Anoxic Carbon Flux in Photosynthetic Microbial Mats As Revealed by Metatranscriptomics

Anoxic carbon flux in photosynthetic microbial mats as revealed by metatranscriptomics

Luke C. Burow1,2,‡, Dagmar Woebken1,2,^‡, Ian P.G. Marshall1, Erika Lindquist3, Brad M. Bebout2, Leslie Prufert-Bebout2, Tori M. Hoehler2, Susannah G. Tringe3, Jennifer Pett-Ridge4, Peter K. Weber4, Alfred M. Spormann1, Steven W. Singer5

1Department of Chemical Engineering and Department of Civil and Environmental Engineering, Stanford University, Stanford, CA; 2Exobiology Branch, NASA Ames Research Center, Moffett Field, CA, USA; 3Joint Genome Institute, Walnut Creek, CA USA; 4Physical and Life Sciences Directorate, Lawrence Livermore National Laboratory, Livermore, CA, USA; 5Earth Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA USA

Supplemental Information

Amplification, sequencing and sequence analysis of 16S rRNA genes and transcripts

For construction of 454 pyrotag amplicon libraries from RNA samples, total RNA was converted to double-stranded cDNA using the SuperScript® Double-Stranded cDNA Synthesis Kit (Invitrogen). For 454 pyrotag libraries from the EN (7:00 am) DNA sample, DNA was extracted using the QIAmp DNA Mini Kit in combination with the above-mentioned homogenization and phenol-chloroform extraction. The V8 region of cDNA or DNA was PCR amplified using the universal primers 926F and 1392R(Lane et al., 1985, Weisburg et al., 1991). The reverse primer included the adaptor sequence and a five-base barcode.Amplicons of the 16S rRNA V8 hypervariable region were constructed and sequenced following standard procedure at the Department of Energy Joint Genome Institute using the GS FLX Titanium Series Reagents (454 Life Sciences, Branford, CT, USA)(Engelbrektson et al., 2010). Pyrosequencing amplicons were processed (sorting, trimming, removing reads of low quality and classification) using the RDP pipeline with standard settings (

Nearly full-length16S rRNA clone libraries were constructed from double-stranded cDNA by amplification with the broadly inclusive bacterial primers 27F and 1391R (Weisburg et al., 1991). PCR products of triplicate PCR reactions were pooled and purified using the Qiagen Min Elute PCR Purification Kit (Qiagen). Ligation, transformation,Sanger sequencing, as well as sequence quality check and contig formation were conducted following standard procedure at the Department of Energy Joint Genome Institute(JGI, Walnut Creek, CA, USA) ( 16S rRNAtranscript sequences were taxonomically classified using the RDP pipeline ( For phylogenetic analysis, sequences were aligned using the SILVA website ( and imported in the ARB program (Ludwig et al., 2004) with the SILVA 94 Ref database. The database was supplemented for the analysis of sequences of particular interest (Cyanobacteria and Chloroflexi) by adding the closest relatives to the Elkhorn Slough sequences as well as sequences from cultured representativesfrom SILVA Ref 104 and NCBI ( Phylogenetic trees were calculated using maximum likelihood, maximum parsimony and neighbor joining algorithms, with and without a 50% position variability filter. Bootstrap values were calculated in Geneious 5.0.3 with the PhyML algorithm, using 100 bootstrap trees.

Additional capture probes used in MICROBEnrich/MICROBExpress Kits

Cyano capture probe 1:

5’-AAAAAAAAAAAAAAAAAATCCCCACTGCTGCCTCCCGTAGG-3’

Cyano capture probe 2:

5’-AAAAAAAAAAAAAAAAAACAACATCTCACGACACGAGCTGA-3’

Cyano capture probe 3:

5’-AAAAAAAAAAAAAAAAAAGGTTCTTTTCACCTTTCCCTCGC-3’

Cyano capture probe 4:

5’-AAAAAAAAAAAAAAAAAAACTTACCCGACAAGGAATTTCGC-3’

Cyano capture probe 5:

5’-AAAAAAAAAAAAAAAAAAGAGCCGACATCGAGGTGCCAAAC-3’

Euk probe 1:

5’-AAAAAAAAAAAAAAAAAAACCGCGGCTGCTGGCACCAGACT-3’

