Metagenomic evaluation of a Utah tar sand microbiota suggests the predominant hydrocarbonoclastic role of Actinobacteria

Microbial community analysis by genomic DNA extraction and 454 pyrosequencing

The University of Alberta Geotechnical Centre, Canada, kindly provided the tar sands sample. All samples were shipped on ice to Florida A&M University where they were processed for metagenomics and CLPP analyses. Genomic DNA was extracted from the samples using the PowerSoil DNA Isolation kit (MoBio Laboratories, Inc., Carlsbad, CA, USA). Bar-coding pyrosequencing of the 16S rRNA gene V4 region was performed using the Earth Microbiome Project (EMP) standard protocols (http://www.earthmicrobiome.org/emp-standard-protocols) and sequenced on a Roche 454 FLX instrument (Roche, Indianapolis, IN, USA) with titanium reagents, following manufacturers recommended procedures. Sequences that passed quality controls were uploaded to MOTHUR v.1.33.3, where tags, low-quality sequences and chimeric reads were removed.

CLPP analysis

CLPP of the tar sand sample was analyzed using BIOLOG™ Ecoplates, as described previously^8,12. Triplicate samples of the BIOLOG™ Ecoplate were used for this experiment. A total of three sets of 1 g samples of the tar sand were shaken in 99 ml of distilled sterile water for 20 min at 20°C. The samples were then incubated at 4°C for 30 min and then centrifuged for 10 min at 500g. The supernatant of the tunes were pooled and thoroughly mixed. In total, 150 μl of the supernatant was inoculated into each well of the Biolog EcoPlates. This ensured that each of the 96-well plate contained three replicates of each sample exposed to 31 different carbon sources. Water blanks were included with each plate and the plates were incubated at 25°C without shaking. Substrate utilization was monitored by measuring absorbance at 595 nm using a BIOTEK UQuant Microplate Spectrophotometer, (BIO-TEK Instruments, Inc., USA). The first measurement was made immediately after inoculation and subsequent readings were taken after every 24 h interval for 7 days. The measurements of individual substrates were corrected for background absorbance by subtracting the absorbance of the control (water) samples. If a negative number was obtained, it was manually set to zero. A well was considered positive if the mean OD_595nm exceeded that of 0.25.

In this study the 31 carbon sources were organized into groups 1–5 as described by Zak et al.¹³. Carbon sources originally grouped as miscellaneous by Zak et al.¹³ or those new to the BIOLOG ™ EcoPlate, were grouped into one of the other five categories - carbohydrates, carboxy acids, amino acids, esters and polymers. Grouping the data into 5 guilds compresses a 31-dimensional data set into 5 dimensions, significantly reducing the complexity of the data and subsequent interpretation¹⁴.

The average well color development (AWCD) was used to assess the microbial response for all the carbon sources. The AWCD was determined as follows:

where OD_i is optical density value from each well, corrected subtracting the blank well (inoculated with water) values from each plate well^8,15. The 96-h data was used for statistical analysis of CLPPs.

The diversity of substrate utilization was calculated using the Shannon’s diversity index (H) and evenness (E)¹⁶. The Shannon-Weaver index is a measure of the capacity of the bacterial community to degrade the carbon sources in the well and can be considered as an index of the physiological diversity of the bacterial community in the tar sands. Higher values of H indicate the ability of the microbial community to degrade the substrates with a high efficiency¹⁶:

where P_i = (OD reading of well I)/(sum of all wells)], based on the OD in the Ecoplates.

The Shannon evenness (E) is a measure of how dissimilar the abundances of the species in a community are from each other¹⁷ and was calculated as:

                                                                              E – H/lnS

where S (Shannon richness) is the number of substrates used by the microbial communities. Further, the cluster analysis of the substrate utilization pattern was constructed using the nearest neighbor method with Euclidian distance to form linkage dendrogram.

