Article

ISSN 0006-2979, Biochemistry (Moscow), 2023, Vol. 88, Nos. 12-13, pp. 1987-1996 © Pleiades Publishing, Ltd., 2023.

Published in Russian in Biokhimiya, 2023, Vol. 88, No. 12, pp. 2375-2386.

1987

A New Mouse Strain

with a Mutation in the NFE2L2 (NRF2) Gene

Evgeniy S. Egorov

, Natalia D. Kondratenko

2,3

, Olga A. Averina

2,4,5

Oleg A. Permyakov

4,5

, Maria A. Emelyanova

4,5

, Anastasia S. Prikhodko

1,2

Ludmila A. Zinovkina

, Petr V. Sergiev

2,4,5

, and Roman A. Zinovkin

2,6,a

Faculty of Bioengineering and Bioinformatics, Lomonosov Moscow State University, 119991 Moscow, Russia

Belozersky Institute of Physico-Chemical Biology, Lomonosov Moscow State University, 119991 Moscow, Russia

Russian Clinical Research Center for Gerontology, Ministry of Health of the Russian Federation,

Pirogov Russian National Research Medical University, 129226 Moscow, Russia

Institute of Functional Genomics, Lomonosov Moscow State University, 119991 Moscow, Russia

Faculty of Chemistry, Lomonosov Moscow State University, 119991 Moscow, Russia

HSE University, 101000 Moscow, Russia

e-mail: roman.zinovkin@gmail.com

Received September 26, 2023

Revised November 22, 2023

Accepted November 23, 2023

Abstract— Transcription factor NRF2 is involved in inflammatory reactions, maintenance of redox balance, metabolism of

xenobiotics, and is of particular interest for studying aging. In the present work, the CRISPR/Cas9 genome editing tech-

nology was used to generate the NRF2

ΔNeh2

mice containing a substitution of eight amino acid residues at the N-terminus

of the NRF2 protein, upstream of the functional Neh2 domain, which ensures binding of NRF2 to its inhibitor KEAP1.

Heterozygote NRF2

wt/ΔNeh2

mice gave birth to homozygous mice with lower than expected frequency, accompanied by their

increased embryonic lethality and visual signs of anemia. Mouse embryonic fibroblasts(MEFs) from the NRF2

ΔNeh2/ΔNeh2

homozygotes showed impaired resistance to oxidative stress compared to the wild-type MEFs. The tissues of homo-

zygous NRF2

ΔNeh2/ΔNeh2

animals had a decreased expression of the NRF2 target genes: NAD(P)H:Quinone oxidoreduc-

tase-1(Nqo1); aldehyde oxidase-1 (Aox1); glutathione-S-transferaseA4 (Gsta4); while relative mRNA levels of the mono-

cyte chemoattractant protein1(Ccl2), vascular cell adhesion molecule1 (Vcam1), and chemokine Cxcl8 was increased.

Thus, the resulting mutation in the Nfe2l2 gene coding for NRF2, partially impaired function of this transcription factor,

expanding our insights into the functional role of the unstructured N-terminus of NRF2. The obtained NRF2

ΔNeh2

mouse

line can be used as a model object for studying various pathologies associated with oxidative stress and inflammation.

DOI: 10.1134/S0006297923120039

Keywords: transcription factor NRF2, transgenic animals, inflammation, oxidative stress

Abbreviations: AOX1,aldehyde oxidase-1; ARE,antioxidant response element; CCL2,monocyte attractant protein; GOx,glu-

cose oxidase; GSTA4,glutathione-S-transferase A4; HMOX1,hemoxygenase-1; KEAP1,Kelch-like ECH-associated protein1;

MEFs,mouse embryonic fibroblasts; Neh,Nrf2-ECH homology; NQO1,NAD(P)H: quinone oxidoreductase; NRF2,nuclear

factor2 related to erythroid factor2; ROS,reactive oxygen species; VCAM-1,vascular cell adhesion molecule1.

* To whom correspondence should be addressed.

