Article

ISSN 0006-2979, Biochemistry (Moscow), 2025, Vol. 90, No. 6, pp. 725-753 © Pleiades Publishing, Ltd., 2025.

Published in Russian in Biokhimiya, 2025, Vol. 90, No. 6, pp. 781-811.

725

REVIEW

Selection of UTRs

in mRNA-Based Gene Therapy and Vaccines

Ilya A. Volkhin

, Anastasia Iu. Paremskaia

2,3#

, Maria A. Dashian

Darya S. Smeshnova

, Roman E. Pavlov

, Olga N. Mityaeva

2,4

Pavel Yu. Volchkov

2,5

, and Andrei A. Deviatkin

2,6,7,a

Faculty of Bioengineering and Bioinformatics, Lomonosov Moscow State University,

119234 Moscow, Russia

Federal Research Center for Innovator and Emerging Biomedical and Pharmaceutical Technologies,

125315 Moscow, Russia

Department of Biomedicine, Pirogov Medical University, 117997 Moscow, Russia

Moscow Center for Advanced Studies, 123592 Moscow, Russia

Center for Personalized Medicine, Loginov Moscow Clinical Scientific Center, 111123 Moscow, Russia

Laboratory of Postgenomic Technologies, Izmerov Research Institute of Occupational Health,

105275 Moscow, Russia

Federal State Budgetary Institution ‘Centre for Strategic Planning and Management

of Biomedical Health Risks’ of the Federal Medical and Biological Agency,

119121 Moscow, Russia

e-mail: devyatkin_aa@academpharm.ru

Received December 25, 2024

Revised May 19, 2025

Accepted May 28, 2025

Abstract— The untranslated regions (UTRs) of messenger RNAs (mRNAs) play a crucial role in regulating

translational efficiency, stability, and tissue-specific expression. The review describes various applications

and challenges of UTR design in the development of gene therapy and mRNA-based therapeutics. UTRs af-

fect critical biological functions, such as mRNA stability, modulation of protein synthesis, and attenuation of

immune response. Incorporating tissue-specific microRNA (miRNA)-binding sites into 3′ UTRs might improve

precise targeting of transgene expression and minimize off-target effects. Nucleotide modifications (pseudou-

ridine, N1-methyladenosine, and N4-acetylcytidine) in mRNA and UTRs in particular, improve mRNA stability

and translational efficiency. At the same time, several challenges remain, such as lack of consensus on UTRs

best suited for certain biomedical applications. Current efforts are focused on integrating high-throughput

screening, computational modeling, and experimental validation to refine UTR-based therapeutic strategies.

The review presents current information on the design of UTRs and their role in therapeutic applications,

with special focus on the possibilities and limitations of existing approaches.

DOI: 10.1134/S0006297924604659

Keywords: gene therapy, mRNA vaccine, UTR, rational design, optimization, RNA secondary structure

* To whom correspondence should be addressed.

# These authors contributed equally to this study.

INTRODUCTION

The emergence of mRNA-based therapeutics has

revolutionized gene therapy and vaccine development

by offering promising solutions in the treatment of

a broad range of diseases, from genetic disorders to

infectious diseases [1, 2]. Currently, there are sever-

al approaches to delivering transgenes into human

cells for the following protein translation on the de-

livered mRNA. mRNA vaccines are typically used for

a short-term transgene expression, while viral vectors

[adeno-associated viruses (AAVs), adenoviruses, len-

tiviruses, etc.] ensure a long-term expression of the

VOLKHIN et al.726

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

transgene. After virus entry to the cell, transcription

of the viral mRNA is regulated by the same processes

as transcription of exogenous mRNAs. However, the

differences between such mRNAs should not be ig-

nored. For example, modifications of ribonucleotides

aimed to reduce the immunogenicity [3] and increase

the translational efficiency [4] of artificial RNAs can

disrupt their translation [5]. Numerous details in the

process of optimizing the structure of viral vectors

has been described in several studies [6-9]. Howev-

er, the issue of UTR design has been rarely discussed

[10-12]. At the same time, the studies describing novel

approaches to the UTR design [13-16] do not address

the challenge of extending such improved UTR design

to viral vectors (e.g., AAV vectors). The introduction

of best practices in the mRNA design phase of AAV

vectors and mRNA vaccines can significantly improve

both technologies.

mRNA vaccine encodes a protein (e.g., surface an-

tigen of a pathogen or cancer epitope) that can trigger

the human immune response. mRNAs can be encapsu-

lated in lipid nanoparticles, lipid-like materials, poly-

meric nanoparticles, hybrid systems, and nanoemul-

sions [1, 17, 18]. These artificial carriers are positively

charged and form polyplexes with negatively charged

mRNAs, which facilitates their release from the par-

ticle into the cell’s cytoplasm for further translation

[19, 20]. It should be noted that mRNAs are relatively

rapidly degraded in the cell. However, their transla-

tion yields heterologous proteins that might induce

the immune response against the target antigen. The

synthesized protein is further processed into smaller

peptide epitopes that are presented on the cell surface

by the major histocompatibility complexes (MHCs)

class  I and class  II to CD8

and CD4

T  cells, leading

to the development of cellular and humoral immu-

nity, respectively, against the antigen encoded in the

mRNA [17,  21]. Due to its high immunogenicity and

short lifespan, mRNA is considered to be the optimal

vaccine against infectious diseases or cancer [17,  20,

22]. In some cases, mRNA is used as a template for

the synthesis of a functional protein that is necessary

for the treatment of a specific disease. For example,

mRNA-based therapeutics against protein deficien-

cy-related metabolic diseases (propionic acidemia,

methylmalonic acidemia, and phenylketonuria) have

recently been successfully tested in preclinical studies

[23]. Hence, the range of possible applications for mR-

NA-based drugs can be extended to the development

of protein replacement therapies.

Viral vectors (adenoviruses, lentiviruses, retro-

viruses, AAVs, and others) are typically used for the

long-term transgene production a with a direct ther-

apeutic effect [24] which is achieved by maintaining

the viral genome through its constant transcription.

This process results in the intracellular synthesis of

mRNA utilized for the production of a desired pro-

tein. The precise dosage of the virus is determined,

in particular, by the efficiency of protein synthesis,

as higher efficiency of transcription and translation

allows to use a lower virus dose, which is safer and

more cost-effective [25]. Due to the high efficacy, AAV

vectors have been used much more commonly than

other viral vectors [26, 27]. For this reason, this re-

view focused mostly on AAV-based gene therapeutics.

Recombinant AAV (rAAV) vectors, which lack the viral

rep and cap genes, carry an expression cassette con-

sisting of the transcription promoter, single-stranded

DNA coding for the transgene and flanked by inverted

terminal repeats, and polyadenylation signal (polyA)

within an icosahedral capsid that can allow tissue-spe-

cific targeting. After endocytosis, viral single-stranded

DNA of the forms episomes in the nucleus and then

converted into the double-stranded DNA transcribed

by the cellular machinery. The mRNA is exported to

the cytosol for the translation of the therapeutic trans-

gene [27, 28]. The lifespan of the transgene introduced

to the cell by this approach is limited only by the lifes-

pan of the cell itself [21  29]. For instance, in tissues

with a low cell turnover (e.g., muscles), expression of

the introduced transgene (factor IX) has been detected

for at least 10 years [30]. Consequently, the use of AAV

vectors is currently the preferred method for produc-

ing intracellular functional proteins to compensate for

defective endogenous proteins in the cytoplasm. The

application of AAVs as delivery vectors for the gene

replacement therapy has been supported by clinical

trials and FDA approvals for their use in the treatment

of ocular, neurological, metabolic, and hematological

disorders [27, 31-35].

The above therapies rely directly (mRNA vaccines)

or indirectly (AAV vectors) on mRNA as a template

for protein synthesis. It should be noted that mRNAs

transcribed on the AAV DNA in the nucleus and then

exported to the cytoplasm are structurally similar

to heterologous mRNAs that enter the cell from the

external environment (Fig.  1). Both types of mRNAs

have similar structural domains that contain all essen-

tial elements required for their translation, namely,

the 5′  cap, 5′  untranslated region (UTR), open read-

ing frame (ORF), 3′ UTR, and poly(A) tail [1, 20, 36].

Atthe same time, the mechanism for exogenous RNA

recognition by the cellular machinery can be differ-

ent from that for endogenous RNAs. RNA synthesized

in vitro may not have the nucleotide composition and

secondary structure of natural RNAs. As a result, ex-

ternal RNAs can trigger the mechanisms of intracellu-

lar immunity after recognition by the corresponding

RNA-binding proteins [37]. Moreover, the process of

RNA synthesis in vitro is not precise and can lead

to the formation of double-stranded RNAs [38] and

truncated RNAs (aborted transcription products) [39].

SELECTION OF UTRs 727

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

Fig. 1. mRNA-based therapeutics: mRNA vaccine (left) and rAAV vector-based drug (right). The mRNA vaccine contains

ready-for-translation mRNA encapsulated in a liposomal particle. Upon release from the liposome inside the cell, the mRNA

is translated by the cell ribosomes, and the synthesized polypeptide serves as an antigen that stimulates the host’s im-

mune response or as a therapeutic protein compensating for a defective cellular protein. The rAAV vector consists of a

single-stranded DNA carrying a transgene inside a viral capsid. After the vector enters the cell via endocytosis, the rAAV

genome is transcribed in the nucleus. The resulting mRNA is exported to the cytoplasm, where it is translated into the

therapeutic protein.

Numerous efforts have been made to minimize these

factors (see review [37]).

After mRNA entry to the cytoplasm, its 3′ poly(A)

tail and 5′ cap interact with the translation initiation

factors (eIFs), leading to the mRNA circularization and

recruitment of the preinitiation complex (PIC) formed

by the 40S small ribosomal subunit, eIFs, initiating me-

thionyl-tRNA, and GTP. The 40S small ribosomal sub-

unit scans the 5′  UTR for the start codon in the Kozak

sequence. After the start codon is found, the eIFs are

released and the 60S ribosomal subunit is recruited

to form the 80S ribosome capable of protein synthe-

sis. The elongation of the amino acid chain continues

until the stop codon is reached. Finally, the nascent

protein gains functionality through post-translational

modifications [1,  17,  40]. It should be noted that trans-

lation requires the presence of all mRNA structural

domains (5′  cap, 5′  UTR, Kozak sequence, ORF, and

3′  UTR). Therefore, altering any mRNA domain can

directly affect the stability, immunogenicity, and trans-

lation of mRNA therapeutics [20, 41]. Currently, there

are a variety of approaches for optimizing the 5′  cap,

nucleotide context of the start codon, and poly(A) tail

[42-58]. For example, the length of the poly(A) tail

can influence the efficiency of mRNA translation and

mRNA degradation rate [45, 50-52]. Interestingly, the

effect of the poly(A) tail on translation depends on the

length of the 3′ UTR [59, 60]. The stability of mRNA

can also be affected by the chemical modifications of

the poly(A) tail [61, 62]. However, this review primar-

ily focuses on the strategies for optimizing the UTRs.

Despite a significant progress in the development

of mRNA-based therapies and vaccines, there is a lack

of consensus on the optimal UTRs to be used in spe-

cific biomedical applications. UTRs, which flank the

coding sequence in mRNA, are crucial for regulating

the translational efficiency and stability of mRNA.

Commonly, the UTRs selected for therapeutic applica-

tions (e.g., UTRs of human β-globin mRNA) provide a

high translational efficiency and mRNA stability. Con-

sequently, the optimization of UTRs has emerged as

an area of focus in the development of mRNA-based

therapeutics [63, 64].

In this review, we analyzed in detail the UTRs

that have shown promise in preclinical and clinical

studies in order to identify the strategies for selection

VOLKHIN et al.728

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

Fig.  2. Regulatory structures in the 5′ UTR and 3′ UTR that influence translational efficiency and mRNA stability. Abbrevi-

ations: uORF, upstream open reading frame; lncRNA, long non-coding RNA; miRBS, miRNA-binding site; miRISC, miRNA-

induced silencing complex; 4E-T, eukaryotic translation initiation factor 4E transporter; CCR4-NOT, carbon catabolite re-

pression – negative on TATA-less; STAU1, staufen double-stranded RNA binding protein 1; ARE, AU-rich element; WPRE,

woodchuck hepatitis virus posttranscriptional regulatory element; RBPs, RNA-binding proteins.

of most effective UTRs for therapeutic applications

and to guide researchers in choosing the optimal UTRs

for their specific needs.

THE EFFECT OF UTRs

ON THE PROTEIN YIELD

The translational efficiency (translation rate) is

often defined as the number of protein molecules

produced per mRNA molecule per unit of time [65].

The mRNA degradation rate is defined as the number

of mRNA molecules degraded per unit of time, reflect-

ing the speed rate at which mRNA is eliminated from

the cell and can be influenced by multiple factors,

such as mRNA sequence elements, RNA-binding pro-

teins (RBPs), microRNAs (miRNAs), and various decay

mechanisms.

The goal behind the generation of synthetic

mRNAs is usually to achieve a higher protein yield,

which depends on both mRNA translation efficiency

and degradation rate. The half-live time differs for dif-

ferent mRNAs, with a median value of approximately

7  h  [66]. There is an inherent trade-off between the

translation rate and mRNA stability  [67]. Although

high ribosome loading increases the translation rate,

it can also make mRNA more susceptible to hydrolysis.

The reasons for this phenomenon remain unknown,

but might involve ribosome crowding and consequent

mRNA degradation. In fact, stable mRNAs, even if they

have a lower translational efficiency, can ultimately

produce more protein over time due to a longer lifes-

pan [67]. Therefore, balancing translational efficiency

and mRNA stability is critical, and the choice of UTRs

plays a pivotal role in modulating these factors.

The efficiency of translation is influenced by the

stability and position of the secondary structure ele-

ments (e.g., hairpins) within the 5′ UTR (Fig. 2). The

rate-limiting step of translation initiation is the bind-

ing of the 43S PIC to mRNA [68]. Since stable second-

ary structures in the 5′ UTR can prevent this process,

it is generally assumed that they suppress translation.

As the stability of the hairpin increases, the transla-

tional efficiency typically decreases. For example, an

increase in the hairpin stability from –25 to –35  kcal/

mol can almost completely halt the translation [68].

It has also been shown that hairpins located close to

the 5′  cap strongly repress translation [69], likely by

interfering with the mRNA recognition or PIC bind-

ing. Furthermore, the guanine–cytosine (GC) content

of the hairpin influences the translational efficien-

cy independently of the hairpin thermal stability or

position [68]. The simplest explanation is that the ri-

bosome does not melt the entire hairpin at once but

unwinds it gradually. The GC pairs are more difficult

to break apart than the AU pairs because of their high-

er stability due to an additional hydrogen bond and,

therefore, can slow down or delay the progress of the

ribosome. The hairpins with the same stability but a

higher GC content have a stronger inhibitory effect

on the translational efficiency. Even though secondary

structures typically reduce the efficiency of transla-

tion, some 5′  UTRs with stable secondary structure el-

ements can support efficient initiation of translation,

for example, the 5′  UTR of dengue virus (DEN2) [67],

presumably, due to the promotion of efficient cap-in-

dependent translation initiation by the secondary

structure [70]. Another probable explanation is that

the 5′  UTR of DEN2 interacts with the cellular protein

La, which acts as an RNA helicase and thus facilitates

translation [71]. In summary, it is usually better to

avoid formation of stable secondary structures in the

5′ UTRs of artificial mRNAs, especially near the start

codon and the cap.

Generally, stable elements of secondary struc-

tures increase RNA stability in solution by protecting

it from hydrolysis [72]. mRNA molecules in the in-

tracellular environment are generally less structured

SELECTION OF UTRs 729

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

than in vitro, due to the interactions with RBPs, RNA

helicases, and ribosomes [73], which highlights the

importance of cellular context when evaluating the

structural features of RNA.

G-quadruplexes (RG4s) are a specific type of tran-

sient four-stranded structures in RNA that can impact

translation [74]. The exact mechanism of their influ-

ence remains unknown, but it has been hypothesized

that RG4s inhibit translation by preventing the bind-

ing of the 43S PIC to mRNA or by slowing down the

scanning [73]. However, there is also evidence of a

positive influence of RG4s on translation (e.g., in the

case of human clAP1 mRNA) [75]. Therefore, the im-

pact of RG4s on translation appears to be multidirec-

tional, and its underlying mechanisms still have to be

elucidated in future studies.

Upstream open reading frames (uORFs) are short

reading frames located upstream of the major coding

sequence; they are common regulatory sequences in

human 5′  UTRs [76-80]. Although ribosomes can reini-

tiate translation downstream of uORFs, these elements

are thought to have a strong inhibitory effect on trans-

lation [81, 82]. The simplest mechanism of such inhibi-

tion is recognition of the uORF instead of the canoni-

cal ORF, which decreases the probability of translation

initiation for the canonical protein. Another mecha-

nism is ribosome stalling of by a peptide encoded by

such uORF, as was shown for human cytomegalovirus

UL4 mRNA  [83]. Beside decrease the translational ef-

ficiency, uORFs cause mRNA degradation via the non-

sense-mediated mRNA decay mechanism [84, 85] initi-

ated by the presence of a premature stop codon [86].

It was demonstrated that an uORF within the 5′ UTR

of mRNA can activate the process, ultimately result-

ing in mRNA degradation [86, 87]. At the same time

some UTRs contain signals that protects mRNA from

degradation despite the presence of uORFs, as it was

shown for yeast mRNAs [88]. The upstream start co-

dons (uAUGs) that do not form complete uORFs within

the 5′  UTR (i.e., lack the in-frame stop codon before

the main AUG) have an even stronger inhibitory effect

on translation than complete uORFs. The reason for

this might be the inability of ribosomes to reinitiate

translation from the main ORF downstream of such

uAUGs [76]. Since uORFs and uAUGs usually have a

strong negative impact on translation, they should be

absent in artificial mRNAs.

