Methods

Reconstructing transcriptional histories by CRISPR acquisition of retron-based genetic barcodes

Bacterial Strains and Growth Conditions

This work uses the following E. coli strains: NEB 5-alpha (NEB C2987), BL21-AI (ThermoFisher C607003), bMS.346, and bSLS.114. bMS.346 was generated from E. coli MG1655 by inactivating exoI and recJ genes with early stop codons as in previous work33. Additionally, the araB::T7RNAP-tetA locus was transferred from BL21-AI by P1 phage transduction34. bSLS.114 (which has been used previously24) was generated from BL21-AI by deleting the retron Eco1 locus by lambda Red recombinase mediated insertion of an FRT-flanked chloramphenicol resistance cassette. This cassette was amplified from pKD335 with homology arms added to the retron Eco1 locus. This amplicon was electroporated into BL21-AI cells expressing lambda Red genes from pKD4635, and clones were isolated by selection on chloramphenicol (10 µg/mL) plates. After genotyping to confirm locus-specific insertion, the chloramphenicol cassette was excised by transient expression of FLP recombinase to leave only an FRT scar. Experimental cultures were grown with shaking in LB broth at 37ºC with appropriate inducers and antibiotics. Inducers and antibiotics were used at the following working concentrations: 2 mg/mL L-arabinose (GoldBio A-300), 1 mM IPTG (GoldBio I2481C), 400 µM erythromycin, 100 ng/mL anhydrotetracycline, 100 µM choline chloride, 1 mM sodium salicylate, 35 µg/mL kanamycin (GoldBio K-120), 25 µg/mL spectinomycin (GoldBio S-140), 100 µg/mL carbenicillin (GoldBio C-103), 25 µg/mL chloramphenicol (GoldBio C-105; used at 10 µg/mL for selection during recombineering). Additional strain information can be found in Supplemental Table 2.

Plasmid Construction

All cloning steps were performed in E. coli NEB 5-alpha. pWUR 1+2, containing Cas1 and Cas2 under the expression of a T7lac promoter, was a generous gift from Udi Qimron12. Eco1 wildtype ncRNA and Eco1 RT, along with Cas1+2, were cloned into pRSF-DUET (Sigma 71341) to generate pSLS.405. Eco1 variant ncRNA sequences v32 and v35 were cloned into pRSF-DUET along with Cas1+2 to generate pSLS.407 and pSLS.408, respectively. Extended a1/a2 v35 ncRNA expression plasmid pSLS.416 was generated from pSLS.408 by site-directed mutagenesis. Retron Eco1 RT and retron Eco4 RT were cloned into pJKR-O-mphR to generate pSLS.402 and pSLS.400, respectively. pJKR-O-mphR was generated previously36 (Addgene plasmid # 62570). Barcoded, extended a1/a2 v35 ncRNA expression plasmids pSBK.009-016 were generated from pSLS.416 by site-directed mutagenesis. Wildtype retron Eco4 ncRNA was cloned into pRSF-DUET along with Cas1+2 to generate SLS.419. pSBK.134 and pSBK.136 were generated in three steps. First, barcoded, extended a1/a2 v35 ncRNA sequences were cloned into the ‘Marionette’ plasmids pAJM.717, pAJM.718, and pAJM.771. pAJM.717, pAJM.718, and pAJM.771 were gifts from Christopher Voigt32 (pAJM.717 - Addgene plasmid # 108517 // pAJM.718 - Addgene plasmid # 108519 // pAMJ.771 - Addgene plasmid # 108534). Then, in two steps, two ncRNA expression cassettes (for barcoded ncRNAs “A” and “B”) from the Marionette plasmids were cloned into pSol-TSF (Lucigen F843213-1) facing in opposite directions. pSBK.079 was generated by cloning the resistance marker AmpR in place of the KanR marker into the plasmid pSLS.425, which was synthesized by Twist biosciences. Additional plasmid information can be found in Supplemental Table 3.