INTRODUCTION

NRF2 (nuclear factor erythroid 2 related factor 2)

is a transcription factor that controls expression of many

genes products of which have antioxidant and anti-

inflammatory properties. In the cytoplasm, NRF2 is as-

sociated with the protein KEAP1 (Kelch-like ECH-as-

sociated protein 1), which under normal conditions pro-

motes permanent proteasomal degradation of NRF2[1].

Simultaneously, KEAP1 serves as a redox-sensitive

EGOROV et al.1988

BIOCHEMISTRY (Moscow) Vol. 88 Nos. 12-13 2023

regulator of NRF2 activity. When oxidizing agents and

electrophiles enter the cell, KEAP1 undergoes thiol

modification of its cysteine amino acid residues [2]. This

modification keeps KEAP1 bound to the NRF2. Due to

the lack of “vacant” negative regulators, the newly syn-

thesized NRF2 accumulates in the cytoplasm and then

moves to the nucleus. In the nucleus, NRF2 in complex

with its coactivators, including small proteins of the Maf

family, recognizes antioxidant response elements(ARE)

sequences in the promoters of its target genes and trig-

gers their transcription [3].

There are seven highly conserved Neh (NRF2-ECH

homology)-domains in the NRF2 structure [4, 5]. The

N-terminal part of the protein contains a Neh2 domain

(amino acids (aa) 16-86) that includes two sequences

known as DLG and ETGE motifs [6]. The NRF2 nega-

tive regulator KEAP1 binds to these sequences. KEAP1,

being an adaptor protein for the E3 ubiquitin ligase

complex Cullin 3 (Cul3), stimulates ubiquitinylation

of seven lysine residues located in the Neh2 domain of

NRF2 between the DLG and ETGE motifs and pro-

motes proteasomal degradation of the latter [6, 7].

The domains Neh1 (aa 435-562), Neh4 (aa 112-134),

Neh5 (aa 183-201), and Neh7 (aa 209-316) are respon-

sible for interaction of NRF2 with its coactivators and

corepressors [3, 8, 9]. The Neh6 domain (aa 338-388)

contains two degron sequences that are recognized by

the E3 ubiquitin-ligase β-TrCP [10, 11]. The C-terminal

part of the protein contains the Neh3 domain (aa 562-

605), which is responsible for recognizing ARE elements

in the promoters of the NRF2 target genes and contains

a VFLVPK motif that helps NRF2 to bind to the CHD6

helicase [12]. While NRF2 is complexly organized, this

protein is partially disordered and its Neh2, Neh7, and

Neh1 domains can only be structured for short periods

of time [13].

NRF2 activates transcription of the genes of the

2nd phase of xenobiotic detoxification responsible for

removal of the modified compounds from the cell.

NRF2 also actively participates in the cell defense

against electrophilic stress [14]. NRF2 also controls ex-

pression of the genes products of which are involved in

glutathione biosynthesis, as well as enzymes that directly

or indirectly neutralize reactive oxygen species (ROS):

NAD(P)H:quinone oxidoreductase (NQO1), hemoxy-

genase-1 (HO1), catalase (CAT). Reduction of ROS,

in turn, contributes to the cessation of inflammatory

responses. When the Nfe2l2 expression is decreased, in-

flammation increases, which can lead to organ and tis-

sue damage [15]. Downregulation of the Nfe2l2 expres-

sion in monocytes results in the increased production of

pro-inflammatory cytokines [16]. Mouse models have

shown increase in the ROS levels leading to prolonged

oxidative stress after brain injury [17].

Gene knockouts are widely used to study functions

of transcription factors. The Nfe2l2 knockout mice lack-

ing the NRF2 transcription factor were obtained more

than a quarter of a century ago [18], and all subsequent

experiments have been performed exclusively with this

line. In these mice, a cistron from the lactose operon

was inserted into the Nfe2l2 gene, resulting in inability

to synthesize functional mRNA and protein product.