Euk probe 2:

5’-AAAAAAAAAAAAAAAAAATTTCCCGTGTTGAGTCAAATTAA-3’

Euk probe 3:

5’-AAAAAAAAAAAAAAAAAAAACTAAGAACGGCCATGCACCAC-3’

Euk probe 4:

5’-AAAAAAAAAAAAAAAAAACAACTTTCCCTCACGGTACTTGT-3’

Euk probe 5:

5’-AAAAAAAAAAAAAAAAAATGTTTTAATTAGACAGTCGGATT-3’

Euk probe 6:

5’-AAAAAAAAAAAAAAAAAATGGTAACTTTTCTGACACCTCTT-3’

Euk probe 7:

5’-AAAAAAAAAAAAAAAAAATCGCAGAATTCACTACGACGCCA-3’

Supplemental Tables

Supplemental Table 1Overview of V8 pyrotag sequences derived from the upper most 2 mm of Elkhorn Slough microbial mats before and after quality assessment.

Supplemental Table 2 Statistics of the metatranscriptome reads of Elkhorn Slough microbial mats at 9:00 pm (12th January 2009, BN) and 7:00 am (13th January 2009, EN).

Supplemental Table 3 Estimations of coverage based on reads assigned to microbial RefSeq sequences and unambiguous assignment by MEGAN to SEED subsystems and NCBI species. BN = 21:00, EN = 7:00

Microbial RefSeq / SEED Functional Roles / MEGAN Species
BN / EN / BN / EN / BN / EN
Total Reads Assigned / 157525 / 98894 / 92370 / 58178 / 55935 / 22466
Observed Clusters / 53478 / 34463 / 1698 / 1522 / 422 / 343
ACE-predicted Clusters / 238381 / 150851 / 1885 / 1719 / 422 / 343
ACE Coverage % / 22.4% / 22.8% / 90.0% / 88.5% / 100.0% / 100.0%
Chao-predicted Clusters / 137799 / 87429 / 1936 / 1724 / 422 / 343
Chao Coverage % / 38.8% / 39.4% / 87.7% / 88.3% / 100.0% / 100.0%

Supplemental Table 4 Gene assignments and SEED functional categories for Microcoleus spp. fermentation pathway and glycogen to polyhydroxyalkonoate conversion by the Chloroflexales.

