Labeling of prokaryotic mRNA in living cells using fluorescent in situ hybridization of transcript-annealing molecular beacons (FISH-TAMB)

Rachel L. Harrisa*, Maggie C. Y. Laua, Esta van Heerdenb, Errol B. Casonb, Jan-G Vermeulenb, Anjali Tanejaa,c, Thomas L. Kieftd, Christina DeCostee, Gary Laevskyf, and Tullis C. Onstotta

aDept. of Geosciences, Princeton University, Princeton, NJ 08544; bDept. of Microbial, Biochemical, and Food Technology, University of the Free State, Bloemfontien, South Africa; cMcCourt School of Public Policy, Georgetown University, Washington, D. C. 20057; dDept. of Biology, New Mexico Institute of

Mining and Technology, Sorocco, NM 87801; eFlow Cytometry Resource Facility, Dept. of Molecular Biology, Princeton University, Princeton, NJ 08544; fConfocal Microscopy Resource Facility, Dept. of Molecular Biology, Princeton University, Princeton, NJ 08544

High-throughput sequencing and cellular imaging have expanded our knowledge of microbial diversity and expression of cellular activity. However, it remains challenging to characterize low-abundance, slow-growing microorganisms that play key roles in biogeochemical cycling. With the goal of isolating transcriptionally active cells of these microorganisms from environmental samples, we developed fluorescent in situ hybridization of transcript-annealing molecular beacons (FISH-TAMB) to label living prokaryotic cells. FISH-TAMB utilizes polyarginine cell-penetrating peptides to deliver molecular beacons across cell walls and membranes. Target cells are fluorescently labeled via hybridization between molecular beacons and messenger RNA of targeted functional genes. FISH-TAMB’s target specificity and deliverance into both bacterial and archaeal cells were demonstrated by labeling intracellular methyl-coenzyme M reductase A (mcrA) transcripts expressed by Escherichia coli mcrA+, Methanosarcina barkeri, and a methanogenic enrichment of deep continental fracture fluid. Growth curve analysis supported sustained cellular viability following FISH-TAMB treatment. Flow cytometry and confocal microscopy detected labeled single cells and single cells in aggregates with unlabeled cells. As FISH-TAMB is amenable to target any functional gene of interest, when coupled with cell sorting, imaging, and sequencing techniques, FISH-TAMB will enable characterization of key uncharacterized rare biosphere microorganisms and of the syntrophically activated metabolic pathways between physically associated microorganisms.

Probe Conformation and Target Hybridization

Flow Cytometry Distinguishes FISH-TAMB-Labeled Cells

Target Specificity in vivo: Fluorescence Lifetime and Sustained Cellular Viability

Next Steps

Acknowledgements

Fig. 1. FISH-TAMB probe conformation and hybridization to encountered messenger RNAs. (A) An oligomer comprised of a 24 base-long complementary mcrA mRNA sequence is flanked by 5 reverse complement nucleotides to form a molecular beacon (MB) loop and stem structure. Cell-penetrating peptides (CPPs) comprising 9 arginine sequences (R9) are non-covalently bound to the MB sequence and are responsible for its delivery across the cell wall and plasma membrane. (B) Fluorescence of Cy5 fluorophore covalently bound to the 5’ end of the MB sequence remains quenched by BHQ3 bound to the 3’ terminus until the MB hybridizes to a target transcript sequence. Hybridization results in the linearization of the MB, subsequently unquenching Cy5 from BHQ3, allowing the fluorophore’s emission upon excitation by a source in the red bandwidth of the visible light spectrum. (C) If the MB encounters an mRNA transcript that is not its intended target, it will retain its hairpin conformation and fluorescence of Cy5 will remain quenched by BHQ3. Images not to scale. *Mechanism of CPP delivery across the cell wall and plasma membrane remains under debate. Intracellular fate of R9 is unknown.

