Efficient in vitro transcription of artificial mRNA with highly modified 5' ends

Abstract

The present invention relates to a molecular complex suitable for in vitro transcription with a T7 RNA polymerase comprising modified tetranucleotide cap analogs, wherein the cap analog may include modified nucleobases. The invention further relates to a complex comprising a specific T7 promoter with a specific trinucleotide initiating sequence and a hybridizing tetranucleotide cap analog. The invention also relates to a use of said molecular complex for in vitro transcription of mRNA with high capping efficiency. Further, the invention relates to a method of producing at least one m7G-capped mRNA molecule. The invention also relates to a pharmaceutical composition for use in a method of treating and / or preventing a disease, comprising said at least one m7G-capped mRNA molecule.

Description

[0001] Munich, 7 October 2025

[0002] Our Ref.: EM 6553-04WO CMC / swo

[0003] Applicant: ExploRNA THERAPEUTICS SP. Z O. O.

[0004] Serial Number: N©W Application

[0005] ExploRNA THERAPEUTICS SP. Z O. O. ul. Zwirki i Wigury 93 / 2157, 02-089 WARSZAWA, POLAND

[0006] Efficient in vitro transcription of artificial mRNA with highly modified 5' ends

[0007] Technical Field

[0008] The present invention relates to a molecular complex suitable for in vitro transcription with a T7 RNA polymerase comprising modified tetranucleotide cap analogs, wherein the cap analog may include modified nucleobases. The invention further relates to a complex comprising a specific T7 promoter with a specific trinucleotide initiating sequence and a hybridizing tetranucleotide cap analog. The invention also relates to a use of said molecular complex for in vitro transcription of mRNA with high capping efficiency. Further, the invention relates to a method of producing at least one m7G-capped mRNA molecule. The invention also relates to a pharmaceutical composition for use in a method of treating and / or preventing a disease, comprising said at least one m7G-capped mRNA molecule.

[0009] Background of the invention

[0010] Introduction

[0011] The advancement of in vitro transcription (IVT) of mRNA has revolutionized the field of molecular biology and therapeutic development, presenting a pivotal technology in modern biotechnology and medicine. The efficient production of mRNA in vitro is paramount for a myriad of applications, including vaccine development, gene therapy, and synthetic biology. As the demand for rapid and scalable production of mRNA grows, the need for highly efficient IVT systems becomes increasingly critical.

[0012] In vitro transcription of mRNA refers to the process by which RNA polymerases synthesize mRNA from a DNA template in a controlled laboratory setting. This technology overcomes the limitations of traditional cellular transcription with the promise for precise control over the mRNA sequence, yield, and purity. IVT systems are particularly crucial in the context of mRNA vaccines, which have demonstrated immense potential in combating infectious diseases, as evidenced by the recent success of mRNA-based COVID-19 vaccines (Polack et al., Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine, N Engl J Med (2020), 383, 27, 2603-2615).

[0013] The efficiency of in vitro mRNA transcription directly impacts the scalability and costeffectiveness of mRNA production. Efficient IVT processes result in higher yields of mRNA, reducing the amount of raw materials and time required. This is especially important for large-scale manufacturing needed to meet the global demand for mRNA therapeutics and vaccines. Additionally, efficient IVT systems minimize the presence of impurities and undesirable by-products, which enhances the safety and efficacy of the final mRNA product.

[0014] Linear mRNAs designed for therapeutic applications must be sequence-optimized and equipped with appropriate stabilizing elements at both ends: a 5' cap structure and a 3' polyA tail. The mRNA cap consists of an "inverted" 7-methylguanosine linked to the mRNA body by a 5 '-5' triphosphate bridge, forming the so-called cap 0. This cap structure can be further modified by additional epitranscriptomic marks: a 2'-O methyl group on the ribose of the first transcribed nucleotide to form cap 1 , on the first two nucleotides to form cap 2, or an N6-methyl group on the 5' terminal adenosine to form an m6Am cap, as illustrated in FIG 1 .

[0015] The biological effects of epitranscriptomic modifications can be leveraged to improve the efficacy of mRNA as a therapeutic agent. In humans, cap 1 and cap 2 are crucial for distinguishing self from non-self RNA, helping to evade innate immune responses mediated by IFIT1. The role of the m6Am cap is still under debate but is likely related to mRNA senescence and evasion of innate immune responses. Notably, N6-methylation of 2'-0-methyladenosine in the cap (to form m6Am) is common epigenetic modification in all types of mammalian cells and is relatively abundant in several tissues. For example, in mice, the fraction of N6-methylated 5' terminal A reaches approximately 70% in the liver, 90% in the heart, and 94% in the brain. It has also been shown that m6Am increases the translational activity of mRNA in certain experimental setups (Sikorski et al., Nucleic Acids Research, Vol. 48 (4), 2020, 1607-1626).

[0016] The first commonly used method involves synthesizing in vitro transcribed mRNA with a cap 0 structure. This can be achieved using a dinucleotide cap analog or a Vaccinia Capping Enzyme (VCE). Following this, the 2'-O-methyl group is added to the first transcribed nucleotide using a 2'-O-methyltransferase. While this approach allows forthe introduction of the cap 1 structure posttranscriptionally, it is often inefficient and labor- intensive. The additional enzymatic step increases the complexity and cost of the process, and achieving consistent high efficiency remains challenging.

[0017] The second method involves the co-transcriptional introduction of a trinucleotide primer that comprises the cap 1 structure. In this strategy, RNA is synthesized by T7 RNA polymerase using trinucleotide cap analogs, such as those falling under the general formula rn7GpppN1mpN2, ribonucleotide 5 '-triphosphates (NTPs, such as ATP, CTP, GTP, and UTP, or their analogs like N1-methylpseudouridine or 5'-methoxyuridine), and a double-stranded DNA (dsDNA) template that contains a transcription promoter followed by a transcription initiation site. This site is characterized by the first nucleobase where the transcription actually starts (+1 position, denoted transcription start site (TSS)), whereas the sequence additionally including nucleobase positions +2 and optionally +3 following the TSS may be referred to as the initiating sequence (+1 , +2), in certain embodiments a trinucleotide initiating sequence (TIS; +1 , +2, +3). The formation of a transcriptionally active molecular complex depends on the specific promoter sequence and the polymerase, optionally including the hybridization of a cap analog in certain IVT reactions to the first transcribed nucleobases.

[0018] During transcription initiation in an IVT reaction, a molecular complex forms between the RNA polymerase, the non-coding (antisense) strand of the unwound region of the dsDNA template, and the cap analog. Nucleotides N1and N2of the cap analog hybridize (fully or partially complementarily) to the non-coding strand of the dsDNA template at or near the trinucleotide initiating sequence +1 and +2 positions (FIG 1) of the template, where the +1 position is defined as the predominant TSS in the absence of a cap analog (i.e., when producing uncapped RNA).

[0019] However, the effectiveness of this method varies widely. Ishikawa et al. demonstrated that the cap analog rn7GpppAmpG can be used as a transcription primer to produce capped mRNA with a specific DNA template containing a <t>6.5 promoter followed by a GGG trinucleotide initiating sequence (TIS). In this case, the cap analog hybridized to the -1 and +1 positions of the dsDNA template (Ishikawa et al., Preparation of eukaryotic mRNA having differently methylated adenosine at the 5'-terminus and the effect of the methyl group in translation, Nucleic Acids Symposium Series No. 53 (2009), p. 129-130). Disadvantageously, this method required using a greater than six-fold excess of the capped initiating oligonucleotide primer, i.e. the cap analog, the most expensive nucleotide component of the transcription reaction, over competing GTP to drive the transcription reaction toward capped mRNA rather than uncapped mRNA. This significantly increases the total cost of synthesizing RNA, making the method economically impractical for large-scale production. In WO2017053297A1 , it was demonstrated that using a specific DNA template, which includes a <t>6.5 promoter followed by an AGG sequence, results in higher capping efficiencies for rn7GpppAmpG and similar compounds. Thus, it was concluded that ensuring complete complementarity of N1and N2to the +1 and +2 positions in the DNA template is crucial for achieving high capping efficiency of the final mRNA product. Additionally, it was claimed that for tetranucleotide cap analogs (such as rn7GpppN1mpN2mpN3), the best results are obtained when there is perfect complementarity between N1, N2, and N3and the +1 , +2, and +3 positions of the DNA template. This finding aligns with the prevailing and compelling belief, reflected in the prior art, that perfect alignment between the trinucleotide initiating sequence and the respective cap nucleotides is an essential prerequisite for efficient mRNA capping.

[0020] However, even with optimal reaction conditions, modifications designed to improve in vitro transcribed mRNA properties, such as the introduction of unnatural chemical groups within the 7-methylguanosine, triphosphate chain, the first phosphodiester bond, and N1nucleoside, have been used in this strategy, often resulted in decreased capping efficiency and overall transcription yield. For example, it was shown that while a benzyl group at the N6-position of 2'-G-methyladenosine may increase its translational activity, it significantly reduces capping efficiency and IVT yield. Achieving acceptable capping efficiency (>90%) for this analog required optimization of transcription conditions, including pH of the buffer, concentration of Mg2+ions, and using a relatively high concentration of the cap analog, thereby limiting the application scope and increasing the overall cost of mRNA manufacturing (Warminski et al., Trinucleotide mRNA Cap Analog A / 6-Benzylated at the Site of Posttranscriptionalm6AmMark Facilitates mRNA Purification and Confers Superior Translational Properties In Vitro and In Vivo, JACS (2024), 146, 12, 8149-8163).

[0021] Furthermore, approximately 50% of eukaryotic mRNAs contain a cap 2 structure, which aids in evading the innate immune system and potentially enhances the efficacy of exogenously delivered mRNAs. mRNAs containing cap 2 can be produced in principle via IVT using tetrameric cap analogs (m7GpppN1mpN2mpN3) as initiators for T7 RNA polymerase (Drazkowska et al., 2'-0-Methylation of the second transcribed nucleotide within the mRNA 5' cap impacts the protein production level in a cell-specific manner and contributes to RNA immune evasion, Nucleic Acids Research (2022), 50, 16, 9051-9071). However, there is currently no effective method for incorporating cap 2 into mRNA. Even with an optimal dsDNA template fully complementary at positions +1 , +2, and +3, only about 50% capping efficiency is achieved (WO2017053297A1). A higher capping efficiency in the final mRNA product, crucial for most biotechnological and therapeutic applications, requires additional costly and time-consuming processing to remove immunogenic 5'-triphosphate RNA. This greatly increases both the cost and time required, which severely limits the economic feasibility of producing therapeutic mRNAs on an industrial scale and would likely fail to meet the prospective high demand for rapidly produced mRNA therapeutics.

[0022] Inagaki et al. (Inagaki, M., Abe, N., Li, Z. et al. Cap analogs with a hydrophobic photocleavable tag enable facile purification of fully capped mRNA with various cap structures. Nat Commun 14, 2657 (2023). https: / / doi.org / 10.1038 / s41467-023-38244-8) has independently shown that incorporation of modified tetranucleotide cap analogs aligned with +1 , +2, and +3 positions of the DNA template is significantly less efficient compared to trinucleotide analogs aligned at positions +1 and +2.

[0023] Therefore, developing mRNAs with highly modified cap structures at their 5' ends, such as rn7GpppBn6AmpN, m7Gpppm6AmpN, or cap 2 and its modified analogs, which are easily accessible and free from uncapped RNA impurities, is highly desirable. These mRNAs hold significant potential for therapeutic applications.

[0024] So far, it has never been systematically studied whether certain cap structures (tri- and tetranucleotide cap structures) and specific promoters, i.e., DNA templates, sequences are best compatible with achieving reproducibly high capping efficiency along with a favourably high translation rate.

[0025] As highly modified artificial cap structures are intricately dependent on the complex interplay between the structure of the DNA template, the trinucleotide initiating sequence, the cap analog, and the specific modifications of the cap analog, and as there is currently no general technical solution to achieve highly efficient IVT by using cap 2 primers, let alone efficiently incorporating a more heavily modified mRNA cap for enhanced translation and immune tolerance, there is consequently a great need in providing suitable cap analog:DNA template complexes that will reliably work in a setting using a specific promoter of interest and a cognate RNA polymerase to define the most cost efficient and save cognate cap analog::DNA template pair for a given setting.

[0026] Summary of the Invention

[0027] The above-identified objectives are solved by the technical teaching as provided with the present invention. In a first aspect there is provided a molecular complex suitable for in vitro transcription with a T7 RNA polymerase comprising (i) a Cap analog according to the general formula wherein Base1, Base2and Base3are independently a natural, unnatural or modified base; R1, R2, R3and R4are independently OH, O-Me, H, F, Cl, O-alkyl, O-aryl, O-arylalkyl, O- acyl; wherein at least one from R1and R2, or both, is / are not OH; and X1, X2, X3are independently O, S, BH3, Se; Y1and Y2are independently O, CH2, CHCI, CHF, CF2, CCI2, NH; W1and W2are independently O, S; Z1, Z2are independently O', S', CH3, BHs' (ii) a DNA template comprising a T7 polymerase promoter and a trinucleotide initiating sequence (TIS); optionally: wherein nucleobase Base1hybridizes to the -1 position and / or preferably wherein nucleobase Base2hybridizes to the +1 position and Base3hybridizes to the +2 position of the trinucleotide initiating sequence; or optionally: provided the DNA template comprises a core promoter of SEQ ID NO: 3 or the promoter comprises a sequence of SEQ ID NOs: 12 or 13, 20 or 21 , 22 or 23 nucleobase Base1hybridizes to the -2 position; nucleobase Base2hybridizes to the -1 position and Base3hybridizes to the +1 position of the trinucleotide initiating sequence. In this embodiment, the TIS may be a shortened sequence, starting with G at +1 and G at +2 (see FIG 4C).

[0028] One embodiment relates to the molecular complex according to the first aspect, wherein the DNA template is selected from a T1 to T8 template as detailed in FIG 4A to 4D, or from a T13, 14 and 16 template as detailed in FIG 38A or wherein the DNA template comprises a sequence as defined by any of SEQ ID NOs: 1 to 25, and / or the complement thereof, or a sequence having at least 99% identity to the respective sequence.

[0029] One embodiment relates to the molecular complex according to the first aspect or the previous embodiment thereof, wherein Base1and Base2of the Cap analog as defined in the first aspect (i) are independently selected from adenine or an analog of adenine, Base3is guanine or an analog of guanine, and wherein the DNA template as defined in the first aspect (ii) comprises a promoter with a core region with a sequence as defined by any one of SEQ ID NO: 1 to 3, is selected from an AGG, an AAG, an ATG, a GGG, a GG, an AGA or a GAG.

[0030] One embodiment relates to the molecular complex according to the first aspect and the first embodiment thereof, wherein Base1is independently adenine, N6-benzyladenine, N6-methyladenine, N6-(2-phenylethyl)adenine, and R1 and / or R2 are independently O- Me, and (i) Base2and Base3are both guanine, or (ii) Base2is adenine and Base3is guanine, or (iii) Base2and Base3are independently guanine or adenine, or (iv) Base2is adenine and Base3is uracil, or (v) Base2is uracil and Base3is guanine, or (vi) Base2is uracil and Base3is adenine, or (vii) Base2is adenine and Base3is adenine, in the Cap analog as defined in the first aspect or any embodiment thereof.

[0031] One embodiment relates to the molecular complex according to the first aspect or any embodiment thereof, wherein (i) Base1isBn6A and R1 is O-Me, Base2and Base3are independently guanine, or (ii) Base1ism6A, R1 is O-Me, Base2and Base3are independently guanine, or (iii) Base1isBn6A, R1 is O-Me, Base2is adenine and Base3is guanine, or (iv) Base1ism6A, R1 is O-Me, Base2is adenine and Base3is guanine, or (v) Base1is A, R1 is O-Me, Base2is G, R1 is O-Me, and Base3is guanine, or (vi) Base1isBn6A, R1 is O-Me, Base2is G, R2 is O-Me, and Base3is G, or (vii) Base1ism6A, R1 is O- Me, Base2is G, R2 is O-Me and Base3is guanine, or (viii) Base1is A, R1 is O-Me, Base2is A, R2=O-Me, and Base3is guanine, or (ix) Base1is Am, Base2is adenine, and Base3is guanine, or (x) Base1isBn6A, R1 is O-Me, Base2is A, R2 is O-Me, and Base3is guanine, or (xi) Base1ism6A, R1 is O-Me, Base2is A, R1 is O-Me and Base3is guanine, or (xii) Base1is A, R1 is O-Me, Base2and Base3are both guanine, (xiii) Base1is N6-(2- phenylethyl)adenine, R1 is O-Me, preferably Base2and Base3are independently guanine, more preferably Base 2 is adenine and Base 3 is guanine, according to the Cap analog as defined in the first aspect.

[0032] One embodiment relates to the molecular complex according to the first aspect or any embodiment thereof, additionally comprising a T7 RNA polymerase and / or a homolog thereof and / or a modified variant thereof.

[0033] A second aspect relates to a vector, or more than one vector, for use in forming a molecular complex according to the first aspect or any embodiment thereof, wherein the at least one vector comprises the at least one DNA template as defined in the first aspect or any embodiment thereof. A third aspect relates to a DNA molecule, or more than one than one DNA molecule, for use in forming a molecular complex according to the first aspect or any embodiment thereof, wherein the at least one DNA molecule comprises the at least one DNA template as defined in the first aspect or any embodiment thereof.

[0034] A fourth aspect relates to a use of the molecular complex according to the first aspect or any embodiment thereof for in vitro transcription, wherein the capping efficiency is at least 80%, 81 %, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or up to 100%.

[0035] A fifth aspect relates to a method of producing at least one m7G-capped mRNA molecule, the method comprising: (i) providing a reaction mixture comprising the Cap analog as defined in the first aspect or any embodiment thereof, the DNA template as defined in the first aspect or any embodiment thereof and / or the vector according to the second aspect and / or the DNA molecule according to the third aspect and at least one T7 RNA polymerase and / or the sequence encoding the same; (ii) allowing in vitro transcription from said DNA template and / or from said vector and obtaining at least one m7G-capped mRNA; (iii) optionally: purifying said at least one m7G-capped mRNA; and / or (iv) optionally: formulating said at least one m7G-capped mRNA molecule yielding a pharmaceutical composition.

[0036] One embodiment relates to the method of the fifth aspect further comprising a step of: (v) introducing or contacting said at least one m7G-capped mRNA molecule obtained in step (ii) into or with at least one target cell in an in vitro cellular system for transfecting said at least one target cell and / or for in vitro studying the functionality of said at least one m7G- capped mRNA.

[0037] A sixth aspect relates to a pharmaceutical composition for use in a method of preventing and / or treating a disease, the pharmaceutical composition comprising the at least one m7G-capped mRNA molecule obtained or obtainable by the method according to the fifth aspect or the embodiment thereof, additionally comprising at least one pharmaceutically acceptable carrier or excipient.

[0038] One embodiment relates to the pharmaceutical composition for use in a method of preventing and / or treating a disease according to the sixth aspect, the pharmaceutical composition additionally comprising at least one lipid nanoparticle, wherein the at least one lipid nanoparticle is selected from the group of at least one liposome (LPs), at least one liposome-like nanoparticle (LLP), at least one solid lipid nanoparticle (SLN), at least one nanostructured lipid carrier (NLC), at least one lipid-polymer hybrid nanoparticle (LPN), at least one lipoprotein particle (LPT), at least one nanoemulsion, at least one cationic nanoemulsion (CNE) and at least one exosome.

[0039] A seventh aspect relates to a kit comprising the Cap analog as defined in the first aspect or any embodiment thereof, optionally comprising the DNA template as defined in the first aspect or any embodiment thereof, and / or the vector according to the second aspect and / or the DNA molecule according to the third aspect, optionally wherein the kit comprises at least one further reagent and / or at least one aqueous solution, preferably a buffer.

[0040] Brief Description of the Figures

[0041] The following detailed description of the embodiments of the invention will be better understood when read in conjunction with the appended drawings. For the purpose of illustrating the invention, shown in the drawings are embodiments, which are presently exemplified. It should be understood, however, that the invention is not limited to the precise arrangement and instrumentalities of the embodiments shown in the drawings.

[0042] In all Figures, the lanes of the gels correspond to the numbers and samples summarized in Table 4 and Tables 6 to 10.

[0043] Figure 1 (FIG 1): Chemical structures of eukaryotic cap structures with different methylation patterns (A) and the general structure of synthetic cap analogs used as transcription initiation primers (B).

[0044] Figure 2 (FIG 2): Synthesis of a representative primer carrying the cap 2 structure, rn7GpppAmpAmpG. Other primers of similar structures (but different sequences) were / can be prepared using similar pathways using different commercially available phosphoramidites.

[0045] Figure 3 (FIG 3): Synthesis of representative primers carrying cap 1 or cap 2 structures along with nucleobase modifications. Other primers with similar structures (but different sequences and nucleobase modifications) were / can be prepared using similar pathways using synthetic and commercially available phosphoramidites.

[0046] Figure 4 (FIG 4A to D): Sequences of the promoters for T7 RNA polymerase and transcription start sites (TSSs) used in the experiments. Underlined nucleotide indicates the first transcribed nucleotide (+1 position; TSS), corresponding to the major initiation sites when uncapped mRNA synthesis is performed (i.e. in the absence of cap analog) (see also Table 5). Figure 5 (FIG 5): Results of in vitro transcription (IVT) with T7 RNA polymerase, in the presence of m7GpppAmpGmpG, m7GpppAmpAmpG (cap type 2 means 2'-O-methylated ribose of the nucleotide N1and N2) or without tetranucleotide cap analog, and the DNA templates T1 ( 6.5-AGG), T2 ( 2.5-AGG), and T3 (<t>6.5-GGG) (see also Table 5). IVT reactions were conducted for 2 h at 37°C in the presence of Bis-T ris buffer pH 6.5 and 25 mM MgCh, 5 mM ATP, CTP and UTP, 4 mM GTP, 10 mM tetranucleotide primer (for reactions providing uncapped mRNAs, cap analog was omitted), 40 ng / pL of linearized plasmid as the DNA template and T7 RNA polymerase (Roche). A: Analysis of the IVT efficiency of the crude transcription mixtures. For a rough comparison of the IVT efficiency the same amount (0.1 pL) of each crude IVT mix was applied on the agarose gel. Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. B: Capping efficiency analysis on the denaturing polyacrylamide gel for purified mRNAs cleaved with 5' UTR-specific ribozyme. Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band (or bands) corresponding to capped RNA and the summarized intensities of capped RNA and ppp-RNA (percentage values shown above the gel).