Abbreviation / Gene Product
(gene abbrieviation) / SEED Functional Categories
LDH / Lactate dehydrogenase
(ldh, ykgF) / L-lactate dehydrogenase (EC 1.1.2.3), D-Lactate dehydrogenase (EC 1.1.1.28), Predicted L-lactate dehydrogenase, Iron-sulfur cluster-binding subunit YkgF
PFL / Pyruvate-formate lyase (pfl) / Pyruvate formate-lyase (EC 2.3.1.54)
GP / Glycogen phosphorylase (glg) / Glycogen phosphorylase (EC 2.4.1.1)
ADH / Alcohol/Acetaldehyde dehydrogenase (adhE) / Alcohol dehydrogenase (EC 1.1.1.1): Acetaldehyde dehydrogenase (EC 1.2.1.10)
ACK / Acetate kinase (ack) / Acetate kinase (EC 2.7.2.1)
PFR / Pyruvate:Ferredoxin oxidoreductase (pfo) / Pyruvate-flavodoxin oxidoreductase (EC 1.2.7.-)
HoxH / Bidirectional [NiFe] hydrogenase(hoxH) / NAD-reducing hydrogenase subunit HoxH (EC 1.12.1.2)
HoxY / Bidirectional [NiFe] hydrogenase (hoxY) / NAD-reducing hydrogenase subunit HoxY (EC 1.12.1.2)
HoxU / Bidirectional [NiFe] hydrogenase (hoxU) / NAD-reducing hydrogenase subunit HoxU (EC 1.12.1.2)
HoxF / Bidirectional [NiFe] hydrogenase (hoxF) / NAD-reducing hydrogenase subunit HoxF (EC 1.12.1.2)
HoxE / Bidirectional [NiFe] hydrogenase (hoxE) / NAD-reducing hydrogenase subunit HoxE (EC 1.12.1.2)
PDC / Pyruvate dehydrogenase complex (pdh) / Pyruvate dehydrogenase E1 component alpha subunit (EC 1.2.4.1), Pyruvate dehydrogenase E1 component beta subunit (EC 1.2.4.1), Dihydrolipoamide acetyltransferase component of pyruvate dehydrogenase complex (EC 2.3.1.12), Dihydrolipoamide dehydrogenase component of pyruvate dehydrogenase complex (EC 1.8.1.4)
PhaA / Acetoacetyl CoA transferase
(phaA) / Acetyl-CoA acetyltransferase (EC 2.3.1.9)
PhaB / Acetoacetyl-CoA reductase
(phaB) / Acetoacetyl-CoA reductase (EC 1.1.1.36)
PhaC / Polyhydroxyalkanoate synthase (phaC) / Polyhydroxyalkanoic acid synthase; poly(R)-hydroxyalkanoic acid synthase, class III, PhaC subunit
ActP / Acetate permease (actP) / Acetate permease ActP (cation/acetate symporter)
LAP / Lactate permease (lldp) / L-lactate permease
PFK / 6-phosphofructokinase (pfk) / 6-phosphofructokinase (EC 2.7.1.11), 6-phosphofructokinase, eukaryotic type (EC 2.7.1.11)
GPI / Glucose-6-phosphate isomerase (gpi) / Glucose-6-phosphate isomerase (EC 5.3.1.9)
FBP / Fructose-1,6-bisphosphatase
(fbp) / Fructose-1,6-bisphosphatase, type I (EC 3.1.3.11), Fructose-1,6-bisphosphatase, GlpX type (EC 3.1.3.11)
FBA / Fructose-bisphosphate aldolase (fba) / Fructose-bisphosphate aldolase class II (EC 4.1.2.13), Fructose-bisphosphate aldolase class I (EC 4.1.2.13)
TPI / Triose phosphate isomerase (tim) / Triosephosphate isomerase (EC 5.3.1.1)
G3P / Glyceraldehyde-3-phosphate dehydrogenase (gap) / NAD-dependent glyceraldehyde-3-phosphate dehydrogenase (EC 1.2.1.12); Glyceraldehyde-3-phosphate dehydrogenase (EC 1.2.1.12) (GAPDH)
PGK / Phosphoglycerate kinase (pgk) / Phosphoglycerate kinase (EC 2.7.2.3)
PGM / Phosphoglycerate mutase(pgm) / Phosphoglycerate mutase (EC 5.4.2.1)
ENS / Enolase (ens) / Enolase (EC 4.2.1.11)
PPS / Phosphoenolpyruvate synthase (pps) / Phosphoenolpyruvate synthase (EC 2.7.9.2)
PYK / Pyruvate kinase (pyk) / Pyruvate kinase (EC 2.7.1.40)

Supplemental Table 5Accession numbers in Microcoleus chthonoplastes PCC 7420 and

Chloroflexales genomes (Oscillochloris trichodes DG6, Chloroflexus aggregans DSM 9485, Roseiflexus castenholzii DSM 13941, Chloroflexus sp. Y-400-fl) for fermentation and PHA production pathways.

