Protocol for eVLP-mediated CRISPR-Cas9 and prime editing in pancreatic cancer cell lines

Summary

Pancreatic ductal adenocarcinoma (PDAC) cell lines resist transfection, limiting gene-editing experiments. Here, we present a protocol for delivering CRISPR-Cas9 or prime editors into PDAC cell lines using retrovirus-derived engineered virus-like particles (eVLPs). We describe steps for generating eVLPs, packaging CRISPR-Cas9 ribonucleoproteins or prime editors with engineered prime-editing guide RNAs (epegRNAs), and transducing human PDAC cells. We use the HEK3 locus and NQO1 (NAD(P)H quinone dehydrogenase 1) exon 6 as example edits and controls, respectively. This protocol adapts readily to other genomic sites.

Graphical abstract

Highlights

• •Steps for generating retrovirus-derived eVLPs for genome editing of PDAC cells
• •Instructions for packaging Cas9 RNPs and prime-editing cargos into eVLPs
• •Guidance on eVLP transduction of SUIT-2, PANC-1, and MIA PaCa-2 cells
• •Instructions for identifying CRISPR/prime editing-mediated indels using PCR

Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics.

Pancreatic ductal adenocarcinoma (PDAC) cell lines resist transfection, limiting gene-editing experiments. Here, we present a protocol for delivering CRISPR-Cas9 or prime editors into PDAC cell lines using retrovirus-derived engineered virus-like particles (eVLPs). We describe steps for generating eVLPs, packaging CRISPR-Cas9 ribonucleoproteins or prime editors with engineered prime-editing guide RNAs (epegRNAs), and transducing human PDAC cells. We use the HEK3 locus and NQO1 (NAD(P)H quinone dehydrogenase 1) exon 6 as example edits and controls, respectively. This protocol adapts readily to other genomic sites.

Before you begin

Efficient delivery of CRISPR-based genome editing systems into PDAC cells remains challenging.¹ Commonly used lipid-based transfection and electroporation approaches show reduced efficiency and increased cytotoxicity compared to viral-derived delivery methods.² Prolonged expression of genome-editing components following plasmid-based delivery can further exacerbate off-target editing and cellular stress in these systems.^3,4

Engineered virus-like particles (eVLPs) enable transient, non-integrating delivery of CRISPR/Cas9 or prime editor ribonucleoprotein (RNP) complexes.^5,6 This approach combines the high delivery efficiency associated with viral systems with the safety profile of RNP-based editing, thereby reducing cytotoxicity and minimizing prolonged nuclease activity.⁷

NAD(P)H quinone dehydrogenase 1 (NQO1) is a detoxification enzyme that is overexpressed in PDAC.^8,9,10 Its levels are influenced by a C609T single nucleotide polymorphism (SNP).^11,12 We recently demonstrated that high PDAC tumour NQO1 levels predict better outcome following adjuvant gemcitabine (Gem) plus capecitabine (GemCap) but not Gem monotherapy.¹³ Testing response to chemotherapeutic drugs in syngeneic cells where NQO1 levels are altered could contribute to our understanding of how the levels of NQO1 influence treatment response.

For CRISPR/Cas9-mediated gene disruption, the protocol described here employs a dual-guide RNA (dual-sgRNA) strategy, in which two sgRNAs target the same exon to promote the excision of a defined genomic fragment. In the context of NQO1 editing, dual sgRNAs targeting exon 6 are used to increase knockout efficiency by favouring large deletions over small insertions or deletions. For effective application of this approach, the selected exon should be constitutively expressed across transcript isoforms and of sufficient length to accommodate two sgRNA target sites with appropriate spacing (typically 50–200 bp) to enable efficient excision. Where possible, exon selection should ensure that deletion of the targeted region disrupts protein function, for example by introducing a frameshift or removing functionally critical domains. This strategy improves the likelihood of functional gene disruption in PDAC cell lines, which often exhibit reduced editing efficiencies due to slow cell-cycle progression.¹⁴

For prime editing applications, both sense and antisense epegRNAs are evaluated for targeting the HEK Site 3 (HEK3) locus. Testing epegRNAs on opposing strands allows empirical determination of optimal editing orientation, as prime editing efficiency can vary substantially depending on DNA sequence composition, local chromatin context, and repair pathway engagement. This strategy is particularly important in PDAC cell lines, where cellular repair dynamics may differ from more commonly used, rapidly dividing model systems.

All cell lines used in this protocol were confirmed to be mycoplasma-negative prior to experimentation. SUIT-2, PANC-1, and MIA PaCa-2 cells were cultured in standard Dulbecco’s Modified Eagle Medium (DMEM) supplemented with foetal bovine serum under standard conditions. Cell lines should have short tandem repeat (STR) profiling where possible to ensure experimental reproducibility.

Innovation

This protocol uses MMLV-based eVLPs as transient, non-integrative delivery systems for CRISPR/Cas9 and prime editing tools in PDAC cell lines. Unlike standard viral transduction, this system packages pre-assembled RNPs without introducing viral genetic material, thereby reducing insertional mutagenesis risks. It offers an efficient means to generate precise edits in difficult-to-transfect/transduce PDAC cells and can be adapted for other research questions simply by changing guide RNA/epegRNA oligonucleotides designs.

Institutional permissions

All experimental procedures were performed in accordance with the University of Liverpool guidelines. Users of the protocol must obtain similar permissions from their respective institutions.

Enhanced prime-editing guide RNA and gRNA design

Timing: 1–2 days

This section describes the design phase for guide RNA experiments, including expansion of plasmid stocks obtained from Addgene, identification of target sequences for CRISPR/Cas9-mediated knockout of NQO1, and design of epegRNAs for prime editing at the HEK3 locus.

• 1.
  Expansion of Plasmids.

  • a.
    Streak bacterial stab cultures obtained from Addgene onto LB agar plates containing the appropriate antibiotic for the desired plasmid.
  • b.
    Incubate plates overnight at 37°C.
  • c.
    Select a single colony and inoculate into 5 mL LB broth supplemented with 100 μg/ mL Carbenicillin.
  • d.
    Incubate overnight (12–18 h) at 37°C with shaking (200–250 rpm).
  • e.
    Isolate plasmid DNA using Monarch® Spin Plasmid Miniprep Kit according to the manufacturer’s instructions.
  • f.
    Prepare glycerol stocks by mixing 0.5 mL bacterial culture with 0.5 mL 50% sterile glycerol and store at −80°C for long-term preservation.