The presence of binding sites for proteins and

long non-coding RNAs (lncRNAs) in the UTRs can

also affect translation due to the interaction of these

lncRNAs and RBPs with the translation machinery

[73,  89]. For example, the antisense Uchl1 lncRNA

promotes the synthesis of mouse ubiquitin carboxy-

terminal hydrolase L1 (UCHL1) due to the enhanced

ribosome binding [90]. The RBP-binding sites can ei-

ther enhance or inhibit translation. An example of the

RBP-mediated translational regulation is the role of

iron regulatory proteins  1 and 2 (IRP1 and 2). These

proteins can bind to specific regions (known as iron

response elements) in the 5′  UTRs of mRNAs coding

for iron metabolism proteins, thus inhibiting their

translation [91]. Another example is regulation of the

msl-2 mRNA translation by the sex-lethal (SXL) protein

in Drosophila, which was found to bind both the 5′

and 3′ UTRs of msl-2 mRNA and to inhibit its trans-

lation. However, the mechanisms of inhibition differ

for the 5′ and 3′  UTRs. In the 3′  UTR, SXL recruits the

UNR (upstream of N-Ras) repressor protein and pre-

vents the formation of PIC. The binding of SXL to the

5′  UTR requires the presence of the binding site and

the uORF and promotes the recognition of uAUG, thus

inhibiting translation of the main ORF [92]. An exam-

ple of the positive effect of RBPs is activation of the

cap-independent translation by the YB-1 protein [93].

Some GC-enriched clusters present in particular hu-

man mRNAs can recruit YB-1 and promote translation.

In summary, the interaction of UTRs with lncRNAs and

RBPs may also play a role in the translational regula-

tion. Therefore, potential interactions of the binding

motifs present in the UTRs should be taken into ac-

count when selecting the UTRs for the use in mRNA

vaccines and mRNA-based agents.

In contrast to the 5′ UTR, the 3′  UTR are believed

to have a lesser influence on the ribosome loading

and, therefore, translation rate [67]. However, the

3′  UTRs may contain some regulatory sequences capa-

ble of interacting with RBPs or microRNAs (miRNAs)

and influencing the efficiency of mRNA translation

and lifespan [66, 94]. Their effect could be explained

by several mechanisms. For example, the CCR4-NOT

exonuclease can inhibit translation and cause mRNA

decay. It is recruited by RBPs and miRNAs interacting

with specific sequence motifs in the 3′  UTR. Hence,

a 3′ UTR lacking these sites will not be degraded by

CCR4-NOT. One of the mechanisms of CCR4-NOT action

is physical displacement of poly(A)-binding proteins

(PABPs) and prevention of translation initiation [95].

Another mechanism is the recruitment of special in-

hibitor proteins, such as eukaryotic translation initia-

tion factor 4E transporter (4E-T), which interacts with

the translation initiation factors and inhibits transla-

tion initiation [96]. CCR4-NOT has also been demon-

strated to induce mRNA decapping and deadenylation,

leading to mRNA degradation.

Another mechanism through which 3′  UTRs ex-

ert their influence on mRNA decay is the presence

of adenylate/uridylate (AU)-rich elements (AREs) in

the 3′  UTRs. These sequences are considered to be

mRNA decay signals [94]. It has been demonstrated

that many RBPs, including tristetraprolin and butyrate

response factors 1 and 2, bind to AREs, resulting in

mRNA destabilization and degradation [97]. However,

VOLKHIN et al.730

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

not all AU-rich elements cause translational repres-

sion. Thus, the human antigen R protein (HuR) binds

AU-rich sequences in the 3′ UTRs, thereby increasing

the stability of the corresponding mRNAs by protecting

them from miRNA binding through steric hindrances

[98]. The precise mechanism of ARE influence on the

mRNA stability depends on the stoichiometry between

AREs and RBPs, as well as the 3′  UTR structure and

arrangement of regulatory elements. Furthermore,

mRNA cleavage can be facilitated by lncRNAs. For in-

stance, 3′  UTRs of some mRNAs contain Alu elements

that can pair with certain lncRNAs, resulting in the

formation of double-stranded RNAs that undergo deg-

radation via the STAU1 (staufen double-stranded RNA

binding protein 1)-mediated mechanism [99].

Other molecules that can promote mRNA degra-

dation are miRNAs which guide the miRNA-induced

silencing complex (miRISC) composed of the miRNA

and Argonaute protein [100]. miRISC binds to special-

ized sites on mRNAs, known as miRNA-binding sites

(miRBSs), and recruits effector protein complexes, such

as PAN2-PAN3 and CCR4-NOT, that inhibit translation

and cause mRNA degradation. Therefore, to ensure a

greater mRNA stability, it is important to avoid the

degradation signals (AREs, Alu elements, and miRBSs)

in the 3′  UTRs. However, the effect of such signals de-

pends on the intricate cellular signaling network and

can be almost unpredictable. Interestingly, Leppek

etal. [67] have shown that the 5′  UTRs have a greater

effect on the mRNA stability than the 3′  UTR. This is

surprising because there is a plethora of mechanisms

by which the 3′ UTR can influence the mRNA lifespan.

To the best of our knowledge, the reason for this phe-

nomenon is unknown. The 3′  UTRs can either inhibit

or promote translation. For instance, the 3′  UTRs of

mRNAs coding for some nuclear-encoded mitochondri-

al proteins (e.g., ATPase subunits) promote translation

due to the interactions with the translation machinery

[60, 101]. Another example is the woodchuck hepatitis

virus post-transcriptional regulatory element (WPRE),

which is frequently used in gene therapy to increase

protein production. While its main function is to pro-

mote the nuclear export of mRNAs [102], Loeb et al.

[103] suggested that WPRE can also activate transla-

tion via an unknown mechanism.

UTR DESIGN FOR TISSUE-SPECIFIC EXPRESSION

Approximately half of human genes produce mul-

tiple mRNA isoforms that can vary in their 3′  UTR

sequences [104]. Most mRNAs with a single isoform

are transcribed in a tissue-restricted manner, whereas

ubiquitously expressed genes are typically transcribed

into multiple mRNA isoforms, depending on the tis-

sue context. This tissue-dependent diversity of 3′  UTRs

can be potentially exploited to develop tissue-specific

gene therapies. Limiting expression of transgenes in

non-target tissues is critical for preventing the toxic

effects of protein overexpression and immunotoxicity.

Theoretically, this can be achieved by incorporating

tissue-specific miRBSs into 3′  UTRs in order to restrict

transgene expression in the cells enriched in the cor-

responding miRNA (Fig.  3). For example, it was re-

ported that adding the binding sites for miR-122 and

miR-1 to rAAV9 promoted protein synthesis in the CNS

while reducing its expression in the liver, heart, and

skeletal muscles. The incorporation of miR-204 and

miR-124 targeted transgene expression to the photo-

receptors and retinal pigment epithelium, respectively

[27, 105-108].

miRNA targeting is an effective way to restrict

the transgene expression to specific tissues based on

different miRNA profiles of different cell lineages.

Ludwig et al. [109] created a comprehensive human

miRNA tissue atlas by determining the abundance of

1997 miRNAs in 61 tissue biopsies of different organs.

One thousand three hundred sixty-four miRNAs were

discovered in at least one tissue, 143 were present in

each tissue. The majority of miRNAs displayed inter-

mediate tissue specificity, with some expressed pre-

dominantly in specific tissues, such as the brain [109].

Recent studies have shown that the tissue tropism of

certain viruses could be eliminated by the incorpora-

tion of miRNA target sequences into their genomes to

prevent viral RNA translation in the tissues expressing

tissue-specific miRNAs. Vesicular stomatitis virus is

considered as a promising recombinant vaccine plat-

form and oncolytic agent. However, it has not yet been

tested in humans, as it was found to cause enceph-

alomyelitis in rodents and primates by almost any

route of administration. Kelly et al. [110] attempted

to eliminate this effect by creating an attenuated vi-

ral strain by inserting neuron-specific miRBSs into the

virus 3′  UTR. The new viral strain retained its original

oncolytic properties, but lost the neurotropic effect,

which allows its use in clinical trials. This approach

has been successfully applied to the poliovirus [111]

and mosquito-borne Japanese encephalitis virus [112].

The modified strains retained their ability to replicate

in non-neuronal tissues, but their neurotropism de-

creased significantly, leading to the development of

safer and more efficient vaccines.

5′ UTRs may display tissue specificity due to the

alternative splicing and formation of multiple vari-

ants of the first exon encoding different 5′  UTRs [113,

114], the use of multiple promoters [115,  116], or a

combination of these mechanisms. While the result-

ing mRNAs have identical coding regions, the pres-

ence of different 5′ UTRs affects the translational ef-

ficiency, leading to the tissue-specific expression of

the same gene. However, this regulatory mechanism

SELECTION OF UTRs 731

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

Fig. 3. Principles of tissue-specific 3′ UTR design. Although this approach has been successfully demonstrated for AAV

vectors, it can also be applied to mRNA therapies. a) The AAV vector expression cassette contains the binding sites for

tissue-specific miRNAs in the 3′ UTR. Incorporation of miRBSs for miRNAs present in the non-target tissues or cells allows

to reduce off-target transgene expression. b) The AAV vector is internalized into the cell and enters the nucleus, where

the transgene mRNA is transcribed. After mRNA transcription and nuclear export, tissue-specific miRNA in the content of

miRISC binds to miRBSs in the synthesized mRNA, resulting in silencing due to the translational repression and/or mRNA

degradation in the off-target cells.

is not directly applicable to the development of mRNA

vaccines or gene therapies, as the 5′ UTRs in such con-

structs are synthetic and are not the subjects to en-

dogenous transcriptional regulation. Once an mRNA

enters the cell, transcription no longer occurs and

translation depends on the cellular environment and

properties of mRNA itself. Although this mechanism

is informative from a biological point of view, it can-

not be used for tissue targeting in the development

of mRNA-based therapeutics.

The major challenge in protein replacement ther-

apy is the immune response triggered by heterologous

proteins, which can undermine the efficacy of thera-

py. For example, AAV vectors can deliver transgenes

not only to target cells, but also to antigen-presenting

cells (APCs). APCs process foreign proteins and ini-

tiate the immune response. A promising strategy to

mitigate this issue involves incorporation of binding

sites for APC-specific miRNAs, such as miR-142 and

miR-652, into the rAAV expression cassette to redirect

the transgene expression away from the APCs. The

miRBSs engineered for the transgene can interact with

cellular miRNAs, leading to the transcript degradation.

In APCs, such constructs can prevent the presentation

of transgene peptides to T cells, thus reducing immu-

nity against the transgene protein [117-119]. PARADE

(Prediction and RAtional DEsign of mRNA UTRs) is a

recently developed promising tool [120] for designing

UTRs regions with a tailored cell type-specific activ-

ity. To develop this generative artificial intelligence

framework, the regulatory activity of 60,000 5′ and

3′  UTRs was measured in six cell types. The PARADE

tool was validated by testing 15,800 de  novo designed

sequences in six cell types. Many sequences showed

superior specificity and activity compared to existing

RNA therapeutics. The PARADE tool can potentially

VOLKHIN et al.732

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

facilitate the design of tissue-specific mRNAs for var-

ious therapies by mitigating the negative effects of

protein expression in non-target organs.

THE IMPACT OF MODIFIED

NUCLEOTIDES IN THE UTRs

The dynamic regulation of post-transcriptional

modifications (PTMs) is mediated by a complex net-

work of ‘writer’, ‘reader’, and ‘eraser’ proteins [121].

PTMs play a critical role in RNA preprocessing in the

nucleus, including alternative splicing and nuclear

export. In the cytoplasm, PTMs modulate important

aspects of the transcript life cycle, such as intracellu-

lar localization, translational regulation, stability, and

degradation [122].

Current studies on the epigenetic regulation of

RNAs by PTMs primarily focus on their impact in

cancer therapy [123] and involvement in the devel-

opment of complex neurological, cardiovascular, and

metabolic diseases [122, 124]. However, PTMs may

also hold significance for the future development of

mRNA-based therapeutics. Thus, Pfizer and Moderna

used PTMs in the development of the COVID-19 mRNA

vaccine [125]. The mRNA component of the vaccine

was composed of the 5′ UTR, the coding sequence for

the spike protein with two consecutive stop codons,

and the 3′ UTR in which all uridines have been re-

placed by N1-methylpseudouridine.

The distribution of nucleotide modifications across

different regions of mRNA varies significantly. For ex-

ample, pseudouridines are more prevalent in the 3′

UTRs and coding regions than in the 5′ UTRs [126].

In  vivo isomerization of uridine to pseudouridine is

carried out by pseudouridine synthases (PUSs). PUSs

catalyze the formation of pseudouridine by breaking

the N1–C1′ glycosidic bond and forming a new C5–

C1′ bond, with the N1 atom becoming an additional

hydrogen bond donor. The resulting conformational

changes contribute to the stabilization of the polynu-

cleotide chain [127]. mRNAs containing pseudouridine

and 1-methylpseudouridine have a reduced ability to

activate the innate immune system because they are

less likely to be detected by receptors, such as the

Toll-like receptor  3 (TLR3) and retinoic acid-inducible

gene  1 (RIG-I) [3,  128,  129]. According to predictions

from the convolutional neural network trained on

unmodified sequences, the incorporation of pseudou-

ridine and 1-methylpseudouridine into the 5′ UTRs

correlates positively with the ribosomal loading due

to the increase in the minimum free energy and alter-

ations in the RNA secondary structure [130]. However,

N1-methylpseudouridine induces ribosome pausing

and leads to a +1 frameshift [5]. Pseudouridylation

is associated with the enhanced mRNA translational

efficiency. Thus, it was found that PUS7 promotes the

translation of the tumor suppressor ALKBH3 in gastric

cancer [131]. The translational efficiency of modified

transcripts depends on the extent of modification:

the transcripts containing over 50% N1‐meth-

ylpseudouridine substitutions yielded less protein than

those with lesser percentage of modification [129]. As

demonstrated in a recent study, N1-methylpseudouri-

dine in a context-dependent manner affected the de-

coding of the mRNA sequence by altering the strength

and pattern of the mRNA–tRNA interactions in the

ribosome  A site [132]. 5-Methoxyuridine (5moU), an

alternative modification of uridine, increased the sta-

bility, reduced immunogenicity, and enhanced protein

yield for in vitro transcribed (IVT) mRNA [133].

When located in different transcript regions, the

same modification can perform different biological

functions. For example, the enrichment of the start

codon, stop codon, and 5′  UTR (particularly at the crit-

ical cap+1 and cap+2 positions) with N

-methylade-

nosine (m

A) enhances the translational efficiency,

whereas modification of the coding sequence and the

3′  UTR reduced the efficiency of translation [134-136].

It was also found that m1A in the 3′  UTR impacts the

interaction between mRNA and miRNAs; however, the

underlying mechanisms of this influence are yet poor-

ly understood [137]. The m

A modification enhances

the effect of N

-methyladenosine (m

A) by inducing

the RNA degradation pathway [138]. The m1A sites at-

tract HRSP12 (heat-responsive protein 12), which facil-

itates the binding of the m

A reader protein YTHDF2

(YTH domain family  2) to m

A sites on the same RNA

molecule. The HRSP12–YTHDF2 complex recruits the

P/MRP endoribonuclease that cleaves mRNA thus sup-

pressing translation [139, 140]. The m

A modification

promotes destabilization of the RNA secondary struc-

ture [141]. Early studies have indicated that its pres-

ence in the 5′  UTR near the start codon facilitates the

cap-independent translation. However, recent findings

show that the thermodynamic changes resulting from

a single modification are insufficient to stabilize the

initiation complex and do not affect the translational

efficiency [142]. The effect of the m

A modification

on the translation initiation may be mediated by the

reader protein YTHDF1. YTHDF1 binds to m

A sites

near the 3′ UTR proximal region and, through loop

formation, interacts with eIF3. YTHDF3 interacts

with YTHDF1 to enhance its translation-promoting

effect [143].

N4-acetylcytidine (ac4C) had been originally iden-

tified in tRNA and rRNA; its presence mRNA was

confirmed later. This PTM plays a critical role in reg-

ulating the transcript stability and gene expression.

Overexpression of NAT10, an enzyme catalyzing ac4C

modification, was found to promote proliferation of

malignant cell in many cancer types [144]. It has been

SELECTION OF UTRs 733

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

repeatedly demonstrated that ac4C, which predomi-

nates in the 3′  UTR and surrounds the stop codon, can

stabilize mRNA [145, 146]. The presence of ac4C in the

coding sequence is associated with the upregulated ex-

pression of the corresponding mRNA [147]. Also, the

presence of ac4C in the 5′ UTR influences translation

initiation by suppressing the usage of strong start co-

dons and facilitating alternative initiation at weaker

sites [148].

NSUN proteins are a family of RNA methyltrans-

ferases that catalyze the formation of 5-methylcytosine

C), thus playing a critical role in RNA metabolism.

C is found throughout the transcripts, with a higher

concentration in the 5′ UTRs. Interestingly, m

C-con-

taining mRNAs demonstrate a reduced translational

efficiency, despite longer half-life [149]. The content of

-methylguanosine (m

G) in the coding sequence and

3′  UTR increases after heat shock and oxidative stress,

while the content in the 5′  UTR decreases, which con-

tributes to the increased translational efficiency of

such mRNAs [150]. m

G recruits initiation factors and

protects RNAs from 5′  exonucleases [151].

Therefore, incorporating modified bases in the

UTRs and coding sequencing, as well as taking into

account the functional properties of modified nucle-

otides, including their ability to trigger or attenuate

the immune response, regulate translation rate and

efficiency, and increase mRNA stability, can improve

the development of therapeutic mRNAs. However, it

is important to point out that the existing data are

limited, and further research is needed for a more

comprehensive understanding of the mechanisms un-

derlying the effect of modified bases.