[0047] Figure 6 (FIG 6A to C): The alignment ofm7GpppA*pA*pG-type andm7GpppA*pG*pG- type tetranucleotide cap analogs with DNA templates T1 (<t>6.5-AGG), T2 (<t>2.5-AGG) or T3 (<t>6.5-GGG) during the major transcription initiation events (schemes on the upper panel). The table in FIG 6C summarizes expected lengths of the capped 5' end of RNA after ribozyme-mediated cleavage. Also, for each cap analog the number of nucleotides complementary to the particular template (T1 , T2 or T3; see also Table 5) and hybridized positions (+1 corresponds to the first transcribed nucleotide) are presented.

[0048] Figure 7 (FIG 7): Results of in vitro transcription (IVT) with T7 RNA polymerase, in the presence of various tetranucleotide cap analog primers (10 mM each, and with Bn6- modified primers indicated with squares; see also Table 4) and template T1 (<t>6.5-AGG) (see also Table 5) comprising of <t>6.5-promoter and AGG transcription start site (i.e. with TTCC deoxyribonucleotides at positions -1 , +1 , +2 and +3 of the non-coding strands of the DNA template; see also FIG 4). Trinucleotide cap analogs under their optimal conditions (described in Example 3) were used in control reactions (lanes 9 and 10). A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1.2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. B: Capping efficiency analysis on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs cleaved with 5' UTR- specific ribozyme. The samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000) Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band (or bands) corresponding to capped RNA and the summarized intensities of capped RNA and ppp-RNA (percentage values shown above the gel). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T1 (<t>6.5-AGG) and without cap analog added to the IVT mix (see FIG 5B). All other conditions and NTP concentrations were as described in Example 3.

[0049] Figure 8 (FIG 8): Results of in vitro transcription (IVT) with T7 RNA polymerase, in the presence of various tetranucleotide cap analog primers (10 mM each, and with Bn6- modified primers indicated with squares; see also Table 4) and template T2 (see also Table 5) comprising of <t>2.5-promoter and AGG transcription start site (i.e. with ATCC deoxyribonucleotides at positions -1 , +1 , +2 and +3 of the non-coding strands of the DNA template; see also FIG 4A to D). Trinucleotide cap analogs under their optimal conditions (described in Example 3) were used in control reactions (lanes 9 and 10). A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1 .2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. B: Capping efficiency analysis on the denaturing polyacrylamide gel foroligo(dT)25-purified mRNAs cleaved with 5' UTR-specific ribozyme. The samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000) Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band (or bands) corresponding to capped RNA and the summarized intensities of capped RNA and ppp- RNA (percentage values shown above the gel). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T2 (<t>2.5-AGG) and without cap analog added to the IVT mix (see FIG 5B). All other conditions and NTP concentrations were as described in Example 3. Figure 9 (FIG 9): Results of in vitro transcription (IVT) with T7 RNA polymerase, in the presence of various tetranucleotide cap analog primers (10 mM each, and with Bn6- modified primers indicated with squares; see also Table 4) and template T3 (see also Table 5) comprising of <t>6.5-promoter and GGG transcription start site (i.e. with TCCC deoxyribonucleotides at positions -1 , +1 , +2 and +3 of the non-coding strands of the DNA template; see also FIG 4). Trinucleotide cap analogs under their optimal conditions (described in Example 3) were used in control reactions (lanes 9 and 10). A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1 .2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. B: Capping efficiency analysis on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs cleaved with 5' UTR-specific ribozyme. The samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000) Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band (or bands) corresponding to capped RNA and the summarized intensities of capped RNA and ppp- RNA (percentage values shown above the gel). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T3 (<t>6.5-GGG) and without cap analog added to the IVT mix (see FIG 5B). All other conditions and NTP concentrations were as described in Example 3.

[0050] Figure 10 (FIG 10): Efficient incorporation of 5 mMm7GpppA*pA*pG-type primers (Bn6- modified primers are indicated with squares; see also Table 4) by in vitro transcription (IVT) with template T1 (<t>6.5-AGG) (see also Table 5). IVT reactions were conducted for 2h at 37°C in the presence of Bis-Tris buffer pH 6.5 and 25 mM MgCh (optimal for tetranucleotide transcription primers and Bn6-containing trinucleotide cap primer) or pH 8.0 and 10 mM MgCh (optimal for trinucleotide cap 1 primer), 5 mM ATP, CTP and UTP, 4 mM GTP, 5 mM cap analog primer, 40 ng / pL of linearized plasmid as the DNA template and T7 RNA polymerase (Roche) (see also Example 3). A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mixwas applied on the 1 xTBE 1 .2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. B: Capping efficiency analysis on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs cleaved with 5' UTR-specific ribozyme. The samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000) Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band (or bands) corresponding to capped RNA and the summarized intensities of capped RNA and ppp-RNA (percentage values shown above the gel). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T1 (<t>6.5-AGG) and without cap analog added to the IVT mix (see FIG 5B). All other conditions and NTP concentrations were as described in Example 3.

[0051] Figure 11 (FIG 11 ): Efficient incorporation of 2 mMm7GpppA*pA*pG-type primers (Bn6- modified primers are indicated with squares; see also Table 4) by in vitro transcription (IVT) with template T1 (<t>6.5-AGG) (see also Table 5). IVT reactions were conducted for 2h at 37°C in the presence of Bis-Tris buffer pH 6.5 and 25 mM MgCh (optimal for tetranucleotide transcription primers and Bn6-containing trinucleotide cap primer) or pH 8.0 and 10 mM MgCh (optimal for trinucleotide cap 1 primer), 5 mM ATP, CTP and UTP, 4 mM GTP, 2 mM cap analog primer, 40 ng / pL of linearized plasmid as the DNA template and T7 RNA polymerase (Roche) (see also Example 3). A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mixwas applied on the 1 xTBE 1 .2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. B: Capping efficiency analysis on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs cleaved with 5' UTR-specific ribozyme. The samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000) Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band (or bands) corresponding to capped RNA and the summarized intensities of capped RNA and ppp-RNA (percentage values shown above the gel). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T1 (<t>6.5-AGG) and without cap analog added to the IVT mix (see FIG 5B). All other conditions and NTP concentrations were as described in Example 3. Figure 12 (FIG 12): Incorporation of unmodified cap 1 and cap 2-typem7GpppA*pU*pG- primers and cap 2 m7GpppAmpAmpU into RNA by in vitro transcription (IVT) using template T4 (<t>6.5-ATG) and T5 (<t>6.5-TGG) (see also Table 5) in comparison to mRNA in vitro transcribed without cap primer. IVT reactions were conducted for 2h at 37°C in the presence of Bis-Tris buffer pH 6.5 and 30 mM MgCh, 5 mM ATP, CTP, UTP and GTP, 8 mM cap analog primer, 20 ng / pL of PCR-generated DNA template and T7 RNA polymerase (Roche) (see also Example 4). A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1.2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. B: Capping efficiency analysis on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs cleaved with 5' UTR-specific ribozyme. The samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000) Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band (or bands) corresponding to capped RNA and the summarized intensities of capped RNA and ppp-RNA (percentage values shown above the gel). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T4 (<t>6.5-ATG) or T5 (<t>6.5-TGG) and without cap analog added to the IVT mix. All other conditions and NTP concentrations were as described in Example 4.

[0052] Figure 13 (FIG 13): The alignment of m7GpppAmpAmpU andm7GpppA*pU*pG-type tetranucleotide cap analogs (m7GpppAmpUpG andm7GpppAmpUmpG) with DNA templates T4 (<t>6.5-ATG) or T5 (<t>6.5-TGG) during the major transcription initiation events (schemes on the upper panel). The table below summarizes expected lengths of the capped 5' end of RNA after ribozyme-mediated cleavage. Also, for each cap analog the number of nucleotides complementary to the particular template (T4 orT5; see also Table 5) and hybridized positions (+1 corresponds to the first transcribed nucleotide) are presented.

[0053] Figure 14 (FIG 14): Incorporation of m7GpppAmpAmpU andm7GpppA*pU*pG-type primers into RNA by in vitro transcription (IVT) with A: template T4 (<t>6.5-ATG) and B: template T5 (<t>6.5-TGG) (see also Table 5). IVT reactions were conducted for2h at 37°C in the presence of Bis-Tris buffer pH 6.5 and 30 mM MgCh (optimal form7GpppA*pU*pG- type tetranucleotide transcription primers andm7GpppA*pU-type trinucleotide primers), 5 mM ATP, CTP, UTP and GTP, 8 mM cap analog primer, 20 ng / pL of PCR-generated DNA template and T7 RNA polymerase (Roche) (see also Example 4). Upper gels: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1.2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gels) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. Bottom gels: Capping efficiency analysis on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs cleaved with 5' UTR- specific ribozyme. The samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000) Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band (or bands) corresponding to capped RNA and the summarized intensities of capped RNA and ppp-RNA (percentage values shown above the gels). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T4 (<t>6.5-ATG) or T5 (<t>6.5-TGG) and without cap analog added to the IVT mix (see FIG 12). All other conditions and NTP concentrations were as described in Example 4.

[0054] Figure 15 (FIG 15): Incorporation of unmodified cap 1 and cap 2-typem7GpppA*pA*pG- primers and their Bn6-modified counterparts (Bn6-modified primers are indicated with squares) into RNA by in vitro transcription (IVT) with template T1 (<t>6.5-AGG; with complementarity of cap nucleotides N1, N2and N3with positions -1 , +1 and +2 of the noncoding strand) and T8 (<t>6.5-AAG; with complementarity of cap nucleotides N1, N2and N3with positions +1 , +2 and +3 of the non-coding strand) (see also Table 5 and FIG 6). Complementarity between positions -1 , +1 and +2 and nucleotides N1, N2and N3of cap primer is essential for obtaining RNA with Bn6-modified 5' cap structure. A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1 .2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. B: Capping efficiency analysis on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs cleaved with 5' UTR-specific ribozyme. The samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band (or bands) corresponding to capped RNA and the summarized intensities of capped RNA and ppp- RNA (percentage values shown above the gel). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T1 (<t>6.5-AGG) or T8 (<t>6.5-AAG) and without cap analog added to the IVT mix. All other conditions and NTP concentrations were as described in Example 5.

[0055] Figure 16 (FIG 16): Incorporation ofm7GpppA*pG*pA-type primers into RNA by in vitro transcription (IVT) with template T6 (<t>6.5-AGA; with complementarity of cap nucleotides N1, N2and N3with positions +1 , +2 and +3 of the non-coding strand) and template T7 (<t>6.5-GAG; with complementarity of cap nucleotides N1, N2and N3with positions -1 , +1 and +2 of the non-coding strand) (see also Table 5) (Bn6-modified primers are indicated with squares). A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1 .2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. B: Capping efficiency analysis on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs cleaved with 5' UTR- specific ribozyme. The samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000) Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band (or bands) corresponding to capped RNA and the summarized intensities of capped RNA and ppp-RNA (percentage values shown above the gel). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T6 (<t>6.5-AGA) or T7 (<t>6.5-GAG) and without cap analog added to the IVT mix. All other conditions and NTP concentrations were as described in Example 5.

[0056] Figure 17 (FIG 17): The alignment ofm7GpppA*pG*pA-type tetranucleotide cap analogs (m7GpppAmpGpA,m7GpppBn6AmpGpA,m7GpppAmpGmpA,m7GpppBn6AmpGmpA) with DNA template T6 (<t>6.5-AGA) orT7 (<t>6.5-GAG) during the major transcription initiation events (schemes on the upper panel). The table below summarizes expected lengths of the capped 5' end of RNA after ribozyme-mediated cleavage. Also, for each cap analog the number of nucleotides complementary to the particular template (T6 orT7; see also Table 5) and hybridized positions (+1 corresponds to the first transcribed nucleotide) are presented. Figures 18 to 23 (FIG 18 to 23): Firefly luciferase (FFLuc) cellular expression as the result of transfection with mRNAs co-transcriptionally capped with m7GpppA*pG*pG or m7GpppA*pA*pG tetranucleotide primers. Transcripts were synthesized with T7 RNA polymerase in the presence of 10 mM of various tetranucleotide cap analog primers (see also Table 4) and template T1 (<t>6.5-AGG) (see also Table 5) comprising of <t>6.5- promoter and AGG trinucleotide initiating sequence (i.e. with TTCC deoxyribonucleotides at positions -1 , +1 , +2 and +3 of the non-coding strands of the DNA template; see also FIG 4). Trinucleotide cap analogs under their optimal conditions (described in Example 3) were used for preparation of reference transcripts. 1 xi 04of HEK293T (FIG 18 and 19) and 5x104(FIG 20 and 21) of primary human (hM <P) or murine (mM <P) (FIG 22 and 23) macrophages seeded in 100 pL of Opti-MEM medium (ThermoFisher Scientific) on 96- well plate incubated at 37°C, 100% humidity, and 5% CO2, were transfected with 20 or 200 ng of HPLC-purified mRNAs using Lipofectamine™ MessengerMAX™ transfecting reagent (ThermoFisher Scientific). The luminescence signal (Relative Light Units, RLU) was measured in the cell lysates collected 4 h, 24 h, 48 h (all cell types) and 72 h (HEK293T and hM <P) post-transfection after BrightGlo Reagent (Promega) addition using EnVision plate reader (Perkin-Elmer). Data points represent biological replicates. Bars represent means ± standard deviation (SD), n=3.

[0057] Figure 24 (FIG 24): Results of in vitro transcription (IVT) of mRNA encoding human erythropoietin (hEPO) prepared for in vivo experiment. Transcripts were synthesized with T7 RNA polymerase in the presence of various tetranucleotide cap primers (5 mM each, and with Bn6-modified primers indicated with squares; see also Table 4) and template T9 (<t>6.5-AGG) (see also Table 5) comprising of <t>6.5-promoter and AGG trinucleotide initiating sequence (i.e. with TTCC deoxyribonucleotides at positions -1 , +1 , +2 and +3 of the non-coding strands of the DNA template; see also FIG 4). Trinucleotide cap analogs (Bn6-containing trinucleotide primer rn7GpppBn6AmpG or unmodified rn7GpppAmpG) under their optimal conditions (described in Example 3) were used in reference reactions (lanes 9 and 10). A: Analysis of the IVT efficiency in the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1.2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. B: Capping efficiency analysis on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs. Analyzed mRNAs were cleaved with 5' UTR-specific ribozyme, and the samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE, stained with SYBR® Gold 5 (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band (or bands) corresponding to capped RNA and the summarized intensities of capped RNA and ppp-RNA (percentage values shown below the gel on the panel A). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T1 (<t>6.5-AGG) and without cap analog added to the IVT mix (see FIG 5B). All other conditions and NTP concentrations were as described in Example 3.

[0058] Figures 25 and 26 (FIG 25 and 26): Human erythropoietin (hEPO) in vivo production as the result of intravenous administration of HPLC-purified mRNAs co-transcriptionally capped with 5 mM m7GpppA*pG*pG or m7GpppA*pA*pG tetranucleotide primers. Transcripts were synthesized with T7 RNA polymerase in the presence of various tetranucleotide cap analog primers (see also Table 4) and template T9 (<t>6.5-AGG) (see also Table 5) comprising of <t>6.5-promoter and AGG trinucleotide initiating sequence (i.e. with TTCC deoxyribonucleotides at positions -1 , +1 , +2 and +3 of the non-coding strands of the DNA template; see also FIG 4). Trinucleotide cap analogs (Bn6-containing trinucleotide primer rn7GpppBn6AmpG or unmodified rn7GpppAmpG) under their optimal conditions (described in Example 3) were used in control reactions. C57BL / 6 mice were intravenously administrated with lipid nanoparticles (mRNA-LNP) containing 1 pg of each mRNA variant encoding hEPO, and the protein concentration was determined using ELISA in the blood serum samples collected 6h and 24h post-injection. Data points represent biological replicates. Bars represent means ± standard deviation (SD), n=5.

[0059] Figures 27 and 28 (FIG 27 and 28): Results of the optimization of co-transcriptional capping with rn7GpppBn6AmpApG tetranucleotide primer. In vitro transcription (IVT) reactions were performed using T7 RNA polymerase for the template T1 (<t>6.5-AGG) (see also Table 5) encoding firefly luciferase (FFLuc) and using various concentrations of rn7GpppBn6AmpApG (0, 2, 4, 5, 6, 8 and 10 mM) at pH 6.5, with 25 mM MgCh and 5 mM NTPs, including UTP or NI-MeMJTP. FIG 27 shows capping efficiencies and normalized IVT yields as the function of rn7GpppBn6AmpApG tetranucleotide primer concentration at pH 6.5 determined for oligo(dT)25-purified mRNAs by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. For the capping efficiency determination analyzed mRNAs were cleaved with 5' UTR-specific ribozyme, and the samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE, stained with SYBR® Gold 5 (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band (or bands) corresponding to capped RNA and the summarized intensities of capped RNA and ppp-RNA. Bands corresponding to ppp-RNA were identified for the sample without cap analog added to the IVT mix. FIG 28 shows capping efficiencies and normalized IVT yields as the function of pH of transcription buffer determined for oligo(dT)25-purified mRNAs (samples preparation is described in Example 14).

[0060] Figures 29 and 30 (FIG 29 and 30): Results of the optimization of co-transcriptional capping with m7GpppBn6AmpAmpG tetranucleotide primer. In vitro transcription (IVT) reactions performed using T7 RNA polymerase for the template T1 (<t>6.5-AGG) (see also Table 5) comprising of <t>6.5-promoter and AGG trinucleotide initiating sequence (i.e. with TTCC deoxyribonucleotides at positions -1 , +1 , +2 and +3 of the non-coding strands of the DNA template; see also FIG 4) encoding firefly luciferase (FFLuc) and using various concentrations of m7GpppBn6AmpAmpG (0, 2, 4, 5, 6, 8 and 10 mM) at pH 6.5, with 25 mM MgCh and 5 mM NTPs, including UTP or NI-MeMJTP. FIG 29 shows capping efficiencies and IVT yields as the function of m7GpppBn6AmpAmpG tetranucleotide primer concentration at pH 6.5 determined for oligo(dT)25-purified mRNAs by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. For the capping efficiency determination analyzed mRNAs were cleaved with 5' UTR-specific ribozyme, and the samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE, stained with SYBR® Gold 5 (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band (or bands) corresponding to capped RNA and the summarized intensities of capped RNA and ppp-RNA. Bands corresponding to ppp-RNA were identified for the sample without cap analog added to the IVT mix. FIG 30 shows capping efficiencies and normalized IVT yields as the function of pH of transcription buffer determined for oligo(dT)25-purified mRNAs (samples preparation is described in Example 14).

[0061] Figure 31 (FIG 31): Comparison of dsRNA content in the oligo(dT)25-purified mRNAs encoding firefly luciferase (FFLuc) co-transcriptionally capped with 10 mM m7GpppA*pG*pG or m7GpppA*pA*pG tetranucleotide. In vitro transcription (IVT) reactions were performed using T7 RNA polymerase, template T1 (<t>6.5-AGG) (see also Table 5), at pH 6.5, with 25 mM MgCL and 5 mM NTPs. Trinucleotide cap analogs (Bn6- containing trinucleotide primer rn7GpppBn6AmpG or unmodified rn7GpppAmpG) under their optimal conditions (described in Example 3) were used for preparation of control transcripts. 5, 25 or 250 ng of each mRNA was immobilized on the nylon membrane (Hybond™-N+, Amersham™), incubated with dsRNA-specific J2 antibody (SCICONS), followed by the incubation with secondary anti-mouse HRP-conjugated antibody (Cell Signaling Technology) and HRP substrate (ECL™ Prime Western Blotting Detection Reagents). Chemiluminescence signal was detected using ImageQuant 800 (Amersham™) imaging system. dsRNA content was estimated using ImageQuantTL software by comparison of the detected signals to the measured intensities of standard dsRNA (Abnova). The sample ID’s represent the respective cap analogs as listed in Table 11.

[0062] Figure 32 (FIG 32): Comparison of IVT yields and capping efficiencies for trinucleotide and tetranucleotide primers using market standard IVT protocol. Transcripts were synthesized with T7 RNA polymerase in the presence of 4 mM of commercially available trinucleotide cap primers (rn7GpppAmpG, rn73 OGpppAmpG and m7 3 OGpppm6AmpG, lanes 1-3, respectively), Bn6-containing trinucleotide primer rn7GpppBn6AmpG (lane 4), Bn6- containing tetranucleotide primer rn7GpppBn6AmpApG or m7GpppBn6AmpAmpG (lanes 5 and 6), and unmodified tetranucleotide primer rn7GpppAmpApG or m7GpppAmpAmpG (lanes 7 and 8) (see also Table 4). IVT mix was supplemented with PCR-generated template T10 (<t>6.5-AGG) being the optimized construct encoding firefly luciferase (FFLuc). In vitro transcription reactions were prepared as described in Example 16. For capping efficiency analysis mRNAs purified by silica-based method were cleaved with 5' UTR-specific ribozyme, and the samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE, stained with SYBR® Gold 5 (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band (or bands) corresponding to capped RNA and the summarized intensities of capped RNA and ppp-RNA (percentage values shown below the gel).

[0063] Figure 33 (FIG 33): Testing distinct in vitro transcription (IVT) protocols to identify optimal one for efficient incorporation of m7GpppBn6A1mpU2mG3(and rn7GpppBn6A1mpU2pG3) primer into RNA using template T12 (<t>6.5-TGA; complementarity of cap nucleotides N1, N2and N3with positions -2, -1 , +1 of the non-coding strand) (see also Table 5). A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1.2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification using silica-based approach by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. M denotes the DNA marker, with bands corresponding to 200, 500, 1000, 1500, 2000, 3000, 4000, and 6000 base pairs (bp), arranged from the bottom to the top. B: Capping efficiency analysis on the denaturing polyacrylamide gel for mRNAs cleaved with 5' UTR-specific ribozyme. Samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band corresponding to capped RNA and summarized intensities of capped RNA and ppp-RNA (percentage values shown above the gel). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T12 (<t>6.5-TGA) and without cap analog added to the IVT mix. All other conditions and NTP concentrations were described in Example 17.