However, the use of knockout animals often results in

the secondary effects that make it difficult to interpret

the obtained results. It is likely that in the case of com-

plete absence of any transcription factor, secondary

effects may be due to the absence of its associated co-

factors. It is important to create new models with muta-

tions in the domains of transcription factors, rather than

compromising integrity of the protein structure. Another

problem with the gene deletion approach is possible re-

moval of one or more noncoding RNAs found in the in-

trons and exons.

In this work, a new mutant mouse line, NRF2

ΔNeh2

carrying an 8-aa substitution at the N-terminus of NRF2,

was obtained; embryonic fibroblasts(MEFs) were char-

acterized, and changes in the mRNA levels of a num-

ber of NRF2 target genes in various tissues of these ani-

mals were determined. This mouse line can be used as a

model object for studying embryonic mortality, various

pathologies accompanied with oxidative stress and in-

flammation, as well as for studying aging processes.

MATERIALS AND METHODS

Animals and work with them. Animals were kept in in-

dividually ventilated cages (IVC system, TECNIPLAST

S.p.A., Italy) with free access to food and water purified

by reverse osmosis, in an environment free of specific

pathogens, with light regime 12/12 (light on at 09:00),

in rooms with air exchange rate more than 15 r/h, at

20-24°C, humidity 30-70%. Wood chips with minimal

dust formation were used as bedding. Shelters and build-

ing materials for nests made of natural materials were

used as enrichment of the environment. All materials

supplied to the animals were sterilized by autoclaving.

Transgenic animal generation. The work with an-

imals was approved by the local bioethics committee

“Institute of Mitoengineering MSU” LLC, protocol #79

of April 28, 2015. Mutation in the Nfe2l2 gene was in-

troduced using CRISPR/Cas9 technology. The guide

RNA (5′-GACTTGGAGAGTTGCCACCGCC) to the

first exon of this gene was selected using the Feng Zhang

laboratory (http://crispr.mit.edu/) service. The corre-

sponding single-guide RNA (sgRNA) was produced by

in vitro transcription of T7 (MEGAscript™ T7 Tran-

scription Kit, Thermo Fisher, USA) on a matrix ob-

tained by PCR amplification of a plasmid pX458 [19]

with forward primer: 5′-TGTAATACGACGACTCAC

TATATAGGGACTTGGAGTTGAGTTGCCACCGC

CGTTTTTTTAGAGAGCTAGAAATAGC and reverse

MOUSE WITH MUTATION IN THE NRF2 GENE 1989

BIOCHEMISTRY (Moscow) Vol. 88 Nos. 12-13 2023

primer: 5′-AGCACCACCGACACTCGGTGCCACT.

The sgRNA was mixed with Cas9 mRNA (GeneArt™

CRISPR Nuclease, ThermoFisher) in a TE buffer

(10mM Tris-HCl, 0.1mM EDTA, pH8).

Single-cell embryos for subsequent microinjection

of the genetic construct were isolated from oviducts

of female mice according to a standard protocol [20].

Female zygote donors were preliminarily hormonally

stimulated according to the scheme described by Averi-

na etal.[21] and fertilized by males of the corresponding

lineage [21].

Microinjection of the genetic construct into the

pronucleus of a fertilized oocyte was performed in

an oocyte washing medium with phenol red, pH 7.4,

without heparin (CooperSurgical, Inc., USA, USA),

surrounded by vaseline oil (JSC Tatkhimfarmprepa-

raty, Kazan, Russia) on an inverted microscope

(ECLIPSE Ti, Nikon, Japan) using two micromanipu-

lators (TransferMan 4R, Eppendorf, Germany), accord-

ing to the previously described protocol [20]. Zygotes

after microinjection were incubated in an atmosphere

of 5% CO

at 37°C in a Sequential Fert™ medium with

phenol red (CooperSurgical, Inc.). After incubation,

surviving embryos were implanted in oviduct funnel

of surrogate females according to the standard proto-

col[20]. After birth and completion of lactation period,

mice were genotyped using tip of the tail, according to

the FELASA guidelines for genotyping transgenic ro-

dents [22]. Tissue samples for genotype identification

were frozen at –20°C until genotyping.

