Plasmid library construction

Generating dual-promoter CiBER-seq plasmid libraries                                                   

Introduction

This protocol details the path to generate a plasmid library that contains a library of gRNAs and two expression cassettes, each expressing their own barcode. Below are benchling files for key plasmids/plasmid libraries on the path to the final dual-barcode expression gRNA plasmid library.

Parent vector: (Available soon on addgene as "pNTI743 dual-barcoded gRNA parent vector") https://benchling.com/s/seq-wDLeGF48lpjl9YB9domW

Barcoded gRNA library: (Available soon on addgene as a pooled library) https://benchling.com/s/seq-hmguFaWniCAcosNWrXKc

Dual promoter driving dual barcode expression with gRNA library (PGK1 and HIS4 promoter as example): https://benchling.com/s/seq-S5cQupYmmAS5YGu3hWZU

Materials

› gRNA oligo pool

› Primers for amplifying gRNA pool

  › NM636: ggctgggaacgaaactctgggagctgcgattggca

  › NM637: gccttattttaacttgctatttctagctctaaaac

› Zymo DNA clean and concentrate kit (Zymo #D4013)

› Q5 polymerase (NEB #M0491L)

› AvrIIrestriction enzyme (NEB #R 0174S)

› AscI restriction enzyme (NEB #R0558S)

› Other restriction enzymes, depended on given downstream workflow (HindIII, PmeI, BglII and XhoI if performing downstream PacBio sequencing on the plasmid library)

› BciVi restriction enzyme (NEB #R0596S) if using Addgene "pNTI743 dual-barcoded gRNA parent vector" for generation of divergent promoter template

› HiFi DNA assembly 2x master mix (NEB #E2621L)

› ElectroMAX DH10B Cells (Invitrogen #18290015)

› LB-carbenicillin or LB ampicillin media

› Spectrophotometer to measure OD 600

› HiSpeed Midiprep kit (Qiagen #12643)

› Ultramers to generate the dual barcode sequence

  › RM720:   CCACATGTGCATTGCCTCGGACACTCTTTCCCTACACGACGCTCTTCCGATCTNNNNNNNNNNNNNNNNNNNNNNNNNNCACGT GCTAACAGTGAGGCGCG

  › RM721:   GCTCGATCCAGTCACTCTGGACACGACGCTCTTCCGATCTNNNNNNNNNNNNNNNNNNNNNNNNNNGGTAAGTGACTCGACTG GCGCGCCTCACTGTTAGCACGTG

› Thermocycler

› Thermomixer

› T5 exonuclease (NEB #M0363)

› Shaking Incubator

Procedure

Goal: Insert gRNA library into parent vector

1. Amplify 100 pg CRISPRi Oligo Pool using primers NM636 & NM637 
NM636: ggctgggaacgaaactctgggagctgcgattggca
NM637: gccttattttaacttgctatttctagctctaaaac

Set up eight 50uL PCR reactions using Q5 polymerase, seven using the standard buffer conditions, and one with the optional Q5 High GC Enhancer as detailed by manufacturer protocols (NEB #M0491L). This ensures that even GC rich gRNAs will be represented in your final gRNA library. I use the following PCR conditions.

2. DNA clean&concentrate the PCR reactions according to manufacturer's instructions (Zymo #D4013). Keep the standard bufferred PCR's separate from the GC-buffered PCR.

3. Mix the standard and GC-enhancer-amplified PCR products in 9:1 molar ratio. The 10% high GC PCR mix ensures guides with a high GC content are represented in the gRNA library pool.

4. Digest the parent starting vector (pNTI743 dual-barcoded gRNA parent vector) with AvrII. I use a decent amount of plasmid at this step since you lose plasmid along the purification steps. (I digest 5-10ug of plasmid if I have it). I set the digestion up for 4 hours to ensure complete digestion. Clean and concentrate with the Zymo kit.

Gibson assemble 1ug of digested backbone with 12x-molar concentration of amplified gRNA library pool. Set this up in a total 100uL reaction. Use HiFi DNA Assembly master mix according to manufacturer's protocol. Do not use standard Gibson master mix as high efficiency is key. Incubate at 50°C for 1hour.

5. Zymo Clean and concentrate the Gibson reaction, elute in 12uL nuclease-free water.

6. Electroporate 1 uL of column-cleaned Gibson reaction in 20uL of ElectroMAX DH10B Cells. Refer to manufacturer’s protocols for more guidelines on electroporation conditions, outgrowth, etc. Each vial contains 100uL of electro- competent cells, so you can do up to 5 electroporations for more library coverage of that one library, or you can electroporate other libraries. I usually get close to 30-million transformants per 20uL electroporation which sufficiently covers the ~60,000 gRNAs in my library with ~500x coverage)

7. Take note of the total volume of the outgrowth and plate a small dilution equal to 1:10,000 and 1:100,000 the total cells respresented in the outgrowth. Use the dilution plating to calculate electroporation efficiency and estimate library size and diversity.