Note: All oligonucleotide sequences and/or plasmid are provided in the key resources table.

• 2.
  sgRNA Design for CRISPR/Cas9 Targeting of NQO1.

  • a.
    Retrieve the human genomic sequence of NQO1 from the Ensemble genome database.
  • b.
    Identify target sequences within exon 6 for dual-guide CRISPR-mediated excision shown in Figure 1A.
  • c.
    Select sgRNAs based on predicted on-target efficiency and minimal off-target activity using CRISPOR online program.
  • d.
    Order complementary oligonucleotides encoding each sgRNA protospacer sequence see Figure 1B.

CRITICAL: All oligonucleotide sequences are provided in the key resources table.

Note: Guides for CRISPR/Cas9 Targeting of NQO1 were designed on the using the PAM sites 16:69,711,235 (AGG) and 16:69,711,219 (TGG).

Note: Dual-guide design increases the likelihood of generating functional knockouts by promoting defined exon excision rather than small indels.¹⁴

• 3.
  ePegRNA Design for Prime Editing at HEK3.

  • a.
    Retrieve the human HEK3 locus sequence.
  • b.
    Use the online software https://optipri.me/, a mechanistic machine learning model, to design epegRNAs targeting the desired edit at the HEK3 locus.¹⁵
  • c.
    Generate and compare candidate epegRNAs in both sense and antisense orientations, where supported by OptiPrime.
  • d.
    Select candidate epegRNAs primarily according to the predicted editing outcomes reported by OptiPrime, while considering PBS length, reverse transcription template length and other design parameters included in the software output.

Note: All oligonucleotide sequences are provided in the key resources table and are newly designed for this protocol. Prime editing efficiency is highly dependent on epegRNA architecture and strand orientation; empirical testing of multiple designs is recommended.¹⁴

Figure 1.

Design and cloning of sgRNA oligonucleotides

(A) Schematic showing the target-specific guide sequence, adjacent protospacer adjacent motif (PAM), and cloning overhangs required for directional insertion into the sgRNA expression vector.

(B) Complementary sgRNA oligonucleotides are designed with vector-compatible 5′ overhangs, annealed, and prepared for ligation. The guide sequence is replaced with the sequence specific to the genomic locus of interest.

(C) The annealed oligonucleotide is ligated into the linearized vector, followed by transformation into competent E. coli. Positive clones are identified by colony screening and confirmed by Sanger sequencing before use in downstream CRISPR/Cas9 experiments. The HEK3-targeting sgRNA is shown as an example.

Cloning of sgRNA and epegRNA constructs for CRISPR-Cas9 and prime editing

Timing: 3–4 days

This section describes the complete workflow for generating guide RNA expression plasmids used in CRISPR/Cas9 and prime editing experiments. The process includes: (1) introduction of BsmBI (Esp3I) restriction sites and the COM-scaffold sequence into the COM-epegRNA-Dnmt1 plasmid to generate the pU6 gRNA Com scaffold; (2) cloning of custom sgRNA guide sequences into the modified pU6 gRNA Com scaffold; (3) cloning of sgRNA sequences for CRISPR/Cas9-mediated knockout of NQO1; and (4) assembly of epegRNA constructs for prime editing at the HEK3 locus.

• 4.
  Introduction of BsmBI Cloning Sites and COM-scaffold sequence into COM-epegRNA-Dnmt1.

  • a.
    Design complementary oligonucleotides containing BsmBI recognition sites and the COM-scaffold sequence that will be inserted downstream of the U6 promoter to create a compatible cloning cassette as shown in Figure 1C.

    CRITICAL: The oligonucleotides should contain the following overhangs to enable subsequent ligation of the oligonucleotide duplex into COM-epegRNA-Dnmt1: Forward oligo 5′-CCGG…-3′ and Reverse oligo 5′-AGCT…C-3'. These overhangs ensure compatibility with XmaI and HindIII digested COM-epegRNA-Dnmt1.

  • b.
    Order complementary oligonucleotides from a commercial synthesis provider e.g., Ultramer oligonucleotides from IDT.
  • c.
    Resuspend lyophilized oligonucleotides in nuclease-free water to a final concentration of 100 μM. Vortex briefly and centrifuge to collect the solution.
  • d.
    Prepare annealing reactions by combining:

    • i.
      1 μL of 100 μM forward oligonucleotide.
    • ii.
      1 μL of 100 μM reverse oligonucleotide.
    • iii.
      1 μL of 10× T4 DNA ligase buffer.
    • iv.
      7 μL nuclease-free water.
  • e.
    Anneal oligonucleotides using a thermocycler with the following program: 95°C for 5 min, then gradually cool to 25°C at a rate of 5°C per minute.
  • f.
    Digest 1–5 μg of the COM-epegRNA-Dnmt1with XmaI and HindIII.
  • g.
    Purify the digested vector by agarose gel electrophoresis and extract the backbone fragment using a gel purification kit. Elute in 10–15 μL of TE buffer to obtain a concentrated eluate.
  • h.
    Set up the ligation reaction:

    • i.
      50 ng of digested vector backbone.
    • ii.
      1 μL of annealed oligonucleotide duplex (diluted 1:200).
    • iii.
      5 μL of 2× quick ligase buffer.
    • iv.
      Nuclease-free water to 10 μL.
    • v.
      1 μL of T4 DNA ligase.
  • i.
    Incubate at room temperature for 10 min.
  • j.
    Transform competent E. coli cells with the ligation reaction and plate onto LB agar containing 100 μg/mL carbenicillin.
  • k.
    Incubate plates overnight at 37°C.
  • l.
    Screen colonies by Sanger sequencing using a U6 promoter-specific primer (5′-GAGGGCCTATTTCCCATGATTCC-3′) to confirm successful insertion of BsmBI sites and COM-scaffold.

    CRITICAL: Verify correct orientation and sequence integrity before proceeding to guide cloning. Store verified plasmid at −20°C.

• 5.
  Cloning of Custom sgRNA Guides.

