SEARCH

SEARCH BY CITATION

Fig. S1. A. Location of the study site off Iquique, Chile (Station #3, boxed; 20°07′S, 70°23′W).B. Vertical distributions of physical and chemical water properties at Station #3 in June 2008. Oxygen (O2), photosynthetically active radiation (PAR; time of day: 1215), and temperature (left panel) were sampled via conductivity-temperature-depth (equipped with a dissolved oxygen sensor) on June 16, at the start of the DNA/RNA sample collection (June 16–17). Water for nitrate (NO3-), nitrite (NO2-) and ammonium (NH4+) analyses (right panel) was collected via rosette during a surveying transect on June 14. DNA/RNA sampling depths are marked with dashed lines.

Fig. S2. Representative frequency distribution of 454 read density per gene (unique NCBI-nr reference sequence) in the 50 m DNA and RNA datasets. The vast majority of genes are represented by one or fewer reads in both datasets.

Fig. S3. Frequency distribution of reads matching the 100 most abundant protein-coding reference sequences recovered in BLASTX searches against the NCBI-nr database. Frequency (y-axis) is shown as a percentage of the total number of reads with matches in NCBI-nr. Values in the legend are the total percentages represented by the 100 most abundant genes. The total numbers of unique reference sequences per dataset are shown in Table 1 (main text).

Fig. S4. Discrepancies between abundance and expression for two prominent OMZ genera. Dashed lines indicate the relative abundance of DNA reads matching Nitrosopumilus (maritimus) and Pelagibacter protein-coding genes (as top hits). Solid lines indicate the expression ratio (RNA/DNA, see main text) averaged across all genes per genus. Pelagibacter dominates in DNA abundance, but has significantly lower per gene expression. Error bars are 95% confidence intervals.

Fig. S5. Taxonomic distribution of 16S ribosomal RNA gene fragments in metagenome (DNA) samples at station #3. Putative rRNA-encoding reads were identified via BLASTN searches against a rRNA database composed of both prokaryotic and eukaryotic small and large subunit rRNA nucleotide sequences (5S, 16S, 18S, 23S and 28S rRNA) from available microbial genomes and sequences in the ARB SILVA LSU and SSU databases (http://www.arb-silva.de). Reads greater than 100 nt in length that aligned with bit scores greater than 50 were designated as rRNA sequences and identified taxonomically using the program greengenes (http://greengenes.lbl.gov) and the NCBI taxonomy hierarchy. Numbers of 16S gene sequences are listed for each sample. DNA data from 15, 65, 500 and 800 m do not have corresponding RNA data and are therefore not the focus of the main text, but are provided here to help contextualize our primary results.

Fig. S6. Abundances of top hits (bit score > 50) within KEGG reference pathways (KEGG 2 hierarchy level) represented in RNA datasets from four depths in the OMZ. Abundance is shown as a percentage of total hits mapping across all KEGG 2 pathways. Orthologues are ordered by rank abundance in the 200 m sample (light blue bars).

Fig. S7. Abundances of top hits (bit score > 50) within KEGG reference pathways (KEGG 3 hierarchy level) represented in RNA datasets from four depths in the OMZ. Abundance is shown as a percentage of total hits mapping across all KEGG 3 pathways. Orthologues are ordered by rank abundance in the 200 m sample (light blue bars).

Fig. S8. Abundances of top hits to KEGG orthologues (bit score > 50) in RNA datasets from four depths in the OMZ, as a percentage of total hits mapping to the KEGG ko hierarchy. The figure shows only orthologues for which the percentage of total hits exceeds 0.25% in at least one of the four datasets. Orthologues are ordered by rank abundance in the 200 m sample (light blue bars). Gray shading marks proteins with putative transport functions. Red marks proteins of nitrogen metabolism or transport. Green marks proteins of photosynthesis and carbon fixation. Note: ammonia monooxygenase subunit genes (amoABC) predominantly matched the archaeal genes of Nitrosopumilus maritimus [Nmar_1500 (amoA), Nmar_1502 (amoC), Nmar_1503 (amoB)], which did not have an associated ko identifier and did not parse automatically to the ko hierarchy; these reads were manually extracted from the KEGG BLAST results and added to the totals shown here.

Fig. S9. Abundance and taxonomic classification of sequencing reads that match genes encoding Amt-like ammonium transporters and the subunits of ammonia monooxygenase (amoA, B, C combined). Abundance is shown as a percentage of total sequencing reads with matches in NCBI-nr. Pie charts show the proportion of bacterial, archaeal, or unclassified amt or amo sequences at each depth, based on annotations in NCBI-nr. Amt and amo are represented by a mixture of bacteria and archaea sequences in the DNA pool, but are dominated by archaea in the expressed gene pool.

Fig. S10. Diversity and abundance of sequencing reads matching narG, encoding the alpha subunit of the dissimilatory nitrate reductase. Colours and numbers (lower right for each chart) reflect distinct narG reference sequences in the NCBI-nr database, with the colours for the four most abundance reference taxa (legend) consistent across charts. Chart area reflects the relative abundance of reads matching narG (as the top hit) as a proportion of the total number of reads (in each dataset) with matches in NCBI-nr, peaking at 1.1% in the 200 m RNA sample from the core of the OMZ.

Fig. S11. Transposase abundance increases with depth. OMZ DNA and RNA reads matching transposase genes in NCBI-nr are shown as a percentage of total reads matching the NCBI-nr database. Transposase content in metagenomes along a 4000 m depth profile in the North Pacific Subtropical Gyre (Station ALOHA, Hawaii Ocean Time-Series, HOTS) is shown for comparison, with transposase abundance as a proportion of total gene content in fosmid libaries (see fig. 3 in Konstantinidis et al., 2009, AEM 75:5345–5355). Note variation in depth scales.

Table S1. Total shared gene content across datatsets.

Table S2. Common NCBI-nr content in 454 datasets.

Table S3. Top 10 most abundant genes (protein-coding RNA) from four prominent OMZ taxa (Fig. 5, main text).

Table S4. Top 10 most highly expressed genes from four prominent OMZ taxa (Fig. 5, main text).

Table S5. Top genes identified via BLASTX comparisons with KEGG orthologue and NCBI-nr databases, ranked by expression ratio (exp) and transcript abundance (abun).

Table S6. Proportional abundance and expression of select nitrogen and sulfur metabolism genes.

Table S7. Taxonomic representation and abundance of dissimilatory APS reductase subunit A (aprA) genes.

Table S8. Taxonomic distribution and abundance of dissimilatory sulfite reductase (dsr) genes.

FilenameFormatSizeDescription
EMI_2400_sm_Fig_S1-S11_Tab_S1-S8.pdf1847KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.