[0064] Figure 34 (FIG 34): Testing distinct in vitro transcription (IVT) protocols to identify optimal one for efficient incorporation of rn7GpppBn6A1mpU2pG3and m7GpppBn6A1mpU2mpG3primers into RNA using template T4 (<t>6.5-ATG; complementarity of cap nucleotides N1, N2and N3with positions +1 , +2 and +3 of the non-coding strand) and T5 (<t>6.5-TGG; complementarity of cap nucleotides N1, N2and N3with positions -2, -1 , +1 of the noncoding strand) (see also Table 5 and FIG 13). Complementarity between positions -2, -1 and +1 of <t>6.5-TGG template and nucleotides N1, N2and N3of cap primers m7GpppBn6A1*pU2*pG3is essential for obtaining RNA with Bn6-modified 5' cap structure initiated withBn6AU sequence. A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1 .2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. M denotes the DNA marker, with bands corresponding to 200, 500, 1000, 1500, 2000, 3000, 4000, and 6000 base pairs (bp), arranged from the bottom to the top. B: Capping efficiency analysis on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs cleaved with 5' UTR-specific ribozyme. Samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band corresponding to capped RNA and summarized intensities of capped RNA and ppp-RNA (percentage values shown above the gel). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T4 (<t>6.5-ATG) or T5 (<t>6.5-TGG) and without cap analog added to the IVT mix. All other conditions and NTP concentrations were described in Example 17. Figure 35 (FIG 35): Optimization of MgOAc concentration for efficient in vitro transcription (IVT) and capping using m7GpppBn6A1mpU2pG3(and m7GpppBn6A1mpU2mpG3) primer and template T5 (<t>6.5-TGG; complementarity of cap nucleotides N1, N2and N3with positions -2, -1 , +1 of the non-coding strand) (see also Table 5 and FIG 13). A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1.2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification using silica-based approach by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. M denotes the DNA marker, with bands corresponding to 200, 500, 1000, 1500, 2000, 3000, 4000, and 6000 base pairs (bp), arranged from the bottom to the top. B: Capping efficiency analysis on the denaturing polyacrylamide gel for mRNAs cleaved with 5' UTR-specific ribozyme. Samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band corresponding to capped RNA and summarized intensities of capped RNA and ppp-RNA (percentage values shown above the gel). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T4 (<t>6.5-ATG) or T5 (<t>6.5-TGG) and without cap analog added to the IVT mix. All other conditions and NTP concentrations were described in Example 17.

[0065] Figure 36 (FIG 36): Optimization of MgOAc concentration for efficient in vitro transcription (IVT) and capping using rn7GpppBn6A1mpU2pG3(and m7GpppBn6A1mpU2mpG3) primer and template T12 (<t>6.5-TGA; complementarity of cap nucleotides N1, N2and N3with positions -2, -1 , +1 of the non-coding strand) (see also Table 5). A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1.2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification using silica-based approach by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. M denotes the DNA marker, with bands corresponding to 200, 500, 1000, 1500, 2000, 3000, 4000, and 6000 base pairs (bp), arranged from the bottom to the top. B: Capping efficiency analysis on the denaturing polyacrylamide gel for mRNAs cleaved with 5' UTR-specific ribozyme. Samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band corresponding to capped RNA and summarized intensities of capped RNA and ppp-RNA (percentage values shown above the gel). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T12 (<t>6.5-TGA) and without cap analog added to the IVT mix. All other conditions and NTP concentrations were described in Example 17.

[0066] Figure 37 (FIG 37): Testing in vitro transcription (IVT) and capping efficiency for PrimeCap T7 RNA polymerase (TaKaRa) at pH 6.5 and various MgOAc concentration using rn7GpppBn6A1mpU2pG3and m7GpppBn6A1mpU2mpG3primers and template T5 (<t>6.5- TGG; complementarity of cap nucleotides N1, N2and N3with positions -2, -1 , +1 of the non-coding strand) or T11 (<t>6.5-TGG with alternative 5' UTR; complementarity of cap nucleotides N1, N2and N3with positions -2, -1 , +1 of the non-coding strand) (see also Table 5). A: Capping efficiency analysis on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs capped with rn7GpppBn6A1mpU2pG3and B: m7GpppBn6A1mpU2mpG3cleaved with 5' UTR-specific ribozyme. Samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band corresponding to capped RNA and summarized intensities of capped RNA and ppp-RNA (percentage values shown below the gels). Bands corresponding to ppp-RNA were identified by the experiment performed for RNA in vitro transcribed using template T5 (<t>6.5-TGG) and without cap analog added to the IVT mix. Bands corresponding to capped RNAs were marked with blue asterisk. IVT efficiencies (values shown below the gels) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm and calculated as mRNA yield obtained from 1 pL of starting reaction volume. All other conditions and NTP concentrations were described in Example 17.

[0067] Figure 38 (FIG 38): A: The alignment ofm7GpppA*pU*pA-type tetranucleotide cap analogs (m7GpppBn6AmpUpA andm7GpppBn6AmpUmpA) with DNA templates T16 (<t>6.5- ATA), T13 (<t>6.5-TAG) or T14 (<t>6.5-TAG with alternative 5' UTR) during the major transcription initiation events. B: The table summarizes expected lengths of the capped 5' end of RNA after ribozyme-mediated cleavage. Also, for each cap analog the number of nucleotides complementary to the particular template (T16, T13 or T14; see also Table 5) and hybridized positions (+1 corresponds to the first transcribed nucleotide when cap is not present in the IVT mix) are depicted. Figure 39 (FIG 39): Testing distinct in vitro transcription (IVT) protocols to identify optimal one for efficient incorporation of m7GpppBn6A1mpU2mpA3primer into RNA using template T13 (<t>6.5-TAG; complementarity of cap nucleotides N1, N2and N3with positions -2, -1 , +1 of the non-coding strand (see also FIG 38)) and T16 (<t>6.5-ATA; complementarity of cap nucleotides N1, N2and N3with positions +1 , +2 and +3 of the non-coding strand) (see also Table 5 and FIG 13). Complementarity between positions -2, -1 and +1 of <t>6.5-TAG template and nucleotides N1, N2and N3of cap primer m7GpppBn6A1mpU2mpA3is essential for obtaining RNA with Bn6-modified 5' cap structure initiated withBn6AU sequence. A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1.2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification using silica-based approach by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. M denotes the DNA marker, with bands corresponding to 200, 500, 1000, 1500, 2000, 3000, 4000, and 6000 base pairs (bp), arranged from the bottom to the top. B: Capping efficiency analysis on the denaturing polyacrylamide gel for mRNAs cleaved with 5' UTR-specific ribozyme. Samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band corresponding to capped RNA and summarized intensities of capped RNA and ppp-RNA (percentage values shown above the gel). IVT conditions and NTP concentrations were described in Example 17.

[0068] Figure 40 (FIG 40): Comparison of the in vitro transcription (IVT) and capping efficiency for wild-type (WT) T7 RNA polymerase (Roche) and PrimeCap T7 RNA polymerase (TaKaRa) at pH 6.5 and various MgOAc concentration using rn7GpppBn6A1mpU2pA3primer and template T13 (<t>6.5-TAG; complementarity of cap nucleotides N1, N2and N3with positions -2, -1 , +1 of the non-coding strand (see also FIG 38 and Table 5)) and T14 (<t>6.5-TAG with alternative 5' UTR; complementarity of cap nucleotides N1, N2and N3with positions -2, -1 , +1 of the non-coding strand (see also FIG 38 and Table 5)). A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1.2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. M denotes the DNA marker, with bands corresponding to 200, 500, 1000, 1500, 2000, 3000, 4000, and 6000 base pairs (bp), arranged from the bottom to the top. B: Capping efficiency analysis on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs cleaved with 5' UTR-specific ribozyme. Samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band corresponding to capped RNA and summarized intensities of capped RNA and ppp-RNA (percentage values shown above the gel). Bands corresponding to capped RNAs were marked with blue asterisk. IVT conditions and NTP concentrations were described in Example 17.

[0069] Figure 41 (FIG 41 ): Comparison of the in vitro transcription (IVT) and capping efficiency for wild-type (WT) T7 RNA polymerase (Roche) and PrimeCap T7 RNA polymerase (TaKaRa) at various MgOAc concentration using m7GpppBn6A1mpU2mpA3primer and template T13 (<t>6.5-TAG; complementarity of cap nucleotides N1, N2and N3with positions -2, -1 , +1 of the non-coding strand (see also FIG 38 and Table 5)) and T14 (<t>6.5-TAG with alternative 5' UTR; complementarity of cap nucleotides N1, N2and N3with positions -2, -1 , +1 of the non-coding strand (see also FIG 38 and Table 5)). A: Analysis of the IVT efficiency of the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1.2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. M denotes the DNA marker, with bands corresponding to 200, 500, 1000, 1500, 2000, 3000, 4000, and 6000 base pairs (bp), arranged from the bottom to the top. B: Capping efficiency analysis on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs cleaved with 5' UTR-specific ribozyme. Samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band corresponding to capped RNA and summarized intensities of capped RNA and ppp-RNA (percentage values shown above the gel). Bands corresponding to capped RNAs were marked with blue asterisk. IVT conditions and NTP concentrations were described in Example 17. Figure 42 (FIG 42): Results of in vitro transcription (IVT) of mRNA encoding codon- optimized firefly luciferase (FFLuc2 - construct with standard 5'UTR and FFLuc2.2 - construct with alternative 5'UTR, see Table 5) prepared for in vitro experiment. Transcripts with Bn6-modification on the 5' end (appropriate tetranucleotides used for IVT are indicated with squares above Lane ID) were synthesized under optimal conditions determined for each primer::DNA template complex (for details see Table 12, Example 17, FIG 33-37 and FIG 39-41). Unmodified trinucleotide cap analogs under their optimal conditions were used in reference reactions (lane 3, 6, 9, 10, 13 and 16). A: Analysis of the IVT efficiency in the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1 .2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. B: Capping efficiency analysis on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs. Analyzed mRNAs were cleaved with 5' UTR-specific ribozyme, and the samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE, stained with SYBR® Gold 5 (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band corresponding to capped RNA and the summarized intensities of capped RNA and ppp-RNA (percentage values shown below the gel on the panel A). Bands corresponding to ppp-RNA were marked with red asterisk. IVT conditions are described in Example 3.

[0070] Figure 43 to 47 (FIG 43 to 47): Firefly luciferase (mRNA encoding codon-optimized firefly luciferase FFLuc2 - construct with standard 5'UTR and FFLuc2.2 - construct with alternative 5'UTR, see Table 5) cellular expression as the result of transfection with mRNAs co-trascriptionally capped with m7GpppBn6A*pU*pG, m7GpppBn6A*pU*pA and m7GpppBn6A*pA*pG primers. Transcripts with Bn6-modification on the 5' end were synthesized under optimal conditions determined for each primer: :DNA template complex (for details see Table 12, Example 3, Example 17, FIG 33-37 and FIG 39-41). Unmodified trinucleotide cap analogs under their optimal conditions were used in reference reactions. 1 X104of HEK293T (FIG 43 and 44) and 5x104(FIG 45 and 46) of primary human (hM<t>) or murine (mMd>) (FIG 47) macrophages seeded in 100 pL of Opti- MEM medium (ThermoFisher Scientific) on 96-well plate incubated at 37°C, 100% humidity, and 5% CO2, were transfected with 20 or 100 ng of HPLC-purified mRNAs using Lipofectamine™ MessengerMAX™ transfecting reagent (ThermoFisher Scientific). The luminescence signal (Relative Light Units, RLU) was measured in the cell lysates collected 4 h and 24 h (all cell types) and additionally 48 h and 72 h (HEK293T and hM<t>) post-transfection after BrightGlo Reagent (Promega) addition using EnVision plate reader (Perkin-Elmer). Data points represent biological replicates. Bars represent means ± standard deviation (SD), n=3.

[0071] Figure 48 (FIG 48) Testing in vitro transcription (IVT) and capping efficiency for various concentrations of m7GpppBn6A1mpA2pU3and m7GpppBn6A1mpA2mpU3primers and template T4 (<t>6.5-ATG; complementarity of cap nucleotides N1, N2and N3with positions +1 , +2 and +3 of the non-coding strand) (see also Table 5). Capping efficiency was analyzed on the denaturing polyacrylamide gel for oligo(dT)25-purified mRNAs cleaved with 5' UTR-specific ribozyme. Samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE and stained with SYBR® Gold (1 :10,000). Intensities of the visualized short capped and uncapped RNA (ppp-RNA) fragments were quantified densitometrically. For each sample capping efficiency was determined as the ratio of the measured intensity of the band corresponding to capped RNA and summarized intensities of capped RNA and ppp-RNA (percentage values shown below the gel). IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm and calculated as mRNA yield obtained from 1 pL of starting reaction volume. IVT conditions were described in Example 19.

[0072] Figure 49 (FIG 49): Results of in vitro transcription (IVT) of mRNA encoding codon- optimized firefly luciferase (construct FFLuc2, see Table 5) prepared for in vitro experiment. Transcripts with Bn6-modification on the 5' end (appropriate tetranucleotides used for IVT are indicated with squares above Lane ID) were synthesized under optimal conditions determined for each primer::DNA template complex (for details see Table 13, Example 17, Example 19, FIG 33-37, FIG 39-41 and FIG 48). Unmodified trinucleotide cap analogs under their optimal conditions were used in reference reactions (lane 3, 7, 9, 15 and 18). A: Analysis of the IVT efficiency in the crude transcription mixtures. The same amount of each IVT mix was applied on the 1 xTBE 1 .2% agarose gel, and each lane contains 0.1 pl of unpurified in vitro transcribed mRNA after DNase I treatment and 2x dilution with 50 mM EDTA (0.05 pl crude IVT per lane). Actual IVT efficiencies (values shown below the gel) were estimated after mRNA purification on the oligo(dT)25 resin by measuring eluate volume and the absorbance at 260 nm. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume. B: Electrophoretic profile of cap primer: :DNA template complex formation based on the denaturing polyacrylamide gel for HPLC-purified mRNAs. Analyzed mRNAs were cleaved with 5' UTR-specific ribozyme, and the samples were run on a 15% polyacrylamide + 7 M urea gel in 1 x TBE, stained with SYBR® Gold 5 (1 :10,000). The lane ID’s represent the respective cap analogs as listed in Table 1. In Table 1 , the 4thcolumn lists the respective sequence of the coding strand comprising the last nucleotide of the core promoter region (-1) followed by TIS (positions +1+2+3). In case of templates with TIS starting with pyrimidine nucleotide (i.e. template T5, <t>6.5- TGG) the numbering of TIS position is shifted, and actual numbering should be as follows: A(-2) T(-1) G(+1) G(+2). The 5thcolumn lists the respective 5' end structure of capped mRNA. Nucleotides in normal font are introduced into the 5' end as an integral part of the cap analog. Nucleotides in bold font are introduced as NTPs (regular polymerization performed by T7 RNA pol.) . “m7Gppp” was omitted for clarity in the single entries. The Template IDs in the 3rdcolumn corresponds to the template nucleotides as defined in Table 5.

[0073] Table 1

[0074] Figure 50 to 54 (FIG 50 to 54): Firefly luciferase (mRNA encoding codon-optimized firefly luciferase FFLuc2, see Table 5) cellular expression as the result of transfection with mRNAs co-trascriptionally capped with m7GpppBn6A*pA*pU, m7GpppBn6A*pU*pG, m7GpppBn6A*pA*pA and m7GpppBn6A*pA*pG tetranucleotide primers. Transcripts with Bn6-modification on the 5' end were synthesized under optimal conditions determined for each primer::DNA template complex (for details see Table 13, Example 3, Example 17, Example 19, FIG 33-37, FIG 39-41 and FIG 48). Unmodified trinucleotide cap analogs under their optimal conditions were used in reference reactions. 1 xi o4of HEK293T (FIG 50 and 51) and 5x104(FIG 52 and 53) of primary human (hM<t>) or murine (mMd>) (FIG 54) macrophages seeded in 100 pL of Opti-MEM medium (ThermoFisher Scientific) on 96-well plate incubated at 37°C, 100% humidity, and 5% CO2, were transfected with 20 or 100 ng of HPLC-purified mRNAs using Lipofectamine™ MessengerMAX™ transfecting reagent (ThermoFisher Scientific). The luminescence signal (Relative Light Units, RLU) was measured in the cell lysates collected 4 h and 24 h (all cell types) and additionally 48 h (HEK293T and hM<t>) post-transfection after BrightGlo Reagent (Promega) addition using EnVision plate reader (Perkin-Elmer). Data points represent biological replicates. Bars represent means ± standard deviation (SD), n=3.

[0075] Brief Description of Sequences

[0076] Notably, SEQ ID NOs: 4 and 5, 6 and 7, 8 and 9, 10 and 11 , 12 and 13, 14 and 15, 16 and 17, 18 and 19, 20 and 21 , and 22 and 23, and 24 and 25 each represent pairs forming a dsDNA template named T1 to T16, respectively, as shown in FIG 4, FIG 6, FIG 13, FIG 17 and FIG 38. As the skilled person understands that each strand (5' to 3'and 3'to 5' notation) in a pair of sequences is fully complementary to each other, reference to one of these sequences only will also be enough for the skilled person to identify a suitable dsDNA template selected from T1 to T16 based on the information related to one (singlestranded) sequence I SEQ ID NO only.

[0077] Definitions

[0078] As used herein, the terms “capping analog”, “cap analog”, and “cap primer” are used interchangeably and refer to a synthetic nucleotide analog designed to initiate transcription in vitro and to be incorporated co-transcriptionally at the 5' end of an RNA molecule. Such analogs / primers are typically di-Ztri- or tetranucleotides (optionally chemically modified, e.g. at the N6-position of adenosine or at the 2'-O position of ribose) that both mimic the natural cap structure (e.g. m7GpppN, where N is a nucleotide), and / or serve as the initiating primer during T7 RNA polymerase-driven transcription. Accordingly, unless expressly indicated otherwise, the term “cap analog” in the present disclosure shall be understood to encompass and be synonymous with “cap primer”. A cap analog comprises a guanosine moiety at the 5' terminus, which is typically 7-methylguanosine (m7G) or a functional analog thereof capable of serving as the initiating cap structure. Such analogs include, but are not limited to, N7-ethylguanosine (et7G), N7-benzylguanosine (Bn7G), N7-(2-phenylethyl)guanosine, or other substituted guanosines at the N7 position. This moiety is connected through a 5'-5' triphosphate or modified triphosphate linkage to one or more subsequent nucleotides, which may be arranged as dinucleotide, trinucleotide or tetranucleotide structures and may contain natural, unnatural, or chemically modified nucleobases, sugars, or phosphate groups. Accordingly, the cap analog encompasses compounds of the general structurem7GpppN1N2(N3) as well as analogs in which the m7G moiety is replaced by a functionally equivalent N7-modified guanosine, provided that the compound retains the ability to initiate transcription and be incorporated into the 5' end of an RNA transcript.

[0079] As used herein, the term “DNA molecule” refers to an isolated or synthetic DNA entity comprising at least one DNA template as defined herein, in linear or circular form. The DNA molecule may be provided directly (e.g. as a PCR product, linearized plasmid, or chemically synthesized oligonucleotide) and serves as the transcription template in an in vitro transcription reaction.

[0080] As used herein, the term “DNA template” refers to a double-stranded DNA molecule comprising a T7 RNA polymerase promoter operably linked to a trinucleotide initiating sequence (TIS), wherein the TIS defines the transcription start site (+1 position) and optionally further nucleotides at positions +2 and +3. The DNA template provides the sequence information for in vitro transcription of an RNA molecule and is configured such that at least one nucleobase of a cap analog may hybridize to one or more positions of the non-coding strand at or near the transcription start site. Representative DNA templates encompassed by this definition include, but are not limited to, the T1 to T16 templates as comprised in the present application.

[0081] Detailed

[0082] The present invention solves the problems in the prior art by providing molecular complexes that reliably initiate in vitro transcription catalysed by T7 RNA polymerase using highly modified caps, achieving remarkably higher capping efficiencies and IVT efficiencies than those deemed optimal in prior art. Contrary to the common belief, we discovered that tetranucleotide primers (m7GpppN1mpN2mpN3) of specific sequences are most effectively incorporated when aligned with a dsDNA template that ensures nucleosides N2and N3align and hybridize with the +1 and +2 positions of the template. These novel complexes enable the production of mRNAs with heavily modified cap structures (e.g., incorporating theBn6Ammodification, cap 2, and their combinations) with capping efficiencies exceeding 90% and reaching up to 100%, without compromising the in vitro transcription yield. In addition, the present invention provides not only a solution to the problem of efficiently incorporating a tetrameric cap structure and its modified derivates into mRNA, but also enables more effective, cost sensitive and thus upscalable synthesis of mRNAs carrying a cap 1 structure.

[0083] To solve the prevailing challenges in the field of synthetic mRNA production, particularly the suboptimal incorporation of modified caps which are crucial for enhanced translation efficiency and reduced immune response, the present invention provides a surprising solution. Traditional methods have struggled with low capping efficiencies and incomplete cap incorporation, which severely limits the widescale application of therapeutic mRNA.

[0084] The present invention introduces a novel approach utilizing tetranucleotide primers that are aligned with a dsDNA template, ensuring that the nucleosides in positions N2and N3of the primer hybridize to the +1 and +2 sites of the transcription template. This alignment is critical because it maximizes the efficiency with which cap analogs, particularly complex and modified structures such as theBn6Ammodification and cap 2 analogs, are incorporated into the mRNA strand.

[0085] So far in the art, no detailed study is available analyzing the importance of the primer sequence for capping efficiency and IVT yield for different promoter structures I DNA templates even though knowledge about these molecular details is of outstanding importance to define versatile primer: :DNA template pairs performing best for a given experiment in vitro and, particularly, in vivo. Additionally, the present invention for the first time provides a highly systematic comparison of tetranucleotide and comparative trinucleotide primers and certain DNA templates corresponding to naturally occurring promoter regions, or DNA templates derived and closely related to naturally occurring promoter regions to identify primer::DNA template complexes that synergistically work with each other by forming a sterically favouring complexthat promotes both: transcription initiation and high translational activity. Thereby, novel and generally applicable complexes are provided that are suitable for various different IVT applications depending on the individual need.

[0086] A key advantage of this invention is its remarkable efficiency in cap incorporation — exceeding 90% and in some instances reaching 100%. Such high levels of efficiency are unprecedented in the prior art relating to IVT and represent a significant technical leap forward. Moreover, these high efficiencies are achieved without compromising the yield of the transcription process itself, which remains robust.

[0087] This method not only addresses the technical difficulties previously encountered with cap incorporation but also enhances the overall cost-effectiveness and scalability of mRNA synthesis. By enabling the reliable production of mRNAs with complex cap structures, this invention opens up new possibilities for the development of mRNA-based therapeutics, offering potential treatments for a wide range of diseases with improved efficacy and reduced side effects. The ability to consistently produce high-quality, capped mRNA in a cost-effective and scalable manner is particularly advantageous for rapid vaccine development and personalized medicine applications.