RT-DNA Purification and PAGE Visualization

Retron RT-DNA was expressed in E. coli bMS.346 and purified in two steps. First, DNA was extracted from cells using a plasmid midiprep kit (Qiagen 12943). This purified DNA was then treated for 30 minutes at 37C with RNAse A/T1 mix (ThermoFisher EN0551) and, if required, DBR1 (OriGene TP300024) and/or RecJf (NEB M0264). This sample was then used as the input for the Zymo Research ssDNA/RNA Clean & Concentrate kit (Zymo D7011). Samples eluted from the ssDNA kit were resolved using TBE-urea PAGE (ThermoFisher EC6885BOX). Gels were stained with SYBR Gold for imaging (ThermoFisher S11494) and imaged on a Bio-Rad Gel Doc imager.

Retron Acquisition Experiments

Cells were transformed sequentially: first with the RT expression plasmid (pSLS.400 or pSLS.402), and second with the ncRNA and Cas1+2 expression plasmid (eg. pSLS.416). For the -RT condition, cells were only transformed with an ncRNA and Cas1+2 expression plasmid (e.g. pSLS.416). For testing acquisition of retron-derived spacers in figures 1e-f, cells with RT, ncRNA, and Cas1+2 expression plasmids were grown overnight (16 hours) in 3 mL LB with antibiotics and inducers IPTG and arabinose, from individual clones on plates. In the morning, 240 uL of overnight culture was diluted into 3 mL fresh media with antibiotics, IPTG, and arabinose and grown for 2 hours. After 2 hours, 320 uL of culture was diluted into 3 mL fresh media with antibiotics and erythromycin (no erythromycin was used in the -RT condition) and grown for 8 hours. After 8 hours, culture was diluted 1:1000 into 3 mL LB with antibiotics and without inducers and grown overnight (16 hours). In the morning, 25 uL of culture was mixed with 25 uL of water, heated to 95C for 5 minutes to lyse cells, cooled, and frozen at -20C for later analysis. For data presented in Figures 2b-d and 3i, cells were grown overnight (16 hours) in 3 mL LB with antibiotics and inducers IPTG and arabinose, from individual clones on plates. In the morning, 240 uL of overnight culture was diluted into 3 mL fresh media with antibiotics, IPTG, and arabinose and grown for 2 hours. After 2 hours, 320 uL of culture was diluted into 3 mL fresh media with antibiotics and erythromycin and grown for 2 (rather than 8) hours. At this point, 25 uL of culture was mixed with 25 uL of water, heated to 95C for 5 minutes to lyse cells, cooled, and frozen at -20C for later analysis.

For the 24-hour time course experiment, the experiment was broken into two halves: the first 9 hours, and the final 15 hours. For the entirety of the time course, cells were grown in media with antibiotics and inducers (arabinose, IPTG, and erythromycin). For the first 9-hour samples, cultures were grown starting from single colonies added to 0.5 mL of media. These cultures were sampled every 1.5 hours until hour 9, with 1 mL of media added at hour 3 and 1.5 mL of media added at hour 6. For the final 15-hour samples, 3 mL of media was inoculated with single colonies from plates and grown for 9 hours. Starting at hour 9, samples were taken every 1.5 hours until hour 24. At hour 16.5, 200 uL of culture was diluted into 1.5 mL of fresh media and the experiment continued in the new tube. At hour 21, 1 mL media was added to the culture.

Oligo Prespacer Feeding

For spacer acquisition experiments using exogeneous DNA prespacers (purified RT-DNA or synthetic oligos), cells containing pWUR1+2 were grown overnight from individual colonies on plates. In the morning, 100 uL of overnight culture was diluted into 3 mL LB with antibiotics, IPTG, and arabinose. Cells were grown with inducers for 2 hours. For each electroporation, 1 mL of culture was pelleted and resuspended in water. Cells were washed a second time by pelleting and resuspension, then pelleted one final time and resuspended in 50 uL of prespacer DNA solution at a concentration of 6.25 uM of single-stranded RT-DNA. All wash steps were done using ice cold water, all centrifugation steps were done in a centrifuge chilled to 4C, and samples kept on ice until electroporation was complete. The cell-DNA mixture was transferred to a 1 mm gap cuvette (Bio-Rad 1652089) and electroporated using a Bio-Rad gene pulser set to 1.8 kV and 25 uF with pulse controller at 200 Ohms. After electroporation, cells were recovered in 3 mL of LB without antibiotics for 2 hours. Then, 25 uL of culture was mixed with 25 uL of water, heated to 95C for 5 minutes to lyse cells, cooled, and frozen at -20C for later analysis.

