CRISPR screens

A library of gRNAs targeting 7,253 known or putatively druggable genes with 5 gRNAs per gene for a total of 36,265 gRNAs was synthesized and cloned into the lentiCRISPR V2 puro vector (Addgene cat# 52961) digested with BsmBI. Sequences are available as Table S1. Pooled virus was prepared by transfecting 293T cells with the library plasmid pool along with the psPax2 (Addgene cat# 12260) and pMD2.G (Addgene cat# 12259) lentiviral packaging vectors. 293T cells were plated 1 day prior to transfection to make virus at 14.4 million cells per 15 cm plate so that the cells are ~80-90% confluent at the time of transfection. For a genome scale library, make 10-20 15 cm plates of virus.

Per 15cm plate to make library virus:

• 3.38 micrograms of psPax2
• 1.69 micrograms of pMD2.G
• 6.76 micrograms of lentiviral CRISPR library plasmid
• 30.4 microliters of PolyJet
• 3.4 mL of OptiMEM

Viral supernatants were harvested at 48 and 72 hours post-transfection, cell debris was removed by centrifuging at 500xG for 10 min, and concentrated with lenti-X concentrator solution (3 volumes of viral supernatant to 1 volume lenti-X) (Clontech cat# 631231) for 48 h at 4C on ice. Lenti-X viral concentrate was centrifuged at 1500xG for 45 min. Viral pellet was then resuspended in PBS for a 40 fold concentration from the original viral supernatant. Virus was subsequently treated with benzonase (Millipore cat# 71206-3) to remove any contaminating plasmid DNA from the transfection as follows.

1)  Add 10X Benzonase Buffer to final concentration of 1X

10X Benzonase Buffer:

500mM Tris-HCl pH 8.0, 10 mM MgCl2, 1 mg/ml bovine serum albumin (BSA)

2)  Add Benzonase to final concentration 500 units/ml.  

3)  Wrap tubes in parafilm to keep sterile

4)  Place on shaker or rotator in 37⁰C incubator.  Incubate 30 min.

5)  Following incubation, remove tubes and thoroughly cleanse with 70% ethanol.  Include multiple small aliquots for titering infection.

6)  Store at -80⁰ C.  

Titer the virus as follows:

• Before beginning the screen, titer the virus you prepared on the cell line(s) that you will be using. I typically plate cells in a 6-well dish at the same density that will be used for infection for the screen. For example, if you are going to have 10 million cells per 15cm dish for the screen, plate 0.59 million cells per well of the 6-well plate (this is a ~17:1 ratio based on the surface area difference between a 6-well plate and a 15cm plate).
• Infect the 6-well plate with 10-fold serial dilutions of the virus to estimate the titer. Start with 10uL of concentrated virus per well and dilute down to 1, 0.1, 0.01, and 0.001uL of virus. Leave one well uninfected as a control for puro selection.
• Add growth media with puromycin 1 or 2 days after infection and select until the uninfected well is completely dead.
• Determine the amount of virus needed that gives between 60-80% killing. This will be a multiplicity of infection (MOI) of around 0.2-0.3 for the screen.

Determining the number of cells needed for the start of the screen

• The size of the library to be screened will determine the number of cells that you’ll need. Determine that number by multiplying the number of reagents in the library (sgRNAs, shRNAs, etc.) by the representation you want to maintain. For example, if the library is 40,000 sgRNAs and you want to maintain representation of 500/sgRNA then you will need at least 20 million cells per replicate.

(40,000 sgRNAs) x (500) = 20 million cells per replicate

• Since you’ll be screening in at least duplicate or triplicate you’ll need to multiply the number of cells by 2 or 3.  For example, if the library is 40,000 sgRNAs at a representation of 500/sgRNA then you’ll need: 

(20 million cells) x (3 replicates) = 60 million cells.

• Lastly, since you want a low MOI to ensure only 1 sgRNA per cell, you’ll need 3-5 times as many cells as virus. So for a library of 40,000 sgRNAs at a representation of 500 in triplicate then you’ll need: 

(40,000 sgRNAs) x (500) x (3 replicates) x (3 fold excess cells) = 180 million cells in total

Infecting the library to begin the screen

• One day prior to infection, plate the required number of cells. Plate the same number per plate that you used to determine the titer of the virus. Plate an extra plate to use as a puromycin killing control plate to determine selection. For example, if you need 180 million cells for the screen at 10 million cells per 15cm plate then you’ll need:

(180 million cells)/(10 million cells per plate) = 18 plates + 1 for puro selection control

This will result in 6 plates per independently infected replicate

• For the virus, determine how much to add based on the titering experiment. If you found that 1uL of virus resulted in ~60-80% killing then you’ll need to add 17uL per 15cm plate

