This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
SUPPLEMENTARY INFORMATION Table of contents Page Materials and Methods 2-14 Supplementary Tables 15-33 Supplementary Table 1: Overview of Illumina Sequencing datasets 15 Supplementary Table 2: FDH and FTHFS Scaffold-Library Datasets 16-25 Supplementary Table 3: Mapping of Illumina RNA-Seq Sequences onto FDH dataset and 26 validation by cDNA libraries and qRT-PCR Supplementary Table 4: Mapping of Illumina RNA-Seq Sequences onto FTHFS dataset 27 Supplementary Table 5: QRT-PCR analysis of termite gut luminal fluid and particle 28 associated pellet Supplementary Table 6: Microfluidic chip experiment details 29 Supplementary Table 7: Microfluidic analysis of untreated gut samples (supernatant) 30 Supplementary Table 8: Microfluidic analysis of treated gut pellets 31 Supplementary Table 9: Microfluidic analysis of treated gut pellets 32 Supplementary Table 10: Microfluidic analysis of treated gut pellets 33 Supplementary Figures 34-46 Supplementary Figure 1: 34 Supplementary Figure 2: 35 Supplementary Figure 3: 36 Supplementary Figure 4: 37 Supplementary Figure 5: 41 Supplementary Figure 6: 42 Supplementary Figure 7: 43 Supplementary Figure 8: 44 Supplementary Figure 9: 45 Supplementary Figure 10: 46 Supplementary References 47
2
Materials and Methods
Termite collection
Worker specimens of the dampwood termite Zootermopsis nevadensis were collected in the Chilao campground
in the San Gabriel Mountains of California. Some termites were processed within 24 hours of harvest, and the
rest were maintained for up to 6 months after collection in plastic boxes at 95% humidity in foil-covered glass
aquaria in the laboratory. The entire gut tracts of ~5 worker termites were preserved in 50 – 200 µl of RNA
stabilization buffer (RNA Protect Bacteria Reagent, QIAGEN, Valencia, CA) at -80°C until nucleic acid
extraction for RNA-Seq and inventory experiments.
Termite gut nucleic acid extraction
100 µl of TE buffer (1 mM Tris-HCl, 0.1 mM EDTA, pH 8.0) was added to an ice-thawed tube containing
worker guts. Guts were then homogenized (3 x 30 seconds) by bead beating with sterile zirconia/silica beads
(0.1 mm) using a MiniBeadbeater-8 (BioSpec Products, Inc., Bartlesville, OK). Lysozyme (Sigma, St. Louis,
MO) was added to the homogenate (1 mg); this mixture was incubated at room temperature for 15 min. DNA
and total RNA were extracted from 150 µl aliquots of gut homogenate using a DNeasy Tissue Kit (QIAGEN)
and RNeasy Kit (QIAGEN), respectively, as previously described (1). Total RNA was used for Illumina RNA-
Seq and cDNA library experiments.
RNA-Seq: Processing and sequencing
Samples from total RNA were prepared as previously described (2). Briefly, libraries were built using the
Illumina protocol for RNA-Seq sample preparation V2 (https://icom.illumina.com). Briefly, total RNA (at least
5 µg) was fragmented using an Ambion RNA fragmentation kit and then converted to single-strand cDNA
using an Invitrogen SuperScript II kit (Invitrogen, Carlsbad, CA). Second Strand Buffer (500 mM Tris-HCl, pH
7.8, 50 mM MgCl2, 10mM DTT), dNTP (0.3 mM), RNaseH (2 U ⋅ µl-1, Invitrogen) and DNA polymerase I
(Invitrogen) were then added to the first-strand reaction to synthesize second strand cDNA (16°C, 2.5 hours).
3
Fragmented second strand cDNA samples were sequenced as 37-mers using the standard Illumina protocol and
pipeline at Caltech’s Sequencing Core Facility (Pasadena, CA).
RNA-Seq Data Analysis
Illumina raw data FASTQ files, obtained using GERALD (a software package within the Illumina pipeline),
was aligned to a FASTA file containing FDH gene sequences (Table 4.4, Appendix) with the Maq short read
aligning program (3). Samples were analyzed for perfect matches only. Signal intensities were visualized
graphically by converting Maq aligned reads into a .BAR file using the Cisgenome software (4) and viewed on
the Cisgenome browser and on the IGB genomic browser (http://www.affymetrix.com).
RNA-Seq results were validated by cDNA phylogenetic libraries and Real-Time quantitative-PCR (qRT-PCR)
(see specifics below).
cDNA inventories
Separate cDNA libraries for fdhFSec and fdhFCys gene variants were generated from gut cDNA. A forward
primer for fdhFSec (Sec427F, Table 4.1) that targets the selenocysteine FDHH active site was designed manually.