UTRs IN RESEARCH

AND CLINICAL TRIALS

The development of gene therapeutics and mRNA

vaccines can be compared to the construction of a

building from different ‘building blocks’, for example,

3′ and 5′  UTRs, which can be tailored and combined

with the coding sequences to optimize the resulting

product.

One strategy to increase the expression efficiency

and stability of IVT mRNAs is the use of UTRs from

highly translated, constitutively expressed, and sta-

ble cellular endogenous genes  [67]. To achieve high

translational efficiency, many mRNA therapies uti-

lize the well-studied 3′ and 5′  UTRs of the human α-

and β-globin genes [68,  152-159] and rabbit β-globin

genes [160], as they improve the translation efficien-

cy and increase the stability of heterologous mRNAs

[16, 161-163].

It is believed that the 5′ UTR of β-globin mRNA

ensures efficient cap-dependent translation due to the

lack of stable secondary structures or translation-sup-

pressing elements [68]. The 3′  UTRs of the α- and β-glo-

bin mRNAs provide a high stability of mRNA due to

association with RBPs [164-168]. For this reason, a

combination of 5′ and 3′  UTRs from the globin mRNAs

is widely used in the studies of translation and stabil-

ity of IVT mRNA [67, 157], creation of vaccines for im-

munotherapy [152, 153, 161, 169], and cell reprogram-

ming into induced pluripotent stem cells (iPSCs) [154].

Two 3′  UTRs the from human β-globin mRNA cloned

head to tail were found to be superior to a single

3′  UTR in improving mRNA translation and stability,

inducing immune responses in mice. The efficiency of

such constructs is currently tested in ongoing clinical

trials (ClinicalTrials.gov: NCT01684241, NCT02035956,

NCT02316457, NCT02410733, and NCT03418480)  [170].

In addition to the widely used α- and β-globin

UTRs, UTRs from other genes can also be incorporated

into mRNAs for mRNA-based therapeutics. Some natu-

ral UTRs have been shown to improve the efficiency

of the corresponding products. For example, the UTRs

of the TSMB10, TMSB4X [171]; HBB, H4C2, Rabb, TPL

[163], mtRNR1, AES  [172]; CYBA  [173], Rps27a  [44],

C3, CYP2E1, and APOA2 [63] genes, viral UTRs (e.g.,

from dengue virus) [174], TEV and TMV 5′ UTRs  [67,

175], and many others have been found to improve

the efficacy of therapeutics and can be potentially

used in developing the UTR design strategies aimed

to improve the mRNA translational efficiency. The

IRES (internal ribosome entry site)-containing 5′  UTR

from the encephalomyocarditis virus enhanced tran-

scription of uncapped mRNAs and increased transgene

expression in human and murine cultured dendritic

cells (DCs). Immunization of mice with DCs transfected

with the IRES-containing mRNA induced high levels

of antigen-specific T  cells, promoted elimination of

antigen-bearing cells, and protected immunized mice

from pulmonary metastasis of melanoma cells injected

intravenously, suggesting a potential use of IRES-con-

taining 5′  UTRs in the antitumor immunotherapy [176].

The IRES was also included in the IVT mRNA used in

the reprogramming of human fibroblasts into plurip-

otent stem cells [177]. Some regulatory motifs, such as

translation initiator of short 5′ UTRs (TISU)  [178], may

also enhance mRNA translation. However, all publi-

cations indicated that the observed effects were cell

type-specific [163]. The list of studied 3′ and 5′ UTRs

and their effects on the translational efficiency and

mRNA stability are summarized in Tables 1 and 2.

Due to the emergence of new computational meth-

ods and machine learning, artificially generated UTRs

with specified properties can be currently developed

to provide an increased mRNA stability and efficient

protein production compared to natural endogenous

UTRs. Thus, the creation of artificial 5′  UTRs ensuring

high translational efficiency was described in [185].

VOLKHIN et al.734

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

Table 1. Studied 5′ UTRs and their impact on the translational efficiency and mRNA stability

Selected 5′ UTR Advantages

5′ UTR

used for

comparison

In vitro/

in vivo

model

Origin and length References

α-globin

α-globin UTR demonstrated

higher translational

efficiency in living cells,

whereas lower values

were observed

in in vitro lysates

β-globin Cos-7 cells

origin:

human α-globin mRNA

length:

34 nucleotides (nt) [179]

[68, 180]

α-globin UTR enhanced

mRNA stability and

protein translation

CYBA

ACE2

A549 and

HepG2 cells

[164]

Human Hsp70

UTR increased the

efficiency of mRNA

translation in cultured

cells and enhanced

antibody response in vivo

Grp 78

VEGF

HepG2 cells

Balb/c mice

origin:

human Hsp70 mRNA

length: 213 nt

[181]

HBB

HSPA1A

Rabb

H4C2

UTRs increased

the translational efficiency

and protein level

5′ UTR of

mRNA-1273

(Moderna)

5′ UTRs of

BNT162b2

(BioNTech)

DC2.4 cells

HBB

origin:

human β-globin mRNA

length: 57 nt

HSPA1A

origin: human heat shock

70 kDa protein 1 mRNA

length: 227 nt

Rabb

origin:

rabbit α-globin mRNA

length: 66 nt

H4C2

origin: histone H4 mRNA

length: 42 nt

[163]

Adenoviral

tripartite leader

sequence (TPL)

UTR strengthened

the antigen-specific

T-cell response and

increased the number

of IFNγ-secreting cells

in immunized mice

5′ UTR of

mRNA-1273

(Moderna)

C57BL/6Cit

and

I/StSnEgYCit

mice

origin: tripartite

leader sequence

of adenovirus mRNA

length: 245 nt

[141]

CYBA

5′ and 3′ UTRs

increased protein levels

independently and

in combination without

affecting the half-life

of mRNA transcripts.

DECR1

GMFG

MAPBPIP

MYL6B

NIH3T3 and

A549 cells

origin: human cytochrome

b-245 alpha chain mRNA

length: 39 nt

[44, 173]

5′ UTR

of S27a-44

transcript

of Rps27a gene

UTR increased the

efficiency of transgene

expression

*S27a-44′ without the TOP

motif improved transgene

expression more efficiently

than the intact S27a-44.

CYBA

α-globin

S27a-45

transcript

of Rps27a

Hep3B and

293T cells

origin: modified UTR from

the ribosomal protein

S27a mRNA

length: different variants

[44]

SELECTION OF UTRs 735

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

Table 1 (cont.)

Selected 5′ UTR Advantages

5′ UTR

used for

comparison

In vitro/

in vivo

model

Origin and length References

NASAR

incorporation of NCA-7d as

the 5′ UTR and S27a as the

3′ UTR with the functional

motif R3U enhanced

antigen translation

CYBA

α-globin

Hep3B and

293T cells

origin:

synthetic sequence

length: 70 nt

[44]

CYP2E1

in vitro transcribed

mRNAs containing these

5′ UTRs demonstrated

improved translational

efficiency regardless

of  the 3′ UTRs present

ASL

ALB

FGA

ORM1

HPX

AGXT

APOA2

HepG2 cells

origin: complement factor

3 mRNA

length: 92 nt

CYP2E1

origin: cytochrome C450

2E1 mRNA

length: 37 nt

[63]

Mouse COL1A2

Hoxa9 P4

Rpl18a TOP

RBCS1A

TMV

TEV

HCV

chimeric fusion

of hHBB 5′  UTR

with TEV or TOP

sequence

DEN2 (alone and

in combination

with DEN2

3′  UTR)

5′ UTRs of these genes

ensured higher ribosome

loading and translational

efficiency than the 5′  UTR

of human β globin gene

hHBB

HEK293T

cells

mouse COL1A2

origin: mouse Col1a2 gene

mRNA

length: 137 nt

Hoxa9 P4

origin: mouse Hoxa9

IRES-like element pairing

element 4

length: 323 nt

Rpl18a TOP

origin: ribosomal protein

18a with 5′ terminal

oligopyrimidine motif

length: the length of

Rpl18a is 89 nt; the TOP

motif is a cis-regulatory

RNA element that begins

directly after the m

G cap

structure and contains

the hallmark invariant

5′-cytidine followed

byan uninterrupted tract

of 4-15 pyrimidines

RBCS1A

origin:

Arabidopsis thaliana

ribulose bisphosphate

carboxylase small chain

1A mRNA

length: 418 nt

TEV

origin: Tobacco etch virus

mRNA

length: 47 nt

TMV DEN

origin: tobacco mosaic

virus UTR and dengue

virus UTR fusion

length: 68 nt for TMV

UTR; 96 nt for DEN UTR

[67]

VOLKHIN et al.736

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

Table 1 (cont.)

Selected 5′ UTR Advantages

5′ UTR

used for

comparison

In vitro/

in vivo

model

Origin and length References

TMSB10

UTR provided higher

levels of gene expression

in cultured cells;

when administered

intramuscularly

to mice as a component

of mRNA vaccine,

the UTR promoted

the transgene

expression in whole

blood, spleen, and liver,

leading to enhanced

specific humoral and

cellular immune

responses

against the antigen

5′ and

3′ UTRs

of TMSB4X

B2M

ACTB

MTRNR2L2

HLA-DRA

HLA-DRB1

FTL

FTH1

DC2.4 cells

origin:

thymosin beta 10 mRNA

length: 78 nt

[171]

R27-UTRs

DC2.4 cells

BALB/c mice

[171]

TISU (translation

initiator of short

5′ UTR)

mRNA with TISU UTR

demonstrated better

expression than mRNAs

containing the Kozak

sequence only,

no 5′  UTR,

or eIF4G aptamer

as a 5′  UTR

invivo translation

efficacy was comparable

to that of mRNA

from the SpikeVax

vaccine with

5′ and 3′  UTRs

Kozak

sequence

eIF4G

aptamer

mRNA

containing

5′ and

3′ UTRs of

SpikeVax®

vaccine

HEK-293

cells

peripheral

blood

mononuclear

cells

human

T cells

C57BL/6 mice

length: minimum 4 nt [178]

Table 2. Studied 3′ UTRs and their effect on the translational efficiency and mRNA stability

3′ UTR Advantages

3′ UTR used

for comparison

In vitro/

in vivo model

Origin

and length

References

HBB-HBB

two 3′ UTRs of human

β-globin fused head to tail

improved the transcript

stability and protein yield,

increased the number of

antigen-specific peptide/MHC

complexes on the cell surface,

and improved expansion

of antigen-specific T cells

in vivo after injection of

transfected bone marrow-

derived DCs (BMDCs)

in mice

single β-globin

3′ UTR

poly-A tail

only, no UTR

human

immature and

mature DCs

EL4 cells

BMDCs

C57Bl/J6 mice

origin:

human

β-globin

mRNA

length:

240 nt [163]

[52, 182]

SELECTION OF UTRs 737

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

Table 2 (cont.)

3′ UTR Advantages

3′ UTR used

for comparison

In vitro/

in vivo model

Origin

and length

References

HBVpA

BGHpA

HBVpA UTR increased the level

of in vitro antigen production

and antibody titer after

immunization of BALB/c mice

more efficiently than BGHpA

and betapA UTRs

HBVpA,

BGHpA, rabbit

β globin

betapA

CHO cells

BALB/c mice

HBVpA

origin:

hepatitis B

virus

length:

50-100 nt

[183]

CYBA

incorporation of 3′ and 5′ UTRs

enhanced protein translation

but did not increase the half-life

of mRNA transcripts.

DECR1

GMFG

MAPBPIP

MYL6B

NIH3T3

and A549 cells

origin:

human

cytochrome

b-245 alpha

chain mRNA

length:

200 nt

[52, 153]

Rps27a (S27a)

UTR increased translational

efficiency and protein expression

CYBA

AAT

HCV

α-globin

Hep3B

and 293T cells

origin:

human 40S

ribosomal

protein S27a

mRNA

length:

170 nt

[44]

NASAR

incorporation of NCA-7d asthe

5′  UTR and S27a as 3′  UTR

with the R3U functional motif

promoted antigen

translation

CYBA

α-globin

Hep3B

and 293T cells

origin:

synthetic

sequence

length:

175-190 nt

[44]

mtRNR1-AES

AES-mtRNR1

UTRs increased protein

production and extended mRNA

half-life in cells, provided

higher splenic expression

of luciferase and stronger

CD8

T cell immune response

after intravenous vaccination

of BALB/c mice, and improved

the reprogramming of human

foreskin fibroblasts (HFFs)

into iPSCs

HBB

HBB-mtRNR1

mtRNR1

HBB-HBB

human DCs

C2C12 cells

HFFs

BALB/c mice

mtRNR1-AES

origin:

synthetic

sequence

length:

300 nt

AES-mtRNR1

origin:

synthetic

sequence

length:

300 nt

[172]

AES-AES

mtRNR-

mtRNR

UTRs provided higher

translational efficiency

AES-HBB

mtRNR-HBB

mRNA-1273

BNT162b2

DC2.4 cells

AES-AES

origin:

synthetic

sequence

length:

360 nt

mtRNR-

mtRNR

origin:

synthetic

sequence

length:

360 nt

[172]

VOLKHIN et al.738

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

Table 2 (cont.)

3′ UTR Advantages

3′ UTR used

for comparison

In vitro/

in vivo model

Origin

and length

References

DEN2 (alone

and in

combination

with DEN

5′  UTR)

UTR promoted higher ribosome

loading and translational

efficiency independently;

combining it with the DEN

5′  UTR had an additive effect

on the translation efficiency

and mRNA stability

HBB

CYBA

HBA1

HEK293T cells

Origin:

dengue virus

Length:

451 nt [184].

[67]

rabbit α-globin Vero cells [174]

3′DENΔSL BHK cells [184]

TMSB10

UTR provided higher levels

of gene expression in cultured

cells, as well as in whole

blood, spleen, and liver after

intramuscular injection to mice,

leading to enhanced specific

humoral and cellular immune

responses against the antigen

5′ and 3′ UTRS

of TMSB4X

B2M

ACTB

MTRNR2L2

HLA-DRA

HLA-DRB1

FTL

FTH1

DC2.4 cells

origin:

human

TMSB10

mRNA

length:

240 nt

[171]

SV40

UTR enhanced

transgeneexpression

Hsp70

HepG2,

Hep3B, and

HEK293 cells

origin:

simian virus

40 large

T antigen

length:

different

variants

[171]

A recent study reported a series of minimal-length

(12-14 nt) synthetic 5′ UTRs that provided higher ex-

pression levels than the full-length human α-globin

5′  UTR in cell lines and in vivo [186]. Deep learning

and massively parallel reporter assay (MPRA) screen-

ing of a large set of different UTRs were used to pre-

dict the effect of different 5′  UTRs on the translational

efficiency to facilitate the design of 5′  UTRs for the

optimal protein production [130,  187]. Furthermore,

UTR optimization resulted in the optimal combination

of endogenous and denovo engineered 5′ and 3′  UTRs

(named NASARs) that were 5-10 times more efficient

than the tested endogenous CYBA and α-globin UTRs

[44]. Three synthetic 5′  UTRs were identified that sig-

nificantly outperformed natural 5′ UTRs in  [15]. Ar-

tificial UTRs were also used to engineer mRNAs for

cell reprogramming and vaccine development [154,

188-190]. Based on the above information, the use of

endogenous UTRs in combination with the de novo

design may be the most efficient approach to UTR

engineering capable of facilitating the development

of mRNA therapeutics [44]. Despite the proven effec-

tiveness of these UTRs, there is a lack of research data

on their comparative efficacy, as well as on selection

of particular UTRs best suited for specific applications.

The absence of standardization and comprehensive

control in the selection of UTRs for gene therapeutics

and mRNA vaccines is a significant challenge. While

some natural UTRs are known to improve the product

quality, their use is often based on precedent rather

than systematic evidence.

Even if a particular UTR outperforms another

UTR with the same coding region, there is no guar-

antee that this will be true for other coding sequence.

This context dependency emphasizes the importance

of interpreting the statements about the “best” UTRs

with caution. So far, no study has systematically com-

pared all possible UTR combinations, which illustrates

a serious gap in understanding the possible approach-

es for optimizing UTR selection for gene therapies

and vaccines. UTRs currently tested in clinical trials

are often derived from natural sequences, especially

those that have shown functional benefits in past re-

search or vaccine development. Commonly used UTRs

include those from the α- and β-globin genes, which

are known for their favorable impact on mRNA stabil-

ity and translational efficiency. These sequences have

been successfully used in approved RNA vaccines,

e.g., vaccines against COVID-19 [191].

However, the selection process for UTRs remains

largely empirical, with many decisions based on a

precedent rather than on rigorous comparative stud-

ies. The lack of a standardized framework or algo-

rithm for the UTR selection represents a significant

SELECTION OF UTRs 739

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

bottleneck in the development of RNA-based thera-

peutics. As UTRs strongly affect the mRNA stability,

translational efficiency, and interactions with regula-

tory molecules (e.g., proteins, miRNAs), their under-

explored potential may hinder innovation in gene

therapy and vaccine development.

Systematic strategies are needed to close this

gap, such as the high-throughput screening of UTR

libraries [15], computational modeling to predict the

UTR behavior in different contexts, and establishment

of standardized criteria for the UTR design. These ef-

forts could help to develop a more rational and in-

formed approach to the UTR selection, which will ul-

timately accelerate the progress in the field of mRNA

therapeutics.

UTR DESIGN

Although regulatory motifs and the secondary

structure of UTRs have no coding function, they sig-

nificantly influence the translational efficiency and

the half-life of transcripts. The design of UTRs for

the development of therapeutic molecules is based

on the inclusion of regulatory elements from natural

sequences flanking highly expressed genes. For exam-

ple, for the development of mRNA vaccines, Pfizer/

BioNTech utilized a 5′  UTR fragment from the human

α-globin mRNA in combination with the modified

Kozak sequence [191]. Although the understanding of

regulatory elements in the UTRs is still incomplete,

it has been shown that an efficient 5′  UTR design re-

quires formation of a less structured region upstream

of the start codon [68, 192, 193] to promote translation

initiation by improving the ribosome accessibility.