[0088] In a first aspect there is provided a molecular complex suitable for in vitro transcription with a T7 RNA polymerase comprising (i) a Cap analog according to the general formula (I),

[0089] wherein Base1, Base2and Base3are independently a natural, unnatural or modified base; R1, R2, R3and R4are independently OH, O-Me, H, F, Cl, O-alkyl, O-aryl, O-arylalkyl, O- acyl; wherein at least one from R1and R2, or both, is / are not OH; and X1, X2, X3are independently O, S, BH3, Se; Y1and Y2are independently O, CH2, CHCI, CHF, CF2, CCI2, NH; W1and W2are independently O, S; Z1, Z2are independently O', S', CH3, BHs' (ii) a DNA template comprising a T7 polymerase promoter and a trinucleotide initiating sequence; optionally: wherein nucleobase Base1hybridizes to the -1 position and / or preferably wherein nucleobase Base2hybridizes to the +1 position and Base3hybridizes to the +2 position of the trinucleotide initiating sequence; or optionally: provided the DNA template comprises a core promoter of SEQ ID NO: 3 or the promoter comprises a sequence of SEQ ID NOs: 12 or 13, 20 or 21 , 22 or 23 nucleobase Base1hybridizes to the -2 position; nucleobase Base2hybridizes to the -1 position and Base3hybridizes to the +1 position of the trinucleotide initiating sequence. In another aspect there is provided a molecular complex suitable for in vitro transcription with a T7 RNA polymerase comprising (i) a Cap analog according to the general formula (II),

[0090] wherein Base1, Base2and Base3are independently a natural, unnatural or modified base; R1, R2, R3and R4are independently OH, O-Me, H, F, Cl, O-alkyl, O-aryl, O-arylalkyl, O- acyl; wherein at least one from R1and R2, or both, is / are not OH; and X1, X2, X3are independently O, S, BH3, Se; Y1and Y2are independently O, CH2, CHCI, CHF, CF2, CCI2, NH; W1and W2are independently O, S; Z1, Z2are independently O', S', CH3, BH3, wherein Base0is defined by the general formula (III), wherein R5is methyl, ethyl, benzyl, substituted benzyl, propargyl, alkyl, or alkylaryl,

[0091] R6and / or R7are independently H, methyl, ethyl, benzyl, substituted benzyl, alkyl, or alkylaryl, optionally wherein alkylaryl is substituted, (ii) a DNA template comprising a T7 polymerase promoter and a trinucleotide initiating sequence (TIS); optionally: wherein nucleobase Base1hybridizes to the -1 position and / or preferably wherein nucleobase Base2hybridizes to the +1 position and Base3hybridizes to the +2 position of the trinucleotide initiating sequence; or optionally: provided the DNA template comprises a core promoter of SEQ ID NO: 3 or the promoter comprises a sequence of SEQ ID NOs: 12 or 13, 20 or 21 , 22 or 23 nucleobase Base1hybridizes to the -2 position; nucleobase Base2hybridizes to the -1 position and Base3hybridizes to the +1 position of the trinucleotide initiating sequence. In this embodiment, the TIS may be a shortened sequence, starting with G at +1 and G at +2 (see FIG 4C).

[0092] Certain embodiments relate to the molecular complex, wherein Base0is defined by the formula (III)

[0093] Certain embodiments relate to the molecular complex, wherein Base0is defined by the formula (IV) Certain embodiments relate to the molecular complex, wherein Base0is defined by the formula (V)

[0094] Certain embodiments relate to the molecular complex, wherein Base0is defined by the formula (VI)

[0095] Chen et al. (Chen et al., Structure-Guided Design, Synthesis, and Evaluation of Guanine- Derived Inhibitors of the elF4E mRNA-Cap Interaction, Journal of Medicinal Chemistry (2012), 55, 8) attempted to mimic eukaryotic initiation factor 4E (elF4E), a key regulator of translation initiation, recognizing and binding the 5' cap structure of mRNA to promote protein synthesis bycpe7G (R5= (p-chlorophenoxy)ethyl, as shown in formula (VI). Certain embodiments relate to the molecular complex, wherein Base0is defined by the formula (VII).

[0096] Certain embodiments relate to the molecular complex, wherein Base0is defined by the formula (Certain embodiments relate to the molecular complex, wherein Base0is defined by the formula (VIII) wherein R6is methyl, benzyl, substituted benzyl, propargyl, alkyl or alkylaryl, optionally wherein alkylaryl is substituted, or wherein R6is defined by formula (X)

[0097] (X), or formula (XI)

[0098] Kurpiejewski et al (Kurpiejewski et al., N2 modified cap analogues as translation inhibitors and substrates for preparation of therapeutic mRNA. Ear Biophys J (2023), 52, 511-519) showed that modified caps, whether in the form of nucleoside monophosphates or dinucleotides, may demonstrate favourable biological properties and a strong ability to inhibit translation in the cell-free system. In particular, N2-modified dinucleotides could be efficiently incorporated into mRNA transcripts and, when oriented correctly, rival or surpass ARCA-type analogs. mRNAs bearing caps modified at the exocyclic amino group exhibit markedly enhanced translational activity.

[0099] Certain embodiments relate to the molecular complex, wherein the cap analog is of the formulam7GpppA*pA*pG,m7Gpppm6A*pA*pG orm7GpppBn6A*pA*pG wherein said cap analog forms said complex with a template comprising a T7 <t>6.5-AGG or T7 <t>6.5-AGA promoter sequence.

[0100] Certain embodiments relate to the molecular complex, wherein the cap analog is selected fromm7GpppAmpApG,m7GpppAmpAmpG,m7Gpppm6AmpApG,m7GpppBn6AmpApG,m7Gpppm6AmpAmpG, orm7GpppBn6AmpAmpG and wherein said cap analog forms said complex with a template comprising a T7 <t>2.5-AGG or T7 <t>6.5-AGA promoter sequence.

[0101] Certain embodiments relate to the molecular complex, wherein the cap analog is selected fromm7GpppAmpApU,m7GpppAmpAmpU,m7Gpppm6AmpApU,m7GpppBn6AmpApU,m7Gpppm6AmpAmpU,m7GpppBn6AmpAmpU, and wherein said cap analog forms said complex with a template comprising a T7 <t>6.5-ATG promoter sequence.

[0102] Certain embodiments relate to the molecular complex, wherein the cap analog is selected fromm7GpppAmpUpG,m7GpppAmpUmpG,m7GpppAmpAmpU,m7Gpppm6AmpUpG,m7Gpppm6AmpUmpG,m7Gpppm6AmpAmpU,m7GpppBn6AmpUpG,m7GpppBn6AmpUmpG,m7GpppBn6AmpAmpU, and wherein said cap analog forms said complex with a template comprising a T7 <t>6.5-TGG promoter sequence.

[0103] Certain embodiments relate to the molecular complex, wherein the cap analog is selected fromm7GpppAmpGmpA,m7GpppAmpGpA,m7Gpppm6AmpGmpA,m7Gpppm6AmpGpAm7GpppBn6AmpGmpA orm7GpppBn6AmpGpA, and wherein said cap analog forms said complex with a template comprising a T7 <t>6.5-GAG promoter sequence.

[0104] Certain embodiments relate to the molecular complex, wherein the cap analog is selected fromm7GpppAmpGpG,m7GpppAmpGmpG,m7Gpppm6AmpGpG,m7Gpppm6AmpGmpG,m7GpppBn6AmpGpG orm7GpppBn6AmpGmpG, and wherein said cap analog forms said complex with a template comprising a T7 <t>6.5-GGG or T7 <t>6.5-GGA promoter sequence. Certain embodiments relate to the molecular complex, wherein the cap analog is selected fromm7GpppAmpUpG,m7GpppAmpUmpG,m7Gpppm6AmpUpG,m7Gpppm6AmpUmpG,m7GpppBn6AmpUpG, orm7GpppBn6AmpUmpG, and wherein said cap analog forms said complexwith a template comprising a T7 <t>6.5-TGG orT7 <t>6.5-TGA promoter sequence.

[0105] Certain embodiments relate to the molecular complex, wherein the cap analog is selected fromm7GpppAmpUpA,m7GpppAmpUmpA,m7Gpppm6AmpUpA,m7Gpppm6AmpUmpAm7GpppBn6AmpUpA, orm7GpppBn6AmpUmpA, and wherein said cap analog forms said complex with a template comprising a T7 <t>6.5-TAG promoter sequence.

[0106] Certain embodiments relate to the molecular complex, wherein the cap analog is selected fromm7GpppBn6AmpUpA orm7GpppBn6AmpUmpA, and wherein said cap analog forms said complex with a template comprising a T7 <t>6.5-TAG promoter sequence.

[0107] Certain embodiments relate to the molecular complex, wherein the cap analog is selected fromm7GpppBn6AmpApA,m7GpppBn6AmpAmpA,m7Gpppm6AmpApA,m7Gpppm6AmpAmpA,m7GpppAmpApA, orm7GpppAmpAmpA, and wherein said cap analog forms said complex with a template comprising a T7 <t>6.5-AGG, T7 <t>6.5-AGA, or T7 <t>6.5-AAG promoter sequence.

[0108] As used herein, cap analogs are sometimes denoted with asterisks (e.g. m7GpppA*pA*pG). The use of an asterisk following a nucleotide symbol indicates that the respective nucleotide is chemically modified in its sugar, base, and / or phosphate moiety. Thus, the notation. m7GpppA*pA*pG does not refer to the unmodified tetranucleotide cap analog m7GpppApApG, but rather to its derivatives bearing specific modifications, such as 2'-0-methylation of the ribose, N6-benzyl or N6-methyl substitution of adenosine, or other permissible modifications described herein.

[0109] The T7 polymerase promoter typically spans from nucleotide positions -18 or -17 to +6 relative to the transcription start site is particularly critical for the binding of T7 RNA polymerase. The T7 RNA polymerase binds to the promoter DNA with high specificity between positions -18 or -17 and -5.

[0110] A “T7 polymerase promoter” as used herein refers to any naturally occurring or slightly amended core promoter region. A T7 promoter that can be recognized by T7 RNA polymerase in the presence of a (natural or artificial) cap primer so that transcription is initiated. This region contains specific sequences that are recognized and bound by the polymerase to ensure correct positioning and orientation for transcription initiation. As part of the initiation process, the double-stranded DNA is melted from positions -4 to +3. This melting creates a single-stranded template necessary for the initiation of RNA synthesis. The melting of the DNA at this site allows the polymerase to access the template strand where RNA synthesis begins at the +1 position (Cheetham et al., Structural basis for initiation of transcription from an RNA polymerase-promoter complex. Nature (1999), 399, 80-83; Cheetham et al., Structure of a transcribing T7 RNA polymerase initiation complex. Science (1999), 286, 2305-2309). In the present invention, positions -17 to -1 for T1 to T4, T6 to T10 and T15 and T16, or positions -18 to -1 for T5, and T11 to T14, respectively, relative to a transcription start site are denoted “core promoter region”. Three preferred core promoter regions are defined by SEQ ID NOs: 1 to 3. The subsequent three nucleobases on positions +1 , +2, +3 relative to a transcription start site are denoted as trinucleotide initiating sequence. The transcription start site is defined as the site were transcription is initiated by formation of a transcription complex comprising a promoter sequence and a polymerase. T1 to T16 each including the +1 , +2, +3 relative to the TSS are represented in SEQ ID NOs: 4 to 25, respectively.

[0111] Natural nucleobases consist of the canonical purines - adenine (A) and guanine (G) - and pyrimidines - cytosine (C), thymine (T), and uracil (U) found in the nucleic acids DNA and RNA. These bases engage in specific Watson-Crick base pairing, where adenine pairs with thymine (in DNA) or uracil (in RNA) and cytosine pairs with guanine, facilitated by hydrogen bonds. These bases are designed e.g., to expand the genetic code by introducing new base pairs that are capable of stable and replicable interactions, yet distinct from the traditional base pairing systems. Unnatural bases may engage in alternative hydrogen bonding patterns or utilize other types of chemical interactions such as hydrophobic forces or metal coordination, thereby enabling the creation of novel nucleic acid architectures with enhanced or novel biochemical properties. Modified nucleobases are derivatives of natural nucleobases that have been chemically altered either post-synthetically or through enzymatic processes within cells. Modified nucleobases such as 5-Methylcytosine (5-mC), 5-hydroxymethylcytosine (5-hmC), N6- methyladenine (m6A), N6-benzyladenine, (Bn6A), N6-(2-phenylethyladenine) (Phet6A), hypoxanthin, but not limited thereof, or the addition of complex groups alter the physicochemical properties of the bases. Such modifications can influence DNA stability, the regulation of gene expression, and the epigenetic landscape, and contribute to processes like DNA repair, replication, and cellular differentiation. Notably, modified nucleobases can be strategically utilized to harness specific biological effects, significantly enhancing the efficacy of mRNA therapeutics.

[0112] One embodiment relates to the molecular complex according to the present invention, wherein the DNA template comprises a sequence as defined by any of SEQ ID NOs: 1 to 25, or a sequence having at least 70%, 71 %, 72%, 73%, 74%, 75%, 76%, 77%, 78%, 79%, 80%, 81 %, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98% or preferably at least 99% sequence identity to any of the SEQ ID NOs: 1 to 25, respectively, wherein the sequence comprises a portion spanning positions -1 to -17 upstream of the initiating sequence (-1 to -18 relative to the TSS +1 site for the T5 and T11 to T14 DNA template sequences with SEQ ID NOs: 12 and 13, 20 and 21 , 22 and 23, respectively).

[0113] One embodiment relates to the molecular complex according to the present invention, wherein Base1and Base2of the Cap analog as defined in the first aspect (i) are independently selected from adenine or an analog of adenine, Base3is guanine or an analog of guanine, and wherein the DNA template (ii) comprises a promoter with a core region with a sequence as defined by any one of SEQ ID NO: 1 to 3, is selected from an AGG, an AAG, an ATG, a GGG, a GG, an AGA or a GAG.

[0114] For certain DNA templates, herein and in the literature, two Gs at the +1 and the +2 site of a promoter / DNA template may be rather difficult for IVT reactions, as this configuration may form some tertiary structures in the presence of metal cations. The present inventors thus systematically studied the influence and importance of the +1

[0115] In certain embodiments, the promoter is a phage-derived promoter, preferably, a <t>2.5- promoter, more preferably a <t>6.5-promoter. Preferable embodiments of these two promoter types are illustrated in FIG 4.

[0116] In certain embodiments, the DNA template comprises at least one synthetic and / or at least one unnatural nucleoside.

[0117] Synthetic nucleoside analogs play vital roles in expanding the applications of DNA in research, diagnostics and therapy. Notable examples include BrdU (5-Bromo-2'- deoxyuridine) and EdU (5-ethynyl-2'-deoxyuridine). Locked Nucleic Acids (LNAs) and Phosphorothioate nucleotides introduce backbone modifications that enhance duplex stability and resistance to nucleases. Moreover, unnatural nucleosides like dNaM, dTPT3, NaM, TPT3, dMMO2Bi0, d5SICS dNaM have been developed to potentially expand the genetic code, enabling the creation of novel proteins and biomolecules in synthetic biology (Fischer et al., New codons for efficient production of unnatural proteins in a semisynthetic organism. Nat Chem Biol (2020), 16, 570-576).

[0118] One embodiment relates to the molecular complex according to the present invention, wherein Base1is independently adenine, N6-benzyladenine, N6-methyladenine, N6-(2- phenylethyl)adenine, and R1 and / or R2 are independently O-Me, and (i) Base2and Base3are both guanine, or (ii) Base2is adenine and Base3is guanine, or (iii) Base2and Base3are independently guanine or adenine, or (iv) Base2is adenine and Base3is uracil, or (v) Base2is uracil and Base3is guanine, or (vi) Base2is uracil and Base3is adenine, or (vii) Base2is adenine and Base3is adenine, in the Cap analog according to the present invention.

[0119] One embodiment relates to the molecular complex according to the present invention, wherein (i) Base1isBn6A and R1 is O-Me, Base2and Base3are independently guanine, or (ii) Base1ism6A, R1 is O-Me, Base2and Base3are independently guanine, or (iii) Base1isBn6A, R1 is O-Me, Base2is adenine and Base3is guanine, or (iv) Base1ism6A, R1 is O- Me, Base2is adenine and Base3is guanine, or (v) Base1is A, R1 is O-Me, Base2is G, R1 is O-Me, and Base3is guanine, or (vi) Base1isBn6A, R1 is O-Me, Base2is G, R2 is O- Me, and Base3is G, or (vii) Base1ism6A, R1 is O-Me, Base2is G, R2 is O-Me and Base3is guanine, or (viii) Base1is A, R1 is O-Me, Base2is A, R2=O-Me, and Base3is guanine, or (ix) Base1is Am, Base2is adenine, and Base3is guanine, or (x) Base1isBn6A, R1 is O- Me, Base2is A, R2 is O-Me, and Base3is guanine, or (xi) Base1ism6A, R1 is O-Me, Base2is A, R1 is O-Me and Base3is guanine, or (xii) Base1is A, R1 is O-Me, Base2and Base3are both guanine, (xiii) Base1is N6-(2-phenylethyl)adenine, R1 is O-Me, preferably Base2and Base3are independently guanine, more preferably Base 2 is adenine and Base 3 is guanine.

[0120] In certain embodiments, the cap ism7GpppA*pA*pG,m7GpppA*pA*pU*,m7GpppA*pA*pC orm7GpppA*pU*pG,m7GpppA*pU*pA, m7GpppA*pA*pA, each independently comprising a natural, unnatural or modified base, but not limited thereof.

[0121] Different types of benzyl derivatives added to the N6-position of 2'-0-methyladenosine may also be biologically active, comprising a para-substituted benzyl group, such as p- methylbenzyl, p-methoxybenzyl, p-nitrobenzyl, p-hydroxybenzyl, or p-fluorobenzyl, a meta-substituted benzyl group, such as m-methylbenzyl or m-chlorobenzyl, an orthosubstituted benzyl group, such as o-methylbenzyl, o-hydroxybenzyl, a di-substituted benzyl group, such as 2,4-dichlorobenzyl, or a heteroaromatic benzyl group, such as thiophenylmethyl or 2-pyridylmethyl, a benzyl homologue, such as homobenzyl (2- phenylethyl), 3-phenylpropyl, or arylalkyl derivatives such as a-napthylmethyl, p- naphtylmethyl, napthylethyl.

[0122] In certain embodiments, the N6-position can also be advantageously modified by an alkylaryl substituent comprising a fluorescent dye such as pyrene, coumarine, fluorescein, rhodamine or cyanine.

[0123] One embodiment relates to the molecular complex according to the present invention, additionally comprising a T7 RNA polymerase and / or a homolog thereof and / or a modified variant thereof. T7 RNA polymerase is a highly specific enzyme from the T7 bacteriophage that is instrumental in transcribing DNA into RNA. It is part of a broader family of single-subunit RNA polymerases (RNAPs) that share structural and functional characteristics. This family includes other phage RNAPs like T3, T6, K11 , SP6, and N4, which are similar to T7 RNA polymerase in terms of their mechanism of action and specificity for their respective phage DNA. Additionally, this family encompasses mitochondrial RNA polymerases, which are crucial for transcription within mitochondria.

[0124] Enzymes, including polymerases, can be artificially modified through several advanced techniques to enhance their properties, such as stability, activity, and specificity, suiting them for various applications in biotechnology and medicine. Traditional DNA polymerases, while efficient, often lack the ability to perform certain desired functions that new biotechnological applications demand. To address this, innovative engineering methods such as mutagenesis and the creation of protein chimeras have been employed. These methods have allowed scientists to expand the functionality of DNA polymerases beyond their natural capacities. For instance, engineered polymerases can now tolerate harsh conditions and the presence of inhibitors, and are capable of replicating DNA templates that contain chemically modified nucleotides. A key aspect of this engineering effort is the development of novel selection techniques, such as using water-in-oil emulsions that link genotype to phenotype more effectively than traditional methods like phage display. This approach has facilitated the exploration of an extended sequence space, leading to the discovery of polymerases that are more resilient and versatile. The continuous expansion of the "polymerase universe" is in step with the increasing chemical diversity of modified nucleotides, enabling the development of DNA polymerases with unnatural or enhanced functions that are pivotal for advancing biotechnological applications (Coulther el al., Engineering polymerases for new functions, Trends in Biotechnology (2019), 37, 10, 1091-1103).

[0125] A second aspect relates to a vector, or more than one vector, for use in forming a molecular complex, wherein the at least one vector comprises the at least one DNA template according to the present invention.

[0126] A vector designed for mRNA production is a DNA construct engineered to facilitate the synthesis of specific mRNA molecules. It comprises certain genetic elements, including a promoter, coding sequence, optionally a polyadenylation signal, and may incorporate other components like selection markers and origins of replication. A vector according to the present invention are based on plasmids. These plasmid backbones contain essential elements required for the replication and maintenance of the vector within the host cell. A third aspect relates to a DNA molecule, or more than one than one DNA molecule, for use in forming a molecular complex, wherein the at least one DNA molecule comprises the at least one DNA template according to the present invention.

[0127] In the production of mRNA through IVT, various templates can be utilized aside from commonly used plasmid vectors, each offering unique advantages depending on the specific needs of the research or production process. PCR-generated templates are often used for rapid and cost-effective mRNA synthesis, involving a DNA fragment amplified by PCR that includes the appropriate promoter upstream of the gene of interest. This method circumvents the need for cloning, expediting the process. Alternatively, plasmid DNA can be linearized using restriction enzymes that cut downstream of the insert sequence to provide a clean and efficient template for IVT.

[0128] Synthetic gene blocks, which are short, double-stranded DNA molecules including a promoter sequence and the gene of interest, can be ordered from commercial suppliers and are ready to use without further preparation. Less commonly, single-stranded DNA (ssDNA) with the necessary promoter sequences can also serve as a template, though additional steps may be required to ensure stability and efficiency. Bacteriophage-derived templates, such as M13 or other filamentous phage DNA, can be engineered to contain the gene of interest under an appropriate promoter, offering another method for producing template DNA for IVT. For small-scale synthesis or short mRNA transcripts, synthetic oligonucleotides that include the necessary promoter and coding region can be used directly in IVT reactions.

[0129] A fourth aspect relates to a use of the molecular complex according to the present invention for in vitro transcription, wherein the capping efficiency is at least 80%, 81 %, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or up to 100%.

[0130] In the present invention, the capping efficiency is measured according to the methodology described in Example 8.

[0131] A fifth aspect relates to a method of producing at least one m7G-capped mRNA molecule, the method comprising: (i) providing a reaction mixture comprising the Cap analog as defined in the first aspect or any embodiment thereof, the DNA template and / or the vector and / or the DNA molecule according to the present invention and at least one T7 RNA polymerase and / or the sequence encoding the same; (ii) allowing in vitro transcription from said DNA template and / or from said vector and obtaining at least one m7G-capped mRNA; (iii) optionally: purifying said at least one m7G-capped mRNA; and / or (iv) optionally: formulating said at least one m7G-capped mRNA molecule yielding a pharmaceutical composition.

[0132] Example 3 describes one preferable method of producing the at least one m7G-capped mRNA molecule.

