Adenovirus-mediated CRISPR/Cas9 gene editing in patient primary fibroblasts to validate potential drug targets in lung fibrosisAnne-Marie Zuurmond1,#, Liesbeth Veugen1, Lieke Geerts1, Barbara Killian2, Isabella Gazzoli1, Dennis Gravekamp1, Monique Hazenoot1, Miranda van der Ham1, Britt Sotthewes1, Brigitta Witte1, Krista Ouwehand1, David Fischer3, Jeroen DeGroot1

1Charles River Leiden, Darwinweg 24, 2333 CR Leiden, The Netherlands; 2Charles River Wilmington, 251 Ballardvale St, Wilmington, MA 01887, USA; 3Charles River Cambridge, Chesterford Research Park, Saffron Walden, Essex CB10 1XL, United Kingdom#Corresponding author: Anne-Marie.Zuurmond@crl.com

1 INTRODUCTIONTargeted gene editing has become a key tool in deciphering the molecular mechanisms underlying disease processes and for target validation in the drug discovery process. For research into new drug targets and compounds for treatment of pulmonary fibrosis, we previously developed a cell-based assay with primary lung fibroblasts derived from idiopathic pulmonary fibrosis (IPF) patients. In this in vitro model, the transition of fibroblast into myofibroblast (FMT) is triggered by TGFβ. Myofibroblasts are thought to play a major role in fibrosis through excessive deposition of extracellular matrix. A key feature of myofibroblasts is the production of α-smooth muscle actin (αSMA).

Figure 1: Cell-based phenotypic assay for pulmonary fibrosis

In this study, we have evaluated the feasibility to validate potential drug targets in fibrosis by CRISPR/Cas9-mediated targeted gene knockout in this assay. For delivery of Cas9 and gRNA the adenoviral system was chosenthat is known for its ability to transduce a wide range of cells with high efficiency, including human primary lungfibroblasts that are difficult to transfect. The ACTA2 gene encoding αSMA was selected as target to demonstratefeasibility of CRISPR/Cas9-mediated knockout in lung fibroblasts. The setup was validated using a small gRNAlibrary with putative drug targets for FMT/IPF emerging from literature.

2 EXPERIMENTAL DESIGNAn adenoviral system for CRISPR/Cas9 delivery was constructed with Cas9 and the gRNA being expressed fromtwo different adenoviruses (AdV). Four gRNAs targeting ACTA2 and ≥2 gRNAs per putative drug target weredesigned and cloned into the adapter vector of the adenoviral system. High titer stocks of adenoviruses expressingeither Cas9 or ACTA2 gRNAs were produced and used in the transduction of primary fibroblasts derived from fourIPF patients. CRISPR/Cas9-mediated knockout of ACTA2 was measured by high content analysis (HCA).

Figure 2. Timeline for a cell-based phenotypic assay for lung fibrosis. Fibroblasts derived from IPF patientswere transduced with AdV-Cas9 and AdV-gRNA one day after seeding. The fibrotic process was triggered byadding 2 ng/µl TGFβ 4 days after adenoviral transduction. The fibroblast to myofibroblast transition (FMT)was quantified by measuring αSMA production by high content analysis (HCA) 2 days after adding theTGFβ trigger.

3 RESULTS

Figure 3. Impaired αSMA production in FMT assay due to CRISPR/Cas9-mediated ATCA2 knockout. IPF patient-derived fibroblasts from four donors (IPF03, -05, -07 and -08) were transduced at a MOI of 24 each for AdV-Cas9 and AdV-gRNA_ACTA2 (ACTA2 _g). TGFβ was added 4 days after transduction and αSMA production was quantified by HCA 2 days after TGFβ trigger (day 7). A non-targeting control adenovirus (AcGFP_g1) was used to normalize the data as adenoviral transduction is known to stimulate αSMA production. All four gRNAs were effective in ACTA2 gene knockout in two independent experiments, however gene editing efficiency was donor-dependent. Typical high content images are shown.

4 CONCLUSION

Table 1: Selection of putative drug targets for modulation of αSMA protein levelsin the FMT process. Candidate genes are selected that are presumed to beinvolved in IPF based on available literature. The ACTA2 gene is selected as anassay control. For each target gene ≥2 gRNAs were designed and cloned into theadapter vector of the adenoviral system.

We conclude that adenoviral transduction of primary fibroblasts for targeting gene editing using CRISPR/Cas9 technology is sufficiently efficient to assess the effects of a gene knockout in a cell-based assay such as the FMT assaywithout the need for selection or cloning. This was validated not only by targeting the ACTA2 gene encoding for αSMA protein directly, but also using a small library of gRNAs targeting genes that are associated with the fibrosis processand/or IPF phenotypes. Using this approach IRIF5, ITGA5, COL3A1 and SMAD4 are identified as modulators of myofibroblast transition in this assay. Such effective gene targeting in a functional assay is a valuable asset for targetvalidation in drug discovery.