  • a.
    Design sgRNA target sequences using CRISPOR or equivalent computational tools. Select guides based on predicted on-target efficiency and minimal off-target activity.

    Note: For dual-guide strategies targeting the same exon, select sgRNAs with appropriate spacing (50–200 bp) to enable efficient excision.

  • b.
    Order complementary oligonucleotides encoding each sgRNA protospacer sequence with the following configuration:

    • i.
      pU6 BsmBI-COM-scaffold Forward oligo: 5′-CACCG[N20]-3′ (where N20 is the protospacer sequence).
    • ii.
      pU6 BsmBI-COM-scaffold Reverse oligo: 5′-AAAC[N20 reverse complement]C-3′.
  • c.
    Anneal complementary oligonucleotides as described in steps 4d above.
  • d.
    Digest the pU6 gRNA Com scaffold (generated in step 4) with BsmBI.
  • e.
    Purify the linearized vector by gel extraction.
  • f.
    Ligate annealed sgRNA oligonucleotides into the BsmBI-digested vector using T4 DNA ligase as described in step 4h.
  • g.
    Transform competent E. coli and plate onto LB agar with carbenicillin.
  • h.
    Incubate overnight at 37°C.
  • i.
    Screen colonies by PCR or restriction digest analysis.
  • j.
    Confirm correct sgRNA insertion by Sanger sequencing using the U6 promoter primer.

    CRITICAL: Verify the absence of unwanted mutations in the sgRNA sequence and scaffold region before proceeding to downstream applications.

• 6.
  Cloning of sgRNA Sequences for CRISPR/Cas9 (NQO1 Targeting).

  • a.
    Digest the sgRNA expression vector (pU6-gRNA-COM-scaffold) with BsmBI (Esp3I).
  • b.
    Purify the linearized vector using a gel extraction kit.
  • c.
    Anneal complementary sgRNA oligonucleotides (sgRNA guide 1 and sgRNA guide 2):

    • i.
      Combine 1 μL of each 100 μM oligonucleotide with 1 μL 10× T4 ligation buffer and nuclease-free water to 10 μL.
    • ii.
      Incubate at 95°C for 5 min.
    • iii.
      Gradually cool to room temperature over 30–60 min.
  • d.
    Dilute the annealed oligonucleotides (e.g., 1:200 in nuclease-free water).
  • e.
    Ligate the annealed oligonucleotides into the BsmBI-digested vector using T4 DNA ligase.
  • f.
    Transform competent E. coli with the ligation reaction and plate onto LB agar containing 100 μg/mL carbenicillin.
  • g.
    Incubate overnight at 37°C.
  • h.
    Screen colonies by PCR or restriction digest and confirm sgRNA insertion by Sanger sequencing.
• 7.
  Cloning of epegRNA Constructs for Prime Editing (HEK3 Targeting).

  • a.
    Digest the pU6-pegRNA-GG vector with BsaI-HFv2.
  • b.
    Gel purify the ∼2.2 kb backbone fragment containing the U6 promoter and selection cassette.
  • c.
    Prepare three oligonucleotide components:

    • i.
      Protospacer-target sequence (HEK site3 guide antisenseF (29-mer)/HEK site3 guide antisenseR (29-mer) and HEK site3 guide senseF (29-mer)/HEK site3 guide senseR (29-mer)).
    • ii.
      epegRNA 3′ extension- PBS + RT template (HEK site 3 PEG-RT_PBS antisense for Tevo (84-mer)/HEK site3 PEG-RT_PBS antisense rev Tevo (84-mer) and HEK site3 PEG-RT_PBS sense for Tevo (74-mer)/HEK site3 PEG-RT_PBS anti rev Tevo (74-mer)).
    • iii.
      sgRNA scaffold fragment (contained with-in pU6-pegRNA-GG).
  • d.
    Anneal complementary oligonucleotides for each component:

    • i.
      Combine 1 μL of each 100 μM oligonucleotide with annealing buffer to 25 μL.
    • ii.
      Incubate at 95°C for 3 min.
    • iii.
      Gradually cool to 22°C at 0.1°C/s.
  • e.
    Phosphorylate sgRNA scaffold oligonucleotides using T4 polynucleotide kinase (PNK).
  • f.
    Assemble epegRNA constructs using a Golden Gate reaction:

    • i.
      Combine digested vector, from step 7a, annealed oligonucleotides (all components), BsaI-HFv2, T4 DNA ligase, and ligase buffer.
    • ii.
      Incubate at room temperature for 10 min or cycle between 16°C and 37°C.
  • g.
    Perform final enzyme inactivation by 80°C incubation.
  • h.
    Transform competent E. coli with 1 μL of the assembly reaction and plate onto LB agar containing carbenicillin (100 μg/mL).
  • i.
    Incubate overnight at 37°C.
  • j.
    Screen colonies and confirm correct epegRNA assembly by Sanger sequencing.

CRITICAL: sgRNA and epegRNA cloning use different restriction enzymes (BsmBI vs BsaI) and cloning strategies. These workflows must not be interchanged.

Note: epegRNA assembly requires correct incorporation of all three components (protospacer, scaffold, and 3′ extension). Failed assembly often results in incomplete constructs lacking editing activity.

Pause point: Verified plasmid DNA can be stored at −20°C for long-term use prior to eVLP assembly.