It should be mentioned that the results of UTR

efficiency tests performed in vivo cannot always be

extrapolated to the in  vitro mRNA transcription sys-

tems. For example, it was shown that the 5′  UTR of

the CYBA mRNA provided a higher transcription of

the reporter gene in vivo compared to its transcrip-

tion in  vitro  [179]. In other words, the same mRNA

can have different properties in cells and cell-free

solutions, which can be explained by the presence in

living systems of a large number of factors that can

interact with mRNA.

To summarize, in both basic and applied research,

ORFs are typically combined with UTRs whose ability

to increase the translation efficiency has been already

demonstrated. In some cases, these constructs are in-

deed more efficient compared to the wild-type mRNAs,

thus expanding the list of UTRs positively affecting the

mRNA productivity. However, this approach is not sys-

tematic; rather it is a naive enumeration of possible

options. New synthetic UTR sequences can be devel-

oped using machine learning approaches, which rely

on training datasets of sequences with different UTRs,

whose efficiency in protein synthesis has been mea-

sured. Based on this information, the machine learn-

ing methods can predict new UTR sequences outper-

forming natural UTRs. The quality of such predictions

strongly depends on the quality of training datasets.

Furthermore, the predicted variants should be validat-

ed experimentally. Recent computational approaches

to the UTR design are briefly described in Table 3.

EXPERIMENTAL VALIDATION

OF COMPUTATIONALLY PREDICTED UTRS

UTR optimization is a crucial step in the devel-

opment of mRNA-based therapeutics, as UTRs signifi-

cantly affect mRNA stability and protein production.

Several approaches have been developed to analyze

and compare the efficiency of UTRs, ranging from

traditional reporter gene systems to the state-of-the-

art high-throughput methods. Luciferase reporter as-

says are among the most commonly used techniques

for analyzing the UTR functionality. By coupling a

reporter gene (Luc2CP) to specific 3′  UTR sequences,

researchers can evaluate the kinetics of mRNA decay

and gain insight into the stability conferred by these

UTRs. This assay measures the functional half-life of

mRNAs by monitoring the activity of luciferase over

time, which ensures a straightforward and reliable

assessment of the UTR-mediated stability [172]. Ribo-

some profiling provides an overall view of translation-

al efficiency by mapping mRNA fragments protected

by ribosomes, thus allowing direct measurement of

UTR-mediated translational regulation. This technique

is especially useful for understanding ribosome stall-

ing, uORFs, and other regulatory elements that influ-

ence the translation rate [200]. Another approach is

polysome profiling [201], which is based on the sepa-

ration of translated mRNAs associated with polysomes

from untranslated mRNAs using a sucrose gradient.

MPRA enables a high-throughput evaluation of

UTR sequences by using synthetic mRNA libraries con-

taining a large number of UTR variants [202]. Each

variant is associated with a unique DNA barcode that

is incorporated into the reporter transcript. Cells are

transfected with the library containing multiple mRNA

variants. [203], the most efficient of which are select-

ed by polysome profiling [130, 204] or other methods

(e.g., translating ribosome affinity purification [205,

206]) to enrich actively translated mRNAs. Next, RNA

sequencing can be performed to identify the variants

with a high translational efficiency of the reporter

transcript. This approach enables simultaneous inter-

rogation of numerous UTRs to identify the sequence

elements that regulate the translational efficiency

and stability of mRNA. In addition, MPRA facilitate

VOLKHIN et al.740

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

Table 3. Computational methods for UTR prediction

Name Purpose Approaches to model development Sequence length limit Availability References

FramePool

MRL (mean ribosome loads) prediction

for 5′  UTR, analysis of human 5′  UTR

variants, prediction of the effects

of translation initiation site located

upstream of the canonical start codon

one-hot encoding, convolution

neural network (CNN)

with framepooling*

any length

https://github.com/

Karollus/5UTR

[194]

Optimus

5-prime

MRL prediction for 5′  UTR,

development of synthetic 5′  UTR

CNN for MRL prediction;

genetic algorithm for evolving

new sequences

50 nt

https://github.com/

pjsample/

human_5utr_modeling

[16, 130]

Fast SeqProp

DNA, RNA, and protein design

(strong polyadenylation signals,

5′  UTRs, and enhancers.)

sequence optimization model with

VAE-regularization using APARENT,

Optimus 5, DragoNN, MPRA-

DragoNN, DeepSEA, and trRosetta

varying

https://github.com/

johli/seqprop

[195]

UTRGAN

development of synthetic 5′  UTR;

MRL and translation efficiency

optimization

Deep generative adversarial

network (GAN); gradient ascent

algorithm for optimization

with FramePool and MTtrans 3;

one-hot encoding

64-128 nt

https://github.com/

ciceklab/UTRGAN

[196]

DEN

development of synthetic 3′  UTRs,

splicing regulatory elements, enhancers

Deep exploration networks (DENs) –

https://github.com/

johli/genesis

[197]

FunUV

prediction of functional 5′  UTR

and 3′  UTR variants

Gradient boosting

decision tree (GBDT)

50 nt for 5′  UTRs

and 100 nt for 3′  UTRs

https://github.com/

Wangxiaoyue-lab/FunUV

[198]

mRNA2vec

improvement of mRNA stability

and protein production

language model transformer

architecture-based embedding

for pre-training (using

5′  UTRs and ORFs

as input sequences):

average length, 459 nt;

for the downstream

task on 5′  UTR data:

average length, 91 nt

– [199]

PARADE

design of 5′  UTRs and 3′  UTRs

for highly stable mRNAs with

a programmable cell type specificity

diffusion model; genetic algorithm;

random sampling;

motif-based design

–

https://github.com/

autosome-ru/parade

[120]

Note. * Frame pooling is a method used in CNNs that deal with multiple frames (such as shifts in nucleotide sequences). In this process, the network processes each

frame individually, typically by applying global pooling operations (such as max pooling and average pooling) to extract the most important features from each frame.

The results of these operations are then combined into a unified representation, preserving individual characteristics of each frame and the overall structure of the

input sequence. This enables the network to better understand patterns in genetic data.

SELECTION OF UTRs 741

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

Table 4. Challenges and prospects of UTR engineering for gene therapy

Issue Challenges Prospects

Precise UTR

engineering

development of optimal UTR sequences

for specific therapeutic applications

is complicated due to the intricate

interactions between UTRs and cellular

factors which can affect mRNA stability,

localization, translational efficiency, and

other factors are critical for the control

of transgene expression

molecular mechanisms that control UTR

function need to be further explored; the

development of advanced computational

tools and invitro models is essential for

the prediction and optimization of UTR

sequences; machine learning algorithms

and high-throughput screening methods

can accelerate the design of UTRs tailored

to specific therapeutic outcomes

Long-term

safety and

efficacy

the long-term safety and efficacy of mRNA-

based gene therapies cannot be assured

based solely on the insilico data, including

predicted UTRs; unintended effects,

such as immune response or expression

of off-target genes, are difficult to predict

and pose a risk to patients

rigorous preclinical and clinical testing

conducted in compliance with the Good

Laboratory Practice and Good Manufacturing

Practice standards is essential to ensure

the safety of mRNA-based therapies; these

standards help to identify potential risks and

optimize the therapy protocols; in the context

of long-term monitoring in clinical trials,

novel technologies in the UTR design enable

the prediction of promising therapeutic

mRNAs at the beginning of selection process,

saving money and resources

Targeted gene

delivery

targeted gene delivery to specific tissues

or cell types is still a major issue; off-target

effects can reduce the therapeutic efficacy

and increase the risk of side effects of gene

therapies

engineering UTRs with sequences that

promote interaction with cellular receptors

or transmission vectors could improve tissue-

specific targeting; for example, inclusion

of tissue-specific miRBSs in the UTRs can

restrict gene expression to the desired

cell types; this approach would minimize

the off-target effects and improve

the therapeutic precision

Regulatory

and ethical

considerations

the regulatory environment for gene

therapies is complex, and UTR engineering

algorithms and approaches must meet

stringent safety and efficacy standards;

ethical considerations, such as equitable

access and potential misuse, must also

be taken into account

collaboration between researchers, regulatory

agencies, and industry stakeholders will

be crucial to manage the regulatory

process; transparent communication about

the benefits and risks of mRNA-based

therapies will help to gain public trust

and ensure ethical implementation of the

developed therapies

the investigation of ribosome-dependent and ribosome-

independent regulatory mechanisms, which makes it

an invaluable tool for the UTR optimization [187, 204].

Each of these experimental approaches has its own

strengths and limitations. Reporter assays are simple

and specific, but have a limited throughput. MPRA of-

fers a high-throughput screening of UTR variants, but

require advanced sequencing technologies. Ribosome

profiling and polysome profiling offer deep insights

into translation dynamics, however, these methods are

technically complex and resource intensive.

Together, these complementary approaches en-

able systematic assessment of UTRs and provide the

basis for rational mRNA design for both therapeutic

and research applications. Future advancements, par-

ticularly integration of machine learning and compu-

tational modeling, promise even greater precision in

the selection and optimization of UTRs.

CHALLENGES AND PROSPECTS

Although UTR engineering is a promising ap-

proach in gene therapy, a number of issues still re-

main to be resolved to fully realize its potential. These

include precise UTR engineering, the long-term safety

and efficacy of gene therapy, targeted gene delivery,

and ethical considerations (Table 4).

VOLKHIN et al.742

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

CONCLUSION

Despite a success in the identification of UTR vari-

ants affecting mRNA translation, the rational design

of UTRs continues to be a subject of active research.

For example, high-throughput screening of 5′  UTR

libraries by recombinase-mediated integration has

shown that certain UTR variants can significantly

increase expression of the transgenic GFP protein in

HEK293T cells [15]. In addition, optimization of UTR

sequences can be used for controlling the translation-

al capacity of mRNA to ensure the optimal protein

synthesis without exceeding the toxicity limits. For

example, altering the miRBSs in 3′  UTRs suppressed

the expression of toxic transgenes, which in turn re-

duced the cytotoxicity and increased the AAV vector

yield in HEK 293 cells  [207]. The modular nature of

mRNA allows to combine the selection of UTRs with

other approaches aimed to improve gene therapies.

For example, a combination of the UTR library screen-

ing and codon optimization in the ORF improved the

gene therapy for phenylketonuria  [208]. In another

study  [209], the inclusion of an expanded promoter

and modification of 3′  UTR elements led to a signif-

icant reduction in the hepatic toxicity in the treat-

ment of Rett syndrome in a mouse model. These re-

sults highlight the potential of using UTR selection

in combination with other gene therapy approaches,

such as promoter design  [210] and codon optimiza-

tion  [211]. The targeting of therapeutic transgenes to

specific tissues also remains a crucial aspect of gene

therapy.

Transcriptional (using tissue-specific promoters)

and transductional (using viral serotypes) targeting

strategies are established methods in the develop-

ment of tissue-specific therapeutics. There are also

significant opportunities to improve the tissue spec-

ificity of vector-mediated gene expression through

the post-transcriptional regulation, including the

miRNA-mediated suppression. Engineered expression

cassettes containing miRBSs in the 3′  UTRs allow tar-

geted degradation of transgene mRNAs in cells ex-

pressing the corresponding miRNAs [108]. The effica-

cy of miRNA-regulated transgene expression systems

depends on the tissue-specific expression of the se-

lected miRNAs. Recently, miRNAs with suitable expres-

sion profiles for a specific tissue have been discovered

and described in [106, 212, 213].

The principle of action of mRNA-based gene

therapeutics and vaccines includes protein synthesis

in a eukaryotic cell. The process of translation large-

ly depends on the UTRs selected for the design of a

particular construct. Therefore, the rational design

of gene therapeutics and vaccines should involve the

stage of UTR selection in order to develop high-effica-

cy therapeutic agents.

Abbreviations. AAVs, adeno-associated viruses;

CCR4-NOT, carbon catabolite repression – negative on

TATA-less; DEN2, dengue virus; m

A, N

-methylade-

nosine; mRNA, messenger RNA; miRNA, microRNA;

miRBSs, miRNA-binding sites; ORF, open reading

frame; PTMs, post-transcriptional modifications; RBPs,

RNA-binding proteins; uORF, upstream open reading

frame; UTRs, untranslated regions.

Contributions. O.N.M., P.Yu.V., and A.A.D. de-

veloped the concept and supervised the study; I.A.V.

wrote the section “The effect of UTRs on the protein

yield”. A.Iu.P. wrote the sections “UTR design” and

“The impact of modified nucleotides in the UTRs”;

M.A.D. wrote the sections “UTR design for tissue-spe-

cific expression”, “UTRs in research and clinical tri-

als”, “Experimental validation of computationally

predicted UTRs”; D.S.S. wrote the sections “Introduc-

tion”, “UTR design for tissue-specific expression”, and

“UTRs in research and clinical trials”, and prepared

the figures; R.E.P. wrote the section “Challenges and

prospects”. O.N.M., P.Yu.V., and A.A.D. edited the man-

uscript. All authors contributed substantially to the

preparation of the article, read and approved the final

version of the manuscript before publication.

Funding. This study was supported by the Rus-

sian Science Foundation (grant No.23-64-00002).

Ethics approval and consent to participate. This

work does not contain any studies involving human

and animal subjects performed by any of the authors

Conflict of interest. The authors of this work de-

clare that they have no conflicts of interest.

REFERENCES

1. Qin, S., Tang,X., Chen, Y., Chen,K., Fan,N., Xiao,W.,

Zheng,Q., Li,G., Teng,Y., Wu,M., and Song,X. (2022)

mRNA-based therapeutics: powerful and versatile

tools to combat diseases, Signal Transduct. Tar-

get. Ther., 7, 166, https://doi.org/10.1038/s41392-022-

01007-w.

2. Chandra,S., Wilson, J.C., Good,D., and Wei, M.Q. (2024)

mRNA vaccines: a new era in vaccine development,

Oncol. Res., 32, 1543-1564, https://doi.org/10.32604/

or.2024.043987.

3. Andries,O., Mc Cafferty,S., De Smedt, S.C., Weiss,R.,

Sanders, N. N., and Kitada, T. (2015) N1-meth-

ylpseudouridine-incorporated mRNA outperforms

pseudouridine-incorporated mRNA by providing

enhanced protein expression and reduced immuno-

genicity in mammalian cell lines and mice, J. Con-

trol. Release, 217, 337-344, https://doi.org/10.1016/

j.jconrel.2015.08.051.

4. Anderson, B. R., Muramatsu, H., Nallagatla, S. R.,

Bevilacqua, P. C., Sansing, L. H., Weissman, D., and

Karikó,K. (2010) Incorporation of pseudouridine into

SELECTION OF UTRs 743

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

mRNA enhances translation by diminishing PKR ac-

tivation, Nucleic Acids Res., 38, 5884-5892, https://

doi.org/10.1093/nar/gkq347.

5. Mulroney, T. E., Pöyry, T., Yam-Puc, J. C., Rust, M.,

Harvey, R. F., Kalmar, L., Horner, E., Booth, L., Fer-

reira, A.P., Stoneley, M., Sawarkar, R., Mentzer, A. J.,

Lilley, K.S., Smales, C.M., von der Haar,T., Turtle,L.,

Dunachie, S., Klenerman, P., Thaventhiran, J. E. D.,

and Willis, A. E. (2024) N 1-methylpseudouridyla-

tion of mRNA causes +1 ribosomal frameshifting,

Nature, 625, 189-194, https://doi.org/10.1038/s41586-

023-06800-3.

6. Keeler, A. M., and Flotte, T. R. (2019) Recombinant

adeno-associated virus gene therapy in Light of Lux-

turna (and Zolgensma and Glybera): where are we,

and how did we get here? Annu. Rev. Virol., 6, 601-

621, https://doi.org/10.1146/annurev-virology-092818-

015530.

7. Nieuwenhuis, B., Laperrousaz, E., Tribble, J. R., Ver-

haagen,J., Fawcett, J.W., Martin, K.R., Williams, P.A.,

and Osborne, A. (2023) Improving adeno-associated

viral (AAV) vector-mediated transgene expression in

retinal ganglion cells: comparison of five promot-

ers, Gene Ther., 30, 503-519, https://doi.org/10.1038/

s41434-022-00380-z.

8. Tabebordbar, M., Lagerborg, K. A., Stanton, A., King,

E. M., Ye, S., Tellez, L., Krunnfusz, A., Tavakoli, S.,

Widrick, J. J., Messemer, K. A., Troiano, E. C.,

Moghadaszadeh, B., Peacker, B. L., Leacock, K. A.,

Horwitz, N., Beggs, A. H., Wagers, A. J., and Sabeti,

P. C. (2021) Directed evolution of a family of AAV

capsid variants enabling potent muscle-directed

gene delivery across species, Cell, 184, 4919-4938.e22,

https://doi.org/10.1016/j.cell.2021.08.028.

9. Goertsen, D., Flytzanis, N. C., Goeden, N., Chuapoco,

M. R., Cummins, A., Chen, Y., Fan, Y., Zhang, Q.,

Sharma, J., Duan, Y., Wang, L., Feng, G., Chen, Y., Ip,

N. Y., Pickel, J., and Gradinaru, V. (2022) AAV capsid

variants with brain-wide transgene expression and

decreased liver targeting after intravenous delivery

in mouse and marmoset, Nat. Neurosci., 25, 106-115,

https://doi.org/10.1038/s41593-021-00969-4.

10. Luo,L., Jea, J.D.Y., Wang,Y., Chao, P.W., and Yen,L.

(2024) Control of mammalian gene expression by

modulation of polyA signal cleavage at 5′ UTR, Nat.

Biotechnol., 42, 1454-1466, https://doi.org/10.1038/

s41587-023-01989-0.