[0133] One embodiment relates to the method of the fifth aspect further comprising a step of: (v) introducing or contacting said at least one m7G-capped mRNA molecule obtained in step (ii) into or with at least one target cell in an in vitro cellular system for transfecting said at least one target cell and / or for in vitro studying the functionality of said at least one m7G- capped mRNA.

[0134] A sixth aspect relates to a pharmaceutical composition for use in a method of preventing and / or treating a disease, the pharmaceutical composition comprising the at least one m7G-capped mRNA molecule obtained or obtainable by the method according to the fifth aspect or the embodiment thereof, additionally comprising at least one pharmaceutically acceptable carrier or excipient.

[0135] In certain embodiments, the pharmaceutically acceptable carrier or excipient comprises at least one bulking agent and / or at least one binder and / or at least one disintegrant and / or at least one coating and / or at least one filler or diluent and / or at least one stabilizer or preservative and / or at least one solvent and / or at least one solubilizer.

[0136] One of the primary functions of a pharmaceutically acceptable carrier or excipient is to provide the necessary bulk and form to create various dosage forms, including suspensions, or injectable solutions, ensuring the integrity of the final product. Additionally, carriers or excipients play a pivotal role in enhancing the stability of the active pharmaceutical ingredient (API), shielding it from factors like light, heat, moisture, or chemical reactions that could lead to degradation.

[0137] A critical aspect of these carrier or excipients is their ability to adjust the pH of the formulation to levels conducive to the stability and efficacy of the API. Importantly, these substances must not interact adversely with the API or other components, ensuring the safety and integrity of the medicinal product. Pharmaceutically acceptable carriers and excipients adhere to stringent quality and safety standards, ensuring they do not introduce impurities, contaminants, or adverse effects into the pharmaceutical product. They are indispensable in drug development, allowing for the creation of formulations suitable for patient administration, ultimately delivering the desired therapeutic effect while upholding safety and stability standards. One embodiment relates to the pharmaceutical composition for use in a method of preventing and / or treating a disease according to the sixth aspect, the pharmaceutical composition additionally comprising at least one lipid nanoparticle, wherein the at least one lipid nanoparticle is selected from the group of at least one liposome (LPs), at least one liposome-like nanoparticle (LLP), at least one solid lipid nanoparticle (SLN), at least one nanostructured lipid carrier (NLC), at least one lipid-polymer hybrid nanoparticle (LPN), at least one lipoprotein particle (LPT), at least one nanoemulsion, at least one cationic nanoemulsion (CNE) and at least one exosome.

[0138] Lipid nanoparticles (LNPs) are fundamental components in the field of mRNA delivery, playing a crucial role in the transportation and protection of mRNA molecules for a variety of biotechnological and therapeutic applications.

[0139] One of the primary functions of LNPs is to provide a protective encapsulation for mRNA. This encapsulation shields the mRNA from degradation by nucleases, ensuring the genetic information remains intact during transportation. Additionally, LNPs facilitate the efficient cellular uptake of mRNA by cells, typically through endocytosis, allowing the encapsulated mRNA to enter the cytoplasm where translation and protein synthesis occur.

[0140] LNPs contribute to the stability and bioavailability of mRNA therapeutics, safeguarding the mRNA cargo from degradation by the recipients’ immune system and enhancing its pharmacokinetics. Thus, an essential attribute of LNPs is their minimal immunogenicity when properly designed, reducing the risk of undesirable immune responses against the delivery system itself. Furthermore, they can be engineered for tissue-specific targeting, enabling the targeting of specific cell types or organs for therapeutic purposes.

[0141] A seventh aspect relates to a kit comprising the Cap analog according to the present invention, optionally comprising the DNA template according to the present invention, and / or the vector and / or the DNA molecule according to the present invention, optionally wherein the kit comprises at least one further reagent and / or at least one aqueous solution, preferably a buffer.

[0142] Whenever the present disclosure relates to the percentage of identity of nucleic acid or amino acid sequences to each other these values define those values as obtained by using the EMBOSS Water Pairwise Sequence Alignments (nucleotide) programme (https: / / www.ebi.ac.uk / idispatcher / psa / emboss water ) nucleic acids or the EM-BOSS Water Pairwise Sequence Alignments (protein) programme (www.ebi.ac.uk / Tools / psa / emboss_water / ) for amino acid sequences. Alignments or sequence comparisons as used herein referto an alignment overthe whole length of two sequences compared to each other. Those tools provided by the European Molecular Biology Laboratory (EMBL) European Bioinformatics Institute (EBI) for local sequence alignments use a modified Smith-Waterman algorithm (see https: / / www.ebi.ac.uk / idispatcher / psa and SMITH, T.F. 5 & WATERMAN, M.S. "Identification of common molecular subsequences" Journal of Molecular Biology, 1981 147 (1 ): 195-197). When conducting an alignment, the default parameters defined by the EMBL-EBI are used.

[0143] Those parameters are (i) for amino acid sequences: Matrix = BLOSUM62, gap open penalty = 10 and gap extend penalty = 0.5 or (ii) for nucleic acid sequences: Matrix = DNAfull, gap open penalty = 10 and gap extend penalty = 0.5. The skilled person is well aware of the fact that, for example, a nucleotide sequence encoding a protein can be “codon optimized” if the respective sequence is to be used in another organism in comparison to the original organism a molecule originates from.

[0144] The present invention is further illustrated by the following, non-limiting examples.

[0145] Examples

[0146] Example 1 Synthesis of tetranucleotide cap analogs.

[0147] Example 1A: General aspects

[0148] The tetranucleotide cap analogs were synthesized by modified previously reported synthesis of di- and trinucleotide cap analogues. Warminski et al. “Trinucleotide mRNA Cap Analogue N6-Benzylated at the Site of Posttranscriptional m6Am Mark Facilitates mRNA Purification and Confers Superior Translational Properties In Vitro and In Vivo” J. Am. Chem. Soc. vol. 146, pp. 8149-8163 (2024).) The multistep synthesis of tetranucleotide cap analogs sodium salts consisted of: solid-phase synthesis (SPS) of 5'- phosphorylated trinucleotide, its activation with imidazole and coupling with 7- methylguanosine 5'-diphosphate analogue followed by ion-exchange DEAD Sephadex chromatography, purification by RP HPLC (C18) and multiple freeze drying with addition of NaHCO3 aqueous solution. The N6-benzyladenosine phosphoramidite and N6- methyladenosine phosphoramidite for solid-phase synthesis were prepared according to previously described protocols. See Warminski et al. “Quick Access to Nucleobase- Modified Phosphoramidites for the Synthesis of Oligoribonucleotides Containing Post- Transcriptional Modifications and Epitranscriptomic Marks” J. Org. Chem. vol. 87, pp. 10333-10348 (2022).

[0149] General procedure for synthesizing 5'-monophosphorylated ribotrinucleotides (A*pG*pG or A*pA*pG): Synthesis was performed in a 50 mL glass solid-phase vessel equipped with a frit filter. First coupling: Primer Support 5G RNA (308 / 303 pmol / g, GE Healthcare) resin was placed in SPS vessel and it was sealed with Precision Seal® rubber septa, and purged with argon. Then a 0.3 M solution of appropriate phosphoramidite (1.5 equiv, G, Gm, A, Am) in dry acetonitrile and BTT activator (2.25 equiv, 0.3 M 5-(benzylthio)-1 H- tetrazole solution in acetonitrile) were added. The suspension was gently shaken for 1 hour at room temperature. The resin was washed with acetonitrile, 3% (w / v) trichloroacetic acid solution in dichloromethane (detritylation), 0.05 M iodine in pyridine / water (9:1) (oxidation), acetonitrile and dried under vacuum for 1 hour. Second and third coupling was performed in the same manner as first, using 1 .5 equiv of A, Am, Bn6-,Am, m6A phosphoramidite or 2.5 equiv of bis(2-cyanoethyl)-N,N-diisopropyl phosphoramidite. After the last coupling, trinucleotides, still attached to the solid support, were washed with 20% (v / v) diethylamine in acetonitrile, acetonitrile and dried under vacuum. T rinucleotides were cleaved from solid support and deprotected by AMA solution (methylamine / ammonium hydroxide 1 :1 (v / v)) at 55 °C for 1 h. Solution was evaporated. The TBDMS were removed using of TEA 3HF (1.55 equiv), TEA (1 .55 equiv) in dry DMSO (0.1 M). Mixture was stirred at 60 °C for 1 h. The mixture was 10x diluted with MQ water. The product was isolated by ion-exchange chromatography on DEAE Sephadex (elution gradient: 0-1 .2 M TEAB buffer) to afford trinucleotides as triethylammonium salts. The synthesis scales, chemical structure, yield, HPLC and LRMS data for particular trinucleotides are summarized in Table 2 and 3.

[0150] Table 2. 5 '-Phosphorylated trinucleotide analogs. Table 3. Structure of synthesized tetranucleotide cap analogs.

[0151]

[0152]

[0153]

[0154]

[0155] Example 1 B: General procedure for synthesis of tetranucleotide cap analogs:

[0156] Step A: Synthesis of phosphorimidazolide intermediate: To a 0.05 M solution of trinucleotide 5'phosphate triethylammonium salt in dry DMF, 2,2'-dithiodipiridine (6 equiv), imidazole (16 equiv) and triethylamine (6 equiv) were added. Mixture was stirred at room temperature for 15 minutes. Then triphenylphosphine (6 equiv) was added. The clear yellow solution was stirred at room temperature till full conversion of the trinucleotide (progress on RP-HPLC, 1-24 h). The product was precipitated by addition of cold solution of NaCIC (10 equiv) in acetone (10 times the volume of DMF used). The precipitate was centrifuged (4 °C), washed twice with cold acetonitrile and dried overnight under vacuum. The phosphorimidazolide intermediate without further purification was used in next step. Step B: Coupling reaction of activated trinucleotide with m7GDP / TEA or m27 2°GDP / TEA: The phosphorimidazolide derivative of trinucleotide (1 equiv) was added to a DMF solution of triethylammonium salt of 7-methylguanosine 5'-diphosphate (m7GDP / TEA) or 2'O-modified 7-methylguanosine 5'-diphosphate (m27 2°GDP / TEA) (1 .2 equiv) and ZnCL (12 equiv). The reaction mixture was stirred for 1-4 hours, then it was quenched by adding a aqueous solution (10 times the volume of DMF used) of N32EDTA (50 mg / mL) and NaHCO3 (25 mg / mL). The product was isolated by ion-exchange chromatography (DEAE Sephadex, gradient elution 0-1 .2 M TEAB) and then purified by RP-HPLC (C18, linear gradient of acetonitrile in aqueous ammonium acetate buffer). After lyophilization tetranucleotide cap analogs were obtained as ammonium salts.

[0157] Step C: Exchange of tetranucleotide cap analogs ammonium salts to sodium salts: To an aqueous solution of ammonium salts of tetranucleotide caps (0.04 M) an aqueous solution of NaHCCh (1 M, 3.5 equiv) was added. Mixture was vortexed, centrifuged and freeze- dried twice to get sodium salts of tetranucleotide cap analogs as colourless powder. rn7Gpppbn6AmpGpG: Total yield: 27%. RP-HPLC: Rt= 7.959 min.1H NMR (500 MHz, D2O, 70 °C) 69.07 (s, 1H), 8.48 (s, 1H), 8.24 (s, 1H), 8.07 (s, 1H), 8.05 (s, 1H), 7.44-7.28 (m, 5H), 6.02 (d, J = 6.2 Hz, 1 H), 5.91 (d, J = 4.3 Hz, 1 H), 5.85 (d, J = 5.4 Hz, 1 H), 5.80 (d, J = 5.9 Hz, 1H), 4.94-4.89 (m, 1H), 4.88-4.84 (m, 3H), 4.82-4.77 (m, 1H), 4.70 (t, J = 5.4 Hz, 1H), 4.62 (t, J = 4.3 Hz, 1H), 4.50-4.40 (m, 6H), 4.25-4.13 (m, 7H), 4.03 (s, 3H), 3.34 (s, 3H).31P NMR (203 MHz, D2O, 70 °C): 6 = 0.82 (s, 1P), 0.40 (s, 1P), -10.06 (d, J = 19.0 Hz, 1 P), -10.18 (d, J = 18.4 Hz, 1 P), -21.52 (dd, J = 19.0, 18.4 Hz, 1 P). LRMS ESI(- ): m / z 789.2 (calcd for C49H59N20O31P52- [M-2H]2' 789.1). rn7GpppAmpGmpG: Total yield: 23%. RP-HPLC: Rt= 5.469 min;1H NMR (500 MHz, D2O) 69.42 (s, 1H), 8.83 (s, 1H), 8.56 (s, 1H), 8.37 (s, 1H), 8.31 (s, 1H), 6.31 (d, J = 5.5 Hz, 1H), 6.23 (d, J = 4.0 Hz, 1H), 6.16 (d, J = 5.4 Hz, 1H), 6.12 (d, J = 6.0 Hz, 1H), 5.20 (ddt, J = 8.0, 5.0, 3.2 Hz, 2H), 5.04 (t, J = 5.4 Hz, 1H), 4.93 (dd, J = 5.0, 4.0 Hz, 1H), 4.91 - 4.88 (m, 1 H), 4.82 - 4.74 (m, 3H), 4.66 - 4.59 (m, 3H), 4.57 - 4.53 (m, 1 H), 4.48 (ddt, J = 17.0, 12.1 , 4.1 Hz, 5H), 4.35 (s, 3H), 3.69 (s, 3H), 3.66 (s, 3H).31P NMR (203 MHz, D2O) 6 -0.49 (s, 1 P), -0.67 (s, 1 P), -10.70 - -11.50 (m, 2P), -22.38 - -22.59 (m, 1 P). LRMS ESI(- ): m / z 750.8 (calcd for C43H55N2o03iP52- [M-2H]2' 751.1). m272'°GpppAmpGmpG: Total yield: 22% RP-HPLC: ft = 5.527 min; 1H NMR (500 MHz, D2O) 69.08 (s, 1H), 8.46 (s, 1H), 8.18 (s, 1H), 8.01 (s, 1H), 7.87 (s, 1H), 5.97 (d, J = 4.8 Hz, 1H), 5.91 (d, J = 3.0 Hz, 1H), 5.86 (d, J = 5.6 Hz, 1H), 5.73 (d, J = 6.0 Hz, 1H), 4.93 - 4.86 (m, 2H), 4.63 (t, J = 5.6 Hz, 1 H), 4.55 (t, J = 5.5 Hz, 1 H), 4.49 (dd, J = 5.2, 3.9 Hz, 1H), 4.45 (d, J = 3.7 Hz, 1H), 4.42 - 4.07 (m, 12H), 4.05 (s, 3H), 3.54 (s, 3H), 3.39 (d, J = 7.2 Hz, 6H), 2.04 (s, 1H), 2.00 (s, 1H).31P NMR (203 MHz, D2O) 60.09 (s, 1P), -0.08 (s, 1P), -10.52 (d, J = 18.6 Hz, 1P), -10.64 (d, J = 17.9 Hz, 1P), -21.91 (t, J = 18.1 Hz, 1P). LRMS ESI(-): m / z 757.8 (calcd for C44H57N2o03iP52- [M-2H]2' 758.1). rn7Gpppbn6AmpGmpG: Total yield: 37%. RP-HPLC: ft = 7.737 min;1H NMR (500 MHz, D2O) 69.36 (s, 1H), 8.75 (s, 1H), 8.33 (s, 1H), 8.30 (s, 1H), 7.68 - 7.52 (m, 6H), 6.31 (d, J = 5.9 Hz, 1H), 6.18 (d, J = 4.0 Hz, 1H), 6.13 (t, J = 6.2 Hz, 2H), 5.21 (dq, J = 8.2, 3.9 Hz, 2H), 5.11 (s, 2H), 5.02 (t, J = 5.4 Hz, 1H), 4.92 - 4.85 (m, 2H), 4.83 - 4.72 (m, 6H), 4.61 (qd, J = 4.6, 1.9 Hz, 3H), 4.56 - 4.43 (m, 5H), 4.31 (s, 3H), 3.69 (s, 3H), 3.66 (s, 3H), 2.28 (s, 1H).31P NMR (203 MHz, D2O) 6 -0.49 (s, 1P), -0.64 (s, 1P), -11.12 (t, J = 18.7 Hz, 2P), -22.51 (t, J = 18.0 Hz, 1P). LRMS ESI(-): m / z 796.3 calcd for C5OH6I N20O31 P52' [M-2H]2' 796.2). rn7GpppApGmpG: Total yield: 36%. RP-HPLC: ft = 4.903 min;1H NMR (500 MHz, D2O, 60 °C) 68.69 (s, 1H), 8.45 (s, 1H), 8.27 (s, 1H), 8.21 (s, 1H), 6.23 (d, J= 6.0 Hz, 1H), 6.20 (d, J= 4.0 Hz, 1H), 6.13 (dd, J= 5.7, 2.5 Hz, 2H), 5.21 (s, 1H), 5.09 - 5.01 (m, 2H), 4.99 (d, J= 5.6 Hz, 1H), 4.90 (t, J= 4.5 Hz, 1H), 4.84 (t, J= 5.5 Hz, 1H), 4.76 (t, J= 4.8 Hz, 4H), 4.60 (s, 4H), 4.48 (s, 8H), 4.33 (s, 3H), 4.22 - 4.09 (m, 1H), 3.69 (s, 3H).31P NMR (203 MHz, D2O, 60 °C) 6 -0.37 (s, 1P), -0.46 (s, 1P), -11.10 (dd, J= 18.2, 8.0 Hz, 2P), -22.55 (t, J= 18.6 Hz, 1P). LRMS ESI(-): m / z 744.5 (calcd for C42H53N2o03iP52- [M- 2H]2' 745.1). rn7Gpppm6ApGmpG: Total yield: 37%. RP-HPLC: Rt= 5.170 min;1H NMR (500 MHz, D2O, 60 °C) 68.70 (s, 1H), 8.50 (s, 1H), 8.42 - 8.21 (m, 2H), 6.28 (dd, J= 15.2, 4.9 Hz, 2H), 6.20 (dd, J= 5.9, 2.6 Hz, 2H), 5.32 - 5.22 (m, 1H), 5.19 - 5.03 (m, 5H), 5.00 - 4.89 (m, 10H), 4.72 - 4.47 (m, 5H), 4.39 (s, 3H), 3.76 (s, 3H), 3.49 (s, 3H).31P NMR (203 MHz, D2O, 60 °C) 60.53 (d, J = 15.7 Hz), -9.71 - -10.53 (m), -21.63 (t, J = 18.6 Hz). LRMS ESI(-): m / z 751.3 (calcd for C43H55N20O31P52- [M-2H]2' 751.5). rrfGpppAmpGpG: Total yield: 23%. RP-HPLC: ft = 4.765 min;1H NMR (500 MHz, D2O, 60 °C) 69.48 (s, 1H), 8.88 (s, 1H), 8.63 (s, 1H), 8.41 (s, 1H), 8.34 (s, 1H), 6.38 (d, J = 5.6 Hz, 1H), 6.30 (d, J = 3.9 Hz, 1H), 6.22 (d, J = 5.3 Hz, 1H), 6.14 (d, J = 5.8 Hz, 1H), 5.28 (dt, J = 8.1, 4.1 Hz, 1H), 5.22 (t, J = 5.6 Hz, 1H), 5.16 (dt, J = 8.4, 4.3 Hz, 1H), 5.08 (t, J = 5.4 Hz, 1 H), 5.00 (t, J = 4.5 Hz, 1 H), 4.85 (p, J = 5.0 Hz, 3H), 4.74 - 4.66 (m, 2H), 4.66 - 4.48 (m, 6H), 4.42 (s, 3H), 3.74 (s, 3H).31P NMR (203 MHz, D2O, 60 °C) 60.74 (s, 1P), 0.30 (s, 1P), -10.17 (dd, J = 26.8, 18.2 Hz, 2P), -21.55 (t, J = 18.1 Hz, 1P). LRMS ESI(-): m / z 743.9 (calcd for C42H53N20O31P52- [M-2H]2' 744.2). nfGpppAmpAmpG: Total yield: 42%. RP-HPLC: ft= 5.842 min;1H NMR (500 MHz, D2O) 68.41 (s, 1H), 8.26 (s, 1H), 8.22 (s, 1H), 7.95 (s, 1H), 7.89 (s, 1H), 6.02 (d, J = 3.7 Hz, 1H), 5.93 (d, J = 4.7 Hz, 1H), 5.74 (dd, J = 6.8, 4.5 Hz, 2H), 4.91 - 4.82 (m, 2H), 4.64 (t, J = 5.2 Hz, 1H), 4.50 (dt, J = 8.7, 4.7 Hz, 2H), 4.43 (ddt, J = 15.9, 10.1, 5.2 Hz, 4H), 4.34 (ddt, J = 13.6, 6.0, 3.8 Hz, 6H), 4.27 - 4.17 (m, 7H), 3.98 (s, 3H), 3.55 (s, 3H), 3.52 (s, 3H).31P NMR (203 MHz, D2O) 6 -0.05 (s, 1P), -0.30 (s, 1P), -10.49 (d, J = 18.6 Hz, 1P), -10.70 (d, J = 17.7 Hz, 1P), -21.83 (t, J = 18.0 Hz, 1P). LRMS ESI(-): m / z 742.8 calcd for C42H53N20O31 P52' [M-2H]2' 744.2). rn7Gpppbn6AmpApG: Total yield: 36%. RP-HPLC: ft= 7.788 min;1H NMR (500 MHz, D2O) 69.05 (s, 1 H), 8.42 (s, 1 H), 8.30 (s, 1 H), 8.24 (s, 1 H), 7.93 (d, J = 7.6 Hz, 2H), 7.24 - 7.15 (m, 2H), 7.18 (s, 3H), 5.92 (dd, J = 13.8, 4.6 Hz, 2H), 5.83 (d, J = 3.7 Hz, 1H), 5.77 (d, J = 5.0 Hz, 1 H), 4.88 (dt, J = 8.3, 4.2 Hz, 1 H), 4.68 - 4.59 (m, 2H), 4.58 - 4.42 (m, 7H), 4.38 - 4.12 (m, 7H), 4.02 - 3.98 (m, 3H), 3.51 (s, 3H).31P NMR (203 MHz, D2O) 60.12 (s, 1P), -0.18 (s, 1P), -10.51 (t, J = 19.3 Hz, 2P), -21.83 (t, J = 18.0 Hz, 1P). LRMS ESI(-): m / z 781.4 (calcd for C49H59N20O30P52- [M-2H]2' 781.1). rn7GpppAmpApG: Total yield: 17%. RP-HPLC: ft = 4.865 min;1H NMR (500 MHz, D2O) 6 9.05 (s, 1 H), 8.48 (s, 1 H), 8.30 (s, 1 H), 8.24 (s, 1 H), 7.99 (s, 1 H), 7.89 (s, 1 H), 6.04 - 5.87 (m, 2H), 5.71-5.67 (m 2H), 4.94 - 4.84 (m, 1 H), 4.67 - 4.60 (m, 1 H), 4.58 - 4.10 (m, 13H), 3.96 (s, 3H), 3.57 (s, 3H).31P NMR (203 MHz, D2O) 6 0.15 (s, 1 P), -0.47 (s, 1 P), - 10.45 (d, J = 18.1 Hz, 1 P), -10.87 (d, J = 17.3 Hz, 1 P), -21.62 (t, J = 17.6 Hz, 2P). LRMS ESI(-): m / z 735.8 (calcd for C42H53N2o03oP52- [M-2H]2' 736.1). rn7Gpppbn6AmpAmpG: Total yield: 16%. RP-HPLC: ft = 8.979 min;1H NMR (500 MHz, D2O) 6 9.37 (s, 1 H), 8.79 (s, 1 H), 8.77 (s, 1 H), 8.64 (s, 1 H), 8.45 (s, 1 H), 8.36 (s, 1 H), 7.64 - 7.45 (m, 5H), 6.37 (d, J = 5.2 Hz, 1 H), 6.28 (d, J = 6.5 Hz, 1 H), 6.19 (d, J = 4.0 Hz, 1 H), 6.13 (d, J = 5.4 Hz, 1 H), 5.25 (dt, J = 8.4, 4.5 Hz, 1 H), 5.19 (ddd, J = 7.5, 4.7, 2.6 Hz, 1 H), 5.08 (s, 2H), 5.01 (t, J = 5.3 Hz, 1 H), 4.89 (t, J = 4.5 Hz, 1 H), 4.85 - 4.74 (m, 5H), 4.60 (dq, J = 5.2, 3.2 Hz, 3H), 4.56 - 4.42 (m, 8H), 4.31 (s, 3H), 3.74 (s, 3H), 3.64 (s, 3H).31P NMR (203 MHz, D2O) 6 -0.09 (s, 1 P), -0.21 (s, 1 P), -10.54 (dd, J = 28.5, 18.5 Hz, 2P), -21.76 (t, J = 16,1 Hz, 1 P). LRMS ESI(-): m / z 788.5 (calcd for C5OH6I N2o030P52- [M-2H]2' 788.2).