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Bacterial and virus strains
Escherichia coli DH5[alpha]	Biosearch Technologies	Cat# 60602-1
Chemicals, peptides, and recombinant proteins
DMEM, High Glucose - W L-Glut, Sodpyruvate And Sod Bicarbonate	Sigma-Aldrich	Cat# D6429
Trypsin-EDTA solution	Sigma-Aldrich	Cat# T3924
PBS, pH 7.2	Gibco	Cat# 20012019
FBS	Gibco	Cat# 16000044
Hank’s Balanced Salt Solution (HBSS)	Gibco	Cat# 14025100
Polyethylenimine	Polysciences	Cat# 23966-2
PEG-it Virus Precipitation Solution	System Biosciences	Cat# LV810A-1
GoTaq® G2 DNA Polymerase	Promega	Cat# M7841
Agarose	Invitrogen	Cat# 16500500
Polybrene	Sigma-Aldrich	Cat# TR-1003-G
TAE Buffer 50×	Geneflow	Cat# B9-0030
T4 ligase	Promega	Cat# ADM1801
dNTP	Promega	Cat# ADU1240
Glycerol	Fisher Scientific	Cat# 10579570
Carbenicillin	Sigma-Aldrich	Cat# C1613
Purple 100bp ladder	New England Biolabs	Cat# N3231S
PEI MAX® Transfection Reagent	Kyfora Bio	Cat# 24765
Tryptone	Melford	Cat# 91079–40–2
Yeast extract	Melford	Cat# 8013-01-2
Potassium Chloride KCl	Fisher Scientific	Cat# 10375810
Glucose	Fisher Scientific	Cat# 10141520
Agar	Sigma-Aldrich	Cat# A1296
Stationary phase for GC, phase Synperonic PE/F68	Supelco	Cat# 81112
Opti-MEM Reduced Serum Medium	Gibco	Cat#31985062
BsaI-HF®v2	New England Biolabs	Cat# R3733S
XmaI	New England Biolabs	Cat# R0180S
HindIII	New England Biolabs	Cat#R0104S
Esp3I (BsmBI)	Thermo Fisher Scientific	Cat# ER0451
Critical commercial assays
Wizard® SV Genomic DNA Purification System	Promega	Cat# A2360
Monarch® Spin Plasmid Miniprep Kit	New England Biolabs	Cat# T1110S
Pierce™ BCA Protein Assay Kits	Thermo Scientific	Cat#23225
Wizard® SV Gel and PCR Clean-Up System	Promega	Cat# A9281
Deposited data
Human NQO1	Ensembl	ENSG00000181019
Human EPHA8 (HEK3)	Ensembl	ENSG00000070886
Experimental models: Cell lines
SUIT-2	Creative Bioarray	Cat# CSC-C6655J
PANC-1	Creative Bioarray	Cat#CSC-C7095J
MIA-PaCa-2	Creative Bioarray	Cat#CSC-C9492L
HEK293T	ATCC	Cat# CRL-3216
Oligonucleotides
HEK site3 guide antisenseF (29-mer): caccgacgccctctggaggaagcagttta	IDT DNA	N/A
HEK site3 guide antisenseR (29-mer): ctcttaaactgcttcctccagagggcgtc	IDT DNA	N/A
HEK site 3 PEG-RT_PBS antisense for Tevo (84-mer): gtgcgggcccagact gagcacggattacaaggatgacgacgataagtgatggcagaggaaaggaagccctgcttcctccagagg	IDT DNA	N/A
HEK site 3 PEG-RT_PBS antisense rev Tevo (84-mer): cgcgcctctggag gaagcagggcttcctttcctctgccatcacttatcgtcgtcatccttgtaatccgtgctcagtctgggccc	IDT DNA	N/A
HEK site3 guide senseF (29-mer): caccggcccagactgagcacgtgagttta	IDT DNA	N/A
HEK site3 guide senseR (29-mer): ctcttaaactcacgtgctcagtctgggcc	IDT DNA	N/A
HEK site 3 PEG-RT_PBS sense for Tevo (74-mer): gtgctggaggaagca gggcttcctttcctctgccatcacttatcgtcgtcatccttgtaatccgtgctcagtct	IDT DNA	N/A
HEK site 3 PEG-RT_PBS anti rev Tevo (74-mer): cgcgagactgagcacgg attacaaggatgacgacgataagtgatggcagaggaaaggaagccctgcttcctcca	IDT DNA	N/A
sgRNA guide 1, 5′- CAC CGA ATG CAG AAT GCC ACT CTG- 3′	IDT DNA	N/A
sgRNA guide 2, 5′- CAC CGC CCG AAT TCA AAT CCT GGA-3′	IDT DNA	N/A
F1 HEK3 5′- CGG ATT ACA AGG ATG ACG ATA AGT G- 3′	IDT DNA	N/A
R3 HEK3 5′- GCA TGG ATG AGA GAA GCC TGG- 3′	IDT DNA	N/A
NQO1 screen F2 5′- TAG TGT GCC TGA GGC CTC CTT ATC A- 3′	IDT DNA	N/A
NQO1 screen R2 5′- TGG CCC ACA GAA AGG CCA AAT TTC T-3′	IDT DNA	N/A
pU6-COM-scaffold forward oligo 5′-ccggAGACGGGTTGTAAA TGAGCACACAAAACGTCTCCgtttaagagctatgctggaaacagcatag caagtttaaat aaggctagtccgttatcaacttggctgaatgcctgcgagcatcccacc caagtggcaccgagtcggtgcTTTTTTTG -3′	IDT DNA	N/A
pU6-COM-scaffold reverse oligo 5′: ccttatttaaacttgctatgctgtttccagca tagctcttaaacGGAGACGTTTTGTGTGCTCATTTACAACCCGTCT agctCAAAAAAAgcaccgactcggtgccacttgggtgggatgctcgcaggcattca gccaagttgataacggactag -3′	IDT DNA	N/A
Recombinant DNA
pBS-CMV-gag-pol	Addgene	Cat# 35614
gag-COM-pol	Addgene	Cat# 211373
gag-P3-pol	Addgene	Cat# 211374
pMD2G	Addgene	Cat# 12259
pCMV-P4-3xNES-Cas9	Addgene	Cat# 181752
P4-PE	Addgene	Cat# 211375
COM-epegRNA-Dnmt1	Addgene	Cat # 211376
pU6 tevopreq1-GG acceptor	Addgene	Cat# 174038
Software and algorithms
CRISPOR	UC Santa Cruz	5.2, https://crispor.gi.ucsc.edu/
OptiPrime	Google	https://optipri.me/
Other
Cell Drop BF bright field cell counter	DeNovix	Cat#I-20103
Refrigerated Centrifuge	Eppendorph	Cat#5810-R
Ultra-Centrifuge	Hitachi	Cat#CP100WX
Micro Plate Reader	BMG Labtech	PHERAstar FSX
Tissue Culture Hood	Thermo Scientific	Cat# 50126286
Evos M5000	Thermo Scientific	Cat# AMF5000SV

Materials and equipment

eVLP plasmid DNA mix

CRITICAL: Make fresh for each use. Do not vortex or pipette to mix, gently flick the tube 10–15 times to mix and centrifuge at the lowest setting to collect at the bottom of the tube.