Sec427F was used with 1045R (1) to amplify fdhFSec from gut cDNA. The fdhFCys cDNA library was
constructed with primers Cys499F1b and 1045R (1). PCR reactions contained 200 nM forward primer
(Sec427F or Cys499F1b), 200 nM 1045R, 1X FAILSAFE Premix D (EPICENTRE, Madison, WI), 0.07 U ⋅ µl-1
of EXPAND High Fidelity polymerase (Roche Applied Science, Indianapolis, IN), and 0.5 ng ⋅ µl-1 gut cDNA.
Thermocycling conditions on a Mastercycler Model 5331 thermocycler (Eppendorf, Westbury, NY) were 2 min
at 95°C, 30 cycles of (95°C for 30 seconds, 60°C for 1 minute, 72°C for 1 minute), followed by 10 min at 72°C.
Amplicon size was checked on 1.5% agarose gels (Invitrogen) and the products were TOPO-TA cloned
(Invitrogen). Plasmids were extracted (QIAprep Spin Miniprep Kit, QIAGEN) from 48 randomly chosen clones
and sequenced (Laragen Inc., Los Angeles, CA).
4
Quantitative PCR
Quantitative RT-PCR for select FDH genotypes (ZnD2sec, ZnB5sec, T. primitia fdhFSec) was performed on
termite gut cDNA and DNA. Quantitative PCR primers for these genotypes were: ZnD2sec (ZnO-1636F, 5’–
ACT ATG ACC GGC AAT TGT CGC CTG TT –3’; ZnO-1729R, 5’– TCA GAC CCA TAT CAC GGC AAA GTT
ZnDP-F1 rRNA Probe targets the deltaproteobacterial small subunit ribosomal RNA (16S) FG18 (Genbank accession number DQ420255) with at least 7 mismatches to other sequences in the BLAST nr/nt database (Supplementary Table S10).
Proteobacteria rRNA Probe targets the proteobacteria small subunit ribosomal RNA (16S) with at least 7 mismatches to a selection of other bacteria (Supplementary Table S10).
ZAS2sec mRNA Two probes targeting the ZAS2sec mRNA were used to validate HCR in a pure culture of Treponema primitia (Genbank accession NC_015578) and Treponema azotonutricium ZAS9 (Genbank accession NC_015577)
9
(Supplementary Fig. S7. Probes 1 and 2 have 100% homology to the sec-containing formate-dehydrogenase gene of T. primitia ZAS2. T. azotonutrucium ZAS9 does not contain an fdhF gene (Supplementary Table S9).
HCR amplifier sequences Each HCR amplifier (12) comprises two hairpins (H1 and H2) that undergo conditional polymerization in response to detection of an initiator sequence (I). /5’-dye-C12/: 5’ Alexa Fluor modification with a C12 spacer /C9-dye-3’/: 3’ Alexa Fluor modification with a C9 spacer HCR2
Escherichia coli fdhF Sec 1 To validate RNA-Seq results, cDNA libraries for fdhFSec and fdhFCys genes were constructed and SYBR-green qPCR assays were performed. Both assays used genotype specific primers designed to limit bias. These experiments focused on the ZnD2sec and ZN2cys phylotypes, which were highly encountered phylotypes in RNA-Seq. Analysis of the fdhFSec cDNA inventory from lab-maintained termite guts indicated the ZnD2sec phylotype accounts for 67% of all clone sequences. This number is similar to the RNA-seq results. Likewise, comparison of ZnD2sec transcription with that of ZnB5sec and T. primitia fdhFSec using SYBR green RT-qPCR are consistent with the cDNA and RNA-Seq transcriptional patterns: Overall, the different methods provided agreement on the abundance of ZnD2sec and Zn2cys (ZnHcys was not observed). The row highlighted in orange associates with a protozoa-associated deltaproteobacterium in microfluidic digital-PCR experiments. The rows highlighted in green all associate with treponemal SSU RNA in microfluidic digital-PCR experiments or in pure culture isolates.
Supplementary Table S10. Mismatches between HCR probes and rRNA sequences for selected bacteria in the BLAST nr/nt database. The number of nucleotides substituted, deleted, and inserted in the target are noted for representative deltaproteobacteria, other proteobacteria, and other bacteria. The first 14 deltaproteobacteriaspecies are taken from Figure 1 in the main text, with organisms closer to the top being closer relatives of ZnDP-F1.