[0158] Example 2: General procedure for templates preparation.

[0159] Wild type (WT) or optimized luciferase gene from firefly (Lampyridae) (FFLuc) subcloned into pJET1.2 plasmid vector was linearized using type IIS restriction enzyme, e.g. Eam1 104l or Aarl (ThermoFisher Scientific) generating 5' overhangs. Completeness of the linearization was verified on the 1 xTBE 1 % agarose gel and the linearized plasmid was purified using a commercial DNA purification kit (Macherey-Nagel).

[0160] PCR templates were prepared using pJET1 ,2-FFLuc WT or pJET1 ,2-FFLuc optimized circular plasmids as a templates, Platinum™ SuperFi™ II DNA Polymerase (Invitrogen™), forward primer covering T7 promoter, TSS and several additional nucleotides downstream 5' end, and reverse primer containing polyT sequence and several additional nucleotides downstream 3'UTR. Next, the PCR mix was purified using GeneJET PCR Purification Kit (ThermoFisher Scientific) and template purity was verified on the 1 xTBE 1 .2% agarose gel.

[0161] Example 3: General procedure for in vitro transcription of mRNAs using tetranucleotide primers m7GpppA*pG*pG and m7GpppA*pA*pG; see also FIG 5, FIG 7-11

[0162] The dsDNA template encoding FFLuc was prepared by plasmid linearization (see also Example 2) at the site following polyA tail-coding region. Three different templates were used for these experiments: 1) template T1 (<t>6.5-AGG), which contained <t>6.5 promoter followed by AGG deoxyribonucleotides at positions +1 , +2 and +3 of the coding strand (in this case non-coding strand contained deoxyribonucleotides TTCC corresponding to the positions -1 , +1 , +2 and +3); 2) template T2 (<t>2.5-AGG), which contained <t>2.5 promoter followed by AGG nucleotides at positions +1 , +2 and +3 of the coding strand (in this case non-coding strand contained deoxyribonucleotides ATCC corresponding to the positions -1 , +1 , +2 and +3); 3) template T3 (<t>6.5-GGG), which contained <t>6.5 promoter followed by GGG nucleotides at positions +1 , +2 and +3 of the coding strand (in this case noncoding strand contained deoxyribonucleotides TCCC corresponding to the positions -1 , +1 , +2 and +3). In vitro transcription reactions for 10, 5 and 2 mM of tetranucleotide primer rn7GpppAmpGmpG, m7GpppAmpAmpG, rn7GpppAmpGpG, rn7GpppAmpApG, m7GpppBn6AmpGmpG, m7GpppBn6AmpAmpG, rn7GpppBn6AmpGpG or rn7GpppBn6AmpApG and for 10, 5 and 2 mM Bn6-containing trinucleotide primer rn7GpppBn6AmpG were prepared using 40 mM Bis-Tris buffer pH 6.5, containing 2 mM spermidine, and 5 mM each of ATP, UTP, CTP, 4 mM GTP, 25 mM MgCh, 10 mM DTT, 40 ng / pL linearized plasmid as the DNA template, 1 U / pL RNase inhibitor (RiboLock, ThermoFisher Scientific), 0.002 U / pL inorganic pyrophosphatase (ThermoFisher Scientific) and 75 U / pL T7 RNA polymerase (Roche). In vitro transcription reactions for 10, 5 and 2 mM of trinucleotide cap 1 primer rn7GpppAmpG were prepared using 40 mM Tris buffer pH 8.0, containing 2 mM spermidine, and 5 mM each of ATP, UTP, CTP, 4 mM GTP, 10 mM MgCh, 10 mM DTT, 40 ng / pL linearized plasmid as the DNA template, 1 U / pL RNase inhibitor (RiboLock, ThermoFisher Scientific), 0.002 U / pL inorganic pyrophosphatase (ThermoFisher Scientific) and 75 U / pL T7 RNA polymerase (Roche). For reactions providing uncapped mRNAs, cap analog was omitted. All mixed components were incubated for 2h at 37°C, and subsequently treated with 0.025 U / pL of DNase I (ThermoFisher Scientific) for 30 min at 37°C. The enzymes in the IVT mixtures were inactivated by addition of one volume of 50 mM EDTA, and full-length mRNAs were purified by affinity chromatography (described in Example 6) with oligo(dT)25 resin (POROS™ Oligo (dT)25 Affinity Resin, ThermoFisher Scientific).

[0163] Example 4: General procedure for in vitro transcription of mRNAs with m7GpppA*pU*pG tetranucleotide primers: see FIG 12, FIG 14

[0164] The dsDNA template encoding firefly luciferase was prepared by PCR (see also Example 2). Two different templates were used for this experiment: 1) template T4 (<t>6.5-ATG), which contained <t>6.5 promoter followed by ATG deoxyribonucleotides at positions +1 , +2 and +3 of the coding strand (in this case non-coding strand contained deoxyribonucleotides TTAC corresponding to the positions -1 , +1 , +2 and +3); 2) template T5 (<t>6.5-TGG), which contained <t>6.5 promoter followed by TGG nucleotides at positions +1 , +2 and +3 of the coding strand (in this case non-coding strand contained deoxyribonucleotides TACC corresponding to the positions -1 , +1 , +2 and +3). In vitro transcription reactions for 8 mM of tetranucleotide primer m7GpppAmpUmpG, m7GpppAmpUpG, m7GpppBn6AmpUmpG, m7GpppBn6AmpUpG, m7GpppAmpAmpU, trinucleotide cap 1 primer m7GpppAmpU and Bn6-modified trinucleotide primer rn7GpppBn6AmpU were prepared using 40 mM Bis-Tris buffer pH 6.5, containing 2 mM spermidine, and 5 mM each of ATP, UTP, CTP and GTP, 30 mM MgCh, 10 mM DTT, 20 ng / pL PCR-generated DNA template, 1 U / pL RNase inhibitor (RiboLock, ThermoFisher Scientific), 0.002 U / pL inorganic pyrophosphatase (ThermoFisher Scientific) and 75 U / pL T7 RNA polymerase (Roche). For reactions providing uncapped mRNAs, cap analog was omitted. All mixed components were incubated for 2h at 37°C, and subsequently treated with 0.025 U / pL of DNase I (ThermoFisher Scientific) for 30 min at 37°C. The enzymes in the IVT mixtures were inactivated by addition of one volume of 50 mM EDTA, and full- length mRNAs were purified by affinity chromatography (described in Example 6) with oligo(dT)25 resin (POROS™ Oligo (dT)25 Affinity Resin, ThermoFisher Scientific).

[0165] Example 5: General procedure for in vitro transcription of mRNAs with m7GpppA*pG*pA tetranucleotide primers: see FIG 15, FIG 16

[0166] The dsDNA template encoding firefly luciferase was prepared by PCR (see also Example 2). Two different templates were used for this experiment: 1) template T6 (<t>6.5-AGA), which contained <t>6.5 promoter followed by AGA deoxyribonucleotides at positions +1 , +2 and +3 of the coding strand (in this case non-coding strand contained deoxyribonucleotides TTCT corresponding to the positions -1 , +1 , +2 and +3); 2) template T7 (<t>6.5-GAG), which contained <t>6.5 promoter followed by GAG nucleotides at positions +1 , +2 and +3 of the coding strand (in this case non-coding strand contained deoxyribonucleotides TCTC corresponding to the positions -1 , +1 , +2 and +3). In vitro transcription reactions for 10 mM of tetranucleotide primer m7GpppAmpGmpA, rn7GpppAmpGpA, rn7GpppBn6AmpGmpA or rn7GpppBn6AmpGpA, and for Bn6-containing trinucleotide primer rn7GpppBn6AmpG were prepared using 40 mM Bis-Tris buffer pH 6.5, containing 2 mM spermidine, and 5 mM each of ATP, UTP, CTP, 4 mM GTP, 25 mM MgCh, 10 mM DTT, 20 ng / pL PCR-generated DNA template, 1 U / pL RNase inhibitor (RiboLock, ThermoFisher Scientific), 0.002 U / pL inorganic pyrophosphatase (ThermoFisher Scientific) and 75 U / pL T7 RNA polymerase (Roche). In vitro transcription reactions for trinucleotide cap 1 primer rn7GpppAmpG were prepared using 40 mM Tris buffer pH 8.0, containing 2 mM spermidine, and 5 mM each of ATP, UTP, CTP, 4 mM GTP, 10 mM MgCL, 10 mM DTT, 20 ng / pL PCR-generated DNA template, 1 U / pL RNase inhibitor (RiboLock, ThermoFisher Scientific), 0.002 U / pL inorganic pyrophosphatase (ThermoFisher Scientific) and 75 U / pL T7 RNA polymerase (Roche). For reactions providing uncapped mRNAs, cap analog was omitted. All mixed components were incubated for 2h at 37°C, and subsequently treated with 0.025 U / pL of DNase I (ThermoFisher Scientific) for 30 min at 37°C. The enzymes in the IVT mixtures were inactivated by addition of one volume of 50 mM EDTA, and full-length mRNAs were purified by affinity chromatography (described in Example 6) with oligo(dT)25 resin (POROS™ Oligo (dT)25 Affinity Resin, ThermoFisher Scientific). resin.

[0167] Each crude IVT mix inactivated with 50 mM EDTA was further diluted with high salt buffer (1200 mM KCI, 10 mM Tris-HCI pH 7.5, 1 mM EDTA) and loaded on the accordingly equilibrated oligo(dT)25 resin. After binding step (10 min, RT), resin was washed with high and low salt buffer (200 mM KCI, 10 mM Tris-HCI pH 7.5, 1 mM EDTA), respectively. mRNA was eluted with RNase-free water heated to 65°C, and prior IVT efficiency estimation and ribozyme treatment it was concentrated by ultrafiltration (Amicon Ultra-0.5 50K, Millipore).

[0168] 7: Determination of in vitro

[0169] IVT efficiencies were estimated spectrophotometrically after mRNA purification on the oligo(dT)25 resin and concentration by ultrafiltration using Amicon Ultra-0.5 50K (Millipore) filtration devices. After recovering concentrated mRNA, sample volume and the absorbance at 260 nm corresponding to mRNA concentration was measured using NanoDrop Onec(ThermoFisher Scientific) spectrophotometer. IVT yield was calculated by multiplying the volume of the concentrated sample and measured mRNA concentration expressed as ng / pL. IVT efficiency was calculated as mRNA yield obtained from 1 pL of starting IVT volume.

[0170] Example 8: Determination of capping efficiency; see FIG 5, FIG 7-12, FIG 14-16, FIG 24, FIG 27-30, FIG 32

[0171] Ribozyme complementary to the 5'UTR sequence was designed and used for cleavage of mRNA close to the 5' end. 10 pL of analyzed sample containing >10 pg (>100 nM) of mRNA was mixed with 1 .5 pL of hybridization buffer containing 100 mM Tris-HCI pH 7.5 and 50 mM NaCI and 2 pL of 10 pM ribozyme. The mixture was incubated at 95°C for 2 min, and subsequently at RT for 10-15 min. Cleavage reaction with the ribozyme hybridized to mRNA was initiated by addition of 1.5 pL of 100 mM MgCh, conducted at 37°C for 1 h, and quenched by addition of 2 pL 100 mM EDTA.

[0172] To analyze cleaved RNA fragments 3 pL of quenched reaction was mixed with equal volume of loading dye (8 M urea, 50% formamide, 20 mM EDTA, 0.03% bromophenol blue, 0.03% xylene cyanol), heat denatured at 95°C for 3 min and loaded onto 15% polyacrylamide gel with 7 M urea and 1 x TBE. The gel after electrophoresis was incubated at RT for 15 min with 50 mL of staining reagent (SYBR® Gold, Invitrogen, diluted 1 :10000), and visualized using GelDoc Go Imaging System (Bio-Rad). Migration of short 5' RNA fragments being the result of ribozyme cleavage differs depending on the cap structure present on the 5' end of mRNA. Additionally, bands intensities correlate with the amount of particular RNA species. The intensity of the bands corresponding with capped and uncapped RNA (ppp-RNA) were quantified densitometrically using ImageQuantTL software (GE Healthcare). Percentage of capped mRNA in each sample was determined as the ratio of capped RNA and the summarized intensities for capped RNA and ppp-RNA. is and Results of IVT yield and

[0173] We have employed three dsDNA templates (T1-T3) encoding Firefly luciferase, carrying either <t>6.5 or <t>2.5 promoter and AGG or GGG sequence at +1 , +2, and +3 positions and tested different initiating primers, respectively (FIG 1). We first tested these templates against two different cap 2 primers (m7GpppAmpGmpG or m7GpppAmpAmpG), whose synthesis is shown in FIG 2. These primers were designed to be fully complementary to positions +1 , +2 and +3 or -1 , +1 and +2, respectively, of the DNA template T1 (<t>6.5- AGG), and are fully or partially complementary to the other templates (FIG 4A to D). Additionally, an IVT reaction with a tetranucleotide cap 1 primer designed by analogy to what Ishikawa has proposed for trinucleotides was performed (Ishikawa et al., 2009, supra). The tetranucleotide primer (rn7GpppAmGG) was designed as complementary (starting from the position -1) to the Ishikawa’s type template i.e. containing <p6.5 promoter followed by GGG (T3).

[0174] The mRNAs were synthesized by in vitro transcription in the presence of 5 mM each of ATP, UTP, CTP, 4 mM GTP, 10 mM cap analogs, and 25 mM MgCh as described in Example 3 The mRNAs were purified by affinity chromatography (described in Example 6). Purified mRNA was cleaved at the 5' ends by a ribozyme, and analyzed by electrophoresis followed by densitometry to determine the capping efficiencies (Example 8), as described previously. (Vlatkovic et al., Pharmaceutics, 2022, 14(2):328; doi: 10.3390 / pharmaceuticsl 4020328)] IVT yields were determined spectrophotometrically after mRNA purification by measuring the volume of concentrated eluate and the absorbance at 260 nm corresponding to mRNA concentration. (Example 7). Uncapped mRNA, which was used as a reference, was synthesized in the same way, except that no cap analog was added to the IVT mix.

[0175] FIG 5 shows representative results of such an experiment. Uncapped mRNAs produced from all templates (T1-T3) produce a major RNA product of 23 nt in length after the ribozyme-mediated cleavage. This confirms that the transcription start sites (+1 positions) for templates T1-T3 are as indicated in FIG 4A and B. Some additional bands corresponding to longer and shorter RNAs are also visible for these samples, which is a common phenomenon for in vitro transcribed RNA obtained with T7 polymerase(Pomerantz et al., Mol Cell, 2006, 20;24(2):245-55)ln the presence of cap analogs, additional bands corresponding to longer RNAs are visible. These bands represent capped RNAs of various lengths (24-26 nt). The lengths of the major observed capped RNA bands correspond well with the anticipated hybridization modes in the transcription initiation complexes (FIG 6). The densitometric quantification of band intensities enabled us to determine the overall capping efficiencies, which are also shown in FIG 5. The results show a clear advantage of using m7GpppAmpAmpG as a transcription primer over m7GpppAmpGmpG for each of the studied templates.

[0176] For template T1 (<t>6.5-AGG), m7GpppAmpGmpG afforded 90% capping efficiency, whereas for m7GpppAmpAmpG, uncapped RNA was virtually not observed (capping efficiency ~100%). Moreover, we observed that using m7GpppAmpAmpG resulted in mRNA of much higher homogeneity of the 5' end. A virtually single band was observed for the capped RNA species obtained with m7GpppAmpAmpG. In contrast, for rn7GpppAmpGmpG two distinctive RNA species were observed, which may result either from initiation of transcription at alternative sites or insertion of additional nucleotides by T7 polymerase (Pleiss et al., RNA, 1998;4(10):1313-7).

[0177] For templates T2 (<t>2.5-AGG) and T3 (<t>6.5-GGG), we generally observed lower capping efficiencies than for template T1 (<t>6.5-AGG), but in both cases m7GpppAmpAmpG afforded 20-30% higher capping efficiency than m7GpppAmpGmpG and higher homogeneity of the capped RNA species, despite comparable or even lower degree of complementarity to the template (FIG 6). This is in contrast to what has been previously claimed by WO2017053297A1 that the complementarity between the template and the primer at positions +1 +2 (+3) and N1, N2(and N3), respectively, is the main prerequisite for achieving high capping efficiency and 5' end homogeneity. Also, it was found that the transcription with m7GpppAmpAmpG from <t>6.5-AGG template T1 (FIG 5; lane 6) significantly improved capping efficiency and 5’ end homogeneity without compromising the IVT yield compared to the results obtained for cap 1 analogs and following the approach previously proposed by Ishikawa (FIG.5, lanes 10 and 11).

[0178] We next expanded this experiment to study various derivatives of m7GpppA*pG*pG or m7GpppA*pA*pG (wherein A* and G* denote adenosine or guanosine moieties, unmodified or modified at the positions 2'-O and Bn6to form various cap 1 and cap 2 analogs) against the same set of templates. The representative results for templates T1 , T2, and T3 (see Table 5) and eight cap analogs (four derivatives of m7GpppA*pG*pG type and four corresponding of m7GpppA*pA*pG type) are shown in FIG 7, 8 and 9, respectively, and the mean results of replicate experiments are shown in Tables 5-7. These experiments confirmed our initial finding that the combination of the template T1 (<t>6.5-AGG) and m7GpppA*pA*pG type initiator gives the best capping efficiency results without compromising the in vitro transcription yield. FIG 7 shows that combination of any of the studied m7GpppA*pA*pG derivatives with template T 1 (<t>6.5-AGG) affords capping efficiency close to 100%. This is the case even for tetranucleotide analogs carrying bulky benzyl group at the N6-position of adenosine, which affects the capping efficiency for the corresponding trinucleotide (maximum capping efficiency is 92% at pH 6.5, and decreases with increasing pH (Warminski et al., Journal of the American Chemical Society 2024 146 (12), 8149-8163). In contrast, the mRNAs obtained in the presence of m7GpppA*pG*pG derivatives as transcription initiators under the same conditions are characterized by higher 5' end heterogeneity and capping efficiencies that do not exceed 90%, and decrease notably for analogs carrying bulkier modifications (81 % and 84% form7GpppBn6AmpGpG andm7GpppBn6AmpGmpG, respectively). Moreover, the presence of the Bn6modification within m7GpppA*pG*pG analogs significantly decreased the in vitro transcription yield. Overall, we discovered that the combination of m7GpppA*pA*pG, but not m7GpppA*pG*pG, with template T1 (<t>6.5-AGG) enables preparation of mRNAs comprising cap 1 and cap 2 structures, either unmodified or carrying additional chemical modifications, with exceptionally high capping efficiency and 5' end homogeneity and with very good in vitro transcription yield.

[0179] Table 4. Tetrameric and trimeric capped oligonucleotides used as the transcription initiation primers in the in vitro transcription experiments.

[0180]

[0181] Table 5. Promoters and trinucleotide initiating sequences of the DNA templates used in the experiments.

[0182] 1TSS - transcription start site (sequence of the coding strand of the dsDNA template). First transcribed nucleotide which corresponds to the position +1 of the template strand is underlined.

[0183] 2nucleotides of the non-coding strand following core promoter region of the dsDNA template; positions +1+2+3 represent the TIS starting with the TSS at +1. Table 6. Summarized results of replicate experiments: in vitro transcription (IVT) and capping efficiency (±SD) for tetrameric and trimeric capped oligonucleotide primers (10 mM) used in the experiments with optimized IVT conditions (pH 6.5) and template T1 (O6.5-AGG, see also FIG 4). Table 7. Summarized results of replicate experiments: in vitro transcription (IVT) and capping efficiency (±SD) for tetrameric and trimeric capped oligonucleotide primers (10 mM) used in the experiments with optimized IVT conditions (pH 6.5) and template T2 (<t>2.5-AGG, see also FIG 4).

[0184] Table 8. Summarized results of replicate experiments: in vitro transcription (IVT) and capping efficiency (±SD) for tetrameric and trimeric capped oligonucleotide primers (10 mM) used in the experiments with optimized IVT conditions (pH 6.5) and template T3 (<t>6.5-GGG, see also FIG 4).