Cas9 nuclease eVLPs			Prime editing eVLPs
Component	ng required	Volume (μL)	Component	ng required	Volume (μL)
pMD2G	400	x	pMD2G	400	X
CMV Gag-Pol	2813	x	CMV Gag-Pol	2813	X
Gag-COM-Pol	2000	x	Gag-COM-Pol	2000	X
Gag-P3-Pol	422	x	Gag-P3-Pol	422	X
P4-Cas9	422	x	P4-Cas9	422	X
Guide 1	2200	x	HEK3 sense TEVO	2200	X
Guide 2	2200	x	HEK3 anti-sense TEVO	2200	X
HBSS	-	x	HBSS	-	X
Total	10457	500	Total	10457	500

Note: Make each up in a 2 mL tubes to ensure enough space for other reagents and for adequate mixing.

PEI- (1 mg/mL)

Measure 5 mg of PEI MAX® Transfection Reagent powder. Dissolve the powder in 5 mL of endotoxin-free dH₂O heated to 80°C. Allow to cool to room temperature. Adjust the pH to 7.0 and filter sterilise (0.22 μm).

Note: Store 1 mL aliquots at −20°C and keep a working 1 mL stock working stock at 4°C for up to 1 month.

Bovine serum albumin standards

CRITICAL: This recipe makes exactly enough standards for one BCA assay plate.

Vial	Volume μL of diluent (PBS)	Volume μL & source of BSA		Final BSA concentration
A	0	83.33	Stock	750
B	41.67	33.33	A	500
C	55.00	25.00	B	250
D	43.33	30.00	C	150
E	50.00	23.33	D	70
F	49.40	23.33	E	33
G	50.60	22.73	F	15
H	26.67	23.33	G	7
I	50.00	0	PBS only	0

SOC broth 1L-

Reagent	Final concentration	Amount
Tryptone	2%	20 g
Yeast extract	0.5%	5 g
NaCl	10 mM	0.5 g
KCl	2.5 mM	0.186 g
MgCl2	10 mM	0.952 g (anhydrous)
MgSO4	10 mM	1.204 g (anhydrous)
Deionized water	/	To 1 L

Autoclave at 121°C (15 psi) for 15–20 min. Once cooled, add 20 mL 1 M glucose (final concentration 20 mM).

Storage- Store at 4°C for up to 1 month. Avoid repeated warming cycles.

LB Broth (Luria–Bertani Broth) 1 L-

Reagent	Final concentration	Amount
Tryptone	1%	10 g
Yeast extract	0.5%	5 g
NaCl	1%	10 g
Distilled water	/	To 1 L

Autoclave at 121°C (15 psi) for 15–20 min.

Storage- Store at room temperature for up to 6 months or at 4°C for extended storage. Inspect for turbidity before use.

Note: For LB agar plates, add 15 g agar per liter prior to autoclaving.

Step-by-step method details

Assembly and purification of eVLPs

Timing: 5 days

This step describes the assembly of eVLPs in HEK293T cells using PEI-mediated plasmid transfection, followed by collection and purification of eVLP-containing supernatant via PEG precipitation. Gentle handling, and carful pellet resuspension are essential for maximizing eVLP yield and integrity.

• 1.
  One day before transfection, remove medium from a near-confluent culture dish by gentle aspiration.

  • a.
    Wash cells once with sterile 1× phosphate-buffered saline (PBS).
  • b.
    Add sufficient 0.05% trypsin–EDTA to cover the cell monolayer and incubate at 37°C for 5 min.

    Note: Trypsin–EDTA: Flask size, 0.5 mL: T-25, 2 mL: T-75, 5 mL: T-125.

  • c.
    Neutralize trypsin with an equal volume of complete medium.
  • d.
    Collect the cell suspension into a sterile 15 mL conical tube.
  • e.
    Centrifuge at 400 × g for 5 min at room temperature (22°C).
  • f.
    Aspirate the supernatant and resuspend the pellet in 10 mL of fresh complete medium (DMEM supplemented with 10% FBS).

    Note: All reference to complete medium refers to DMEM supplemented with 10% FBS.

  • g.
    Determine the cell concentration using a haemocytometer or automated cell counter.
  • h.
    Seed 4.0 × 10⁶ cells into a 10 cm tissue culture dish containing 10 mL of complet medium.
  • i.
    Incubate the cells overnight (18–24 h) at 37°C and 5% CO₂ to allow attachment and recovery.

    CRITICAL: Optimal transfection efficiency occurs when cells are around 70% confluent.

• 2.
  Prepare a fresh eVLP plasmid DNA mixture according to the experimental design.

  • a.
    In a separate sterile microcentrifuge tube, prepare the PEI dilution:
  • b.
    Add 470 μL Hank’s Balanced Salt Solution (HBSS).
  • c.
    Add 30 μL PEI (1 mg/mL stock).
  • d.
    Vortex for 5 s to mix thoroughly.
• 3.
  Add the PEI–HBSS mixture to the eVLP plasmid DNA mixture.

  • a.
    Mix by gently inverting the tube 10 time.
  • b.
    Briefly centrifuge at the lowest speed.
• 4.Incubate the DNA–PEI mixture at room temperature for 15 min to allow complex formation.
• 5.Aspirate the complete growth medium from the HEK293T cells cultured in a 10 cm dish.
• 6.Add 9 mL of complete growth to the cells.

Note: Alternative serum-free media (e.g., Opti-MEM) may be used.

• 7.
  Add 1 mL of the incubated DNA–PEI complex dropwise to the cells.

  • a.
    Gently distribute the mixture by moving the plate in a figure-of-eight motion.
• 8.Incubate the cells for 12–14 h at 37°C and 5% CO₂.
• 9.Replace the serum-free medium with 10 mL of complete growth medium.
• 10.Incubate the cells for an additional 60 h at 37°C and 5% CO₂ to allow eVLP production.
• 11.
  eVLP Harvesting and PEG Precipitation.