Temporal Recordings

Cells were transformed sequentially, first with pSBK.134 or pSBK.136 and then with pSBK.079. For recording, single colonies were picked from plates and grown overnight in 3 mL of LB with antibiotics and without inducers. In the morning, 150 uL of culture was diluted into 3 mL of LB with antibiotics and appropriate inducers (Fig. 4) and grown for 8 hours. After 8 hours, 60 uL of culture was diluted into 3 mL of LB with appropriate inducers and grown overnight (16 hours). In the morning, 150 uL of culture was diluted into 3 mL of LB with appropriate inducers (for second day of expression) and grown for 8 hours. Samples were collected at this 24-hour timepoint. 25 uL of culture was mixed with 25 uL of water, heated to 95C for 5 minutes to lyse cells, cooled, and frozen at -20C for later analysis. After 8 hours, 60 uL of culture was diluted into 3 mL of LB with appropriate inducers and grown overnight (16 hours). In the morning, 25 uL of culture was mixed with 25 uL of water, heated to 95C for 5 minutes to lyse cells, cooled, and frozen at -20C for later analysis.

Analysis of Spacer Acquisition

Analysis of spacer acquisition was conducted by sequencing a library of all CRISPR arrays in an experimental population using an Illumina MiSeq instrument. Libraries were created by amplifying a region of the genomic CRISPR array using PCR, then indexed using custom indexing oligos. Up to 192 conditions were run per flow cell. A list of oligo prespacers and primers can be found in Supplemental Table 4.

Processing and Analysis of MiSeq Data

Sequences were analyzed using custom Python software, which will be available on GitHub upon peer-reviewed publication. In brief, newly acquired spacer sequences were extracted from array sequences based on their position between identifiable repeats and compared to preexisting spacers in the array. In this preliminary analysis, metrics were collected including number of expansions in arrays (unexpanded, single, double, and triple expanded) and proportion of each present in the library. Sequenced arrays were sorted into subcategories based on these characteristics (e.g. doubly expanded with first three repeats identifiable) for further analysis. Next, to determine number of retron-derived spacers and the order of spacers in multiply expanded arrays, two different analyses were used: one strict and one lenient. In the strict analysis (used in figures 1, 2, and 3) a retron-derived spacer is defined to be a spacer which contains the 23-base core region of the hypothetical prespacer structure from a given retron (with three mismatches or indels allowed). In the lenient analysis (used in figures 4 and 5) a retron-derived spacer is defined to be a spacer which contains an 11-base region of the hypothetical prespacer consisting of the 7-base barcode region and 2 bases on either side (with one mismatch or indel allowed). The order of spacers in multiply expanded arrays is then reported (e.g. Leader-NNA) and these data are used to complete the ordering rule analysis.

Article TitleReconstructing transcriptional histories by CRISPR acquisition of retron-based genetic barcodes

Abstract

Biological processes depend on the differential expression of genes over time, but methods to make true physical recordings of these processes are limited. Here we report a strategy for making time-ordered recordings of transcriptional events into living genomes. We do this via engineered RNA barcodes, based on prokaryotic retrons, which are reverse-transcribed into DNA and integrated into the genome using the CRISPR-Cas system. This approach enables the targeted recording of time-ordered transcriptional events in cells. The unidirectional integration of barcodes by CRISPR integrases enables reconstruction of transcriptional event timing based on a physical record via simple, logical rules rather than relying on pre-trained classifiers or post-hoc inferential methods.


Login or Signup to leave a comment
Find your community. Ask questions. Science is better when we troubleshoot together.
Find your community. Ask questions. Science is better when we troubleshoot together.

Have a question?

Contact support@scifind.net or check out our support page.