[0185] We additionally tested the same set of initiating nucleotides with template T2 (<t>2.5-AGG) under analogous conditions (FIG 8). The experiments revealed that transcriptions with primers derived from m7GpppA*pA*pG structure (m7GpppAmpAmpG, rn7GpppAmpApG, rn7GpppBn6AmpAmpG and rn7GpppBn6AmpApG) result in mRNAs characterized by higher capping efficiencies than transcriptions with m7GpppA*pG*pG structure, despite the fact that only the latter is fully complementary to the template (as shown in FIG 6). Also, the transcriptions performed in the presence of m7GpppA*pG*pG, especially Bn6-modified and cap 2 variants, resulted in significantly lower overall IVT yields and higher 5' end heterogeneities. This suggested that the alignment of the m7GpppA1*pA2*pG tetranucleotide primer with the DNA template wherein the position A1* of the primer is aligned with position -1 of the DNA template (even if not complementary to the template) is favorable over the alignment of m7GpppA*pG1*pG2tetranucleotide wherein the position A* of the primer is aligned with the position +1 of the template (even if complementary to the template). The first case produces mRNA of higher capping efficiency and higher 5' end homogeneity and with higher overall yield, suggesting that the complexes formed between the T7 RNA polymerase, m7GpppA1*pA2*pG-type cap analog, and the DNA template are more effective for transcription initiation.

[0186] Finally, we tested the same set of primers with template T3 (<t>6.5-GGG) (FIG 9), which provides incomplete complementarity for m7GpppA*pA*pG-type primers (2 out of 3 complementary base pairs) and differs in their alignment against the template (FIG 6). In this case we again observed that the m7GpppA*pA*pG-type primers provided significantly higher capping efficiencies than m7GpppA*pG*pG-type primers. Notably, the m7GpppA*pG*pG-type primers used with template T3 (<t>6.5-GGG) had an analogous alignment as m7GpppA*pA*pG-type with template T1 (<t>6.5-AGG), but yielded mRNAs characterized by significantly lower capping efficiencies and higher heterogeneities. This indicates that not only alignment position of the primer against the template, but also the sequence of the primer / template are a relevant factor for obtaining the most effective transcription initiation complexes. More importantly, it is ultimately the synthesis and the provision of a suitable synthetically modified tetranucleotide cap analogs and the targeted combination of these highly artificial cap analogs with a suitable cognate DNA template that will guarantee a high capping efficiency together with a high translational activity.

[0187] It was further demonstrated that the combination of m7GpppA*pA*pG-type primers and template T1 (<t>6.5-AGG) enables the preparation of high quality in vitro transcribed mRNA under more-restrictive conditions, particularly at decreased cap analog concentration. It is currently estimated that the cost of capping reagents is appr. 40% of the cost of production of in vitro transcribed mRNA. Therefore decreasing the concentration of cap analog in the IVT mix without compromising the capping efficiency, RNA integrity or IVT yield is highly desirable. As such we performed IVT reactions with template T1 (<t>6.5-AGG) in the presence of cap analogs at concentrations 5 mM or 2 mM. The representative results are shown in FIG 10 (for 5 mM cap analog concentration) and FIG 11 (for 2 mM cap analog concentration), respectively, and the mean results of replicate experiments are shown in Table 9 and 10. The experiments revealed excellent capping efficiencies for all studies tetranucleotide m7GpppA*pA*pG-type primers (cap 1 and cap 2 derivatives), despite cap analog concentration being decreased to 5 mM (capping efficiency 96%-100%) or even 2 mM (capping efficiency 90%-94%). Notably, form7GpppA*pG*pG-type primers, capping efficiency is decreased to 76-86% (5 mM) and 60-76% (2 mM). Taking into account that the synthesis of tetranucleotide primers is not significantly more labor-intensive and reagent-consuming compared to trinucleotide capping reagents, the transcription initiation complexes comprising tetranucleotide cap 1 analogs according to this invention are also a viable alternative to complexes comprising trinucleotide cap 1 analogs.

[0188] Table 9. Summarized results of replicate experiments: in vitro transcription (IVT) and capping efficiency (±SD) for tetrameric and trimeric capped oligonucleotide primers (5 mM) used in the experiments with optimized IVT conditions (pH 6.5) and template T1 (<t>6.5-AGG, see also FIG 4A)

[0189] Table 10. Summarized results of the replicate experiments: in vitro transcription (IVT) and capping efficiency (±SD) for tetrameric and trimeric capped oligonucleotide primers (2 mM) used in the experiments with optimized IVT conditions (pH 6.5) and template T1 (<t>6.5-AGG, see also FIG 4A)

[0190] Example 10: Analysis and Results of IVT yield and capping efficiency analysis for AUG cap analogs.

[0191] To explore alternative TSS combined with appropriate tetranucleotide cap primers we have employed two PCR-generated templates encoding FFLuc, carrying <t>6.5 promoter followed by ATG (template T4) or TGG (template T5) seguence (FIG 4C and D). Particularly, ATG trinucleotide initiating seguence is used in the constructs encoding selfamplifying mRNAs (saRNAs), being a promising improvement of the first generation of the mRNA-based vaccines. This solution provides comparable therapeutic effect obtained for decreased vaccination dose, which reduces manufacturing costs and potential harmful effects related to application of exogenous RNA [https: / / www.mdpi.eom / 2076-393X / 12 / 3 / 318] It is expected that a similar therapeutic effect could be obtained for even lower dosages of saRNA capped with cap analogs carrying modifications increasing translational activity (such as Bn6), therefore, prolonging persistence of saRNA in cellular environment. As we discovered, co-transcriptional incorporation of such cap analogs by T7 RNA polymerase occurs only under the specific combination of transcription initiating primer and the DNA template containing particular TSS.

[0192] We first tested cap analogs m7GpppAmpUmpG and rn7GpppAmpUpG as the tetranucleotide primers designed to be fully complementary to templates T4 (<t>6.5-ATG) and T5 (<t>6.5-TGG), but differing in respective alignments. Additionally, rn7GpppAmpAmpU tetranucleotide primer was used with both templates (T4 and T5) to reflect full or partial complementarity previously observed for combination of m7GpppA*pA*pG-type primers with template T1 (<t>6.5-AGG) ortemplate T3 (<t>6.5-GGG), respectively (FIG 6A to C, 7 and 9). The mRNAs were synthesized by in vitro transcription in the presence of 5 mM each of ATP, UTP, CTP and GTP, 8 mM cap analogs and 30 mM MgCh as described in Example 4. The mRNAs were purified by affinity chromatography (described in Example 6). Purified mRNAs were cleaved at the 5' ends by a ribozyme, and analyzed by electrophoresis followed by densitometry to determine the capping efficiencies (Example 8), as described previously. [https: / / pubmed.ncbi.nlm.nih.gov / 35214060 / ] IVT efficiencies were determined spectrophotometrically after mRNA purification by measuring the volume of concentrated eluate and the absorbance at 260 nm corresponding to mRNA concentration (Example 7). Uncapped mRNA, which was used as a reference, was synthesized in the same way, except that no cap analog was added to the IVT mix.

[0193] FIG 12 depicts the obtained results. A first intriguing observation that we made was that uncapped RNAs obtained from these templates differed in length indicating most likely for different transcription initiation start sites. The result of ribozyme-mediated cleavage of uncapped RNA indicated that major initiation site for T5 was one nucleotide shorter than in case of the template T4 (<t>6.5-ATG) (FIG 12B, lane 7 vs. lane 8). In other words, the core promoter region was extended. Without wishing to be bound by theory, this confirms that the TSS (always numbered as +1 herein) for a given pair of primer::DNA template is dependent on the template and cannot be assumed to be in the same place for every template likely in view of steric aspects associated with the artificial tri- and tetranucleotide cap primers. For template T4, we observed the expected 23 nt-long uncapped product, analogously as in the case of IVT reactions performed for templates T1-T3. In the case of template T5 the product was 1 nt shorter indicating that the major transcription start site (+1) position has shifted to the guanine as indicated in FIG 4. Such phenomenon is observed for the templates containing pyrimidine deoxyribonucleotide at position +1 of the coding strand, following core promoter region (Pomerantz et al., Mol Cell, 2006, 20;24(2):245-55).

[0194] However, bands on the denaturing gel for the transcripts obtained using template T5 (<t>6.5-TGG) and capped with tetranucleotides m7GpppA*pU*pG indicated formation of the initiation complex aligned analogously as shown in FIG 13 with mRNA’s 5' end being one nucleotide longer than in case of m7GpppA*pU*pG-capped transcripts obtained using template T4 (<t>6.5-ATG) (FIG 12B, lane 1 vs. lane 4 and lane 2 vs. lane 5). This result suggests that in the presence of tetranucleotide primer m7GpppA1*pU2*pG3T7 RNA polymerase recovers the expected register of transcription bubble and allows nucleotides U2* and G3of the primer being fully hybridized with the template T5 (<t>6.5-TGG) of the non-coding strand.

[0195] Analyzing the results obtained in the presence of cap analogs we found that full complementarity between nucleotides N1, N2, and N3of the tetranucleotide primer rn7GpppAmpUmpG or rn7GpppAmpUpG with positions +1 , +2 and +3 of the non-coding strand of template T4 (<t>6.5-ATG) guarantees high capping efficiency (100-95%). However, to the IVT yields for reaction were relatively low (especially for rn7GpppAmpUpG). The IVT efficiency was higher for m7GpppAmpAmpU analog, which is fully aligned with positions -1 , +1 and +2 of the template T4 (<t>6.5-ATG), which also provided an excellent capping efficiency of 97%. The same set of tetranucleotide primers was efficiently (84-86%) incorporated into 5' ends of mRNAs encoded by the template T5 (<t>6.5-TGG) and allowed obtaining high transcription yield, even for previously ineffective m7GpppAmUpG cap 1 analog.

[0196] We further expanded this experiment to investigate incorporation efficiency of heavily modified tetranucleotide analogs m7GpppA1*pU2*pG3(wherein A* and U* denote adenosine or uridine moieties, unmodified or modified at the positions 2'-O and Bn6to form various cap 1 and cap 2 analogs) using the same set of DNA templates. These experiments confirmed that it is possible to co-transcriptionally introduce bulky substitutions (such as benzyl group at the N6-position of the adenosine A1*) into 5' end of mRNA initiating with AUG sequence. However, to obtain the transcripts with heavily modified 5' ends, it is essential to design tetranucleotide cap primer::DNA template complex providing alignment of Bn6-modified adenosine A1* with the position -1 (or -2 for T5) of the non-coding template strand (FIG 14). This observation combined with the previous findings described herein suggests that for efficient initiation of transcription with cap primers carrying bulky substitution, they should be located within T7 RNA polymerase binding pocket corresponding to -1 position of the non-coding DNA strand. Otherwise, when heavily modified cap analog is inappropriately aligned with the template, the transcriptional activity of the enzyme can be almost completely abolished (FIG 14B vs.

[0197] FIG 14A).

[0198] Overall, we discovered that the combination of m7GpppA*pU*pG and m7GpppAmpAmpU with the template T5 (<t>6.5-TGG) enables preparation of mRNAs comprising cap 1 and cap 2 structures, either unmodified or carrying additional chemical modifications, with very good IVT yield, excellent 5' end homogeneity and with satisfactory capping efficiency, possibly improvable by additional optimization of the in vitro transcription reaction (e.g. by increasing cap analog concentration in the IVT mix).

[0199] 11 : Analysis and Results of IVT yield and modified tetranucleotide

[0200] We discovered that for introduction of heavily modified cap structures into 5' end of RNA it is crucial to use appropriately designed tetranucleotide cap primerDNA template complex, i.e. with Bn6-modified nucleotide N1being aligned with the position -1 (-2 for T5) of the non-coding template strand. To confirm these findings, we performed two additional experiments.

[0201] For this purpose, in the first experiment (FIG 15) we used previously validated combination of tetranucleotide primers m7GpppA*pA*pG (wherein A* and G* denote adenosine or guanosine moieties, unmodified or modified at the positions 2'-O and Bn6to form various cap 1 and cap 2 analogs) with the template T1 (<t>6.5-AGG) (see Table 5). Additionally, the same set of tetranucleotide cap primers was applied for co- transcriptional capping using a newly designed template T8 (<t>6.5-AAG), which contained <t>6.5 promoter followed by AAG nucleotides at positions +1 , +2 and +3 of the coding strand (in this case non-coding strand contained deoxyribonucleotides TTTC corresponding to the positions -1 , +1 , +2 and +3) (see Table 5). In the second experiment (FIG 16) we used the combination of two newly designed templates, i.e. T6 (<t>6.5-AGA) and T7 (<t>6.5-GAG), and variously compatible m7GpppA*pG*pA tetranucleotide cap analogs. Based on the discoveries described herein, we expect to observe full complementarity of m7GpppA*pG*pA cap primers with the positions +1 , +2 and +3 of the template T6 (<t>6.5-AGA), and with the positions -1 , +1 and +2 of the template T7 (<t>6.5- GAG), respectively.

[0202] In both experiments the mRNAs were synthesized as described in the Example 11. In vitro transcription was performed in the presence of 5 mM each of ATP, UTP, CTP, 4 mM GTP, 10 mM cap analogs, and 25 mM MgCh. The mRNAs were purified by affinity chromatography (described in Example 6). Purified mRNA was cleaved at the 5' ends by a ribozyme, and analyzed by electrophoresis followed by densitometry to determine the capping efficiencies (Example 8), as described previously. [https: / / pubmed.ncbi.nlm.nih.gov / 35214060 / ] IVT efficiencies were determined spectrophotometrically after mRNA purification by measuring the volume of concentrated eluate and the absorbance at 260 nm corresponding to mRNA concentration (Example 7). Uncapped mRNA, which was used as a reference, was synthesized in the same way, except that no cap analog was added to the IVT mix.

[0203] As previously, m7GpppA*pA*pG tetranucleotide primers efficiently initiate transcription when nucleotides N1, N2and N3are aligned with positions -1 , +1 and +2 of the template T1 (<t>6.5-AGG) (FIG 15, lanes 5-8, and FIG 6A to C). Extraordinary capping efficiency and high IVT yield was obtained regardless bulky substituent present on the nucleotide N1. On the other hand, for the template T8 (<t>6.5-AAG), which for m7GpppA*pA*pG-type analogs provides full complementarity of nucleotides N1, N2and N3with positions +1 , +2 and +3 of the non-coding strand, efficiently capped RNA product was obtained only for tetranucleotides without Bn6modification, i.e. rn7GpppAmpApG and m7GpppAmpAmpG (FIG 15, lane 2 and 3). Analogously, a satisfactory capping efficiency was obtained for all variants of m7GpppA*pG*pA tetranucleotides combined with template T7 (<t>6.5-GAG), providing alignment of nucleotides N1, N2and N3with positions -1 , +1 and +2 of noncoding strand (FIG 16B, lanes 5-8 and FIG 17), but transcription initiation with T6 (<t>6.5- AGA) template was efficient only for m7GpppA*pG*pA tetranucleotides lacking Bn6 modification (FIG 16B, lanes 1-4). Additionally, the 5' end of T7 (<t>6.5-GAG)-templated RNAs was more homogenous than for transcripts obtained using template T6 (<t>6.5- AGA), which enables hybridization of nucleotides N1, N2and N3with position +1 , +2 and +3 of the template (FIG 16B, lanes 5-8 vs. lanes 1-4, respectively).

[0204] Overall, it was confirmed for various tetranucleotide cap primerDNA template complexes (Example 9, Example 10, Example 11), that for efficient co-transcriptional capping, especially using heavily modified analogs (e.g. containingBn6Ammodification, cap 2, and their combinations), it is essential to provide alignment of cap ribonucleotides N1and N2(N3) with position -1 and +1 (and +2) of the template strand (or -2, -1 and +1 for T5). Such hybridization pattern guarantees good capping efficiency, high IVT yield and increases homogeneity of the mRNA’s 5' end.

[0205] Example 12: In vitro translational properties of the mRNA synthesized using tetranucleotide primer for co-transcriptional capping; see FIG 18 to 21 ■.

[0206] Next, it was verified if the biological activity of mRNAs obtained using initiating complexes according to the invention is unperturbed compared to mRNAs obtained in the presence of trinucleotide primers, and particularly if the superior biological activity ofBn6Am-rnodified caps is retained when incorporated by means of tetranucleotide primers. To assess the translational activity of mRNAs containing m7GpppA*pG*pG or m7GpppA*pA*pG 5' cap structures we have transfected selected cell types with purified transcripts encoding firefly luciferase (FFLuc). In vitro transcription reactions in the presence of 10 mM each of the tetranucleotide primers: m7GpppAmpGmpG, m7GpppAmpAmpG, rn7GpppAmpGpG, rn7GpppAmpApG, m7GpppBn6AmpGmpG, m7GpppBn6AmpAmpG, rn7GpppBn6AmpGpG or rn7GpppBn6AmpApG and for reference primers: Bn6-containing trinucleotide primer - rn7GpppBn6AmpG or unmodified - rn7GpppAmpG, were prepared as described in Example 3. For that purpose, we have employed template T1 (<t>6.5-AGG), which comprised <t>6.5 promoter followed by AGG deoxyribonucleotides at positions +1 , +2 and +3 of the coding strand (i.e. non-coding strand contained deoxyribonucleotides TTCC corresponding to the positions -1 , +1 , +2 and +3). The synthesized mRNAs were initially purified by affinity chromatography (described in Example 6) with oligo(dT)25 resin (POROS™ Oligo (dT)25 Affinity Resin, ThermoFisher Scientific), and the capping efficiencies were determined densitometrically for the products of the 5' terminal cleavage with ribozyme (described in Example s, and based on Vlatkovic et al., Pharmaceutics, 2022, 14(2):328; doi: 10.3390 / pharmaceuticsl 4020328). IVT yields were determined spectrophotometrically after mRNA purification by measuring the volume of concentrated eluate and the absorbance at 260 nm corresponding to mRNA concentration (described in Example 7). To reduce influence of dsRNA or low molecular weight contaminants, such as short RNA species, transcripts were further purified by RP-HPLC prior in vitro transfection. Each sample was applied on RNASep™ Semi-Prep column (ADS Biotec) combined with Agilent Infinity 1260 II HPLC system, and eluted with the linear gradient (10-14.5%) of acetonitrile in 0.1 M TEAA pH 7.0 at 55°C. Fractions containing mRNA of the highest purity confirmed by electrophoresis (using 1 xTBE 1.2% agarose gel) were combined, desalted (by ultrafiltration using Amicon Ultra-15 50K, Millipore) and precipitated overnight at -20°C (with 0.3 M NaOAc pH 5.2, one volume of isopropanol, and washed with 80% ethanol). Prior quality control and in vitro experiment mRNA pellets were resuspended in RNase-free water to obtain concentration of >1 pg / pL.

[0207] 50x103murine and human primary macrophages and 10x103HEK293T were seeded on 96 well plate. Next day cells were transfected with 20 or 200 ng mRNA with the use of Lipofectamine MessengerMax. Medium was collected 24 h after transfection. Flue activity was measured with BrightGlo at 4, 24, 48 h. Viability of the cells was measured with CellTiter Blue 24, 48 h after transfection.

[0208] 10x103HeLa, HepG2, A549 cells were seeded on 96 well plate. Next day cells were transfected with 20 or 200 ng mRNA with the use of Lipofectamine MessengerMax. Flue activity was measured with BrightGlo at 4, 24, 48 h. Viability of the cells was measured with CellTiter Blue 24, 48 h after transfection.

[0209] Translation activity was measured in HEK293T cell line and human or murine macrophages for in vitro transcribed and purified mRNAs 5'-capped with one of previously investigated tetranucleotide primers, i.e. m7GpppAmpGmpG, m7GpppAmpAmpG, rn7GpppAmpGpG, rn7GpppAmpApG, m7GpppBn6AmpGmpG, m7GpppBn6AmpAmpG, rn7GpppBn6AmpGpG or rn7GpppBn6AmpApG and for Bn6-containing trinucleotide primer rn7GpppBn6AmpG or unmodified rn7GpppAmpG, as the controls. At the initial time point of 4 h the luminescence signal was similar for all m7GpppA*pA*pG 5'-capped mRNAs, regardless applied mRNA dose (20 ng or 200 ng), but subsequent measurements 24 h and 48 h post-transfection indicated reduced luminescence for the transcripts containing m7GpppA*pG*pG modification when compared to m7GpppA*pA*pG 5'-capped counterparts (FIG 18 and 19). However, the most significant differences in the luminescence were observed in the primary cells transfected with the same set of mRNAs (FIG 20 and 21 and FIG 22 and 23). For human macrophages (FIG 20 and 21) the highest luminescence signal was measured for the mRNAs co-transcriptionally capped with Bn6-modified m7GpppA*pA*pG tetranucleotide primers, i.e. m7GpppBn6AmpAmpG or rn7GpppBn6AmpApG, and with rn7GpppBn6AmpG trinucleotide primer, and the superiority of Bn6-modification was emphasized for analogous measurements performed for transfected murine macrophages (FIG 22 and 23). Both experiments with primary cells pointed Bn6-modification as the key player for efficient and long lasting mRNA’s translatability upon transfection and have shown that the incorporation ofBn6Ammodification by means of tetranucleotide primers does not supress its biological advantages, compared to incorporation using trinucleotide primers. of the mRNA tetranucleotide primer m7' for co-l

[0210] ; see FIG 24 to 26.

[0211] To compare efficiencies of protein production in vivo we have prepared set of the transcripts encoding human erythropoietin (hEPO). The mRNAs were co-transcriptionally 5 '-capped analogously to FLuc mRNAs described above, with the exception that the IVT were run in the presence of two fold lower concentration (5 mM) of each of the tetranucleotide primers: m7GpppAmpGmpG, m7GpppAmpAmpG, rn7GpppAmpGpG, rn7GpppAmpApG, m7GpppBn6AmpGmpG, m7GpppBn6AmpAmpG, rn7GpppBn6AmpGpG, rn7GpppBn6AmpApG. For reference trinucleotide primers: rn7GpppAmpG and rn7GpppBn6AmpG 10 mM concentration was used, because lower primer concentrations did not yield satisfactory capping efficiency (below 90%). IVT reactions were run as described in Example 3 using template T9 (<t>6.5-AGG), which contained <t>6.5 promoter followed by AGG deoxyribonucleotides at positions +1 , +2 and +3 of the coding strand (in this case non-coding strand contained deoxyribonucleotides TTCC corresponding to the positions -1 , +1 , +2 and +3). The synthesized mRNAs were initially purified by affinity chromatography (described in Example 6) with oligo(dT)25 resin (POROS™ Oligo (dT)25 Affinity Resin, ThermoFisher Scientific), and the capping efficiencies were determined densitometrically for the products of the 5' end cleavage with ribozyme (described in Example s, and based on Vlatkovic et al., Pharmaceutics, 2022, 14(2):328; doi: 10.3390 / pharmaceuticsl 4020328) (FIG 24B). IVT yields were determined spectrophotometrically after mRNA purification by measuring the volume of concentrated eluate and the absorbance at 260 nm corresponding to mRNA concentration (described in Example 7) (FIG 24A). To reduce influence of dsRNA or low molecular weight contaminants, such as short RNA species, transcripts were further purified by RP-HPLC prior in vitro transfection. Each sample was applied on RNASep™ Semi-Prep column (ADS Biotec) combined with Agilent Infinity 1260 II HPLC system, and eluted with the linear gradient (10-14.5%) of acetonitrile in 0.1 M TEAA pH 7.0 at 55°C. Fractions containing mRNA of the highest purity confirmed by electrophoresis (using 1 xTBE 1 .2% agarose gel) were combined, desalted (by ultrafiltration using Amicon Ultra-15 50K, Millipore) and precipitated overnight at -20°C (with 0.3 M NaOAc pH 5.2, one volume of isopropanol, and washed with 80% ethanol). Prior quality control and in vitro experiment mRNA pellets were resuspended in RNase-free water to obtain concentration of >1 pg / pL.