Microbial community analysis of tar sands

In this study, the compositions of bacterial community of the tar sands was investigated by a pyrosequencing-based analysis of the 16S rRNA gene sequences. The gram-positive Actinobacteria (50%) were dominant in the tar sand followed by Betaproteobacteria (27%), Alphaproteobacteria (7%), Gammaproteobacteria (7%) and Acidobacteria (2%) (Figure 1). Actinobacteria are known for their role in the biodegradation of a variety of different pollutants including petroleum hydrocarbons^18,19. The predominant genus identified in these tar sands was Arthrobacter, followed by Dietzia, Janibacter, Nocardioides, Microbacterium, Agrococcus and Salinibacterium (Figure 2). Most of these genera have also been previously shown to degrade aromatic hydrocarbons, indicating that tar sands are a very active repository of hydrocarbonoclastic microorganisms. When the taxonomic affiliation of the obtained metagenomic sequences were investigated, we found that the gram-positive Arthrobacter spp. from the Actinobacteria phyla were predominant, which comprised approximately half of the total microbial community assemblage in this particular Utah tar sand. Arthrobacter species possess a significant hydrocarbonoclastic potential as demonstrated by previous studies and we recommend that Actinobacteria, native to the tar sand habitats, should be targeted for future research on this area.

Figure 1. Taxonomic distribution of the obtained metagenome sequences shown at the phylum level.

The gram-positive Actinobacteria were dominant in the tar sand followed by Betaproteobacteria, Alphaproteobacteria, Gammaproteobacteria and Acidobacteria.

Figure 2. Taxonomic distribution of the obtained metagenome sequences shown at the genera level.

The gram-positive Arthrobacter spp. from the Actinobacteria phyla were dominant in the tar sand microbiota.

Microbial community metabolic analyses

Biolog® Ecoplates were used to evaluate the functional diversity of the microbial community of the tar sands. The AWCD revealed sigmoidal relationships between the OD₅₉₀ and time for all carbon sources (polymers, carbohydrates, amino acids, esters) except the carboxylic acids (Figure 3). This pattern for the polymers, carbohydrates, amino acids, and esters is fundamentally similar to the bacterial growth curve. Yao et al²⁰ suggested that color development reflected species metabolic activity and the ability of the bacterial community to respond to substrates. The AWCD for all the substrates in the Biolog® Eco-microplates was 0.25. Less than 24h was needed for the first color development the microplate wells containing the polymers whereas the duration of the lag phase of the carboxylic acids was prolonged. The first color development of the carboxy acids (AWCD_0.2) was recorded at 48 h (Figure 3). The highest utilization was observed in the microplate wells containing the polymers (AWCD_0.78) after 96 hours. The observed differences in the substrate utilization pattern of five major carbon groups (Figure 3) might be due to the presence of different functional groups, such as carbohydrates R-C=O), amino acids (-NH₂ and -COOH), carboxylic acid (-COOH), esters, amines and phosphorylated acids (-COOR’, -NH₂), and polymers (–(CH₂-CH₂)–(R)_n). The high utilization of the recalcitrant polymers suggest that the polymers could be more easily utilized by the indigenous microbial community in the tar sands. After 96 h, there was a decrease in the rate of AWCD; thus, the analysis of functional diversity of microbial community of tar sands was completed at 96 h. The decreased rate may be due to the change of metabolic activity of active microbial communities utilizing the C-substrates. Cluster analysis of the substrates revealed a systematic grouping of the substrates based on their utilization pattern (Figure 4).

Figure 3. Dendrogram for hierarchical clustering of Biolog® Ecoplates substrates.

Figure 4. Substrate Utilization Pattern shown by Utah tar sand soils.

The Shannon-Wiener index is an indication of the spread or distribution of carbon source utilization by the microbial community²¹. Typical values of the Shannon-Wiener Index usually lies between 1.5 and 3.5, and rarely exceed 4.0²². In this study, the Shannon-Weiner index, H, of the microbial community in the tar sands was 3.072 and the Shannon evenness, E, was 1.009. These values indicate a more diverse microbial community and an even distribution of the species within the tar sands.