GeoChip as a metagenomics tool to analyze the microbial gene diversity along an elevation gradient.

To examine microbial responses to climate change, we used a microarray-based metagenomics tool named GeoChip 4.0 to profile soil microbial functional genes along four sites/elevations of a Tibetan mountainous grassland. We found that microbial communities differed among four elevations. Soil pH, temperature, NH4 (+)-N and vegetation diversity were four major attributes affecting soil microbial communities. Here we describe in details the experiment design, the data normalization process, soil and vegetation analyses associated with the study published on ISME Journal in 2014 [1], whose raw data have been uploaded to Gene Expression Omnibus (accession number GSM1185243).

Direct link to deposited data

Experimental design, materials and methods

Description of the sites

This experiment was conducted at an alpine meadow in the Haibei Alpine Meadow Ecosystem Research Station of Chinese Academy of Science, which is located in a large valley surrounded by the Qilian Mountain of the northeast of Qinghai-Tibet Plateau (37°37′N, 101°12E′)in Qinghai province. It has a typical highland continental climate with cold and long winter but cool and short summer. The annual mean air temperature recorded at the station is − 1.7 °C [2]. The day/night temperature variation is substantial due to strong sun radiation. The annual mean precipitation is 560 mm and 85% of rainfall is within the growing season from May to September [3].

The dominant soil type at the station is Mat Cryic Cambisols (a typical alpine grassland soil) and its pH values are 7.3 and 7.4 at depths of 10 and 20 cm, respectively. Aboveground plant biomass increases from May to July, reaches the maximum level in late July and early August, and withers in early October. Over 80% of vegetation species use C3 photosynthetic pathway for carbon fixation [3].

This experiment, designed to study the effects of climate changes with the space-substitutes-time strategy, was set at four sites/elevations of 3200, 3400, 3600 and 3800 m in May 2006. The spatial distances between adjacent sites are 6.2 km (3200–3400 m), 4.2 km (3400–3600 m) and 1.3 km (3600–3800 m), respectively.

The sites have typical vegetation and soil attributes for their respective elevations. The alpine meadow plant community at 3200 m is largely dominated by Kobresia humilis, Elymus nutans, Stipa aliena, Potentilla anserine and Thalictrum alpinum. The plant community at 3400 m is primarily dominated by Potentilla fruticosa shrub meadow and grass species of K. humilis, E. nutans and Festuca ovina. The plant community at 3600 m site is dominated by K. humilis, Potentilla nivea, Thalictrum alpinum, Carex atrofusca, Poa crymophila and P. fruticosa. At the 3800 site, the plant community is dominated by K. humilis, P. crymophila, Androsace mariae, Polygonum macrophyllum and Kobresia pygmaea. Due to the short growth period, aboveground plant biomass has low primary production and diversity.

Three 1.0 × 1.0 × 0.3 m3 plots were fenced at each elevation/site to prevent disturbance. The distance between two adjacent plots was roughly 0.6 m. In August 2009, soil at a depth of 0–20 cm was collected from all plots. Briefly, soil samples were collected randomly at five locations of every plot to ensure homogeneity. Then soil cores were mixed thoroughly on a clean tray. After materials such as roots, stones, pebbles and gravels were removed, soil was combined into a composite sample. Soil was sieved through a 2 mm sieve and stored at 4 °C for soil attribute measurements or − 80 °C until DNA extraction. All tools were sterilized with 70% alcohol.

DNA extraction

Soil metagenomic DNA was extracted using a FastDNA spin kit for soil (MP Biomedical, Carlsbad, CA, USA) following the manufacturer's instructions and precipitated with 100% ethanol and 0.3 M NaOAc. DNA purity was assessed by UV absorbance ratios of A260/A280 (> 1.8) and A260/A230 (> 1.7), and DNA concentrations were measured with a PicoGreen method [4].

GeoChip 4.0 experiment

The labeling and hybridization of soil DNA were conducted as previously prescribed [5]. A total of 2 μg extracted DNA was mixed with 20 μl random primers, containing 2.5 μl deoxynucleoside triphosphate (dNTP) (5 mM dATP/dGTP/dCTP, 2.5 mM dTTP), 1 μl Cy5 dUTP (Amersham, Piscataway, NJ) and 80 U of the large Klenow fragment (Invitrogen, Carlsbad, CA). Then DNA mixture was treated at 99.9 °C for 5 min and chilled immediately to denature DNA, followed by addition of 2.5 μl of water and incubation at 37 °C for 3 h. Finally, the mixture was heated at 95 °C for 3 min to terminate DNA labeling.

Labeled DNA was purified using the QIA quick purification kit (Qiagen, Valencia, CA, USA) following manufacturer's instructions and measured by NanoDrop ND-1000 spectrophotometer to assess labeling efficiency. DNA was then dried in the SpeedVac (ThermoSavant, Milford, MA, USA) at 45 °C for 45 min.