Abbreviation / M chthonoplastes Accession Number(s) / Chloroflexales Accession Numbers
LDH / ZP_05029329.1, ZP_05027132.1, ZP_05024030.1 / ZP_07687075.1, YP_002464022.1, YP_001430946.1, YP_002568646.1, YP_001278741.1
PFL / ZP_05031046.1
GP / ZP_05024411.1, ZP_05024271.1 / YP_002464170.1, YP_002568551.1, YP_001634362.1, YP_001431030.1, ZP_07687229.1, YP_001278475.1
ADH / ZP_05027137.1
ACK / ZP_05027154.1
PFR / ZP_05026241.1 / YP_001275570.1, YP_001434103.1, ZP_07686769.1
HoxH / ZP_05027024.1 / YP_002569037.1, YP_001634797.1, YP_002463799.1, YP_001431870.1, YP_001276440.1, ZP_07687015.1
HoxY / ZP_05027037.1
HoxU / ZP_05027023.1
HoxF / ZP_05027028.1
HoxE / ZP_05027052.1
PDC / ZP_05028143.1, ZP_05027952.1, ZP_05025210.1, ZP_05028044.1, ZP_05025437.1, ZP_05025524.1 / YP_002462952.1, YP_002569853.1, YP_001635576.1, ZP_07685943.1, YP_002569855.1, YP_001635578.1, YP_002462954.1, ZP_07684644.1, YP_001275667.1, YP_001431750.1
PhaA / YP_001278635.1, YP_001430842.1, ZP_07684826.1, YP_002569328.1, YP_001635078.1, YP_002463281.1
PhaB / YP_002569329.1, YP_001635079.1, YP_002463282.1, YP_001430841.1, YP_001278634.1, ZP_07684827.1
PhaC / YP_002571218.1, YP_002462002.1, YP_001636840.1, YP_001278834.1, YP_001430259.1, ZP_07684148.1
ActP / ZP_05025509.1 / ZP_07686543.1, YP_002461629.1, YP_002571351.1, YP_001636967.1, ZP_07684384.1
LAP / YP_001431699.1, YP_001276400.1
PFK / ZP_05027350.1, ZP_05028256.1 / YP_001432611.1, YP_001276170.1, ZP_07684184.1, YP_002568694.1, YP_001634491.1, YP_002464006.1, ZP_07684183.1, YP_001431984.1, YP_001277056.1, ZP_07684183.1
GPI / ZP_05030931.1
FBP / ZP_05023982.1, ZP_05027047.1 / YP_002571277.1, YP_001636895.1, YP_002462025.1, YP_001431328.1, YP_001278423.1, ZP_07685583.1
FBA / ZP_05028766.1
TPI / ZP_05029410.1 / YP_001278117.1, YP_001431118.1, ZP_07684429.1, YP_002571815.1, YP_001637392.1, YP_002462281.1
G3P / ZP_05024661.1 / ZP_07687131.1, ZP_07687132.1, YP_002464998.1, YP_002571711.1, YP_001637296.1
PGK / ZP_05030467.1 / YP_002570760.1, YP_002462015.1, YP_001636411.1, YP_001275819.1, YP_001432174.1, ZP_07685368.1
PGM / ZP_05023660.1 / YP_002463031.1, ZP_07684561.1, YP_002569529.1, YP_001635261.1, YP_001434297.1, YP_002463496.1, ZP_07684870.1, YP_001430323.1, YP_001278867.1, YP_002569059.1, YP_001634818.1, ZP_07685097.1, YP_002464852.1, YP_002568144.1, YP_001633993.1
ENS / ZP_05031224.1 / YP_001637375.1, ZP_07686911.1, YP_001430229.1, YP_001278813.1, YP_002465011.1
PPS / ZP_05028882.1, ZP_05025507.1, ZP_05029015.1, ZP_05029154.1, ZP_05029107.1
PYK / ZP_05024576.1, ZP_05026799.1, ZP_05023969.1, ZP_05023802.1 / ZP_07684115.1, YP_002571078.1, YP_001636715.1, YP_001431802.1, YP_001275776.1, YP_002461694.1

Supplemental Table 6 Number of metatranscriptomic reads recruited to M. chthonoplastesand Chloroflexales genes for fermentation and PHA production pathways.