•  The molecular beacon (MB) utilized in this study comprised a GC-rich 5-base pair stem and a 24-mer nucleotide probe sequence (5'-[Cy5]cctggCGTTCATBGCGTAGTTVGGRTAGTccagg[BHQ3]-3')

•  Similarity of MB sequence and mcrA gene verified in silico via BLAST (https:/blast.ncbi.nlm.nih.gov/Blast.cgi) against

nucleotide database

3’ 5’

5’ 3’

5’ 3’

Complex denaturation

i

ii

TTemperature (˚C) 29.4 34.0 44.0 64.0 54.0 74.0 84.0 94.0

4.000 E+2

9.000 E+2

1.400 E+3

1.900 E+3

2.400 E+3

2.900 E+3

3.400 E+3

Fluo

resc

ence

Bound state

Unbound state

B AA

C

D

E

10 µm

A

B

C

D

T = 0 min T = 20 min T = 120 min T = 240 min

5 µm

Conformational Stability and in vitro Specificity

Fig. 2. MB and FISH-TAMB probe target specificity in vitro. (A) 0.4 µM MB in 1x PBS. (B) 0.4 µM MB + pmoA target oligo. (C) 0.4 µM MB + mcrA target oligo. Images taken after 20 minutes incubation with a Typhoon 9410 Variable Mode Imager®. Excitation 633 nm. Emission 675/10 nm Exposure time 5 minutes.

Fig. 3. Heat stability of molecular beacon sequence. In vitro melting curve analysis of MB in 5x PCR buffer containing 7.5 mM MgCl2. (i) mcrA MB hybridized to target oligonucleotide sequence. (ii) Unbound mcrA MB.

Abstract

3. Assess optimal molar ratio of carrier peptide to molecular beacon for formation of stable FISH-TAMB probes.

Events 0.5 µm microspheres

Cell-sized objects

Cy5 cell-sized objects

Population

Cy5-A (700/25 nm)

F420

-A (4

50/5

0 nm

)

Cell aggregates

FITC

-A (5

30/3

0 nm

)

D

F420

-A (4

50/5

0 nm

)

B E

A

F420

-A (4

50/5

0 nm

)

Cy5-A (700/25 nm)

C F

Cy5-A (700/25 nm)

Table 1. Flow cytometry data of FISH-TAMB-treated cells.

Sample Total Cells

(# mL-1)

FISH-TAMB labeled cells

(# mL-1)

% FISH-TAMB labeled

cells

FISH-TAMB probes/cell p-value* t-value*

E. coli pmoA+ (n=3) 3.9 ± 0.7 x 106 7.9 ± 2.4 x 102 0.02% 1.6 ± 0.3 x 107 0.25 1.6

E. coli mcrA+ (n=3) 1.6 ± 0.9 x 106 2.5 ± 1.5 x 104 2% 3.8 ± 0.2 x 107 0.03** 5.2

M. barkeri (n=2) † 1.5 ± 0.2 x 107 5.4 ± 0.2 x 104 32% (†~22%) 3.5 ± 0.1 x 108 0.01** 9.6 BE326 BH2-Conc (n=3) † 4.7 ± 1.5 x 105 5.0 ± 1.0 x 103 1% (†~3%) 1.4 ± 0.4 x 108 0.01** 9.2

* Student's t-test of average FISH-TAMB labeled cell concentration versus uninocculated control with 1x DPBS + growth media +

FISH-TAMB probes (two-tailed distribution, equal variance assumed). **Significant p-values (p<0.05). † M. barkeri and BE326 BH2-Conc data corrected to account for cell aggregates initially enumerated as individual events.

Fig. 9. Flow cytometry distinguishes FISH-TAMB labeled cells. Flow cytometry reveals single cells demonstrating distinctive Cy5 fluorescence patterns in cells treated with 1 µM FISH-TAMB probes (D-F) relative to untreated cells (A-C). (A,D) E. coli mcrA+. (B,E) M. barkeri. (C,F) BE326 BH2-Conc. Positively labeled cells indicated in red. FISH-TAMB-labeled cells are gated with respect to Cy5 fluorescence intensity (X-axis) and target cell autofluorescence properties (Y-axis).