11. Zhong, G., Wang, H., He, W., Li, Y., Mou, H., Tickner,

Z.J., Tran, M.H., Ou,T., Yin,Y., Diao,H., and Farzan,

M. A. (2020) reversible RNA on-switch that controls

gene expression of AAV-delivered therapeutics in vivo,

Nat. Biotechnol., 38, 169-175, https://doi.org/10.1038/

s41587-019-0357-y.

12. Luoni, M., Giannelli, S., Indrigo, M. T., Niro, A.,

Massimino, L., Iannielli, A., Passeri, L., Russo, F.,

Morabito, G., Calamita, P., Gregori, S., Deverman, B.,

and Broccoli, V. (2020) Whole brain delivery of an

instability-prone Mecp2 transgene improves be-

havioral and molecular pathological defects in

mouse models of Rett syndrome, Elife, 9, e52629,

https://doi.org/10.7554/eLife.52629.

13. Chu, Y., Yu, D., Li, Y., Huang, K., Shen, Y., Cong, L.,

Zhang, J., and Wang, M. (2024) A 5’ UTR language

model for decoding untranslated regions of mRNA

and function predictions, Nat. Mach. Intell., 6,

449-460, https://doi.org/10.1038/S42256-024-00823-9.

14. Castillo-Hair, S.M., and Seelig,G. (2022) Machine learn-

ing for designing next-generation mRNA therapeutics,

Acc. Chem. Res., 55, 24-34, https://doi.org/10.1021/

acs.accounts.1c00621.

15. Cao, J., Novoa, E. M., Zhang, Z., Chen, W. C. W.,

Liu, D., Choi, G. C. G., Wong, A. S.L., Wehrspaun, C.,

Kellis,M., and Lu, T.K. (2021) High-throughput 5′ UTR

engineering for enhanced protein production in non-

viral gene therapies, Nat. Commun., 12, 4138, https://

doi.org/10.1038/s41467-021-24436-7.

16. Castillo-Hair, S., Fedak, S., Wang, B., Linder, J.,

Havens, K., Certo, M., and Seelig, G. (2024) Opti-

mizing 5’UTRs for mRNA-delivered gene editing us-

ing deep learning, Nat. Commun., 15, 5284, https://

doi.org/10.1038/S41467-024-49508-2.

17. Hajj, K. A., and Whitehead, K. A. (2017) Tools for

translation: non-viral materials for therapeutic

mRNA delivery, Nat. Rev. Mater., 2, 17056, https://

doi.org/10.1038/natrevmats.2017.56.

18. Kon, E., Elia, U., and Peer, D. (2022) Principles for

designing an optimal mRNA lipid nanoparticle vac-

cine, Curr. Opin. Biotechnol., 73, 329-336, https://

doi.org/10.1016/j.copbio.2021.09.016.

19. Park, K.S., Sun,X., Aikins, M.E., and Moon, J.J. (2021)

Non-viral COVID-19 vaccine delivery systems, Adv.

Drug Deliv. Rev., 169, 137-151, https://doi.org/10.1016/

j.addr.2020.12.008.

20. Lu, R. M., Hsu, H. E., Perez, S. J. L. P., Kumari, M.,

Chen, G.H., Hong, M.H., Lin, Y.S., Liu, C.H., Ko, S.H.,

Concio, C. A. P., Su, Y. J., Chang, Y. H., Li, W. S., and

Wu, H.C. (2024) Current landscape of mRNA technol-

ogies and delivery systems for new modality thera-

peutics, J.Biomed. Sci., 31, 89, https://doi.org/10.1186/

s12929-024-01080-z.

21. Wadhwa, A., Aljabbari, A., Lokras, A., Foged, C., and

Thakur,A. (2020) Opportunities and challenges in the

delivery of mRNA-based vaccines, Pharmaceutics, 12,

102, https://doi.org/10.3390/pharmaceutics12020102.

22. Jones, C. H., Androsavich, J.R., So, N., Jenkins, M. P.,

MacCormack,D., Prigodich, A., Welch,V., True, J. M.,

and Dolsten,M. (2024) Breaking the mold with RNA–

a “RNAissance” of life science, NPJ Genomic Med.,

9, 2, https://doi.org/10.1038/s41525-023-00387-4.

23. Baek, R., Coughlan, K., Jiang, L., Liang, M., Ci, L.,

Singh,H., Zhang,H., Kaushal,N., Rajlic, I.L., Van,L.,

Dimen,R., Cavedon,A., Yin,L., Rice, L., Frassetto,A.,

VOLKHIN et al.744

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

Guey, L., Finn, P., and Martini, P. G. V. (2024) Char-

acterizing the mechanism of action for mRNA ther-

apeutics for the treatment of propionic acidemia,

methylmalonic acidemia, and phenylketonuria, Nat.

Commun., 15, 3804, https://doi.org/10.1038/s41467-

024-47460-9.

24. Lundstrom, K. (2023) Viral vectors in gene therapy:

where do we stand in 2023? Viruses, 15, 698, https://

doi.org/10.3390/v15030698.

25. Kishimoto, T.K., and Samulski, R.J. (2022) Addressing

high dose AAV toxicity – “one and done” or “slower

and lower”? Expert Opin. Biol. Ther., 22, 1067-1071,

https://doi.org/10.1080/14712598.2022.2060737.

26. Zwi-Dantsis, L., Mohamed, S., Massaro, G., and

Moeendarbary, E. (2025) Adeno-associated virus

vectors: principles, practices, and prospects in gene

therapy, Viruses, 17, 239, https://doi.org/10.3390/

V17020239.

27. Wang, J. H., Gessler, D. J., Zhan, W., Gallagher, T. L.,

and Gao, G. (2024) Adeno-associated virus as a de-

livery vector for gene therapy of human diseases,

Signal Transduct. Target. Ther., 9, 78, https://doi.org/

10.1038/s41392-024-01780-w.

28. Li, C., and Samulski, R. J. (2020) Engineering ade-

no-associated virus vectors for gene therapy, Nat. Rev.

Genet., 21, 255-272, https://doi.org/10.1038/s41576-

019-0205-4.

29. Von der Haar, T., Mulroney, T. E., Hedayioglu, F.,

Kurusamy, S., Rust, M., Lilley, K. S., Thaventhiran,

J. E., Willis, A. E., and Smales, C. M. (2023) Transla-

tion of in vitro-transcribed RNA therapeutics, Front.

Mol. Biosci., 10, 1128067, https://doi.org/10.3389/

fmolb.2023.1128067.

30. Buchlis, G., Podsakoff, G. M., Radu, A., Hawk, S. M.,

Flake, A.W., Mingozzi,F., and High, K.A. (2012) Fac-

tor IX expression in skeletal muscle of a severe he-

mophilia B patient 10 years after AAV-mediated gene

transfer, Blood, 119, 3038-3041, https://doi.org/10.1182/

blood-2011-09-382317.

31. Fischer, M. D., Simonelli, F., Sahni, J., Holz, F. G.,

Maier,R., Fasser, C., Suhner, A., Stiehl, D.P., Chen,B.,

Audo,I., and Leroy, B.P. (2024) Real-world safety and

effectiveness of Voretigene Neparvovec: results up

to 2 years from the prospective, registry-based PER-

CEIVE study, Biomolecules, 14, 122, https://doi.org/

10.3390/biom14010122.

32. Strauss, K. A., Farrar, M. A., Muntoni, F., Saito, K.,

Mendell, J.R., Servais,L., McMillan, H.J., Finkel, R.S.,

Swoboda, K.J., Kwon, J.M., Zaidman, C.M., Chiriboga,

C. A., Iannaccone, S. T., Krueger, J. M., Parsons, J. A.,

Shieh, P. B., Kavanagh, S., Tauscher-Wisniewski, S.,

McGill, B.E., and Macek, T. A. (2022) Onasemnogene

abeparvovec for presymptomatic infants with two

copies of SMN2 at risk for spinal muscular atrophy

type 1: the phase III SPR1NT trial, Nat. Med., 28,

1381-1389, https://doi.org/10.1038/s41591-022-01866-4.

33. Ferreira, V., Twisk, J., Kwikkers, K., Aronica, E.,

Brisson, D., Methot, J., Petry, H., and Gaudet, D.

(2014) Immune responses to intramuscular admin-

istration of alipogene tiparvovec (AAV1-LPLS447X)

in a phase II Clinical trial of lipoprotein lipase de-

ficiency gene therapy, Hum. Gene Ther., 25, 180-188,

https://doi.org/10.1089/hum.2013.169.

34. Ozelo, M. C., Mahlangu, J., Pasi, K. J., Giermasz, A.,

Leavitt, A. D., Laffan, M., Symington, E., Quon, D. V.,

Wang, J.-D., Peerlinck, K., Pipe, S. W., Madan, B.,

Key, N.S., Pierce, G. F., O’Mahony,B., Kaczmarek, R.,

Henshaw, J., Lawal, A., Jayaram, K., Huang, M.,

Yang, X., Wong, W. Y., and Kim, B. (2022) Valoctoco-

gene roxaparvovec gene therapy for hemophilia A,

N.Engl. J. Med., 386, 1013-1025, https://doi.org/10.1056/

nejmoa2113708.

35. Sekayan, T., Simmons, D. H., and von Drygalski, A.

(2023) Etranacogene dezaparvovec-drlb gene therapy

for patients with hemophilia B (congenital factor IX

deficiency), Expert Opin. Biol. Ther., 23, 1173-1184,

https://doi.org/10.1080/14712598.2023.2282138.

36. Xu, S., Yang, K., Li, R., and Zhang, L. (2020) mRNA

vaccine era – mechanisms, drug platform and clin-

ical prospection, Int. J. Mol. Sci., 21, 6582, https://

doi.org/10.3390/ijms21186582.

37. Mu,X., and Hur,S. (2021) Immunogenicity of invitro-

transcribed RNA, Acc. Chem. Res., 54, 4012-4023,

https://doi.org/10.1021/acs.accounts.1c00521.

38. Gholamalipour, Y., Karunanayake Mudiyanselage, A.,

and Martin, C. T. (2018) 3′ end additions by T7 RNA

polymerase are RNA self-templated, distributive and

diverse in character– RNA-Seq analyses, Nucleic Ac-

ids Res., 46, 9253-9263, https://doi.org/10.1093/nar/

gky796.

39. Milligan, J. F., Groebe, D. R., Witherell, G. W., and

Uhlenbeck, O. C. (1987) Oligoribonucleotide synthe-

sis using T7 RNA polymerase and synthetic DNA

templates, Nucleic Acids Res., 15, 8783-8798, https://

doi.org/10.1093/nar/15.21.8783.

40. Pelletier,J., and Sonenberg, N. (2019) The organizing

principles of eukaryotic ribosome recruitment, Annu.

Rev. Biochem., 88, 307-335, https://doi.org/10.1146/

annurev-biochem-013118-111042.

41. To, K. K. W., and Cho, W. C. S. (2021) An overview

of rational design of mRNA-based therapeutics and

vaccines, Expert Opin. Drug Discov., 16, 1307-1317,

https://doi.org/10.1080/17460441.2021.1935859.

42. Shanmugasundaram, M., Charles, I., and Kore, A. R.

(2016) Design, synthesis and biological evaluation

of dinucleotide mRNA cap analog containing prop-

argyl moiety, Bioorganic Med. Chem., 24, 1204-1208,

https://doi.org/10.1016/j.bmc.2016.01.048.

43. Kore, A. R., Shanmugasundaram, M., Charles, I.,

Vlassov, A. V., and Barta, T. J. (2009) Locked nucleic

acid (LNA)-Modified dinucleotide mRNA cap an-

alogue: synthesis, enzymatic incorporation, and

SELECTION OF UTRs 745

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

utilization, J.Am. Chem. Soc., 131, 6364-6365, https://

doi.org/10.1021/ja901655p.

44. Zeng, C., Hou, X., Yan, J., Zhang, C., Li, W., Zhao, W.,

Du,S., and Dong,Y. (2020) Leveraging mRNA sequenc-

es and nanoparticles to deliver SARS-CoV-2 antigens

in vivo, Adv. Mater., 32, e2004452, https://doi.org/

10.1002/adma.202004452.

45. Mockey,M., Gonçalves,C., Dupuy, F.P., Lemoine, F.M.,

Pichon, C., and Midoux, P. (2006) mRNA transfection

of dendritic cells: synergistic effect of ARCA mRNA

capping with Poly(A) chains in cis and in trans for

a high protein expression level, Biochem. Biophys.

Res. Commun., 340, 1062-1068, https://doi.org/10.1016/

j.bbrc.2005.12.105.

46. Strenkowska, M., Kowalska, J., Lukaszewicz, M.,

Zuberek, J., Su, W., Rhoads, R. E., Darzynkiewicz, E.,

and Jemielity, J. (2010) Towards mRNA with superi-

or translational activity: synthesis and properties of

ARCA tetraphosphates with single phosphorothioate

modifications, New J. Chem., 34, 993-1007, https://

doi.org/10.1039/b9nj00644c.

47. Grudzien, E., Stepinski, J., Jankowska-Anyszka, M.,

Stolarski, R., Darzynkiewicz, E., and Rhoads, R. E.

(2004) Novel cap analogs for in vitro synthesis of

mRNAs with high translational efficiency, RNA, 10,

1479-1487, https://doi.org/10.1261/rna.7380904.

48. Grudzien-Nogalska, E., Jemielity, J., Kowalska, J.,

Darzynkiewicz,E., and Rhoads, R.E. (2007) Phospho-

rothioate cap analogs stabilize mRNA and increase

translational efficiency in mammalian cells, RNA, 13,

1745-1755, https://doi.org/10.1261/rna.701307.

49. Kuhn, A. N., Diken, M., Kreiter, S., Selmi, A.,

Kowalska,J., Jemielity,J., Darzynkiewicz,E., Huber,C.,

Türeci, Ö., and Sahin, U. (2010) Phosphorothioate

cap analogs increase stability and translational effi-

ciency of RNA vaccines in immature dendritic cells

and induce superior immune responses in vivo,

Gene Ther., 17, 961-971, https://doi.org/10.1038/

gt.2010.52.

50. Peng, J., Murray, E. L., and Schoenberg, D. R. (2008)

In vivo and in vitro analysis of poly(a) length effects

on mRNA translation, Methods Mol. Biol., 419, 215-

230, https://doi.org/10.1007/978-1-59745-033-1_15.

51. Chang, H., Lim, J., Ha, M., and Kim, V. N. (2014)

TAIL-seq: Genome-wide determination of poly(A) tail

length and 3’ end modifications, Mol. Cell, 53, 1044-

1052, https://doi.org/10.1016/j.molcel.2014.02.007.

52. Holtkamp, S., Kreiter, S., Selmi, A., Simon, P.,

Koslowski, M., Huber, C., Türeci, Ö., and Sahin, U.

(2006) Modification of antigen-encoding RNA increas-

es stability, translational efficacy, and T-cell stimula-

tory capacity of dendritic cells, Blood, 108, 4009-4017,

https://doi.org/10.1182/blood-2006-04-015024.

53. Karikó, K., Buckstein, M., Ni, H., and Weissman, D.

(2005) Suppression of RNA recognition by toll-like

receptors: the impact of nucleoside modification

and the evolutionary origin of RNA, Immunity, 23,

165-175, https://doi.org/10.1016/j.immuni.2005.06.008.

54. Karikó, K., Muramatsu, H., Welsh, F. A., Ludwig, J.,

Kato,H., Akira, S., and Weissman, D. (2008) Incorpo-

ration of pseudouridine into mRNA yields superior

nonimmunogenic vector with increased translational

capacity and biological stability, Mol. Ther., 16, 1833-

1840, https://doi.org/10.1038/mt.2008.200.

55. Kozak, M. (1986) Point mutations define a sequence

flanking the AUG initiator codon that modulates

translation by eukaryotic ribosomes, Cell, 44, 283-292,

https://doi.org/10.1016/0092-8674(86)90762-2.

56. Kozak, M. (1997) Recognition of AUG and alterna-

tive initiator codons is augmented by G in position

+4 but is not generally affected by the nucleotides in

positions +5 and +6, EMBO J., 16, 2482-2492, https://

doi.org/10.1093/emboj/16.9.2482.

57. Hernández,G., Osnaya, V. G., and Pérez-Martínez, X.

(2019) Conservation and variability of the AUG ini-

tiation codon context in eukaryotes, Trends Bio-

chem. Sci., 44, 1009-1021, https://doi.org/10.1016/

j.tibs.2019.07.001.

58. Hernández, G., García, A., Weingarten-Gabbay, S.,

Mishra, R. K., Hussain, T., Amiri, M., Moreno-

Hagelsieb, G., Montiel-Dávalos, A., Lasko, P., and

Sonenberg, N. (2024) Functional analysis of the AUG

initiator codon context reveals novel conserved

sequences that disfavor mRNA translation in eu-

karyotes, Nucleic Acids Res., 52, 1064-1079, https://

doi.org/10.1093/nar/gkad1152.

59. Gallie, D.R. (1991) The cap and poly(A) tail function

synergistically to regulate mRNA translational efficien-

cy, Genes Dev., 5, 2108-2116, https://doi.org/10.1101/

gad.5.11.2108.

60. Tanguay, R. L., and Gallie, D. R. (1996) Translational

Efficiency Is Regulated by the Length of the 3′ Un-

translated Region, Mol. Cell. Biol., 16, 146-156, https://

doi.org/10.1128/mcb.16.1.146.

61. Aditham,A., Shi,H., Guo,J., Zeng,H., Zhou,Y., Wade,

S. D., Huang, J., Liu, J., and Wang, X. (2022) Chemi-

cally modified mocRNAs for highly efficient protein

expression in mammalian cells, ACS Chem. Biol.,

17, 3352-3366, https://doi.org/10.1021/acschembio.

1c00569.