Abbreviation / Microcoleus / Cyanobacteria / Chloroflexales / Total
BN Reads / EN Reads / BN Reads / EN Reads / BN Reads / EN Reads / BN Reads / EN Reads
LDH / 4 / 0 / 9 / 1 / 1 / 4 / 23 / 12
PFL / 116 / 27 / 177 / 53 / 0 / 0 / 334 / 198
GP / 43 / 2 / 56 / 11 / 31 / 40 / 118 / 77
ADH / 125 / 61 / 276 / 125 / 0 / 0 / 300 / 138
ACK / 3 / 0 / 7 / 1 / 0 / 0 / 20 / 11
PFR / 458 / 264 / 512 / 296 / 17 / 21 / 731 / 494
HoxH / 20 / 11 / 41 / 18 / 0 / 1 / 45 / 34
HoxY / 29 / 4 / 32 / 6 / 0 / 0 / 36 / 18
HoxU / 4 / 10 / 4 / 11 / 0 / 0 / 24 / 22
HoxF / 31 / 21 / 40 / 41 / 0 / 0 / 83 / 94
HoxE / 15 / 17 / 16 / 18 / 0 / 0 / 23 / 26
PDC / 32 / 5 / 60 / 16 / 3 / 2 / 139 / 62
PhaA / 0 / 0 / 0 / 0 / 8 / 5 / 36 / 45
PhaB / 0 / 0 / 0 / 0 / 26 / 19 / 49 / 48
PhaC / 0 / 0 / 6 / 29 / 41 / 28 / 74 / 110
ActP / 1 / 0 / 2 / 3 / 22 / 8 / 132 / 69
LAP / 0 / 0 / 1 / 0 / 1 / 0 / 19 / 12
PFK / 10 / 5 / 28 / 11 / 11 / 9 / 66 / 48
GPI / 1 / 0 / 4 / 1 / 0 / 0 / 21 / 20
FBP / 24 / 2 / 59 / 7 / 8 / 6 / 81 / 24
FBA / 7 / 1 / 271 / 57 / 0 / 0 / 342 / 79
TPI / 3 / 0 / 5 / 1 / 21 / 30 / 44 / 41
G3P / 12 / 3 / 17 / 4 / 8 / 4 / 121 / 45
PGK / 12 / 0 / 49 / 2 / 6 / 16 / 104 / 49
PGM / 2 / 1 / 5 / 2 / 4 / 14 / 18 / 36
ENS / 10 / 1 / 26 / 2 / 3 / 1 / 68 / 38
PPS / 641 / 201 / 1056 / 370 / 0 / 0 / 1117 / 410
PYK / 16 / 3 / 38 / 11 / 5 / 9 / 79 / 32

Supplemental Figure Legend

Supplemental Figure 1. Diel production of H2 or organic acids in Elkhorn Slough mats from January 12-13, 2009. Mat core incubations in vials commenced at dawn with H2 and organic acids allowed to accumulate over a diel cycle with N2 flushing of the vial at the end of the day and night periods. Vials in which organic acids were measured were reset at dawn by replacement with fresh field site water. Vertical error bars indicate the standard deviation of three replicates. (a) H2 production in different layers of microbial mat. Mat cores were sectioned at different depths, 0-2 mm, 2-4 mm and 4-15 mm. Yellow diamonds denote natural solar irradiance. E, microeinsteins. (b) Organic acid production in intact mat cores (0-15 mm).

Supplemental Figure 2. Collector’s curves generated by MOTHUR for reads assigned to self-BLAST clusters (reads overlap by 200 or more base pairs with 90% or greater sequence identity), RefSeq sequences (a), and unambiguous MEGAN assignments to SEED subsystems and species (b).

Supplemental Figure 3.Classification of reads for Chloroflexi populations. Within the metatranscriptome reads affiliated with the phylum Chloroflexi, 96% were assigned to the class Chloroflexi at both time points, and ca. 85% of those reads further to the order Chloroflexales (BN; n = 18,769 and EN; n = 26,981, respectively). About 3% of the reads within the class Chloroflexi were assigned to the order Herpetosiphonales, which are non-phototrophic members of the Chloroflexi, while the remaining ~12% of sequences in the class Chloroflexi could not be assigned further. b) In both samples, within the Chloroflexales, ~31% of the reads were assigned to the Chloroflexaceae and further to the genera Chloroflexus and Roseiflexus, while ~12% were assigned to the Oscillochloridaceae and further to the genus Oscillochloris. All the reads within the Oscillochloris could be assigned to Oscillochloris trichoides, while the majority of the reads within the Chloroflexus and Roseiflexus could not be assigned any further than to the genus level. Furthermore, 57% of the Chloroflexales assigned reads could not be classified further than to the order.

Supplemental Figure 4. Transcripts classified within different SEED subsystems (functional categories) from the Begin Night (BN; 21:00, January 12, 2009) and End Night (EN; 07:00, January 13, 2009) libraries. Percentage abundances of transcripts within different SEED subsystems were calculated for different taxonomic group: all organisms, Chloroflexales or Microcoleus spp.. SEED subsystems with <1% of transcripts for a given taxonomic group are not shown. Unassigned reads represent transcripts that had matches to the RefSeq microbial database but could not be classified within a specific SEED category. See materials and methods for more information regarding taxonomic and functional classifications of transcripts.

Supplemental Figure 1

Supplemental Figure 2

Supplemental Figure 3

Supplemental Figure 4