  • a.
    Collect the culture supernatant into a sterile tube.
  • b.
    Centrifuge at 500 × g for 5 min at room temperature to remove cell debris.
  • c.
    Filter supernatant using a 0.45 μm filter into a sterile 15 mL conical tube to remove any excess debris.
  • d.
    Add 3 mL PEG-It (equilibrated to room temperature) to the filtered supernatant.
  • e.
    Mix by gently inverting the tube 10–15 times.

    Note: PEG-It is highly viscous when stored at 4°C. Allow it to reach room temperature before pipetting to ensure accurate volume transfer.

  • f.
    Incubate the mixture on a rotator at 4°C overnight (minimum 12 h).
  • g.
    Centrifuge at 1600 × g for 60 min at 4°C to pellet the eVLPs.
  • h.
    Carefully aspirate and discard the supernatant without disturbing the pellet.
  • i.
    Resuspend the eVLP pellet in 100 μL sterile 1× PBS.

    CRITICAL: Do not disturb or dislodge the pellet during aspiration. eVLPs are highly concentrated at this stage, and pellet disruption can result in significant material loss.

    Note: The pellet is typically large and cloudy white in appearance.

    Alternatives: eVLPs can also be purified by ultracentrifugation, set at 75,000 × g at 4°C for 2 h.

• 12.Aliquot 50 μL of eVLP per cell type for transduction, keeping 25–50 μL to one side for a BCA assay.

CRITICAL: BCA assay is an optional supplementary quality-control step that estimates the total protein concentration recovered in the eVLP preparation. Because BCA is not eVLP-specific. It is not a measure of eVLP number nor of the functionality of the particles. Where validation is required, confirm eVLP production using an appropriate capsid or particle-specific assay. No validation step for PEG-IT extraction has been tested.

• 13.
  Prepare bovine serum albumin (BSA) standards.

  • a.
    Load 25 μL of standard samples of known (BSA) in wells of a clear flat-bottomed 96 well plate.
  • b.
    Load 25 μL of eVLP sample to empty wells as shown below.
  • c.
    Calculate the volume of WR required: (((No. standards) + (No. eVLP samples)) + 4) ∗ 200.
  • d.
    Create working reagent by mixing reagents A and B 50:1.
  • e.
    Add 200 μL of WR to each well containing standards and eVLP samples.
  • f.
    Cover plate in foil and mix for 30s on a plate mixer.
  • g.
    Incubate for 30 min at 37°C.
  • h.
    Measure the absorbance at 570 nm wavelength and calculate a standard curve plotting the unknown eVLP samples.

Preparation of SUIT-2, PANC-1, and MIA-PaCa-2 cells

Timing: 24 h

This step describes the preparation of SUIT-2, PANC-1, and MIA-PaCa-2 PDAC cell lines for transduction in 24-well plates. Consistent seeding density and confluency levels are essential for reproducible transduction efficiency across replicates. This step also describes the transduction of cells with eVLPs.

• 14.
  One day before transduction, remove medium from a near-confluent culture dish/flask by gentle aspiration.

  • a.
    Aspirate the spent medium and wash three times with 1× PBS.
  • b.
    Add sufficient 0.05% trypsin–EDTA to cover the cells and incubate for 5 min at 37°C.
  • c.
    Neutralize trypsin by adding an equal volume of complete medium.
  • d.
    Collect the cells in a sterile 15 mL conical tube and centrifuge at 400 × g for 5 min at room temperature.

CRITICAL: Each PDAC cell line exhibits distinct growth characteristics. Always monitor morphology and confluency before plating.

Note: Trypsin–EDTA: Flask size, 0.5 mL: T-25, 2 mL: T-75, 5 mL: T-125. All three cell types can adhere strongly; you may need to adjust the time spent incubating with trypsin–EDTA. The higher the confluency the more adherent the cell become.

• 15.Aspirate the supernatant and resuspend the cell pellet in 5 mL of complete medium.
• 16.Determine the cell concentration using a hemacytometer or automated cell counter.
• 17.
  Seed cells into a 24-well plate as follows:

  • a.
    SUIT-2: 1 × 10⁵ cells per well.
  • b.
    PANC-1: 1 × 10⁵ cells per well.
  • c.
    MIA PaCa-2: 1 × 10⁵ cells per well.
• 18.Add complete medium so total volume is 500 μl per well.

Note: Seed three replicate wells for each cell type to ensure technical reproducibility and each cell type has a control.

• 19.Gently rock the plate in a figure-of-8 pattern to distribute cells evenly across the well surface.
• 20.Incubate plates (18–24 h) at 37°C and 5% CO₂ to allow attachment and recovery.
• 21.
  Verify that each culture reaches the optimal confluency before transduction:

  • a.
    SUIT-2: 70%–80% confluency.
  • b.
    PANC-1: 60%–70% confluency.
  • c.
    MIA PaCa-2: 70%–80% confluency.

CRITICAL: Over-confluency can significantly reduce transduction efficiency and underconfluency may lead to toxicity.

Note: Use cells between passages 5–20 to maintain consistent growth and transduction performance. PANC-1 cells often grow more slowly; allow extra time to reach target confluency if needed.

Transduction of pancreatic cancer cell lines

Timing: 3 days

This step describes the transduction of pancreatic cancer cell lines plated in 24-well plates. eVLPs prepared by either ultracentrifugation or PEG precipitation are combined with transduction enhancers and added to cells under serum-free conditions to maximize uptake and transduction efficiency.

• 22.In a tissue culture hood, label sterile 0.5 mL microcentrifuge tubes, one for each well to be transduced, including all experimental and control conditions.
• 23.
  Prepare the transduction enhancer mix by adding the following reagents to each labelled tube:

  • a.
    Add 10 μL Synperonic F108 (1,000 μg/mL).
  • b.
    Add 0.8 μL polybrene (10 mg/mL).

Note: Both these compounds are toxic to cells at higher concentrations. Higher-passage cells often exhibit increased sensitivity and reduced transduction efficiency compared with lower-passage cells.

• 24.
  Add eVLPs to the tubes as follows:

  • a.
    For ultracentrifugation harvested VLPs, add 25 μL VLPs and 22.2 μL 1× PBS.
  • b.
    For PEG precipitation–harvested VLPs, add 50 μL VLPs and 0.2 μL 1× PBS.
  • c.
    For negative control wells, add 50.2 μL 1× PBS.
• 25.Gently mix each tube by gently vortexing.
• 26.Aspirate the culture medium from the cells in the 24-well plate.
• 27.Gently wash each well three times with pre-warmed 1× PBS.

