertilizer field experiment

Appl Environ Microbiol

4215

–

4224

(

1998

)

Organic matter and enzyme activity in particle-size fractions of soils obtained after low-energy sonication

Soil Biol Biochem

–

(

1986

)

Influence of soil mineral colloids on metabolic processes, growth, adhesion, and ecology of microbes and viruses

Interactions of Soil Minerals with Natural Organics and Microbes: Special Publication No 17

( , eds), pp.

305

–

428

Soil Science Society of America

Madison, WI

(

1990

)

Amplifiation and direct sequencing of fungal ribosomal RNA genes for phylogenetics

PCR Protocols: A Guide to Methods and Applications

( , eds), pp.

315

–

322

Academic press, Inc.

San Diego, CA

(

2005

)

Group-specific primer and probe sets to detect methanogenic communities using quantitative real-time polymerase chain reaction

Biotechnol Bioeng

670

–

679

(

2007

)

Changes in microbial community structure and function within particle size fractions of a paddy soil under different long-term fertilization treatments from the Tai Lake region, China

Colloids Surf B Biointerfaces

264

–

270

(

2007

)

Effect of long-term fertilization on C mineralization and production of CH₄ and CO₂ under anaerobic incubation from bulk samples and particle size fractions of a typical paddy soil

Agric Ecosyst Environ

120

129

–

138

Author notes

Editor: Cindy Nakatsu

German Research Foundation

TE 383/3-2

What Are Artificial Soil Mixes Made of

Source: https://academic.oup.com/femsec/article/90/3/770/540824

What Are Artificial Soil Mixes Made of

Abstract

Introduction

Materials and methods

Artificial soil preparation

Particle size fractionation

DNA extraction

Quantitative PCR (qPCR) of microbial domains

Genetic profiling of microbial communities

Measurement of carbon and nitrogen content

Statistical analysis

Results

Carbon and nitrogen contents and microbial colonisation of the artificial bulk soils

Carbon, nitrogen and microorganisms associated with the particle size fractions

Responses of the microbial community structure to the soil variants

Discussion

Acknowledgements

References

Author notes

0 Response to "What Are Artificial Soil Mixes Made of"

Post a Comment

Iklan Atas Artikel

Iklan Tengah Artikel 1

Iklan Tengah Artikel 2

Iklan Bawah Artikel