8. Innoculate the rest of the outgrowth into 500mL of LB-carbenicillin or LB-ampicillin media in a large Erlenmeyer Flask.

9. Incubate the 500mL liquid culture shaking at 250rpm in 30°C. The 30°C is to make sure the E.coli don’t grow too fast. We want to harvest at an OD close to 2 and avoid overgrowing, which can favor jackpot winners in the library pool. It generally takes close to 12 hours to reach an OD close to 2. I would start monitoring OD about 10 hours after starting the 30°C incubation. When the cells reach an OD of 2, spin down and freeze E.coli pellets.

10. Check on plate dilutions (these can be kept at 37C to permit faster growth). If the coverage looks good, (I aim for at least 100x coverage of library size. So for 60,000 guides, at least 6 million) harvest liquid culture and extract plasmid library. If you fall short of the coverage needed, you can still extract plasmid library and add it to a repeat of the transformation to get more coverage.

11. Extract plasmid library from pellets via midiprep according to manufacturer's intructions (Qiagen #12643)

12. Sanger sequence through the gRNA insertion. You can do this both on a collection of individual colonies from your dilution plate, and on the extracted plasmid pool as a sanity check to make sure you’ve assembled what you’re aiming for. I recommend sequencing ~20 colonies to get a decent estimate of percentage successful gRNA insertion.

Goal: Insert dual barcodes into gRNA plasmid library, with an average of ~4 barcodes per gRNA.

14. Anneal and extend the following oligos. Set up eight 50uL Q5 PCR reactions with a final concentration of 0.5uM for each oligo.

RM720:CCACATGTGCATTGCCTCGGACACTCTTTCCCTACACGACGCTCTTCCGATCTNNNNNNNNNNNNNN NNNNNNNNNNNNCACGTGCTAACAGTGAGGCGCG RM721:GCTCGATCCAGTCACTCTGGACACGACGCTCTTCCGATCTNNNNNNNNNNNNNNNNNNNNNNNNNN GGTAAGTGACTCGACTGGCGCGCCTCACTGTTAGCACGTG

.

Use the following thermocycler conditions: 98C 30s

98C 15s

[68C 15s, 72C 5s] x6

12C hold

.

This specifically excludes the denaturing step in the six cycles to avoid amplifying any specific barcode sequence. You want to uniquely identify each guide RNA with a unique barcode.

15. Zymo column clean the annealed and extended barcode library. This product contains your dual 26N barcode with Gibson homology arms.

16. Determine which restriction enzymes you will use in downstream plasmid library preparation, the in vitro transcription portion of the eventual sequencing library preparation (if preparing DNA barcode libraries), the PacBio sequencing workflow, and any other downstream workflows you have planned. Because the random nucleotide barcode can contain restriction enzyme sites by chance, some barcodes can be lost from the population during downstream restriction digestion. Thus, digesting your dual barcode with any downstream workflow restriction enzymes will purge your dual barcode library of restriction enzyme barcode artifacts.

For my set-up, I use AscI for downstream plasmid library preparation and HindIII, PmeI, BglII and XhoI for PacBio sequencing (Additional details about how these enzymes are used for PacBio library preparation can be found in "Barcode to gRNA assignment").

17. Digest your dual barcode library with these 5 restriction enzymes (Or your chosen enzymes for your given workflow) in a 200uL digestion reaction and suitable NEB digestion buffer that optimizes reaction conditions for the set of restriction enzymes. Zymo column clean.

18. Digest 10ug of gRNA plasmid library from Part 1 with AscI. I set the digestion up for 4 hours to ensure complete digestion. Column clean.

19. Refer to Part 1 steps 5-8 to Gibson assemble your dual barcode library into the digested gRNA plasmid library and electroporate into competent cells. Use the sample molar ratio of dual barcode to plasmid backbone.

20. At this point, you want to control the average number of barcodes that are assigned to a specific gRNA. We chose to generate an average of 4 barcodes per guide RNA, but you can generate other averages. You do so by inoculating several flasks with varying amounts of the transformation and then choosing the set closest to target library size after determining number of total transformants from the plates.

As an example, from the transformation, say we have 1mL of outgrowth with a currently unknown number of transformants in it. Given a gRNA library size of 60,000 we want 240,000 barcoded gRNAs to achieve an average of 4 barcodes per gRNA. I inoculated several flasks with the following percent fractions of that 1mL: 4%, 6%, 8%, 22%, and 44%. Thus, for example, if the dilution plates indicate 2 million transformants in that 1mL, I can combine the 4% (80,000 barcoded gRNA plasmids) and the 8% (160,000 barcoded gRNA plasmids) to reach my target 240,000 barcoded gRNA plasmids.