Motivation: Connecting Taxonomy to Biogeochemical Cycling

Method Pros Cons

Culturework Deepcharacteriza2onpoten2al Cul2va2onresistantorganisms

Metagenomics Iden2fiesrarebiosphereMDM Taxonomy≠metabolicroledeep-sequencingcoverage;metabolicpoten2alonly

Metatranscriptomics Insitumetabolicac2vity Metabolicfunc2on≠taxonomy;community-levelonly

16SrRNAFISH+isotopelabeling Iden2fysinglecellsfromenvironment Requiresknowledgeoftaxonomy

mRNAFISH Singlecellmetabolicac2vity Requirescellularfixa2on

Singlecellgenomics Sequencegenomesfromindividualcells Rarebiospherelimita2ons;Metabolicpoten2alonly

Question: How can we assess the metabolic and genomic potential of active and geochemically important microorganisms?

Solution: Target messenger RNA (mRNA) encoding for key enzymes involved in nutrient cycling in living cells, one cell at a time.

0:1 5:1 10:1 15:1 20:1 25:1 30:1

Fig. 4. Complexation between R9 cell-penetrating peptide and mcrA MB sequences in fixed R9:MB molar ratios. Unbound MB shows up as white bands in the gel and complete complexation of all MB to R9 is evidenced by the lack of a physical band.

1. Assess specificity of MB to target oligonucleotide sequence and compare to background fluorescence of unbound MB.

2. Assess stability of FISH-TAMB probe as a function of temperature and salinity: Temperature range: 25˚- 95˚C

Salt Buffers: Takara PCR buffer containing 1.5 mM MgCl2 (1x), 7.5 mM MgCl2 (5x), or 15 mM MgCl2 (10x).

In pure culture methanogens and a methanogenic enrichment from the deep biosphere:

Fig. 5. FISH-TAMB-treated methanogens labels mcrA-transcribing cells. Spinning disk photomicrographs of diverse surface morphologies in FISH-TAMB-treated methanogenic enrichments. (A) M. barkeri aggregate treated with 1 µM FISH-TAMB probes. Cellular autofluorescence and Cy5 fluorescence were excited with 405 nm and 647 nm lasers, respectively. Image snapped using 518 nm and 670 nm emission filters. (B-E) Methanogenic enrichments from borehole BE326-BH2 located 1.39 km below land surface in Beatrix Gold Mine, South Africa. Enrichments were incubated for 15 minutes with 1 µM FISH-TAMB probes and imaged every minute for 14 hours. Sample images are from the first minute of monitoring. Cy5 fluorescence was captured according to above-described parameters and integrated in real time with bright field image collection. FISH-TAMB demonstrates indiscriminate hybridization to (B) single cells, (C) physically-associated cells, (D) labeled and unlabeled cell pair, and (E) cell aggregate. Positively labeled cells indicated in red. All samples were maintained under a 100% CO2 atmosphere during imaging. 100x magnification.

Fig. 8. FISH-TAMB viability assessment by growth curve analysis. E. coli mcrA+, E. coli pmoA+, and M. barkeri cultures (~106 cells ml-1) were incubated with 1 µM FISH-TAMB probes and inoculated into their respective growth media. Growth was measured spectrophotometrically (OD600 for E. coli, OD550 for M. barkeri) and growth rates compared to untreated cultures.

Fig. 7. Fluorescence lifetime of Cy5 in FISH-TAMB-hybridized cells. B E 3 2 6 B H 2 - C o n c m e t h a n o g e n i c e n r i c h m e n t s w e r e incubated anaerobically with 1 µM FISH-TAMB probes and subsequently imaged via spinning disk p h o t o m i c r o s c o p y . Samples were excited with a 647 nm laser line and analyzed at 670 nm under a 100% CO2 a t m o s p h e r e . M i c r o g r a p h s w e r e snapped every minute f o r 1 4 h o u r s . M i c r o g r a p h s h e r e represent the first four hours of observation. (A) S i n g l e c e l l s . ( B ) Phys ica l ly-associated cells. (C) Cell pair in which an unlabeled cell b e c o m e s l a b e l e d between 20 and 120 m i n u t e s . ( D ) C e l l a g g r e g a t e . 1 0 0 x magnification.