Supplementary Figure 1. Schematic of gene inventory, RNA-Seq, microfluidic PCR work-flow. FDHH gene inventories and NCBI database sequences serve as scaffolds for RNA-Seq read mapping and data analysis. RNA-Seq based identifications of candidate genotypes belonging to transcriptionally important organisms can be corroborated using independent transcriptomic methods (cDNA gene inventory, qRT-PCR). Microfluidic, multiplex digital PCR on single cells can then be employed to obtain more genetic information on these important organisms. Illumina transcript reads of gut community RNA were first mapped to gene inventory and pure culture sequence data. This first step identified highly transcribed acetogenic gene markers (fdhF genotypes seen as arrows leading to 1+2a in Figure SUPP 1). These results were corroborated with two independent methods (cDNA libraries and qPCR). Next, microfluidic PCR was used to uncover the identity of organisms encoding highly transcribed fdhF genotypes (arrows leading to 3 in SUPP Figure 1). Finally, HCR-FISH probes were used to target a highly expressed gene phylotype from an organism identified to be a part of large gut particles by microfluidic and qPCR analyses.
35
Supplementary Figure 2.
Supplementary Figure 2. Microfluidic-dPCR links FDH phylotypes to bacterial hosts. Microfluidic-chip samples are labeled “ZnChp(Chip number)-sample”, bolded, and in colored font (blue, green, orange). Pure culture sequences are bolded. fdhF sequences outlined by black boxes were highly encountered in RNA-Seq and cDNA datasets. Lines connecting sequences highlight 16S rRNA - fdhF colocalizations (duplex gene pairs for all but ZnChp1-1, ZnChp1-2 samples, which contained 16S rRNA, fdhFSec and fdhFCys gene products amplified together in triplex-PCR). Line thickness corresponds to the number of repeated co-localizations and indicates our confidence in the observed associations. Only repeatable associations are reported. Tree construction parameters and sequence accession numbers are reported in supplementary materials and methods.
36
Supplementary Figure 3.
Supplementary Figure 3: FTHFS linked to bacterial SSU Phylotypes Microfluidically associated FTHFS sequences are on the left, connected via line to the corresponding 16S sequences.
37
Supplementary Figure S4
Supplementary Figure S4. Additional examples of co-localization of ZnD2sec mRNA, ZnDP-F1 rRNA, and the rRNA of all bacteria in lysed protozoal preparations (cf. Fig. 2). One sample per page. (A) Channel 1: Signal for ZnD2sec mRNA. (B) Channel 2: Signal for ZnDP-F1 rRNA. (C) Channel 3: Signal for rRNA of all bacteria. (D-F) Composite of each pair of channels with phase. (G-I) White pixels are above a background threshold in a given channel. White pixels within the orange and blue rectangles are used for the scatter plots of panels J-L. (J-L) Pixel intensities for each pair of channels for bacterial clusters with contrasting morphologies and co-staining properties. Scale bar = 5 µm.
38
Supplementary Figure S4 (continued).
39
Supplementary Figure S4 (continued).
40
Supplementary Figure S4 (continued).
41
Supplementary Figure S5.
Supplementary Figure S5. Additional examples of co-localization of ZnD2sec mRNA and ZnDP-F1 rRNA in association with termite gut protozoa (cf. Fig. 3). One sample per column. (A) Channel 1: Signal for ZnD2sec mRNA. (B) Channel 2: Signal for ZnDP-F1 rRNA. (C) Composite of ZnD2sec mRNA signal and ZnDP-F1 rRNA signal with phase contrast. (D) Channel 3: Signal for rRNA of all bacteria. (E) Composite of ZnDP-F1 rRNA signal and all-bacteria rRNA signal with phase contrast. Scale bar = 10 µm.
42
Supplementary Figure S6.
Supplementary Figure S6. Autofluorescence (AF) and non-specific-amplification (NSA) for in situ HCR in termite gut protozoa (cf. Fig. 3 and Supplementary Fig. S2). These control experiments illustrate two sources of background: autofluorescence (AF) and non-specific amplification (NSA). In termite gut samples, wood chips are a major source of autofluorescence. Non-specific amplification arises if HCR hairpins are retained in the sample that are not contained within polymers tethered to specifically bound probes. AF is characterized by using the in situ protocol while leaving out the probes and the HCR amplifiers. AF + NSA is characterized by using the in situ protocol while leaving out the probes. Scale bar = 10 µm.
43
Supplementary Figure S7.
Supplementary Figure S7. Testing ZnD2sec probe specificity in cultured termite-gut acetogenic Treponema primitia (expressing ZAS2sec but not ZnD2sec) and non-acetogenic Treponema azotonutricium spirochetes (expressing neither ZAS2sec nor ZnD2sec). Treponema primitia cells grown under homo-acetogenic conditions stained (red) using (A) ZAS2 mRNA probes (probe 1 and probe 2) or (B) ZnD2sec mRNA probes (probe 1 and probe 2). Non-acetogenic Treponema azotonutricium stained (red) by (C) ZAS2sec mRNA probes or (D) ZnDsec mRNA probes. Note the presence of signal (red) in panel A and the absence of signal in panels B, C, D. Cells in all panels stained by DAPI (blue). Scale bar = 10 µm.