CRITICAL: Aspirate carefully to avoid detaching loosely adherent cells, particularly MIA PaCa-2 cells.

• 28.Add 439 μL Opti-MEM to each well.
• 29.Add the prepared eVLP mixture dropwise to the corresponding wells.
• 30.Gently distribute the transduction mixture by moving the plate in a figure-of-eight motion.
• 31.Incubate the cells at 37°C and 5% CO₂ for 24 h.
• 32.After 24 h, aspirate the Opti-MEM/eVLP mixture from each well.
• 33.Add 500 μL of complete growth medium to each well.
• 34.Incubate the cells at 37°C and 5% CO₂ for a minimum additional 48 h before downstream analysis.

Validation of editing success

Timing: 1 week

This step details the process of assessing the success of the edit attempt, including Genomic DNA extraction, PCR amplification and purification as well as preparation of sanger sequence samples.

• 35.
  Ensure that cells have recovered and their confluency is adequate to passage them.

  • a.
    Aspirate the spent medium and wash three times with 1× PBS.
  • b.
    Add sufficient 0.05% trypsin–EDTA to cover the cells and incubate for 5 min at 37°C.
  • c.
    Neutralize trypsin by adding an equal volume of complete medium.
  • d.
    Collect the cells in a sterile 2 mL centrifuge tube and centrifuge at 400 × g for 5 min at room temperature.

CRITICAL: If the cells have not reached an adequate confluency aspirate the spent medium and replace with a medium with a higher FBS concentration (15%–20%). Check them daily until the cells have recovered.

• 36.Resuspend the cells in 1 mL of complete media and determine the cell concentration using a hemacytometer or automated cell counter.
• 37.Remove 1 × 10⁵ cells and re-pellet by centrifugation at 400 × g for 5 min at room temperature.
• 38.Replate the remaining cells in a 12 well plate ensuring the working volume is 1 mL and no more than 2 mL.
• 39.Extract the gDNA from the pelleted cells using the Wizard® SV Genomic DNA Purification System using the manufactures protocol.
• 40.Determine the concentration of gDNA in your samples.
• 41.
  Perform a 3 step 50 μL PCR reaction following the manufactures guidelines.

  • a.
    For CRISPR/Cas9 knockout use NQO1 screen F2 and NQO1 screen R2 primers.
  • b.
    For HEK3 prime edit use F1 HEK3 and R3 HEK3.
• 42.Separate PCR generated species on 1% agarose gel for 30–50 min dependent on length of gel (Figures 2A and 2B).
• 43.Gel purify positive bands using Wizard® SV Gel and PCR Clean-Up System using the manufactures protocol.

Note: For further validation, confirm the DNA sequences of the PCR products through Sanger sequencing. This can be achieved by gel purification of the PCR product and utilizing a commercial company that provides Sanger sequencing options.

Figure 2.

Gel electrophoresis images of PCR validation of editing machinery

(A) PCR validation of CRISP/Cas9 knockout in HEK293T SUIT-2, MIA-PaCa-2 and PANC-1.

(B) PCR validation of Prime Editing of HEK3, positive in SUIT-2 and MIA-PaCa-2 and negative in PANC-1. All positive controls were generated using the HEK293T cells used to produce VLPs for the respective edit. KO= Knockout, WT= Wild-type.

Expected outcomes

Using this protocol, efficient transient delivery of CRISPR/Cas9 and prime editing ribonucleoprotein complexes into PDAC cell lines is achieved, as demonstrated by successful eVLP production and genome editing in PDAC cells.

For prime editing applications, targeted modification of the HEK3 locus is achieved in SUIT-2 and MIA-PaCa-2 cells following transduction with eVLP-packaged prime editing components. Successful editing results in precise installation of the intended nucleotide change with minimal undesired editing events. It is difficult to predict the expected editing levels. Several factors can impact upon this, and each experiment would require empirical optimisation. Examples of these factors include the prime editing oligonucleotide itself (e.g., length of the primer binding site (balancing stable primer binding against product release), the length and sequence composition of the reverse transcription template, and the stability of the pegRNA). Intrinsic cellular factors can also affect the efficiency of the prime editing, including activation of the mismatch repair pathway, which can be overcome to some extent by inhibiting the MMR pathway by utilising PE4/PE5 techniques. However, with careful optimisation of the steps involved, Zhao et al. reported editing efficiency of up to 80% in multiple cell lines and across multiple loci, and up to 50% in hPSCs.¹⁶

For CRISPR/Cas9-mediated gene disruption, transduction with dual sgRNAs targeting exon 6 of NQO1 results in efficient deletion of the genomic region between the two guide RNAs in SUIT-2 and MIA-PaCa-2 cells. The dual-guide strategy increases the likelihood of functional gene disruption by generating defined exon deletions rather than relying solely on small insertions or deletions.

Collectively, these results demonstrate that eVLP-mediated delivery enables reproducible genome editing in PDAC cell lines that exhibit low transfection efficiency using conventional approaches. The protocol supports both precise prime editing and dual-guide CRISPR-mediated gene disruption, providing a flexible platform for functional genomic studies in pancreatic cancer models.

If editing is unsuccessful, the target locus remains indistinguishable from untreated control cells, indicating inefficient delivery or genome editing. Reduced editing efficiency may also be accompanied by decreased cell viability or poor recovery following transduction. In such cases, eVPLdose, guide RNA design, envelope glycoprotein choice, and cell confluency at the time of transduction should be re-evaluated. Where low editing efficiencies are suspected, more sensitive downstream analytical methods may be required to accurately quantify editing events. Repeated eVLP dosing may enhance editing by increasing the cumulative delivery of transient editing complexes. Two PE-eVLP doses administered 24 h apart achieved 25% editing at the HEK3 locus in proliferating human myoblasts,¹⁷ while sequential eVLP treatments produced stepwise increases in editing in H1975 cells.¹⁸ The number of doses and interval between treatments should be optimized for each cell type and target because repeated treatment may also reduce cell viability.