62. Chen,H., Liu,D., Guo,J., Aditham,A., Zhou,Y., Tian,J.,

Luo,S., Ren, J., Hsu,A., Huang, J., Kostas,F., Wu, M.,

Liu, D. R., and Wang, X. (2024) Branched chemical-

ly modified poly(A) tails enhance the translation ca-

pacity of mRNA, Nat. Biotechnol., 43, 194-203, https://

doi.org/10.1038/s41587-024-02174-7.

63. Asrani, K. H., Farelli, J. D., Stahley, M. R., Miller,

R. L., Cheng, C. J., Subramanian, R. R., and Brown,

J. M. (2018) Optimization of mRNA untranslated

regions for improved expression of therapeutic

mRNA, RNA Biol., 15, 756-762, https://doi.org/10.1080/

15476286.2018.1450054.

VOLKHIN et al.746

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

64. Pardi, N., Hogan, M. J., and Weissman, D. (2020) Re-

cent advances in mRNA vaccine technology, Curr.

Opin. Immunol., 65, 14-20, https://doi.org/10.1016/

j.coi.2020.01.008.

65. Schwanhüusser, B., Busse, D., Li, N., Dittmar, G.,

Schuchhardt, J., Wolf, J., Chen, W., and Selbach, M.

(2011) Global quantification of mammalian gene

expression control, Nature, 473, 337-342, https://

doi.org/10.1038/nature10098.

66. Mayr,C. (2017) Regulation by 3′-untranslated regions,

Annu. Rev. Genet., 51, 171-194, https://doi.org/10.1146/

annurev-genet-120116-024704.

67. Leppek, K., Byeon, G. W., Kladwang, W., Wayment-

Steele, H.K., Kerr, C. H., Xu, A.F., Kim, D.S., Topkar,

V. V., Choe, C., Rothschild, D., Tiu, G. C., Wellington-

Oguri, R., Fujii,K., Sharma, E., Watkins, A. M., Nicol,

J. J., Romano, J., Tunguz, B., Diaz, F., Cai, H., Guo, P.,

Wu, J., Meng, F., Shi, S., Participants, E., Dormitzer,

P.R., Solórzano,A., Barna,M., and Das,R. (2022) Com-

binatorial optimization of mRNA structure, stability,

and translation for RNA-based therapeutics, Nat.

Commun., 13, 1536, https://doi.org/10.1038/s41467-

022-28776-w.

68. Babendure, J. R., Babendure, J. L., Ding, J. H., and

Tsien, R. Y. (2006) Control of mammalian transla-

tion by mRNA structure near caps, RNA, 12, 851-861,

https://doi.org/10.1261/rna.2309906.

69. Kozak, M. (1989) Circumstances and mechanisms of

inhibition of translation by secondary structure in eu-

caryotic mRNAs, Mol. Cell. Biol., 9, 5134-5142, https://

doi.org/10.1128/mcb.9.11.5134-5142.1989.

70. Edgil, D., Polacek, C., and Harris, E. (2006) Dengue

virus utilizes a novel strategy for translation initi-

ation when cap-dependent translation is inhibited,

J. Virol., 80, 2976-2986, https://doi.org/10.1128/jvi.

80.6.2976-2986.2006.

71. García-Montalvo, B. M., Medina, F., and Del Angel,

R. M. (2004) La protein binds to NS5 and NS3 and

to the 5′ and 3′ ends of Dengue 4 virus RNA, Virus

Res., 102, 141-150, https://doi.org/10.1016/j.virus-

res.2004.01.024.

72. Wayment-Steele, H. K., Kim, D. S., Choe, C. A., Nicol,

J. J., Wellington-Oguri, R., Watkins, A. M., Sperberg,

R. A. P., Huang, P.-S., Participants, E., and Das, R.

(2020) Theoretical basis for stabilizing messenger

RNA through secondary structure design, bioRxiv,

https://doi.org/10.1101/2020.08.22.262931.

73. Leppek, K., Das, R., and Barna, M. (2018) Function-

al 5′ UTR mRNA structures in eukaryotic translation

regulation and how to find them, Nat. Rev. Mol.

Cell Biol., 19, 158-174, https://doi.org/10.1038/nrm.

2017.103.

74. Huang,X., and Du, Z. (2024) Possible involvement of

three-stemmed pseudoknots in regulating translation-

al initiation in human mRNAs, PLoS One, 19, e0307541,

https://doi.org/10.1371/journal.pone.0307541.

75. Singha Roy, A., Majumder, S., and Saha, P. (2023)

Stable RNA G-quadruplex in the 5′-UTR of human

cIAP1 mRNA promotes translation in an IRES-inde-

pendent manner, Biochemistry, 63, 475-486, https://

doi.org/10.1021/acs.biochem.3c00521.

76. Iacono, M., Mignone, F., and Pesole, G. (2005) uAUG

and uORFs in human and rodent 5′-untranslated

mRNAs, Gene, 349, 97-105, https://doi.org/10.1016/

j.gene.2004.11.041.

77. Wethmar, K. (2014) The regulatory potential of up-

stream open reading frames in eukaryotic gene ex-

pression, Wiley Interdiscip. Rev. RNA, 5, 765-778,

https://doi.org/10.1002/wrna.1245.

78. Dever, T.E., Ivanov, I.P., and Hinnebusch, A.G. (2023)

Translational regulation by uORFs and start codon

selection stringency, Genes Dev., 37, 474-489, https://

doi.org/10.1101/gad.350752.123.

79. Manske, F., Ogoniak, L., Jürgens, L., Grundmann, N.,

Makałowski, W., and Wethmar, K. (2023) The new

uORFdb: integrating literature, sequence, and varia-

tion data in a central hub for uORF research, Nucleic

Acids Res., 51, 1029-1042, https://doi.org/10.1093/nar/

gkac899.

80. Orr, M. W., Mao, Y., Storz, G., and Qian, S. B. (2021)

Alternative ORFs and small ORFs: Shedding light on

the dark proteome, Nucleic Acids Res., 48, 1029-1042,

https://doi.org/10.1093/NAR/GKZ734.

81. Hinnebusch, A. G., Ivanov, I. P., and Sonenberg, N.

(2016) Translational control by 5′-untranslated re-

gions of eukaryotic mRNAs, Science, 352, 1413-1416,

https://doi.org/10.1126/science.aad9868.

82. Araujo, P. R., Yoon, K., Ko, D., Smith, A. D., Qiao, M.,

Suresh, U., Burns, S. C., and Penalva, L. O. F. (2012)

Before it gets started: Regulating translation at the

5′UTR, Comp. Funct. Genomics, 2012, 475731, https://

doi.org/10.1155/2012/475731.

83. Bottorff, T. A., Park, H., Geballe, A. P., and

Subramaniam, A. R. (2022) Translational buffering

by ribosome stalling in upstream open reading

frames, PLoS Genet., 18, e1010460, https://doi.org/

10.1371/journal.pgen.1010460.

84. Matsui, M., Yachie, N., Okada, Y., Saito, R., and

Tomita,M. (2007) Bioinformatic analysis of post-tran-

scriptional regulation by uORF in human and mouse,

FEBS Lett., 581, 4184-4188, https://doi.org/10.1016/

j.febslet.2007.07.057.

85. Russell, P. J., Slivka, J. A., Boyle, E. P., Burghes,

A.H.M., and Kearse, M.G. (2023) Translation reiniti-

ation after uORFs does not fully protect mRNAs from

nonsense-mediated decay, RNA, 29, 735-744, https://

doi.org/10.1261/rna.079525.122.

86. Brogna, S., and Wen, J. (2009) Nonsense-mediated

mRNA decay (NMD) mechanisms, Nat. Struct. Mol.

Biol., 16, 107-113, https://doi.org/10.1038/nsmb.1550.

87. Oliveira, C.C., and McCarthy, J.E.G. (1995) The rela-

tionship between eukaryotic translation and mRNA

SELECTION OF UTRs 747

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

stability. A short upstream open reading frame

strongly inhibits translational initiation and greatly

accelerates mRNA degradation in the yeast Saccha-

romyces cerevisiae, J.Biol. Chem., 270, 8936-8943,

https://doi.org/10.1074/jbc.270.15.8936.

88. Ruiz-Echevarría, M. J., and Peltz, S. W. (2000) The

RNA binding protein Pub1 modulates the stabili-

ty of transcripts containing upstream open reading

frames, Cell, 101, 741-751, https://doi.org/10.1016/

S0092-8674(00)80886-7.

89. Kim, S. C., Sekhon, S. S., Shin, W. R., Ahn, G., Cho,

B. K., Ahn, J. Y., and Kim, Y. H. (2022) Modifications

of mRNA vaccine structural elements for improving

mRNA stability and translation efficiency, Mol. Cell.

Toxicol., 18, 1-8, https://doi.org/10.1007/s13273-021-

00171-4.

90. Carrieri, C., Cimatti, L., Biagioli, M., Beugnet, A.,

Zucchelli,S., Fedele,S., Pesce,E., Ferrer,I., Collavin,L.,

Santoro, C., Forrest, A. R. R., Carninci, P., Biffo, S.,

Stupka, E., and Gustincich, S. (2012) Long non-cod-

ing antisense RNA controls Uchl1 translation through

an embedded SINEB2 repeat, Nature, 491, 454-457,

https://doi.org/10.1038/nature11508.

91. Muckenthaler, M. U., Galy, B., and Hentze, M. W.

(2008) Systemic iron homeostasis and the iron-re-

sponsive element/iron-regulatory protein (IRE/

IRP) regulatory network, Annu. Rev. Nutr., 28, 197-

213, https://doi.org/10.1146/annurev.nutr.28.061807.

155521.

92. Medenbach, J., Seiler, M., and Hentze, M. W. (2011)

Translational control via protein-regulated upstream

open reading frames, Cell, 145, 902-913, https://

doi.org/10.1016/j.cell.2011.05.005.

93. Evdokimova, V., Tognon, C., Ng, T., Ruzanov, P.,

Melnyk, N., Fink, D., Sorokin, A., Ovchinnikov, L. P.,

Davicioni, E., Triche, T. J., and Sorensen, P. H. B.

(2009) Translational activation of Snail1 and other

developmentally regulated transcription factors by

YB-1 promotes an epithelial-mesenchymal transi-

tion, Cancer Cell, 15, 402-415, https://doi.org/10.1016/

j.ccr.2009.03.017.

94. Mayya, V. K., and Duchaine, T. F. (2019) Ciphers and

executioners: how 3 0 -untranslated regions deter-

mine the fate of messenger RNAs, Front. Genet., 10,

6, https://doi.org/10.3389/fgene.2019.00006.

95. Zekri, L., Kuzuoǧlu-Öztürk, D., and Izaurralde, E.

(2013) GW182 proteins cause PABP dissociation from

silenced miRNA targets in the absence of deadenyla-

tion, EMBO J., 32, 1052-1065, https://doi.org/10.1038/

emboj.2013.44.

96. Kamenska,A., Simpson,C., Vindry,C., Broomhead,H.,

Bénard, M., Ernoult-Lange, M., Lee, B. P., Harries,

L.W., Weil,D., and Standart,N. (2016) The DDX6-4E-T

interaction mediates translational repression and

P-body assembly, Nucleic Acids Res., 44, 6318-6334,

https://doi.org/10.1093/nar/gkw565.

97. Lai, W.S., Carballo,E., Thorn, J.M., Kennington, E.A.,

and Blackshear, P.J. (2000) Interactions of CCCH Zinc

finger proteins with mRNA, J.Biol. Chem., 275, 17827-

17837, https://doi.org/10.1074/jbc.m001696200.

98. Brennan, C. M., and Steitz, J. A. (2001) HuR and

mRNA stability, Cell. Mol. Life Sci., 58, 266-277, https://

doi.org/10.1007/PL00000854.

99. Gong, C., and Maquat, L. E. (2011) LncRNAs trans-

activate STAU1-mediated mRNA decay by duplexing

with 39 UTRs via Alu eleme, Nature, 470, 329-333,

https://doi.org/10.1038/nature09701.

100. O’Brien, J., Hayder, H., Zayed,Y., and Peng, C. (2018)

Overview of microRNA biogenesis, mechanisms of

actions, and circulation, Front. Endocrinol., 9, 402,

https://doi.org/10.3389/fendo.2018.00402.

101. Liegro, C. M. D., Bellafiore, M., Izquierdo, J. M.,

Rantanen,A., and Cuezva, J.M. (2000) 3′-Untranslated

regions of oxidative phosphorylation mRNAs function

in vivo as enhancers of translation, Biochem. J., 352,

109-115, https://doi.org/10.1042/bj3520109.

102. Popa, I., Harris, M. E., Donello, J. E., and Hope, T. J.

(2002) CRM1-dependent function of a cis-acting RNA

export element, Mol. Cell. Biol., 22, 2057-2067, https://

doi.org/10.1128/mcb.22.7.2057-2067.2002.

103. Loeb, J. E., Cordier, W. S., Harris, M. E., Weitzman,

M. D., and Hope, T. J. (1999) Enhanced expression

of transgenes from adeno-associated virus vectors

with the woodchuck hepatitis virus posttranscrip-

tional regulatory element: implications for gene

therapy, Hum. Gene Ther., 10, 2295-2305, https://

doi.org/10.1089/10430349950016942.

104. Lianoglou, S., Garg, V., Yang, J. L., Leslie, C. S., and

Mayr, C. (2013) Ubiquitously transcribed genes use

alternative polyadenylation to achieve tissue-spe-

cific expression, Genes Dev., 27, 2380-2396, https://

doi.org/10.1101/gad.229328.113.

105. Xie, J., Xie, Q., Zhang, H., Ameres, S. L., Hung, J. H.,

Su, Q., He, R., Mu, X., Seher Ahmed, S., Park, S.,

Kato, H., Li, C., Mueller, C., Mello, C. C., Weng, Z.,

Flotte, T. R., Zamore, P. D., and Gao, G. (2011)

MicroRNA- regulated, systemically delivered rAAV9:

A step closer to CNS-restricted transgene expression,

Mol. Ther., 19, 526-535, https://doi.org/10.1038/mt.

2010.279.

106. Qiao, C., Yuan, Z., Li, J., He, B., Zheng, H., Mayer, C.,

Li, J., and Xiao, X. (2011) Liver-specific microR-

NA-122 target sequences incorporated in AAV vec-

tors efficiently inhibits transgene expression in the

liver, Gene Ther., 18, 403-410, https://doi.org/10.1038/

gt.2010.157.

107. Karali, M., Manfredi, A., Puppo, A., Marrocco, E.,

Gargiulo, A., Allocca, M., de Corte, M., Rossi, S.,

Giunti, M., Bacci, M. L., Simonelli, F., Surace, E. M.,

Banfi,S., and Auricchio,A. (2011) MicroRNA-Restrict-

ed transgene expression in the retina, PLoS One, 6,

e22166, https://doi.org/10.1371/journal.pone.0022166.

VOLKHIN et al.748

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

108. Geisler,A., and Fechner,H. (2016) MicroRNA-regulated

viral vectors for gene therapy, World J. Exp. Med., 6,

37-54, https://doi.org/10.5493/wjem.v6.i2.37.

109. Ludwig, N., Leidinger, P., Becker, K., Backes, C., Fe-

hlmann, T., Pallasch, C., Rheinheimer, S., Meder, B.,

Stähler, C., Meese, E., and Keller, A. (2016) Distribu-

tion of miRNA expression across human tissues, Nu-

cleic Acids Res., 44, 3865-3877, https://doi.org/10.1093/

nar/gkw116.

110. Kelly, E. J., Nace, R., Barber, G. N., and Russell, S. J.

(2010) Attenuation of vesicular stomatitis virus en-

cephalitis through microRNA targeting, J. Virol., 84,

1550-1562, https://doi.org/10.1128/jvi.01788-09.

111. Barnes,D., Kunitomi,M., Vignuzzi,M., Saksela,K., and

Andino,R. (2008) Harnessing endogenous miRNAs to

control virus tissue tropism as a strategy for devel-

oping attenuated virus vaccines, Cell Host Microbe,

4, 239-248, https://doi.org/10.1016/j.chom.2008.08.003.

112. Yen, L.C., Lin, Y.L., Sung, H.H., Liao, J.T., Tsao, C.H.,

Su, C. M., Lin, C. K., and Liao, C. L. (2013) Neurovir-

ulent flavivirus can be attenuated in mice by incor-

poration of neuron-specific microRNA recognition

elements into viral genome, Vaccine, 31, 5915-5922,

https://doi.org/10.1016/j.vaccine.2011.09.102.

113. Turner, J. D., and Muller, C.P. (2005) Structure of the

glucocorticoid receptor (NR3C1) gene 5′ untranslat-

ed region: identification, and tissue distribution of

multiple new human exon 1, J. Mol. Endocrinol., 35,

283-292, https://doi.org/10.1677/jme.1.01822.

114. Hirahara, Y., Tsuda, M., Wada, Y., and Honke, K.

(2000) cDNA cloning, genomic cloning, and tis-

sue-specific regulation of mouse cerebroside sulfo-

transferase, Eur. J. Biochem., 267, 1909-1917, https://

doi.org/10.1046/j.1432-1327.2000.01139.x.

115. Wortman, M. J., Hanson, L. K., Martínez-Sobrido, L.,

Campbell, A. E., Nance, J. A., García-Sastre, A., and

Johnson, E. M. (2010) Regulation of PURA gene tran-

scription by three promoters generating distinctly

spliced 5-prime leaders: a novel means of fine con-

trol over tissue specificity and viral signals, BMC Mol.

Biol., 11, 81, https://doi.org/10.1186/1471-2199-11-81.

116. Roberts, S. J., Chung, K.-N., Nachmanoff, K., and

Elwood, P. C. (1997) Tissue-specific promoters of the

α human folate receptor gene yield transcripts with

divergent 5′ leader sequences and different transla-

tional efficiencies, Biochem. J., 326, 439-447, https://

doi.org/10.1042/bj3260439.