DNA hybridization

Labeled DNA was dissolved in 50 μl hybridization buffer (40% formamide, 25% SSC, 5 μg of unlabeled herring sperm DNA [Promega, Madison, WI], and 0.1% SDS) and 2 μl universal standard DNA (0.2 pmol μl− 1) labeled with fluorescent dye Cy5. The samples were then mixed by vortexing, incubated at 95 °C for 5 min, and maintained at 50 °C until hybridization. Microarrays were scanned by a NimbleGen MS 200 Microarray Scanner (Roche NimbleGen, Madison, WI) for approximately 16 h at 42 °C. Then scanned images were quantified by NimbleScan software as previously described [6].

Raw data processing

Data of signal intensities were uploaded to the laboratory's Microarray Data Manager System (http://ieg.ou.edu/microarray/) [1], [5], [6]. Then we processed them in the following steps: (i) spots of poor quality were removed, which were flagged as 1 or 3 by ImaGene (Arrayit, Sunnyvale, CA, USA) or with a signal to noise ratio of less than 2.0; (ii) the relative abundance of each sample was calculated by dividing the total intensity of the microarray, then multiplying by a constant and applying natural logarithm transformation; and (iii) probes detected in only one out of three replicates were removed to improve data quality.

Statistical analysis

Principal component analysis (PCA) was used to measure the overall functional gene structure. Bray–Curtis distance was used to obtain dissimilarity matrices in the adonis algorithm of the dissimilarity test for comparing GeoChip data of four elevations. The similarity test, Mantel test, Canonical correspondence analysis (CCA) and Variation partitioning analysis (VPA) were used to evaluate the linkages between microbial gene compositions and environmental attributes. In the similarity test, Euclidean distance was used to calculate the distance between samples, followed by calculation of Pearson correlation coefficient. To select attributes in CCA modeling, we used variation inflation factors (VIF) to examine whether the variance of canonical coefficients was inflated by the presence of correlations with other attributes. If an attribute had a variation inflation factor value larger than 20, we deemed it to depend on other attributes and consequently removed it from the CCA modeling. Correlation coefficients (r) were calculated using Pearson's correlation. The normalized total gene abundance for each functional gene was the average of the total gene abundance from all the replicates and all data are presented as mean ± s.e. The least significant difference (LSD) test was used to compare the significance of differences in relative abundance among four elevations. All of the analyses were performed with the Vegan package (v.1.15-1) using R, version 2.8.1 (R Foundation for Statistical Computing, Vienna, Austria).

Discussion

Here we describe a dataset of GeoChip 4.0 for profiling functional potentials of microbial community along four elevations in a grassland of the Tibetan plateau (Table 1). GeoChip is comprised of approximately 82,000 probes covering 410 functional gene families related to microbial carbon, nitrogen, sulfur, phosphorus cycling and others [6]. We showed that microbial gene abundances were correlated with greenhouse gas emissions. Therefore, it is possible to assess soil biogeochemical cycles based on measurements of microbial gene abundance. Furthermore, it can be used to predict the impact of further climate changes in this region on functional potentials of microbial communities.

Table 1

Number of detected genes at four elevations.

Gene categories	3200	3400	3600	3800	All elevationsa
Antibiotic resistance	903	1668	1548	1495	1818
Bacteria phage	158	424	358	336	468
Bioleaching	192	389	341	329	431
Carbon cycling	2979	5953	5485	5384	6476
Energy process	258	516	479	471	559
Metal Resistance	2832	5352	4999	4887	5757
Nitrogen	2110	4181	3889	3820	4520
Organic Remediation	5926	10,692	10,333	10,122	11,490
Other category	561	1192	1101	1069	1324
Phosphorus	370	764	687	682	829
Stress	5389	11,010	9828	9616	11,958
Sulfur	773	1724	1591	1553	1894
Virulence	918	1820	1643	1574	1996
Total	23,369	45,685	42,282	41,338	49,520

The number of genes detected at any of all four elevations.

Conflict of interest

The authors declare that there is no conflict of interest on any work published in this paper.

Specifications
Organism	Uncultured bacterium
Sequencer or array type	GeoChip 4.0
Data format	Raw data: TXT, normalized data: TXT
Experimental factors	Soil samples were collected from four elevations: 3200 m, 3400 m, 3600 m and 3800 m.
Experimental features	Profiling microbial functional potentials with a microarray-based metagenomics tool named GeoChip 4.0 along an elevation gradient in a Tibetan grassland.
Consent	n/a
Sample source location	The Haibei Alpine Meadow Ecosystem Research Station (37°37′N, 101°12E′), Qinghai, China,