Fig. 6. Epifluorescence microscopy of E. coli mcrA+. FISH-TAMB-associated fluorescence indicated byred arrows. Composite FITC and TexasRedfiltercubes.Imagesnappedat100Xmagnifica2on.

InE.coliexpressionclonescontaininganmcrAgeneinsert

Fig. 10. FISH-TAMB specificity to mcrA in E. coli. (A) E. coli pmoA+ cells lacking the mcrA gene do not demonstrate Cy5 fluorescence when treated with 1 µM FISH-TAMB probes. (B) Active E. coli mcrA+ cells demonstrate Cy5 fluorescence when treated with 1 µM FISH-TAMB probes. FISH-TAMB-labeled cells are indicated in red. FISH-TAMB-labeled cells are gated with respect to Cy5 fluorescence intensity (X-axis) and E. coli autofluorescence properties (Y-axis).

FISH-TAMB treated Untreated

E. coli mcrA

+ M

. barkeri

Deep biosphere

methanogenic enrichm

ent

Major Findings: 1.  FISH-TAMB treated cells containing mcrA

gene exhibit a significant increase in fluorescence on Cy5 filter relative to

untreated and E. coli pmoA+ cell populations (Student t-test, t ≥ 2.5, p < 0.05).

2.  Recovery % of FISH-TAMB labeled cells in BE326-Conc is greater fraction than expected

0.4 – 0.5% of methanogens and ANMEs reported by 16S rRNA amplicon and

metagenomics data (1,2). 3.  Recovery % of FISH-TAMB labeled cells

from flow cytometry is consistent with microscopy counts.

4.  Insert toxicity of mcrA gene in E. coli verified by low recovery of FISH-TAMB labeled cells.

Organism Ecotype Relative Abundance (16S rDNA amplicons)

Methanogen 0.3%

Anaerobic Methanotroph (ANME)

0.1% – 0.2%

Methanotroph 1.5% - 2.0%

(1) Simkus et al., 2016.

Methane-cycling microbes in BE326-BH2

•  Assess alternative E. coli expression clones and vector systems more tolerant of toxic gene inserts to improve E. coli mcrA+ recovery

•  Target endogenous genes (e.g. lacZ) to avoid toxicity woes all together? •  Fluorescence-activated cell sorting (FACS) of FISH-TAMB labeled cells

•  Cultivation of sorted populations to better constrain viability of FISH-TAMB labeled cells vs. untreated cells

•  Single cell genome sequencing of sorted BE326-BH2 cells to search for novel methanogens and anaerobic methanotrophs (ANMEs)

•  Assess sensitivity and detection threshold of FISH-TAMB method via qPCR and RT-qPCR •  qPCR to obtain mcrA gene copy counts •  RT-qPCR to obtain mcrA mRNA copy counts

•  Sibanye Gold, Ltd. and the staff at the Beatrix Gold Mine for granting us continued access to the BE326-BH2 borehole •  Mike Pullin, Gilbert Tetteh, Sarah Hendrickson, and Olukayode Kuloyo for their field assistance •  Peter Jaffe and Melany Ruiz Uriguen for sharing access to lab equipment •  National Science Foundation GRFP grant DGE-1148900 to R.L.H. •  NSF grants DEB-1441717, EAR-1528492, DEB-1442059, and DEB-1441646 grants awarded to T.C.O. •  Pioneer funding from the Dept. of Geosciences at Princeton University to A. T.

References: (1) Simkus DN, et al., 2016. Geochim Cosmochim Acta 173:264–283. (2) Magnabosco C., et al., 2014. Front Microbiol 5.