(1uL for 60-80% killing in 6-well) x (17 times as many cells for 15cm plate) x 18 total plates = 306uL of virus needed for infection

• Make a master mix of virus, polybrene, and growth media for infection:

(20mL of media per plate x 18 plates) = 360 mL of growth media

360 mL growth media + 306uL of virus + 360uL 8mg/mL polybrene

• Aliquot ~20mL of the viral master mix to each 15cm plate of cells and incubate overnight. Add fresh media with no virus to the puro selection control plate
• 1 or 2 days after infection, add growth media with puromycin to the cells to start selection. Keep cells in puromycin until the puro selection control plate is completely dead (usually 2-4 days).
• Once the control plate is dead, harvest each replicate and determine the total number of cells you have. In the above example of a library of 40,000 sgRNAs with a representation of 500 you’ll need to keep at least 20 million cells growing for each replicate to maintain the library and you’ll need to take a cell pellet for each replicate of at least 20 million cells for your initial time point for genomic DNA isolation. Try to freeze an excess number of cells i.e. if you need at least 20 million cells then try to take a cell pellet of at least 30 million. Store cell pellets for genomic DNA extraction/sequencing in -80C until ready. This is also where you will inject the mice. Make sure you have enough cells to inject mice and also keep cells in culture to monitor in vitro growth.
• You can also freeze an aliquot of the library infected cells in case you want to thaw them later without having to reinfect the library. Make sure to freeze at least representation plus extra since some will die during thaw.
• Passage the cells as necessary for your screen, maintaining at least representation throughout. For example, if representation is 20 million cells then each time you count and replate cells for the screen you need to replate at least 20 million cells.
• Once the desired number of population doublings or time after treatment has passed, take a final time point cell pellet of at least representation plus extra. Once tumors reach ~2 cubic cm, sac the mice, harvest tumors, and store at -80C until Genomic DNA extraction.

Genomic DNA isolation for sequencing

- Gather cell pellets isolated during the screen that you want to sequence. We use phase-lock tubes and to ease genomic DNA isolation from phenol:chloroform extractions. Some labs use genomic DNA isolation kits but we’ve found that you lose a fraction of your genomic DNA in most kits so you’ll need to make sure you start with more than enough cells if you decide to go this route.

1. Resuspend cell pellet in TE (10mM Tris pH 8.0, 10mM EDTA)

• To final concentration 2 million – 10 million cells/mL of TE
• Break up pellet really well by pipetting

1. Add SDS + Proteinase K

• To a final concentration of 0.5% SDS, 0.5mg/mL proteinase K
• 10mL TE => 0.55mL SDS, 0.27mL proteinase K

1. Incubate @ 55°C o/n

• Go back a few times to shake over the course of one hour, invert a few times to help digest the cell

1. Add NaCl to final conc 0.2M for each 

• Make sure it’s homogenous/clear the next morning (no cell-like floaty bits) to ensure complete digestion.

1. Phenol-chloroform/chloroform extraction

• 1 round phenol chloroform extraction
• 1 round chloroform extraction
• Use PLG tubes (phase lock gel tubes, Qiagen Maxtract 15 mL catalog # 129065) – traps organic layer “locks” phase, pour off aqueous phase – use 15mL tube – 10mL sample in 2 tubes
• Pre-spin tubes according to manual
• Mix equal parts phenol:chloroform with sample in PLG tube
• Shake for 1 full minute to extract
• 5 min spin – DNA aqueous phase now on top 
• Repeat for chloroform extraction

1. Incubate tubes cap open 50°C incubator 1h to evaporate chloroform (optional)

1. RNase A o/n treatment 37°C

• Add to final concentration 25mg/mL RNase A 

1. Phenol-chloroform/chloroform extraction in PLG tubes

• 1 round phenol-chloroform extraction
• 1 round chloroform extraction

1. If you used multiple PLG tubes per sample then pool all aliquots for each replicate at each time point.

1. Precipitate DNA with ethanol o/n @ -20°C in 50mL conicals

• or -80°C 3h
• Add 1/10 v/v 3M sodium acetate pH 5.2 + 2 volumes 100% ethanol

1. Spin 30-45 min 4500rpm 4°C
2. Wash 1x 70% EtOH 1-1.5mL and transfer DNA pellet to an eppendorf tube
3. Wash 2 more times with 70% ethanol
4. Dry open cap 37°C >10-20 min to evaporate ethanol
5. Resuspend in ~1mL EB/TE, minimize shear or put @ 55°C to resuspend by dissolving
6. Store @ -20°C long-term or 4°C short-term

Genomic DNA is now ready for PCR amplification for sequencing. Sequencing primers and conditions will vary depending on the library screened.