CRISPR-Cas9 was used under license to granted and pending US and international patents from the Broad Institute and ERS Genomics Limited

TGFβ

Myofibroblast

DAPI aSMA

Fibroblast

Refresh medium

D7

Fix and stain cells

D1

AdV transduction

D2 D5

Add TGFβ

αSMA and DAPI immunostaining measured by HCA

Seed IPF cells

Unt

reat

ed (n

on-tr

igge

red)

Unt

reat

ed (t

rigge

red)

Cru

de ly

sate

(trig

gere

d)

AcG

FP_g

1 (tr

igge

red)

AC

TA2_

g1 (t

rigge

red)

AC

TA2_

g2 (t

rigge

red)

AC

TA2_

g3 (t

rigge

red)

AC

TA2_

g4 (t

rigge

red)

0.0

0.5

1.0

1.5

αSM

A / N

ucle

i(n

orm

aliz

ed to

AcG

FP_g

1)

IPF08 (Exp A)IPF03 (Exp B)IPF05 (Exp B)

IPF07 (Exp A) non-triggered

triggered

crude lysate

AcGFP_g1

ACTA2_g1

ACTA2_g2

ACTA2_g3

ACTA2_g4

controls gRNA

Gene Full name Rationale

ACTA2 Actin alpha 2, smooth muscle Involved in vascular contractility, is used as a FMT marker

IRF5 Interferon regulatory factor 5 Transcription factor modulating cell growth and differentiation

ITGA5 Integrin subunit alpha 5Involved in the formation of the fibronectin receptor, cell surface adhesion

and signaling

ITGB7 Integrin subunit beta 7Adhesion receptor involved in cellular signaling with the extracellular matrix

and MAPK-ERK pathway

ITGB1 Integrin subunit beta 1Involved in the formation of the fibronectin receptor and tissue repair

processes

COL3A1 Collagen type III alpha 1 chainFibrillary collagen found in extensible connective tissue, is used as a FMT

marker

TGFBR3Transforming growth factor beta

receptor 3Membrane proteoglycan often functioning as a co-receptor with other TGFβ

receptor superfamily members

SMAD4 SMAD family member 4Signal transduction protein involved in TGFβ signaling regulating

transcription processes

MAPK8 Mitogen-activated protein kinase 8Regulation of kinases signaling processes involved with cell proliferation and

differentiation associated with renal fibrosis and fatty liver disease

CXCL10 C-X-C motif chemokine ligand 10Ligand for the receptor CXCR3 and inflammation marker, involved with

adhesion molecule expression

S100A8 S100 calcium binding protein A8Involved in the regulation cell cycle progression and differentiation

associated with cystic fibrosis 0.0

0.5

1.0

1.5

ITGB7_g4ITGB7_g1

MAPK8_g3MAPK8_g2MAPK8_g4

CXCL10_g3CXCL10_g1CXCL10_g2CXCL10_g4

ITGB1_g1ITGB1_g2ITGB1_g3ITGB1_g4

TGFBR3_g3TGFBR3_g2TGFBR3_g4TGFBR3_g1S100A8_g4S100A8_g2S100A8_g1S100A8_g3ITGA5_g4 ITGA5_g2ITGA5_g3

IRF5_g2IRF5_g1 IRF5_g4IRF5_g3

SMAD4_g1SMAD4_g2SMAD4_g3SMAD4_g4

COL3A1_g1COL3A1_g2COL3A1_g3

ACTA2_g4ACTA2_g3ACTA2_g2ACTA2_g1AcGFP_g1

Crude lysateUntreated (triggered)

Untreated (non-triggered)

IPF03 (Exp B)

αSM

A /

Nuc

lei

(nor

mal

ized

to A

cGFP

_g1)

0.0

0.5

1.0

1.5

IPF05 (Exp B)

0.0

0.5

1.0

1.5

IPF07 (Exp C)

0.0

0.5

1.0

1.5

IPF08 (Exp C)

Controls

SMAD4

IRF5

ITGA5

S100A8

TGFBR3

ITGB1

CXCL10

MAPK8

ITGB7

COL3A1Not tested

Not tested

Not tested

Not tested

Not tested

Not tested

Not tested

Not tested

Not tested

Not tested

Not tested

Not tested

Not tested

Not tested

Not tested

Not tested

Not tested Not tested

100% 18% 72% 59% 50% 42% 48%

AcGFP_g1 ACTA2_g4 IRF5_g3 COL3A1_g2 COL3A1_g3 SMAD4_g4ITGA5_g3

Figure 5. Representative high content images of CRISPR/Cas9-mediated ATCA2, IRIF5, ITGA5, COL3A1 and SMAD4 knockout using the top-performing gRNAs indonor IPF07. Relative percentage αSMA / nuclei levels from high content analysis is given normalized to the non-targeting control adenovirus (AcGFP_g1).

Figure 4: CRISPR/Cas9-mediated downregulation of αSMA protein in FMTassay was observed using gRNA directed against IRIF5, ITGA5, COL3A1 andSMAD4. The relative effect size for modulation of each of these targets wasdonor- and gRNA-dependent and varied between the two experiments.Percentage αSMA / nuclei levels is given normalized to the non-targetingcontrol adenovirus (AcGFP_g1). Non-triggered, TGFβ-triggered, crudelysate/TGFβ-treated and ACTA2 gRNA-exposed fibroblasts were taken along ascontrols. Dark green: ≤50%; light green: 51-75%; grey: ≥76%.