117. Xiao, Y., Muhuri, M., Li, S., Qin, W., Xu, G., Luo, L.,

Li,J., Letizia, A.J., Wang, S. K., Chan, Y. K., Wang, C.,

Fuchs, S.P., Wang,D., Su,Q., Abu Nahid,M., Church,

G. M., Farzan, M., Yang, L., Wei, Y., Desrosiers, R. C.,

Mueller,C., Tai, P.W.L., and Gao,G. (2019) Circumvent-

ing cellular immunity by miR142-mediated regulation

sufficiently supports rAAV-delivered OVA expression

without activating humoral immunity, JCI Insight,

4, e99052, https://doi.org/10.1172/jci.insight.99052.

118. Majowicz, A., Maczuga, P., Kwikkers, K. L., van der

Marel,S., van Logtenstein,R., Petry,H., van Deventer,

S.J., Konstantinova,P., and Ferreira,V. (2013) Mir-142-

3p target sequences reduce transgene-directed immu-

nogenicity following intramuscular adeno-associated

virus 1 vector-mediated gene delivery, J. Gene Med.,

15, 219-232, https://doi.org/10.1002/jgm.2712.

119. Muhuri, M., Zhan, W., Maeda, Y., Li, J., Lotun, A.,

Chen, J., Sylvia, K., Dasgupta, I., Arjomandnejad, M.,

Nixon,T., Keeler, A.M., Manokaran,S., He,R., Su,Q.,

Tai, P. W. L., and Gao, G. (2021) Novel combinatori-

al microRNA-binding sites in AAV vectors synergis-

tically diminish antigen presentation and transgene

immunity for efficient and stable transduction,

Front. Immunol., 12, 674242, https://doi.org/10.3389/

fimmu.2021.674242.

120. Khoroshkin,M., Zinkevich,A., Aristova,E., Yousefi,H.,

and Sean, B. (2024) A generative framework for en-

hanced cell-type specificity in rationally designed

mRNAs, bioRxiv, https://doi.org/10.1101/2024.12.

31.630783.

121. Flamand, M.N., Tegowski,M., and Meyer, K.D. (2023)

The proteins of mRNA modification: writers, readers,

and erasers, Annu. Rev. Biochem., 92, 145-173, https://

doi.org/10.1146/annurev-biochem-052521-035330.

122. Khan, F. A., Nsengimana, B., Awan, U. A., Ji, X. Y.,

Ji,S., and Dong,J. (2024) Regulatory roles of N

-meth-

yladenosine (m

A) methylation in RNA process-

ing and non-communicable diseases, Cancer Gene

Ther., 31, 1439-1453, https://doi.org/10.1038/S41417-

024-00789-1.

123. Yu,X., Zhao,H., Wang,R., Chen,Y., Ouyang,X., Li,W.,

Sun,Y., and Peng,A. (2024) Cancer epigenetics: from

laboratory studies and clinical trials to precision med-

icine, Cell Death Discov., 10, 28, https://doi.org/10.1038/

s41420-024-01803-z.

124. Wilinski,D., and Dus,M. (2023) N 6-adenosine meth-

ylation controls the translation of insulin mRNA,

Nat. Struct. Mol. Biol., 30, 1260-1264, https://doi.org/

10.1038/s41594-023-01048-x.

125. Morais, P., Adachi, H., and Yu, Y.-T. (2021) The criti-

cal contribution of pseudouridine to mRNA COVID-19

vaccines, Front. Cell Dev. Biol., 9, 789427, https://

doi.org/10.3389/fcell.2021.789427.

126. Song,B., Tang,Y., Wei,Z., Liu,G., Su,J., Meng,J., and

Chen, K. (2020) PIANO: a web server for pseudou-

ridine-Site (Ψ) identification and functional anno-

tation, Front. Genet., 11, 88, https://doi.org/10.3389/

fgene.2020.00088.

127. Kierzek,E., Malgowska,M., Lisowiec,J., Turner, D.H.,

Gdaniec, Z., and Kierzek, R. (2014) The contribution

of pseudouridine to stabilities and structure of RNAs,

Nucleic Acids Res., 42, 3492-3501, https://doi.org/

10.1093/nar/gkt1330.

128. Karikó, K., Muramatsu, H., Keller, J. M., and Weiss-

man, D. (2012) Increased erythropoiesis in mice

SELECTION OF UTRs 749

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

injected with submicrogram quantities of pseudou-

ridine-containing mRNA encoding erythropoie-

tin, Mol. Ther., 20, 948-953, https://doi.org/10.1038/

mt.2012.7.

129. Chen, S., Liu, Z., Cai, J., Li, H., and Qiu, M. (2024)

N1-methylpseudouridine modification level cor-

relates with protein expression, immunogenicity,

and stability of mRNA, MedComm, 5, e691, https://

doi.org/10.1002/MCO2.691.

130. Sample, P. J., Wang, B., Reid, D. W., Presnyak, V.,

McFadyen, I. J., Morris, D. R., and Seelig, G. (2019)

Human 5′ UTR design and variant effect prediction

from a massively parallel translation assay, Nat. Bio-

technol., 37, 803-809, https://doi.org/10.1038/s41587-

019-0164-5.

131. Chang,Y., Jin,H., Cui,Y., Yang,F., Chen,K., Kuang,W.,

Huo,C., Xu,Z., Li,Y., Lin,A., Yang,B., Liu,W., Xie,S.,

and Zhou, T. (2024) PUS7-dependent pseudouridyla-

tion of ALKBH3 mRNA inhibits gastric cancer pro-

gression, Clin. Transl. Med., 14, e1811, https://doi.org/

10.1002/CTM2.1811.

132. Monroe,J., Eyler, D.E., Mitchell,L., Deb,I., Bojanows-

ki,A., Srinivas,P., Dunham, C.M., Roy,B., Frank, A.T.,

and Koutmou, K. S. (2024) N1-Methylpseudouridine

and pseudouridine modifications modulate mRNA

decoding during translation, Nat. Commun., 15, 8119,

https://doi.org/10.1038/s41467-024-51301-0.

133. Moradian, H., Roch, T., Anthofer, L., Lendlein, A.,

and Gossen, M. (2022) Chemical modification of uri-

dine modulates mRNA-mediated proinflammatory

and antiviral response in primary human macro-

phages, Mol. Ther. Nucleic Acids, 27, 854-869, https://

doi.org/10.1016/j.omtn.2022.01.004.

134. Li, X., Xiong, X., Zhang, M., Wang, K., Chen, Y.,

Zhou, J., Mao, Y., Lv, J., Yi, D., Chen, X.W., Wang, C.,

Qian, S. B., and Yi, C. (2017) Base-resolution map-

ping reveals distinct m1A methylome in nucle-

ar- and mitochondrial-encoded transcripts, Mol.

Cell, 68, 993-1005.e9, https://doi.org/10.1016/j.molcel.

2017.10.019.

135. Zhou, H., Kimsey, I. J., Nikolova, E. N., Sathyamoor-

thy, B., Grazioli, G., McSally, J., Bai, T., Wunderlich,

C.H., Kreutz,C., Andricioaei,I., and Al-Hashimi, H.M.

(2016) M1A and m1G disrupt A-RNA structure through

the intrinsic instability of Hoogsteen base pairs,

Nat. Struct. Mol. Biol., 23, 803-810, https://doi.org/

10.1038/nsmb.3270.

136. Qi, Z., Zhang, C., Jian, H., Hou, M., Lou, Y., Kang, Y.,

Wang, W., Lv, Y., Shang, S., Wang, C., Li, X., Feng,S.,

and Zhou,H. (2023) N1-Methyladenosine modification

of mRNA regulates neuronal gene expression and ox-

ygen glucose deprivation/reoxygenation induction,

Cell Death Discov., 9, 159, https://doi.org/10.1038/

s41420-023-01458-2.

137. Zhang, C., Yi, X., Hou, M., Li, Q., Li, X., Lu, L., Qi, E.,

Wu,M., Qi, L., Jian,H., Qi, Z., Lv, Y., Kong,X., Bi, M.,

Feng, S., and Zhou, H. (2023) The landscape of m1A

modification and its posttranscriptional regulato-

ry functions in primary neurons, Elife, 12, e85324,

https://doi.org/10.7554/elife.85324.

138. Boo, S.H., Ha,H., and Kim, Y.K. (2022) m1A and m

modifications function cooperatively to facilitate rap-

id mRNA degradation, Cell Rep., 40, 111317, https://

doi.org/10.1016/j.celrep.2022.111317.

139. Park, O.H., Ha,H., Lee,Y., Boo, S.H., Kwon, D.H., Song,

H.K., and Kim, Y.K. (2019) Endoribonucleolytic cleav-

age of m

A-containing RNAs by RNase P/MRP com-

plex, Mol. Cell, 74, 494-507.e8, https://doi.org/10.1016/

j.molcel.2019.02.034.

140. Lee,Y., Choe,J., Park, O.H., and Kim, Y.K. (2020) Mo-

lecular mechanisms driving mRNA degradation by

A modification, Trends Genet., 36, 177-188, https://

doi.org/10.1016/j.tig.2019.12.007.

141. Liu, N., Dai, Q., Zheng, G., He, C., Parisien, M., and

Pan, T. (2015) N6-methyladenosine-dependent RNA

structural switches regulate RNA-protein interac-

tions, Nature, 518, 560-564, https://doi.org/10.1038/

nature14234.

142. Guca, E., Alarcon, R., Palo, M. Z., Santos, L., Alonso-

Gil,S., Davyt,M., de Lima, L.H.F., Boissier,F., Das,S.,

Zagrovic,B., Puglisi, J.D., Hashem,Y., and Ignatova,Z.

(2024) N6-methyladenosine in 5′ UTR does not pro-

mote translation initiation, Mol. Cell, 84, 584-595.e6,

https://doi.org/10.1016/j.molcel.2023.12.028.

143. Meyer, K. D. (2019) m

A-mediated translation reg-

ulation, Biochim. Biophys. Acta Gene Regul. Mech.,

1862, 301-309, https://doi.org/10.1016/j.bbagrm.

2018.10.006.

144. Zhang, S., Liu, Y., Ma, X., Gao, X., Ru, Y., Hu, X., and

Gu, X. (2024) Recent advances in the potential role

of RNA N4-acetylcytidine in cancer progression,

Cell Commun. Signal., 22, 49, https://doi.org/10.1186/

s12964-023-01417-5.

145. Dominissini, D., and Rechavi, G. (2018) N4-acetyl-

ation of cytidine in mRNA by NAT10 regulates sta-

bility and translation, Cell, 175, 1725-1727, https://

doi.org/10.1016/j.cell.2018.11.037.

146. Liu, Y., Wang, X., Liu, Y., Yang, J., Mao, W., Feng, C.,

Wu,X., Chen,X., Chen,L., and Dong,P. (2023) N4-ace-

tylcytidine-dependent GLMP mRNA stabilization

by NAT10 promotes head and neck squamous cell

carcinoma metastasis and remodels tumor micro-

environment through MAPK/ERK signaling path-

way, Cell Death Dis., 14, 712, https://doi.org/10.1038/

s41419-023-06245-6.

147. Arango, D., Sturgill, D., Alhusaini, N., Dillman, A. A.,

Sweet, T.J., Hanson,G., Hosogane,M., Sinclair, W.R.,

Nanan, K. K., Mandler, M. D., Fox, S. D., Zengeya,

T.T., Andresson,T., Meier, J.L., Coller,J., and Oberdo-

erffer,S. (2018) Acetylation of cytidine in mRNA pro-

motes translation efficiency, Cell, 175, 1872-1886.e24,

https://doi.org/10.1016/j.cell.2018.10.030.

VOLKHIN et al.750

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

148. Arango, D., Sturgill, D., Yang, R., Kanai, T., Bauer, P.,

Roy, J., Wang, Z., Hosogane, M., Schiffers, S., and

Oberdoerffer, S. (2022) Erratum: direct epitranscrip-

tomic regulation of mammalian translation initiation

through N4-acetylcytidine, Mol. Cell, 82, 2797-2814.

e11, https://doi.org/10.1016/j.molcel.2022.06.022.

149. Schumann, U., Zhang, H. N., Sibbritt, T., Pan, A.,

Horvath, A., Gross, S., Clark, S. J., Yang, L., and Pre-

iss,T. (2020) Multiple links between 5-methylcytosine

content of mRNA and translation, BMC Biol., 18, 40,

https://doi.org/10.1186/s12915-020-00769-5.

150. Malbec,L., Zhang,T., Chen, Y.S., Zhang,Y., Sun, B.F.,

Shi, B.Y., Zhao, Y. L., Yang, Y., and Yang, Y. G. (2019)

Dynamic methylome of internal mRNA N

-methyl-

guanosine and its regulatory role in translation,

Cell Res., 29, 927-941, https://doi.org/10.1038/s41422-

019-0230-z.

151. Dong, R., Wang, C., Tang, B., Cheng, Y., Peng, X.,

Yang, X., Ni, B., and Li, J. (2024) WDR4 promotes

HCC pathogenesis through N7-methylguanosine by

regulating and interacting with METTL1, Cell. Sig-

nal., 118, 111145, https://doi.org/10.1016/j.cellsig.

2024.111145.

152. Schlake, T., Thess, A., Fotin-Mleczek, M., and Kallen,

K.-J. (2012) Developing mRNA-vaccine technolo-

gies, RNA Biol., 9, 1319-1330, https://doi.org/10.4161/

rna.22269.

153. Carralot, J. P., Weide, B., Schoor, O., Probst, J.,

Scheel,B., Teufel,R., Hoerr,I., Garbe,C., Rammensee,

H. G., and Pascolo, S. (2005) Production and charac-

terization of amplified tumor-derived cRNA libraries

to be used as vaccines against metastatic melanomas,

Genet. Vaccines Ther., 3, 6, https://doi.org/10.1186/

1479-0556-3-6.

154. Warren, L., Manos, P. D., Ahfeldt, T., Loh, Y. H.,

Li, H., Lau, F., Ebina, W., Mandal, P. K., Smith, Z. D.,

Meissner, A., Daley, G. Q., Brack, A. S., Collins, J. J.,

Cowan, C., Schlaeger, T. M., and Rossi, D. J. (2010)

Highly efficient reprogramming to pluripotency and

directed differentiation of human cells with synthet-

ic modified mRNA, Cell Stem Cell, 7, 618-630, https://

doi.org/10.1016/j.stem.2010.08.012.

155. Weng, Y., Li, C., Yang, T., Hu, B., Zhang, M., Guo, S.,

Xiao, H., Liang, X.-J., and Huang, Y. (2020) The chal-

lenge and prospect of mRNA therapeutics land-

scape, Biotechnol. Adv., 40, 107534, https://doi.org/

10.1016/j.biotechadv.2020.107534.

156. Malone, R. W., Felgner, P. L., and Verma, I. M. (1989)

Cationic liposome-mediated RNA transfection., Proc.

Natl. Acad. Sci. USA, 86, 6077-6081, https://doi.org/

10.1073/pnas.86.16.6077.

157. Karikó,K., Kuo,A., and Barnathan, E.S. (1999) Over-

expression of urokinase receptor in mammalian cells

following administration of the in vitro transcribed

encoding mRNA, Gene Ther., 6, 1092-1100, https://

doi.org/10.1038/sj.gt.3300930.

158. Kreiter, S., Selmi, A., Diken, M., Sebastian, M.,

Osterloh, P., Schild, H., Huber, C., Türeci, Ö., and

Sahin, U. (2008) Increased antigen presentation ef-

ficiency by coupling antigens to MHC class I traf-

ficking signals, J. Immunol., 180, 309-318, https://

doi.org/10.4049/jimmunol.180.1.309.

159. Kozak, M. (1994) Features in the 5′ non-coding se-

quences of rabbit α and β-globin mRNAs that affect

translational efficiency, J. Mol. Biol., 235, 95-110,

https://doi.org/10.1016/S0022-2836(05)80019-1.

160. Annweiler, A., Hipskind, R. A., and Wirth, T. (1991)

A strategy for efficient in vitro translation of cDNAs

using the rabbit β-globin leader sequence, Nucleic

Acids Res., 19, 3750, https://doi.org/10.1093/nar/

19.13.3750.

161. Conry, R.M., LoBuglio, A. F., Wright, M., Sumerel,L.,

Pike, M. J., Johanning, F., Benjamin, R., Lu, D., and

Curiel, D. T. (1995) Characterization of a messenger

RNA polynucleotide vaccine vector., Cancer Res., 55,

1397-1400.

162. Rohner,E., Yang,R., Foo, K.S., Goedel,A., and Chien,

K. R. (2022) Unlocking the promise of mRNA ther-

apeutics, Nat. Biotechnol., 40, 1586-1600, https://

doi.org/10.1038/s41587-022-01491-z.

163. Reshetnikov,V., Terenin,I., Shepelkova,G., Yeremeev,V.,

Kolmykov,S., Nagornykh,M., Kolosova,E., Sokolova,T.,

Zaborova, O., Kukushkin,I., Kazakova, A., Kunyk,D.,

Kirshina, A., Vasileva, O., Seregina, K., Pateev, I.,

Kolpakov, F., and Ivanov, R. (2024) Untranslated re-

gion sequences and the efficacy of mRNA vaccines

against tuberculosis, Int. J. Mol. Sci., 25, 888, https://

doi.org/10.3390/ijms25020888.

164. Yu, J., and Russell, J. E. (2001) Structural and func-

tional analysis of an mRNP complex that mediates

the high stability of human β-globin mRNA, Mol. Cell.

Biol., 21, 5879-5888, https://doi.org/10.1128/mcb.21.17.

5879-5888.2001.

165. Wang, Z., Day, N., Trifillis, P., and Kiledjian, M.

(1999) An mRNA stability complex functions with

poly(A)-binding protein to stabilize mRNA in vitro,

Mol. Cell. Biol., 19, 4552-4560, https://doi.org/10.1128/

mcb.19.7.4552.