CRISPR library PCR protocol

PCR1 Primer sequences:

LC353F

AAT GGA CTA TCA TAT GCT TAC CGT AAC TTG AAA GTA TTT CG

LCR2L

TCT ACT ATT CTT TCC CCT GCA CTG TTG TGG GCG ATG TGC GCT CTG

PCR1 reaction info:

You need to have each cell represented in the PCR. So if you have a representation of 500 and a library of 100,000 sgRNAs then you’ll need to have enough genomic DNA to cover 500 * 100,000 = 50 million cells. Diploid cells have 6.6 pg of DNA but your tumor cells are likely aneuploid so you’ll need to take that into account.

6.6pg of DNA/cell * 50,000,000 cells = 330 micrograms of DNA needed per condition

We do 12 micrograms of DNA per 100uL PCR1 reaction. This means you’ll need 28 reactions per condition

400 microgram / 12 microgram per reaction = 27.5 total PCR1 reactions

We use Q5 polymerase from NEB for our PCR reactions for screening.  The catalog number is M0493L. https://www.neb.com/products/m0493-q5-hot-start-high-fidelity-dna-polymerase#Product%20Information

Alternatively you can use Q5 2X mastermix (https://www.neb.com/products/m0492-q5-high-fidelity-2x-master-mix#Product%20Information

)

For each 100uL PCR1, assemble the following components:

Or if you use the 2X master mix:

The thermal cycler conditions for PCR1 are as follows:

Combine the PCR1 reactions for each condition (in the above example, put the 28 PCR1 reactions together in 1 tube). Do a PCR cleanup on a portion of the combined PCRs (200-500uL is more than enough). Run a gel to verify correct band size.

PCR2

For PCR2, use 100-500ng of PCR1 from the cleanup step. Here you’ll only have 1 reaction per screening condition.

Stagger cocktails for PCR2 forward. Stagger primers help to generate diversity on the flow cell so it can more accurately define clusters.

Make an equimolar ratio cocktail of the following:

KMN_stagger_PCR2_F01

ACACTCTTTCCCTACACGACGCTCTTCCGATCTTCTTGTGGAAAGGACGAAACACCG

KMN_stagger_PCR2_F02

ACACTCTTTCCCTACACGACGCTCTTCCGATCTcTCTTGTGGAAAGGACGAAACACCG

KMN_stagger_PCR2_F03

ACACTCTTTCCCTACACGACGCTCTTCCGATCTagTCTTGTGGAAAGGACGAAACACCG

KMN_stagger_PCR2_F04

ACACTCTTTCCCTACACGACGCTCTTCCGATCTgagTCTTGTGGAAAGGACGAAACACCG

KMN_stagger_PCR2_F05

ACACTCTTTCCCTACACGACGCTCTTCCGATCTcgagTCTTGTGGAAAGGACGAAACACCG

KMN_stagger_PCR2_F06

ACACTCTTTCCCTACACGACGCTCTTCCGATCTtcgacTCTTGTGGAAAGGACGAAACACCG

KMN_stagger_PCR2_F07

ACACTCTTTCCCTACACGACGCTCTTCCGATCTatcaacTCTTGTGGAAAGGACGAAACACCG

KMN_stagger_PCR2_F08

ACACTCTTTCCCTACACGACGCTCTTCCGATCTgaacgaaTCTTGTGGAAAGGACGAAACACCG

PCR2 reverse primer

KM3_LCV2_PCR2R

GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTTCTACTATTCTTTCCCCTGCACTGT

PCR2 thermal cycler conditions:

PCR3

There’s no need to gel extract or purify after PCR2. Simply use 2uL of PCR2 as a template for PCR3

KMN_LCV2_PCR3F

AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCT

PCR3 reverse are our standard indexing primers

PCR3 thermal cycler conditions (2uL of PCR2 as template):

Example indexing primers:

CAAGCAGAAGACGGCATACGAGATtcgcaggGTGACTGGAGTTCAGACGTGT

CAAGCAGAAGACGGCATACGAGATctctgcaGTGACTGGAGTTCAGACGTGT

CAAGCAGAAGACGGCATACGAGATcctaggtGTGACTGGAGTTCAGACGTGT

where the lowercase red portion serves as a barcode to allow for multiplexing of samples.

After PCR3, combine the reactions in equimolar ratios and submit for sequencing.

Trim sequencing reads to the 20 nt protospacer sequence, align sequencing reads to CRISPR library, and generate read counts. Analyze read counts via MAGeCK, edgeR, etc.