The resulting heterozygous mice (Nrf2

wt/ΔNeh2

) were

crossed to the inbred line C57BL/6J to avoid poten-

tial effects of secondary mutations. Homozygous indi-

viduals of Nrf2

ΔNeh2/ΔNeh2

and wild-type Nrf2

WT/WT

mice

were obtained from crosses between heterozygous pairs

of Nrf2

wt/ΔNeh2

Mice genotyping. Alkaline extraction method was

used to isolate genomic DNA from tissue samples [23]

followed by purification of DNA by phenol-chloro-

form extraction. An Encyclo Plus PCR kit (Evrogen,

Russia) with 50 ng of genomic DNA was used for ge-

notyping. PCR1 with primers mNrf-F476 (5′-GCAG

GCTATCTCTCCTAGTTCT) and mNrf-R668 (5′-CG

GCTTCTTGGCACAG) and PCR2 with primers

mNrf-F476 and mNrf-R1153 (5′-GACAGGCGTGATC

TTACAG) were performed on the DNA template. PCR

regime: 95°C for 5 min, followed by 35 cycles (95°C for

25 s, 60°C for 25 s, 72°C for 25 s). PCR products were

analyzed electrophoretically in a 1.5% agarose gel.

Mouse embryonic fibroblasts (MEF). Embryos were

isolated at 10-14

day post coitus from heterozygous

pregnant females (Nrf2

wt/ΔNeh2

) crossed with a male het-

erozygote. MEF isolation was performed according to

the method described in [24]. The cells were cultured on

a Dulbecco’s modified Eagle medium (DMEM) (Pan-

Eco, Russia).

RNA isolation, reverse transcription, and real-time

PCR. Total cellular RNA was isolated using a QuickRNA

MiniPrep reagent kit (ZymoResearch, USA) accord-

ing to the manufacturer’s protocol. RNA concentration

was determined using a Nanodrop ND-1000 spectro-

photometer (Thermo Scientific, USA), RNA quality

was confirmed electrophoretically. Reverse transcrip-

tion was performed using a SuperScript III kit (Ther-

moFischer, USA) as described previously[25] for sub-

sequent analysis by real-time PCR with an EvaGreenI

intercalating dye (Syntol, Russia). The following

primers were used for real-time PCR: Hmox1 (for-

ward: 5′-CACGCATATATATACCCGCTACC; reverse:

5′-TCATCTCCAGAGAGTGTTCATTCG); Nqo1 (for-

ward: 5′-GTCCTCCATCAAGATTCG; reverse: 5′-GC

TAACTGCTAACTGCTAACTGCTAA); Aox1 (forward:

5′-CATAGGTCAGGTTGAAGGTGAAGGT; reverse:

5′-GGCAGGAATCTTGTATTGG); Gsta4 (forward:

5′-AGCAACATTCCTACAATTAAGAAGT; reverse: 5′-TC

CTGACCACCACCTCAACATAG); Vcam1 (forward:

5′-CCCTCCACACAAACCAAGCC; reverse: 5′-CCATTC

CAGTCACTTCAACG); Il-6 (forward: 5′-ACCGCTA

TGAAGTTCCTCTCTCTC; reverse: 5′-CTCTGTGAA

GTCTCCTCCTCTCC); Cxcl8 (forward: 5′-ACTTCAA

GAACATCCAGAGC; reverse: 5′-CTTTCCAGGTCA

GTTAGTTAGCC). Nucleotide sequences of primers for

Ccl2 and reference genes Rpl32 and Gapdh are given in

Ref.[26].

Western blot. MEFs were lysed in a hot SDS buf-

fer (62.5 mM Tris-HCl, pH 6.8; 2% SDS; 10% glycerol;

50 mM DTT, 0.01% bromophenol blue) for 5 min at

94°C. Proteins were separated by electrophoresis using

12% SDS-PAGE, transferred to a PVDF membrane (Bio-

Rad, USA), and sequentially incubated with rabbit anti-

bodies against NRF2 (Invitrogen, USA) and horseradish

peroxidase-labeled goat anti-rabbit immunoglobulin sec-

ondary antibodies (Sigma Aldrich, USA). AWest Dura

Extended Duration Substrate (Thermo Fisher) was used

as a substrate for peroxidase, and images were acquired

using a ChemiDoc gel documentation system (Bio-Rad).