Limitations

Despite the successful application of this protocol for genome editing in pancreatic cancer cell lines, several limitations should be considered. First, variable protocol efficiency was observed across different cell lines, as shown in Figure 2B with no evidence of editing in the PANC-1 cell line. Differences in proliferation rate, membrane composition, and intrinsic antiviral responses can influence eVLP uptake. For example, SUIT-2, PANC-1, and MIA-PaCa-2 cells exhibit distinct growth kinetics and slower-dividing cultures may demonstrate reduced apparent editing frequencies.^1,19,20 Cell confluency at the time of transduction and passage number further contribute to variability in editing outcomes.³

Second, editing efficiency is strongly dependent on guide RNA design and functional eVLP concentration.¹⁴ Low eVLP production yield or degradation during purification can reduce effective dose and compromise editing efficiency.^2,3,4 Careful validations of guide sequences and consistent eVLP production are therefore critical determinants of success.

Third, prime editing efficiency is generally lower than Cas9-mediated indel formation under comparable conditions.⁷ This observation is consistent with published literature cited through this protocol and reflects the increased mechanistic complexity of prime editing, which relies on reverse transcription and precise repair pathway engagement. As a result, prime editing outcomes may require higher eVLP input, optimization of epegRNA orientation (sense versus antisense), or additional screening to identify optimal editing configurations.

Together, these limitations underscore the importance of empirical optimization when applying this protocol to new pancreatic cancer models or genomic targets.

Troubleshooting

Problem 1

HEK293T producer cells detach easily during eVLP assembly (related to step 1–13).

HEK293T cells are semi-adherent and may detach during medium changes or plasmid complex addition. Excessive detachment can reduce eVLP production and lead to lower functional titters.

Potential solution

Handle HEK293T cells gently during all washing and medium replacement steps. Aspirate medium slowly from the edge of the plate avoiding contact with the cell layer. Ensure, cells are 70%–80% confluent at the time of transfection; over-confluent or under-confluent cultures detach more readily. Lastly, use low-passage HEK293T cells and avoid over-trypsinization during routine maintenance.

Problem 2

When PEG-based precipitation is used, eVLP yield cannot be directly quantified, making it difficult to determine whether failure results from poor eVLP production or inefficient genome editing. (related to steps 22–34).

Potential solution

PEG-based eVLP preparations do not allow direct quantification of functional titter without additional assays. If repeated editing failure occurs, consider assessing eVLP production by p24 ELISA, Western blot for capsid protein, or switching to ultracentrifugation-based concentration to better estimate yield.

Problem 3

Low prime editing efficiency compared to Cas9-mediated indel formation (related to steps 22–34).

Potential solution

Prime editing efficiency may vary substantially between pancreatic cancer cell lines due to differences in DNA repair pathway activity. In addition to the editor architecture used in this protocol, state-of-the-art prime editing systems may be considered as potential optimisation strategies. For example, PE8 prime editors and pegRNA designs incorporating PRISM motifs may improve editing activity by enhancing prime editor performance and pegRNA function. Their suitability should, however, be validated empirically in each PDAC cell line, taking into account editor expression, eVLP packaging capacity, delivery efficiency, and potential effects on cell viability. Inhibition of mismatch repair pathways, such as MLH1, has been reported to enhance prime editing efficiency by reducing correction of edited intermediates.²¹ Additional optimisation of epegRNA design, including primer binding site (PBS) length, reverse transcription template length, and strand orientation (sense versus antisense), and the incorporation of improved pegRNA motifs, can improve editing outcomes. Increasing eVLP input may improve editing efficiency, although a plateau may be reached at higher doses.⁶ Where appropriate, repeat dosing can increase the cumulative delivery of transient editing complexes.¹⁷ For prime editing, actively proliferating cells may support higher editing efficiencies than less proliferative or differentiated cells; however, these effects should be optimised empirically for each cell line, target locus, and eVLP formulation.²²

Note: Modulation of DNA repair pathways may improve editing efficiency but requires careful optimisation, as disruption of these pathways can impact cell viability and genomic stability. Prime editing is inherently less efficient than Cas9-mediated editing due to its reliance on multiple coordinated enzymatic steps and precise DNA repair mechanisms. PE8 editors and PRISM-containing pegRNAs were not evaluated in the present workflow, but they represent relevant state-of-the-art strategies for future optimisation.

CRITICAL: Lower prime editing efficiency relative to Cas9-mediated editing is expected and does not necessarily indicate failure of eVLP delivery.²³

Resource availability

Lead contact

All requests for further information should be directed to the lead contact, Eithne Costello (ecostell@liverpool.ac.uk).

Technical contact

Questions on execution of this protocol should be directed to the technical contact, Mark Glenn (maglenn@liverpool.ac.uk).

Materials availability

All un-manipulated plasmids/recombinant DNA can be found on Addgene. All sequences in this protocol can be found in the key resources table.

Data and code availability

This manuscript did not report new datasets or code.

Acknowledgments

This article is based upon work from COST Action “Identification of biological markers for prevention and translational medicine in pancreatic cancer (TRANSPAN),” CA21116, supported by COST (European Cooperation in Science and Technology). E.C. is supported by Cancer Research UK, grant no. C7690/A26881. C.T.-Y. is supported by funding from the University of Liverpool’s Douglas Endowment PhD prize. M.C. is supported by grant RYC2021-031003I funded by MICIU/AEI/https://doi.org/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR. E.G. is supported by KWF/Dutch Cancer Society and AIRC/Associazione Italiana per la Ricerca sul Cancro. A.G. is supported by Spanish Ministry of Science, Innovation and Universities PID2024-157599OB-I00 and CNS2023-144109. The graphical abstract was created using BioRender.com.

Author contributions

C.T.-Y. contributed to the protocol creation and troubleshooting and drafted the manuscript. M.C., E.G., D.A., and A.G. provided intellectual input, technical advice, and critical feedback on the manuscript. E.C. provided supervision and protocol revisions and edited the manuscript. M.G. provided supervision, contributed to the protocol creation and troubleshooting protocol revisions, and edited the manuscript.

Declaration of interests

M.C. has received honoraria/consultancy contracts from EQA Certificados and OCA Global honoraria contracts from Elsevier B.V. and Springer Nature for editorial board membership. These external engagements had no influence on the design, execution, interpretation, or conclusions of the present study.