44
Supplementary Figure S8.
Supplementary Figure S8. Two-channel redundant detection of the ZnD2sec mRNA in lysed protozoal preparations using two probes with differing degrees of selectivity. (A) Channel 1: signal from probe 1. (B) Channel 2: signal from probe 2. (C) Composite of both channels with phase contrast. (D-E) White pixels are above a background threshold in a given channel. White pixels within the orange and blue rectangles are used for the scatter plot of panel F. (F) Pixel intensities for both channels for different bacterial clusters. Scale bar = 10 µm.
45
Supplementary Figure S9.
Supplementary Figure S9. Two-channel redundant detection of the ZnDP-F1 rRNA in lysed protozoal preparations using one probe targeting ZnDP-F1 rRNA and one probe targeting the rRNA of proteobacteria. (A) Channel 1: Signal for ZnDP-F1 rRNA. (B) Channel 2: Signal for Proteobacteria rRNA. (C) Composite of both channels with phase contrast. (D-E) White pixels are above a background threshold in a given channel. White pixels within the orange and blue rectangles are used for the scatter plot of panel F. (F) Pixel intensities for both channels for different bacterial clusters. Scale bar = 10 µm.
46
Supplementary Figure S10
Supplementary Figure S10. Testing ZnDP-F1 rRNA probe specificity in a pure culture of the deltaprotiobacterium Desulfovibrio alaskensis whose rRNA contains 18 substitutions and one insertion within the probe target window (see Table S10). (A) Phase. (B) Channel 1: Signal for ZnDP-F1 rRNA probe. (C) Channel 2: Signal for all-bacterial rRNA probe. (D) Composite of both channels with phase contrast. The ZnDP-F1 probe yields minimal staining.
47
Supplementary References 1. X. Zhang, E. G. Matson, J. R. Leadbetter, Genes for selenium dependent and independent formate
dehydrogenase in the gut microbial communities of three lower, wood-feeding termites and a wood-feeding roach. Environ. Microbiol. 13, 307 (2011).
2. A. Z. Rosenthal, E. G. Matson, A. Eldar, J. R. Leadbetter, RNA-seq reveals cooperative metabolic interactions between two termite-gut spirochete species in co-culture. ISME J 5, 1133 (2011).
3. H. Li, J. Ruan, R. Durbin, Mapping short DNA sequencing reads and calling variants using mapping quality scores. . Genome Res. 18, 1851 (2008).
4. H. Ji et al., An integrated software system for analyzing ChIP-chip and ChIP-seq data. Nat. Biotechnol. 26, 1293 (2008).
5. S. Rozen, H. J. Skaletsky, in Bioinformatics Methods and Protocols: Methods in Mol. Biol., S. Krawetz, S. Misener, Eds. (Humana Press, Totowa, NJ, 2000), pp. 365-386.
6. E. A. Ottesen, J. W. Hong, S. R. Quake, J. R. Leadbetter, Microfluidic digital PCR enables multigene analysis of individual environmental bacteria. Science 314, 1464 (2006).
7. E. A. Ottesen, J. W. Hong, S. R. Quake, J. R. Leadbetter, Microfluidic digital PCR enables multigene analysis of individual environmental bacteria. Science 314, 1464 (2006).
8. P. D. Schloss, J. Handelsman, Introducing DOTUR, a computer program for defining operational taxonomic units and estimating species richness. Appl. Environ. Microbiol. 71, 1501 (2005).
9. W. Ludwig et al., ARB: a software environment for sequence data. Nucl. Acids Res. 32, 1363 ( 2004). 10. E. Pruesse et al., SILVA: a comprehensive online resource for quality checked and aligned ribosomal
RNA sequence data compatible with ARB. Nucl. Acids Res. 35, 7188 (2007). 11. S. Guindon, F. Lethiec, P. Duroux, O. Gascuel, PHYML Online--a web server for fast maximum
likelihood-based phylogenetic inference. Nucl. Acids Res. 33, W557 (2005). 12. Choi, H.M.T., Chang, J. Y., Trinh, L. A., Padilla, J. E., Fraser, S. E. & Pierce, N. A. Programmable in .
. situ amplification for multiplexed imaging of mRNA expression. Nat. Biotechnol. 28, 1208 (2010). 13. Amann R. I., Binder B. J., Olson R. J., Chisholm S. W., Devereux R. and Stahl D. A. Combination of . .