21. Innoculate the flasks with LB-carbenicillin or LB-ampicillin, using an amount of media proportional to the fraction of the transformation inoculated. For example, 500mL media for 1mL transformation outgrowth, 20mL media for the 4% inoculation, etc.

22. Incubate in 250rpm shaking flask at 30C until reaches OD close to 2. Harvest liquid cultures, pool those that add up to desired barcode library size, and extract plasmid library via midiprep.

23. As an optional step, I sanger sequence through the barcode. You can do this both on a collection of individual colonies from your dilution plate, and on the extracted plasmid pool as a sanity check to make sure you’ve assembled what you’re aiming for.

24. Refer to "Barcode to gRNA assignment" protocol for details on preparing a PacBio sequencing sample to assign barcodes to gRNAs.

Goal: Insert divergent reporter (in my case, P(PGK1)-P(HIS4)) into your barcoded gRNA plasmid library pool

25. Generate a DNA fragment that contains the divergent promoters and ORFs of interest with Gibson homology arms that match the barcoded gRNA library on either side of the AscI site (in between the dual barcodes). This can be done by PCR amplification, or restriction digest from a template plasmid. The P(PGK1)-P(HIS4) template plasmid is available on Addgene (pNTI743 dual-barcoded gRNA parent vector) for this purpose and digestion with BciVi will generate a fragment that contains the necessary Gibson homology arms at each end. Make sure you get one clean DNA product. Primer dimers or PCR amplification of shorter products can still insert if they have the same Gibson homology arms.

As an optional consideration, your DNA fragment can include a 5nt identifier at each end such that the 5nt identifier is placed just upstream of the 3’UTR barcodes. This allows one to use the same paired-end sequencing information, but to insert different reporters into the barcoded-gRNA library and run them in the same pooled experiment. With different nucleotide identifiers, the single-end reads can sequence through the 26N barcode as well as the 5nt reporter ID.

26. Digest 20ug of barcoded gRNA library pool with AscI. Asc1 cuts in between the two random nucleotide barcodes such that an inserted divergent promoter will drive expression of each separately. I set the digestion up for 4 hours to ensure complete digestion. After the AscI digestion, add 0.2 units of T5 exonuclease (NEB) and digest for 20mins at 30°C. (AscI and T5 conveniently use the same buffering conditions). The T5 exonuclease chews back the overhangs left by AscI and prevents re-ligation during the Gibson reaction. Clean and concentrate the backbone.

27. Gibson assemble 1ug of digested backbone with 5x-molar concentration of amplified divergent reporter insert. I set this up in a total 50uL reaction. Use HiFi DNA Assembly master mix. Do not use standard Gibson master mix. Incubate at 50°C for 1hour.

28. Clean and concentrate the Gibson reaction, elute in 6uL nuclease-free water to maximize concentration, which should be close to 100ng/uL (theoretical yield).

29. Electroporate 1 uL of column-cleaned Gibson reaction into 20uL of ElectroMAX DH10B Cells. Refer to manufacturer’s protocols for more guidelines. Each vial contains 100uL of electro-competent cells, so you can do up to 5 electroporations for more library coverage of that one library, or you can electroporate other libraries. I usually get close to 30-million transformants per 20uL electroporation which sufficiently covers the ~270,000 barcodes)

30. Plate a 1:100,000 and 1:1,000,000 dilution of the electroporation outgrowth to calculate electroporation efficiency and estimate library size and diversity.

31. Innoculate the rest of the electroporation into 400mL of LB-carbenicillin in a large Erlenmeyer Flask.

32. Incubate the 400mL liquid culture shaking at 250rpm in 30°C. The 30°C is to make sure the E.coli don’t grow too fast. We want to harvest at an OD between 2-4 and avoid overgrowing, which can favor jackpot winners in the library pool. It generally takes close to 11 hours to reach an OD close to 2. I would start monitoring OD about 10 hours after starting the 30°C incubation. (I would avoid using either of the 6th floor Brun 30C shakers. I’ve had disasters with both of them)

Check on plate dilutions. If the coverage looks good, (I aim for at least 20x coverage of library size. So for 270,000 guides, at least 6 million) harvest liquid culture and extract plasmid library. If you fall short of the coverage needed, you can still extract plasmid library and add it to a repeat of the transformation to get more coverage.

33. I sanger sequence through the barcode and into the inserted divergent reporter. You can do this both on a collection of individual colonies from your dilution plate, and on the extracted plasmid pool as a sanity check to make sure you’ve assembled what you’re aiming for. This plasmid library is ready for library experiments with yeast expressing dCas9-Mxi1