166. Aviv,H., Voloch,Z., Bastos,R., and Levy,S. (1976) Bio-

synthesis and stability of globin mRNA in cultured

erythroleukemic friend cells, Cell, 8, 495-503, https://

doi.org/10.1016/0092-8674(76)90217-8.

167. Waggoner, S.A., and Liebhaber, S.A. (2003) Regulation

of α-globin mRNA stability, Exp. Biol. Med., 228, 387-

395, https://doi.org/10.1177/153537020322800409.

168. Holcik, M., and Liebhaber, S. A. (1997) Four highly

stable eukaryotic mRNAs assemble 3′ untranslated

region RNA-protein complexes sharing cis and trans

components, Proc. Natl. Acad. Sci. USA, 94, 2410-2414,

https://doi.org/10.1073/pnas.94.6.2410.

169. Wollner, C.J., Richner,M., Hassert, M.A., Pinto, A.K.,

Brien, J.D., and Richner, J.M. (2021) A Dengue Virus

SELECTION OF UTRs 751

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

Serotype 1 mRNA-LNP Vaccine Elicits Protective

Immune Responses, J. Virol., 95, e02482-20, https://

doi.org/10.1128/jvi.02482-20.

170. Sahin, U., Derhovanessian, E., Miller, M., Kloke,

B. P., Simon, P., Löwer, M., Bukur, V., Tadmor,

A. D., Luxemburger, U., Schrörs, B., Omokoko, T.,

Vormehr, M., Albrecht, C., Paruzynski, A., Kuhn,

A. N., Buck, J., Heesch, S., Schreeb, K. H., Müller, F.,

Ortseifer,I., Vogler,I., Godehardt,E., Attig,S., Rae,R.,

Breitkreuz, A., Tolliver, C., Suchan, M., Martic, G.,

Hohberger, A., Sorn, P., Diekmann, J., Ciesla, J.,

Waksmann, O., Brück, A. K., Witt, M., Zillgen, M.,

Rothermel, A., Kasemann, B., Langer, D., Bolte, S.,

Diken, M., Kreiter, S., Nemecek, R., Gebhardt, C.,

Grabbe,S., Höller, C., Utikal, J., Huber, C., Loquai, C.,

and Türeci, Ö. (2017) Personalized RNA mutanome

vaccines mobilize poly-specific therapeutic immu-

nity against cancer, Nature, 547, 222-226, https://

doi.org/10.1038/nature23003.

171. Ding, X., Zhou, Y., He, J., Zhao, J., and Li, J. (2024)

Enhancement of SARS-CoV-2 mRNA vaccine efficacy

through the application of TMSB10 UTR for superior

antigen presentation and immune activation, Vaccines,

12, 432, https://doi.org/10.3390/vaccines12040432.

172. Orlandini von Niessen, A. G., Poleganov, M. A.,

Rechner, C., Plaschke, A., Kranz, L. M., Fesser, S.,

Diken, M., Löwer, M., Vallazza, B., Beissert, T.,

Bukur,V., Kuhn, A.N., Türeci,Ö., and Sahin,U. (2019)

Improving mRNA-based therapeutic gene delivery by

expression-augmenting 3′ UTRs identified by cellu-

lar library screening, Mol. Ther., 27, 824-836, https://

doi.org/10.1016/j.ymthe.2018.12.011.

173. Ferizi, M., Aneja, M. K., Balmayor, E. R., Badieyan,

Z.S., Mykhaylyk,O., Rudolph,C., and Plank,C. (2016)

Human cellular CYBA UTR sequences increase mRNA

translation without affecting the half-life of recom-

binant RNA transcripts, Sci. Rep., 6, 39149, https://

doi.org/10.1038/srep39149.

174. Chiu, W.-W., Kinney, R. M., and Dreher, T. W. (2005)

Control of translation by the 5′- and 3′-terminal re-

gions of the dengue virus genome, J.Virol., 79, 8303-

8315, https://doi.org/10.1128/jvi.79.13.8303-8315.2005.

175. Gallie, D. R., Tanguay, R. L., and Leathers, V. (1995)

The tobacco etch viral 5′ leader and poly(A) tail

are functionally synergistic regulators of trans-

lation, Gene, 165, 233-238, https://doi.org/10.1016/

0378-1119(95)00521-7.

176. Tan, X., and Wan, Y. (2008) Enhanced protein ex-

pression by internal ribosomal entry site-driven

mRNA translation as a novel approach for in vitro

loading of dendritic cells with antigens, Hum. Im-

munol., 69, 32-40, https://doi.org/10.1016/j.humimm.

2007.11.009.

177. Yakubov,E., Rechavi,G., Rozenblatt,S., and Givol, D.

(2010) Reprogramming of human fibroblasts to plu-

ripotent stem cells using mRNA of four transcription

factors, Biochem. Biophys. Res. Commun., 394, 189-

193, https://doi.org/10.1016/j.bbrc.2010.02.150.

178. Mamaghani, S., Penna, R. R., Frei, J., Wyss, C.,

Mellett, M., Look, T., Weiss, T., Guenova, E., Kündig,

T. M., Lauchli, S., and Pascolo, S. (2024) Synthetic

mRNAs containing minimalistic untranslated regions

are highly functional in vitro and in vivo, Cells, 13,

1242, https://doi.org/10.3390/cells13151242.

179. Ma,Q., Zhang,X., Yang,J., Li,H., Hao,Y., and Feng,X.

(2024) Optimization of the 5′ untranslated region

of mRNA vaccines, Sci. Rep., 14, 19845, https://

doi.org/10.1038/s41598-024-70792-x.

180. Schrom, E., Huber, M., Aneja, M., Dohmen, C.,

Emrich, D., Geiger, J., Hasenpusch, G., Herrmann-

Janson, A., Kretzschmann, V., Mykhailyk, O.,

Pasewald, T., Oak, P., Hilgendorff, A., Wohlleber, D.,

Hoymann, H.G., Schaudien,D., Plank,C., Rudolph, C.,

and Kubisch-Dohmen,R. (2017) Translation of angio-

tensin-converting enzyme 2 upon liver- and lung-tar-

geted delivery of optimized chemically modified

mRNA, Mol. Ther. Nucleic Acids, 7, 350-365, https://

doi.org/10.1016/j.omtn.2017.04.006.

181. Vivinus,S., Baulande,S., Van Zanten,M., Campbell,F.,

Topley, P., Ellis, J. H., Dessen, P., and Coste, H.

(2001) An element within the 5′ untranslated re-

gion of human Hsp70 mRNA which acts as a gener-

al enhancer of mRNA translation, Eur. J. Biochem.,

268, 1908-1917, https://doi.org/10.1046/j.1432-1327.

2001.02064.x.

182. Zhao,Y., Moon,E., Carpenito,C., Paulos, C.M., Liu,X.,

Brennan, A. L., Chew, A., Carroll, R. G., Scholler, J.,

Levine, B. L., Albelda, S. M., and June, C. H. (2010)

Multiple injections of electroporated autologous T

cells expressing a chimeric antigen receptor medi-

ate regression of human disseminated tumor, Cancer

Res., 70, 9053-9061, https://doi.org/10.1158/0008-5472.

CAN-10-2880.

183. Zinckgraf, J. W., and Silbart, L. K. (2003) Modulat-

ing gene expression using DNA vaccines with dif-

ferent 3′-UTRs influences antibody titer, seroconver-

sion and cytokine profiles, Vaccine, 21, 1640-1649,

https://doi.org/10.1016/S0264-410X(02)00740-5.

184. Holden, K. L., and Harris, E. (2004) Enhancement

of dengue virus translation: Role of the 3′ untrans-

lated region and the terminal 3′ stem-loop domain,

Virology, 329, 119-133, https://doi.org/10.1016/j.virol.

2004.08.004.

185. Linares-Fernández, S., Moreno, J., Lambert, E.,

Mercier- Gouy, P., Vachez, L., Verrier, B., and Exposito,

J. Y. (2021) Combining an optimized mRNA template

with a double purification process allows strong ex-

pression of in vitro transcribed mRNA, Mol. Ther.

Nucleic Acids, 26, 945-956, https://doi.org/10.1016/

j.omtn.2021.10.007.

186. Trepotec, Z., Aneja, M. K., Geiger, J., Hasenpusch, G.,

Plank, C., and Rudolph, C. (2019) Maximizing the

VOLKHIN et al.752

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

translational yield of mRNA therapeutics by mini-

mizing 5′-UTRs, Tissue Eng. Part A, 25, 69-79, https://

doi.org/10.1089/ten.tea.2017.0485.

187. Jia, L., Mao, Y., Ji, Q., Dersh, D., Yewdell, J. W., and

Qian, S.B. (2020) Decoding mRNA translatability and

stability from the 5′ UTR, Nat. Struct. Mol. Biol., 27,

814-821, https://doi.org/10.1038/s41594-020-0465-x.

188. Mandal, P. K., and Rossi, D.J. (2013) Reprogramming

human fibroblasts to pluripotency using modified

mRNA, Nat. Protoc., 8, 568-582, https://doi.org/10.1038/

nprot.2013.019.

189. Lee, J., Dykstra, B., Spencer, J. A., Kenney, L. L.,

Greiner, D. L., Shultz, L. D., Brehm, M. A., Lin, C. P.,

Sackstein, R., and Rossi, D. J. (2017) MRNA-mediated

glycoengineering ameliorates deficient homing of

human stem cell-derived hematopoietic progenitors,

J.Clin. Invest., 127, 2433-2437, https://doi.org/10.1172/

JCI92030.

190. Richner, J. M., Himansu,S., Dowd, K. A., Butler, S. L.,

Salazar, V., Fox, J. M., Julander, J. G., Tang, W. W.,

Shresta,S., Pierson, T.C., Ciaramella,G., and Diamond,

M. S. (2017) Modified mRNA vaccines protect against

Zika virus infection, Cell, 168, 1114-1125.e10,

https://doi.org/10.1016/j.cell.2017.02.017.

191. Xia, X. (2021) Detailed dissection and critical eval-

uation of the Pfizer/BioNTech and Moderna mRNA

vaccines, Vaccines, 9, 734, https://doi.org/10.3390/

vaccines9070734.

192. Ringnér, M., and Krogh, M. (2005) Folding free en-

ergies of 5′-UTRs impact post-transcriptional regu-

lation on a genomic scale in yeast, PLoS Comput.

Biol., 1, e72, https://doi.org/10.1371/journal.pcbi.

0010072.

193. Pelletier, J., and Sonenberg, N. (1985) Insertion mu-

tagenesis to increase secondary structure within the

5′ noncoding region of a eukaryotic mRNA reduc-

es translational efficiency, Cell, 40, 515-526, https://

doi.org/10.1016/0092-8674(85)90200-4.

194. Karollus,A., Avsec,Ž., and Gagneur,J. (2021) Predict-

ing mean ribosome load for 5’UTR of any length us-

ing deep learning, PLoS Comput. Biol., 17, e1008982,

https://doi.org/10.1371/journal.pcbi.1008982.

195. Linder, J., and Seelig, G. (2021) Fast activation max-

imization for molecular sequence design, BMC Bio-

informatics, 22, 510, https://doi.org/10.1186/s12859-

021-04437-5.

196. Barazandeh, S., Ozden, F., Hincer, A., Seker, U. O. S.,

and Cicek, A. E. (2023) UTRGAN: learning to gen-

erate 5’ UTR sequences for optimized translation

efficiency and gene expression, bioRxiv, https://

doi.org/10.1101/2023.01.30.526198.

197. Linder, J., Bogard,N., Rosenberg, A.B., and Seelig,G.

(2020) A generative neural network for maximiz-

ing fitness and diversity of synthetic DNA and pro-

tein sequences, Cell Syst., 11, 49-62.e16, https://

doi.org/10.1016/j.cels.2020.05.007.

198. Li, G., Wu, J., and Wang, X. (2024) Predicting func-

tional UTR variants by integrating region-specific fea-

tures, Brief. Bioinform., 25, bbae248, https://doi.org/

10.1093/bib/bbae248.

199. Zhang,H., Gao,X., Zhang,J., and Lai,L. (2025) mRNA-

2vec: mRNA embedding with language model in the

5’UTR-CDS for mRNA design, Proc. Thirty-Ninth AAAI

Conf. Artif. Intell., 1, 1057-1065, https://doi.org/10.1609/

aaai.v39i1.32092.

200. Brar, G.A., and Weissman, J.S. (2015) Ribosome pro-

filing reveals the what, when, where and how of pro-

tein synthesis, Nat. Rev. Mol. Cell Biol., 16, 651-664,

https://doi.org/10.1038/nrm4069.

201. Chassé, H., Boulben,S., Costache, V., Cormier, P., and

Morales, J. (2017) Analysis of translation using poly-

some profiling, Nucleic Acids Res., 45, e15, https://

doi.org/10.1093/nar/gkw907.

202. Zheng, Y., and VanDusen, N. J. (2023) Massively

parallel reporter assays for high-throughput

in vivo analysis of cis-regulatory elements, J. Car-

diovasc. Dev. Dis., 10, 144, https://doi.org/10.3390/

jcdd10040144.

203. Gallego Romero, I., and Lea, A. J. (2023) Leveraging

massively parallel reporter assays for evolutionary

questions, Genome Biol., 24, 26, https://doi.org/10.1186/

s13059-023-02856-6.

204. Griesemer, D., Xue, J. R., Reilly, S. K., Ulirsch, J. C.,

Kukreja,K., Davis, J.R., Kanai, M., Yang, D.K., Butts,

J. C., Guney, M. H., Luban, J., Montgomery, S. B.,

Finucane, H.K., Novina, C.D., Tewhey,R., and Sabeti,

P. C. (2021) Genome-wide functional screen of 3′UTR

variants uncovers causal variants for human dis-

ease and evolution, Cell, 184, 5247-5260.e19, https://

doi.org/10.1016/j.cell.2021.08.025.

205. Lagunas,T., Plassmeyer, S.P., Fischer, A.D., Friedman,

R. Z., Rieger, M. A., Selmanovic, D., Sarafinovska, S.,

Sol, Y.K., Kasper, M.J., Fass, S.B., Aguilar Lucero, A.F.,

An, J. Y., Sanders, S. J., Cohen, B. A., and Dougherty,

J. D. (2023) A Cre-dependent massively parallel re-

porter assay allows for cell-type specific assessment

of the functional effects of non-coding elements in

vivo, Commun. Biol., 6, 1151, https://doi.org/10.1038/

s42003-023-05483-w.

206. Reynoso, M. A., Juntawong, P., Lancia, M., Blanco,

F.A., Bailey-Serres,J., and Zanetti, M.E. (2015) Trans-

lating ribosome affinity purification (TRAP) followed

by RNA sequencing technology (TRAP-SEQ) for quan-

titative assessment of plant translatomes, Methods

Mol. Biol., 1284, 185-207, https://doi.org/10.1007/978-

1-4939-2444-8_9.

207. Blahetek, G., Mayer, C., Zuber, J., Lenter, M., and

Strobel, B. (2024) Suppression of toxic transgene ex-

pression by optimized artificial miRNAs increases

AAV vector yields in HEK-293 cells, Mol. Ther. Meth-

ods Clin. Dev., 32, 101208, https://doi.org/10.1016/

j.omtm.2024.101280.

SELECTION OF UTRs 753

BIOCHEMISTRY (Moscow) Vol. 90 No. 6 2025

208. Perez-Garcia, C.G., Diaz-Trelles,R., Vega, J.B., Bao,Y.,

Sablad, M., Limphong, P., Chikamatsu, S., Yu, H.,

Taylor,W., Karmali, P.P., Tachikawa,K., and Chivuku-

la, P. (2022) Development of an mRNA replacement

therapy for phenylketonuria, Mol. Ther. Nucleic Acids,

28, 87-98, https://doi.org/10.1016/j.omtn.2022.02.020.

209. Gadalla, K. K. E., Vudhironarit, T., Hector, R. D.,

Sinnett, S., Bahey, N. G., Bailey, M. E. S., Gray, S. J.,

and Cobb, S. R. (2017) Development of a novel AAV

gene therapy cassette with improved safety features

and efficacy in a mouse model of Rett syndrome,

Mol. Ther. Methods Clin. Dev., 5, 180-190, https://

doi.org/10.1016/j.omtm.2017.04.007.

210. Artemyev,V., Gubaeva,A., Paremskaia, A.I., Dzhioeva,

A. A., Deviatkin, A., Feoktistova, S. G., Mityaeva, O.,

and Volchkov, P.Y. (2024) Synthetic promoters in gene

therapy: design approaches, features and applications,

Cells, 13, 1963, https://doi.org/10.3390/cells13231963.

211. Paremskaia, A. I., Kogan, A. A., Murashkina, A.,

Naumova, D.A., Satish, A., Abramov, I. S., Feoktistova,

S.G., Mityaeva, O.N., Deviatkin, A.A., and Volchkov,

P. Y. (2024) Codon-optimization in gene therapy:

promises, prospects and challenges, Front. Bioeng.

Biotechnol., 12, 1371596, https://doi.org/10.3389/fbioe.

2024.1371596.

212. Gentner, B., and Naldini, L. (2012) Exploiting

microRNA regulation for genetic engineering, Tis-

sue Antigens, 80, 393-403, https://doi.org/10.1111/

tan.12002.

213. Geisler,A., Jungmann,A., Kurreck,J., Poller,W., Katus,

H. A., Vetter, R., Fechner, H., and Müller, O. J. (2011)

MicroRNA122-regulated transgene expression in-

creases specificity of cardiac gene transfer upon in-

travenous delivery of AAV9 vectors, Gene Ther., 18,

99-209, https://doi.org/10.1038/gt.2010.141.

Publisher’s Note. Pleiades Publishing remains

neutral with regard to jurisdictional claims in published

maps and institutional affiliations. AI tools may have

been used in the translation or editing of this article.