Cytotoxic test. Resazurin test was performed ac-

cording to the standard protocol as described previ-

ously[25]. Experiments to study the effect of oxidative

stress on survival of MEFs were performed when cells

were exposed to 250 μM H

(Ecotex, Russia) and

3 units/ml glucose oxidase (GOx) (Sigma-Aldrich®,

USA) for 3h, n=3.

DNA sequencing. DNA sequencing was performed

using an ABI PRISM® BigDye™ Terminator reagent

kit v.3.1 followed by analysis of reaction products on an

Applied Biosystems3730 DNA Analyzer.

Statistical analysis. The results of crosses between

mice were analyzed using the χ

criterion. Differences in

the levels of gene expression between groups, as well as

in survival of MEFs, was determined using the unpaired

Student’s t-test.

EGOROV et al.1990

BIOCHEMISTRY (Moscow) Vol. 88 Nos. 12-13 2023

RESULTS

Transgenic mice with a mutation in the Nrf2 gene.

Using CRISPR/Cas9 technology and guide RNA cor-

responding to the first exon of the Nfe2l2 gene, a female

F0 mouse was made with a 284nt deletion affecting the

boundary of the first exon and intron of the Nfe2l2 gene

(Fig.1). To eliminate the influence of possible second-

ary mutations caused by the nonspecific effect of ge-

nomic editing, sequential crosses of heterozygotes with

inbred mice of the C57BL/6J line were performed for

ten generations. Heterozygote crosses then produced

homozygous animals, as well as wild-type and heterozy-

gous animals of different sexes.

The observed deletion in the region affecting the

splicing site (5′-CAG|GTGCTGCCC) between the first

exon and intron of the Nfe2l2 gene could, theoretical-

ly, prevent splicing of the Nfe2l2 pre-mRNA and lead to

the subsequent degradation of this transcript. Thus, one

would expect to obtain an animal knockout of Nfe2l2.

However, cDNA sequencing of the Nfe2l2 gene re-

vealed deletion of the codon encoding 8aa residues at

the 3′-end region of the first exon (PPGLQSQQQ),

which were replaced by two aa residues (RW) formed

during translation of the non-coding region of the first

intron (Fig. 1b). At the same time, rest of the NRF2

cDNA remained unchanged. Based on these data, we

concluded that alternative splicing occurs in the ob-

tained NRF2

ΔNeh2

mouse line due to activity of the hid-

den donor splice site (5′-TGG|GTGGGGGAGGC) in

the first intron of the Nfe2l2 gene. Although we had

not obtained the Nfe2l2 knockout mouse, work on this

unique mouse model with a mutation in this gene was

continued.

NRF2

ΔNeh2/ΔNeh2

homozygous mice have increased

embryonic lethality. A total of 22 wild-type (37%),

34heterozygotes (57%), and 3 homozygotes (5%) were

obtained from the crosses between heterozygous mice.

Fig. 1. Structure of Nfe2l2 gene in wild-type and in the NRF2

ΔNeh2

mice. a)WT pre-mRNA sequence of Nfe2l2 gene; b)pre-mRNA sequence

ofthe Nfe2l2 gene with deletion obtained using CRISPR/Cas9 and guide RNA (gRNA). The deleted region of the gene is marked in gray, and

alternative splicing site in the first intron of the gene is also depicted; c)sequence of the mature spliced mRNA of the Nfe2l2 gene with the in-

troduced mutation (substitutions are marked in black); d)domains of the NRF2 protein and comparison of the aa sequences between NRF2

ΔNeh2

andWT. The mutation is located at the N-terminus of NRF2 and represents deletion of 8aa (PPGLQSQQQ) substituted by two aa (RW) that were

translated from the non-coding region of the first intron. The mutation affects positions 8-15, which are adjacent to the Neh2 domain responsible

for binding of NRF2 to its negative regulator KEAP1.