[0212] FIG 19 shows the results of IVT efficiency and capping efficiency analysis for these hEPO mRNAs obtained with T9. In agreement to what was previously observed forthe template encoding Flue (T1), the combination of T1 with primers of m7GpppA*pA*pG sequence, ensure very good IVT yields and capping yields (98-98%), even for heavily modified caps such as those comprising Bn6Ammodifications. Worth noticing, the trinucleotide primer comprising Bn6Ammodification afforded lower capping efficiency and IVT yield, despite the fact it was used at a two-fold higher concentration. This exemplifies again how the novel transcription initiation complexes disclosed herein enable cost-effective production of mRNAs comprising heavily modified 5’-ends. mRNA lipid nanoparticles (mRNA-LNPs) were prepared using SM-102 lipid purchased from BroadPharm (USA). The 1 ,2 dimyristoyl-rac-glycero-3-methoxypolyethylene glycol- 2000 (DMG-PEG2k), cholesterol (from ovine wool) and 1 ,2-dioctadecanoyl-sn-glycero-3- phosphocholine (DSPC) were purchased from Avanti Polar Lipids (USA). The stock solutions of lipids were prepared in absolute ethanol (Thermo Fisher Scientific) at the concentration of 100 mg / mL. The lipid mixes were prepared by combining SM-102, DSPC, cholesterol and DMG-PEG2k at molar ratio of 50:10:38.5:1 .5 in absolute ethanol at total concentration of 15 mM for SM-102 lipid mix. Stock solution of mRNA was diluted in 100 mM sodium citrate buffer pH 4.0 at the mRNA final concentration of 102 ng / pL. The lipid mix and mRNA solution were combined together in a microfluidic device (NanoAssemblr® IgniteTM, Precision NanoSystems) equipped with NxGen Cartridge (cat#NIN0002) at a flow ratio of 4:1 with a total flow rate of 12 mL / min. The final N / P ratio was 6, where N / P represents the ratio of ionizable nitrogen atoms to phosphate groups in the mixture. All resulting lipid nanoparticles (LNPs) were diluted 20-40 times in 1 x PBS sterile buffer (w / o calcium and magnesium ions) and concentrated by ultrafiltration using Amicon ultracentrifugal tubes (Merck Millipore) with 50kDa MWCO (2000 x g, 20°C). Final LNPs were stored at 4°C and diluted in 1 x PBS (w / o calcium and magnesium ions) sterile buffer before application into mice. Size distribution and polydispersity index were determined using dynamic light scattering (DLS) on Malvern Zetasizer Ultra Red (Malvern, UK) in PBS buffer at 25°C in back scatter mode. Encapsulation efficiency and concentration of mRNA entrapped in LNPs was determined using the Quant-iT Ribogreen RNA assay (Thermo Fisher Scientific, USA) by compering fluorescence intensities in the presence or absence of 0.1 % (w / v) Triton X-100.

[0213] Efficiency of hEPO protein production was monitored in the blood serum collected from mice C57BL / 6 at two timepoints: 6 h and 24 h after intravenous administration of 1 pg of hEPO-encoding mRNA-LNPs into mice. mRNAs co-transcriptionally capped with Bn6- modified m7GpppA*pA*pG tetranucleotide primers, i.e. m7GpppBn6AmpAmpG and rn7GpppBn6AmpApG, and with rn7GpppBn6AmpG trinucleotide primer, yielded highest hEPO levels at both monitored timepoints (6 h and 24 h) (FIG 25 and 26). Importantly, the activity of mRNA comprising tetranucleotide primers modified withBn6Amwas comparable to the activity ofBn6Amincorporated using trinucleotide primers, again confirming that the use of tetranucleotide primers does not abolish the beneficial effects ofBn6Ammodification. All Bn6-modified mRNAs (including co-transcriptionally capped with rn7GpppBn6AmpApG or rn7GpppBn6AmpAmpG tetranucleotide primer) had higher translational activity over current industry standard (rn7GpppAmpG).

[0214] Example 14: Comparison of capping efficiencies and in vitro transcription yields in the broad range of reaction conditions for mRNAs 5' capped with tetranucleotide primers m7GpppA*pA*pG; see FIG 27 to 30.

[0215] After confirming the biological compatibility of mRNAs obtained from initiating complexes according to the invention we further investigated the compatibility of these complexes with different transcription protocols. The limitation of heavily modified trinucleotide primers such as rn7GpppB6AmpG is that they require application of high cap concentration and strictly defined buffering conditions to achieve good IVT yield and capping efficiency (10 mM cap concentration, pH 6.5; Warminski et al., Trinucleotide mRNA Cap Analog N6- Benzylated at the Site of Posttranscriptionalm6AmMark Facilitates mRNA Purification and Confers Superior Translational Properties In Vitro and In Vivo, JACS (2024), 146, 12, 8149-8163). Such restrictions increase the cost of mRNA manufacturing and may not be acceptable for many industry applications. Initiating complexes yielding high capping efficiencies at lower cap concentrations and enabling more flexibility regarding IVT conditions are thus desired. Therefore, we have investigated how different conditions affect the IVT process in the presence of initiating complexes according to the present invention. In these experiments, template T1 (<t>6.5-AGG) was used again (see also Example 2). First, in vitro transcription reactions were run in the presence 2, 4, 5, 6, 8 or 10 mM of tetranucleotide primers m7GpppBn6AmpAmpG or rn7GpppBn6AmpApG in 40 mM Bis-Tris buffer pH 6.5, containing 2 mM spermidine, and 5 mM each of ATP, CTP, GTP and UTP or N1 -MeMJTP, 25 mM MgCI2, 10 mM DTT, 40 ng / pL template T1 , 1 U / pL RNase inhibitor (RiboLock, ThermoFisher Scientific), 0.002 U / pL inorganic pyrophosphatase (ThermoFisher Scientific) and 75 U / pL T7 RNA polymerase (Roche). Second, in vitro transcription reactions at pH different than 6.5 were run in 40 mM Tris buffer pH 7.0 (marked as 7.0T), 7.5 or 8.0 and 40 mM HEPES buffer pH 7.0 (marked as 7. OH), containing 2 mM spermidine, and 5 mM each of ATP, CTP, GTP and UTP or NI-MeMJTP, 25 mM MgCh, 10 mM DTT, 40 ng / pL linearized plasmid as the DNA template, 1 U / pL RNase inhibitor (RiboLock, ThermoFisher Scientific), 0.002 U / pL inorganic pyrophosphatase (ThermoFisher Scientific) and 75 U / pL T7 RNA polymerase (Roche). Reference uncapped mRNAs were prepared analogously, except cap analog was omitted. All mixed components were incubated for 2 h at 37°C for reaction in the presence of UTP or 3h at 37°C if NI-MeMJTP was used instead of UTP, and subsequently treated with 0.025 U / pL of DNase I (ThermoFisher Scientific) for 30 min at 37°C. The enzymes in the IVT mixtures were inactivated by addition of one volume of 50 mM EDTA, and full-length mRNAs were purified by affinity chromatography (described in Example 6) with oligo(dT)25 resin (POROS™ Oligo (dT)25 Affinity Resin, ThermoFisher Scientific). Capping efficiencies were determined densitometrically for the products of the 5' terminal cleavage with ribozyme (described in Example 8, and based on Vlatkovic et al., Pharmaceutics, 2022, 14(2):328; doi: 10.3390 / pharmaceuticsl 4020328). IVT yields were determined spectrophotometrically after mRNA purification by measuring the volume of concentrated eluate and the absorbance at 260 nm corresponding to mRNA concentration (described in Example 7).

[0216] The results of the experiments revealed a broad range of the useful concentrations of rn7GpppBn6AmpApG or m7GpppBn6AmpAmpG tetranucleotides applicable for in vitro transcription in the presence of T1. Capping efficiency ~90% as observed even for the lowest tested tetranucleotide primer concentration, i.e. 2 mM, and the IVT yields remained constant at the whole analyzed range of rn7GpppBn6AmpApG or m7GpppBn6AmpAmpG concentrations (0-10 mM) (FIG 27 and 29). Also the change of pH value of the IVT buffer did not dramatically affect capping efficiency or IVT yield within the tested range, demonstrating high flexibility of initiating complexes according to the invention in regards to IVT conditions (FIG 29 and 30). However, for the 5 mM or lower tetranucleotide primer concentration lowering the pH is beneficial for obtaining highly capped transcripts (FIG 28 and 30).

[0217] Example 15: Comparison of dsRNA content in the oliqofdThs-purified mRNAs co- transcriptionally capped with m7GpppA*pG*pG or m7GpppA*pA*pG tetranucleotide; see FIG 31.

[0218] Unexpectedly, we have also noticed that some initiating complexes according to the invention often produce fewer dsRNA impurities compared to others. To compare dsRNA contamination profiles for mRNAs co-transcriptionally capped with various m7GpppA*pG*pG or m7GpppA*pA*pG tetranucleotide primers we applied semi- quantitative dot-blot analysis with anti-dsRNA antibody. 5, 25 or 250 ng of each mRNA initially purified by affinity chromatography (POROS™ Oligo (dT)25 Affinity Resin, ThermoFisher Scientific) was immobilized on the nylon membrane (Hybond™-N+, Amersham™), incubated with dsRNA-specific J2 antibody (SCICONS), followed by the incubation with secondary anti-mouse HRP-conjugated antibody (Cell Signaling Technology) and HRP substrate (ECL™ Prime Western Blotting Detection Reagents). Chemiluminescence signal was detected using ImageQuant 800 (Amersham™) imaging system. dsRNA content was estimated using ImageQuantTL software by comparison of the detected signals to the measured intensities of standard dsRNA (Abnova) in the range of 0.078-10 ng (see Table 11). The analysis performed for various conditions (5 or 10 mM cap concentration in the IVT mix) or mRNA length and sequence composition (optimized or not optimized FFLuc or hEPO) provided similar results, indicating that the presence of theBn6Ammodification on the 5' end of mRNA significantly reduce dsRNA formation during IVT process. Dot-blot analysis of the mRNA samples with Bn6-modified and corresponding unmodified cap structures are presented on FIG 31 and determined dsRNA content is listed in Table 11 , indicating that:m7GpppBn6AmpGpG contains less dsRNA thanm7GpppAmpGpG (sample 1 vs. 8),m7GpppBn6AmpApG contains less dsRNA thanm7GpppAmpApG (sample 2 vs. 6),m7GpppBn6AmpGmpG contains less dsRNA thanm7GpppAmpGmpG (sample 4 vs. 3)m7GpppBn6AmpAmpG contains less dsRNA thanm7GpppAmpAmpG (sample 7 vs. 5), andm7GpppBn6AmpG contains less dsRNA thanm7GpppBAmpG (sample 10 vs. 9).

[0219] Table 11 : dsRNA content in the oligo(dT)25-purified mRNAs encoding firefly luciferase (FFLuc) co-transcriptionally capped with 10 mM m7GpppA*pG*pG or m7GpppA*pA*pG tetranucleotide (see also Example 3, Example 15 and FIG 31). mRNAs were analyzed by dot-blot using anti-dsRNA J2 antibody (SCICONS) and compared to dsRNA standard in the range of 0.078-10 ng (Abnova).

[0220] Table 11

[0221] 1contamination with dsRNA is expressed as a per mille of dsRNA in 5 ng of analyzed mRNA sample [<0.078 ng dsRNA / 5 ng of blotted mRNA (<15.6%o)]

[0222] 2contamination with dsRNA is expressed as a per mille of dsRNA in 25 ng of analyzed mRNA sample [<0.078 ng dsRNA / 25 ng of blotted mRNA (<3.12%o)]3contamination with dsRNA is expressed as a per mille of dsRNA in 250 ng of analyzed mRNA sample [<0.078 ng dsRNA / 250 ng of blotted mRNA (<0.31 %o)]

[0223] 4dsRNA content in 5 ng of the analyzed mRNA sample is not present in the sample or dsRNA content is below the detection limit of the method (<15.6%o), which is determined based on the dsRNA standard in the range of 0.078-10 ng (Abnova) Example 16: Comparison of IVT yields and capping efficiencies obtained with a standard IVT protocol recommended for co-transcriptional capping with market-standard trinucleotide primers: see FIG 32

[0224] To validate compatibility of tetranucleotide primers m7GpppA*pA*pG with market standard IVT protocol for co-transcriptional capping with rn7GpppAmpG we have prepared set of IVT reactions containing 4 mM of commercially available trinucleotide primers: rn7GpppAmpG, rn7’3 OGpppAmpG or rn7’3 OGpppm6AmpG, (FIG 32, lanes 1-3); Bn6- comprising trinucleotide primer rn7GpppBn6AmpG (FIG 32, lane 4); Bn6-modified tetranucleotide primers: rn7GpppBn6AmpApG or m7GpppBn6AmpAmpG (FIG 32, lanes 5 and 6), and unmodified tetranucleotide primers: rn7GpppAmpApG or m7GpppAmpAmpG (FIG 32, lanes 7 and 8) (see also Table 4). IVT mix was supplemented with 20 ng / pL of the PCR-generated template T10 (<t>6.5-AGG) (see also Table 5) comprising of <t>6.5- promoter and AGG trinucleotide initiating sequence (i.e. with TTCC deoxyribonucleotides at positions -1 , +1 , +2 and +3 of the non-coding strands of the DNA template; see also FIG 4), and encoding codon-optimized firefly luciferase (construct FFLuc2). Tetranucleotide primers m7GpppA*pA*pG pair with -1 +1 and +2 positions of T10, whereas trinucleotide primers such as rn7GpppAmpG pair with +1 and +2 positions of T10. In vitro transcription reactions were prepared using 40 mM Tris buffer pH 8.0, containing 2 mM spermidine, 16.5 mM MgOAc, 10 mM DTT, 0.002% (v / v) Triton X-100, and 5 mM each of ATP, CTP, GTP and UTP, 1 U / pL RNase inhibitor (RiboLock, ThermoFisher Scientific), 0.002 U / pL inorganic pyrophosphatase (ThermoFisher Scientific) and 75 U / pL T7 RNA polymerase (Roche). Synthesized mRNAs were purified by affinity chromatography (described in Example 6) with oligo(dT)25 resin (POROS™ Oligo (dT)25 Affinity Resin, ThermoFisher Scientific) or using silica-based approach (Monarch® RNA Cleanup Kit, New England Biolabs) according to the manufacturer’s instruction, and the capping efficiencies were determined densitometrically for the products of the 5' terminal cleavage with ribozyme (described in Example s, and based on Vlatkovic et al., Pharmaceutics, 2022, 14(2):328; doi: 10.3390 / pharmaceuticsl 4020328). IVT yields were determined spectrophotometrically after mRNA purification by measuring the volume of concentrated eluate and the absorbance at 260 nm corresponding to mRNA concentration (described in Example 7).

[0225] Obtained results (FIG 24) demonstrate that while unmodified trinucleotide, rn7GpppAmpG, yields good capping efficiency under these conditions, further modification of its structure result in lowered capping efficiency (to 93%, 72%, and 67% for rn73 OGpppAmpG, m73-°Gpppm6AmpG, and rn7GpppBn6AmpG, respectively). In contrast, all the tested tetranucleotide primers (m7GpppA*pA*pG) that form with T10 initiating complexes according to this invention (rn7GpppBn6AmpApG, m7GpppBn6AmpAmpG, rn7GpppAmpApG and rn7GpppAmpAmpG) are compatible with IVT protocol recommended for efficient co- transcriptional capping with rn7GpppAmpG as evidenced by high capping efficiencies and excellent IVT yields. Lowering cap analog and Mg2+concentration in comparison to previously tested IVT conditions for tetranucleotide primers (see Example 3 and Example 9) affected neither efficiency of co-transcriptional capping nor IVT yields. Additionally, production parameters for the transcripts capped with tetranucleotide primers m7GpppA*pA*pG, suggest that they can functionally replace rn7GpppAmpG in the IVT mix without any further optimizations of mRNA synthesis protocols (FIG 32, lanes 5- 8 vs. lanes 1 -3). This compatibility also applies to heavily-modified tetranucleotide primers, i.e. rn7GpppBn6AmpApG and m7GpppBn6AmpAmpG, providing straightforward access to mRNAs with superior biological activity. Consequently, the initiating complexes according to the invention allow convenient incorporation of various modifications providing superior biological properties, including

[0226] Bn6Am, [https: / / pubmed.ncbi.nlm.nih.gov / 38442005 / ] but alsom6Amand their derivatives (see also Example 12 and Example 13 and FIG 18 to 23, and FIG 25 and 26). Moreover, the same IVT protocol is suitable for efficient incorporation of cap 2 structure into mRNA when using tetranucleotide primer m7GpppAmpAmpG and m7GpppBn6AmpAmpG, which is not possible with standard trinucleotide primers available on the market.

[0227] 17: n of tetranucleotide at the 5' end. IVT yield and

[0228] As it was initially stated in Example 10, satisfying result of co-transcriptional capping with heavily modified m7GpppBn6A1*pU2*pG3tetranucleotide primers (wherein A1* and U2* denote adenosine or uridine moieties corresponding to the first transcribed nucleotides, i.e. N1and N2, and the asterisk indicates possible methylations at the positions 2'-O to form various cap 1 and cap 2 analogs) requires DNA template carrying <t>6.5 promoter followed by TGG trinucleotide initiating sequence (template T5, see Table 5) (FIG 4C, FIG 13 and FIG 14B). However, additional modifications have been implemented for a further improved capping efficiency with rn7GpppBn6AmpUpG cap 1 analog.

[0229] To that end, we have employed PCR-generated DNA templates encoding codon- optimized firefly luciferase (construct FFLuc2,), carrying <t>6.5 promoter followed by TGG (template T5), TGG followed by an alternative 5' UTR (template T11), and TGA sequence (template T12) followed by 5' UTR analogous to template T5 (see Table 5). mRNAs were synthesized by in vitro transcription based on the procedure described in Example 4, including the following modifications:

[0230] IVT buffer x1 containing 2 mM spermidine and 40 mM Bis-Tris pH 6.5 or 40 mM Tris pH 8.0;

[0231] 4, 5, or 10 mM tetranucleotide primer rn7GpppBn6AmpUpG or m7GpppBn6AmpUmpG and 8 mM trinucleotide primer rn7GpppBn6AmpU or 4 mM unmodified trinucleotide rn7GpppAmpU;

[0232] - 5 mM each of ATP, UTP, CTP and GTP;

[0233] 16.5-33 mM MgOAc or 25 mM MgCh;

[0234] 20 ng / pL of PCR-generated DNA template;

[0235] 75 U / pL T7 RNA polymerase (Roche) or 10 U / pL PrimeCap T7 RNA polymerase (TaKaRa) (details of the usage of particular conditions are denoted on respective figures). The mRNAs were purified by affinity chromatography (as described in Example 6) or using silica-based approach (Monarch® RNA Cleanup Kit, New England Biolabs) according to the manufacturer’s instructions. Purified mRNAs were cleaved at the 5' ends using ribozyme, separated on the denaturing polyacrylamide gel with 7 M urea and analyzed by densitometry to determine the capping efficiencies (Example 8), as described previously (Vlatkovic et al., Ribozyme Assays to Quantify the Capping Efficiency of In Vitro-Transcribed mRNA. Pharmaceutics. (2022), 14(2):328). IVT efficiencies were determined spectrophotometrically after mRNA purification by measuring the volume of the eluate and the absorbance at 260 nm corresponding to mRNA concentration (Example 7). Uncapped mRNA, which was used as a reference, was produced the same way, except that no cap analog was added to the IVT mix (uncapped mRNA in vitro transcribed using template T5 (<t>6.5-TGG) is analyzed in lane 8 in FIG 12, and using template T12 (<t>6.5-TGA) is analyzed in lane 0 in FIG 33B).

[0236] In the case of template T5 (<t>6.5-TGG), the coding strand of TIS starts with a pyrimidine deoxyribonucleotide, which shifts position +1 (being the actual transcription start site (TSS)), to the first deoxynucleotide containing a purine (in this case, guanine, see FIG. 13). As previously, we observed that for efficient initiation of transcription using template T5 (<t>6.5-TGG) and tetranucleotide cap primers carrying bulky substitution on the nucleotide N1, i.e. rn7GpppBn6A1mpU2pG3and m7GpppBn6A1mpU2mpG3, the Bn6-modified residue is aligned with position -2 of the <t>6.5 promoter (FIG 13, lanes 1 and 3 in FIG 14B and lanes 13-16 in FIG 34). The position -2 is located at the very end of the core promoter region (see FIG 13) and corresponds to position -1 of DNA templates with TIS initiated with nucleotide containing purine, e.g. template T 1 (<t>6.5-AGG) or T4 (<t>6.5-ATG). In the case of the template T4 (<t>6.5-ATG) the alignment of Bn6-modified adenosine (nucleotide N1) with position +1 of the non-coding DNA strand (it is also the first nucleotide of TIS) almost completely inhibited activity of T7 RNA polymerase under all tested IVT conditions (lanes 1 and 3 in FIG 14A vs. lanes 1 and 3 in FIG 14B and lanes 1 -4 and 9-12 in FIG 34). Additionally, the performed experiments showed that single-nucleotide substitution at position +2 of the template T5 (<t>6.5-TGG), which provides template T12 (<t>6.5-TGA) can elevate capping efficiency for rn7GpppBn6AmpUpG from 69% to 78% (compare lanes 9-15 in FIG 35 with lanes 5-8 in FIG 36). However, regardless of maintaining satisfying IVT efficiency, capping with tetranucleotide rn7GpppBn6AmpUpG was still below 80%. An analogous approach also moderately increased efficiency of cap 2 analog rn7GpppBn6AmpUmpG incorporation (compare lanes 5-8 in FIG 34 with lanes 2-4 in FIG 33), but at the same time, the IVT yield dropped significantly.

[0237] Therefore, further modification of IVT conditions was implemented by replacing the wildtype T7 RNA polymerase (Roche) with the commercially available enzyme designed for the reduction of dsRNA formed during in vitro transcription and increasing capping efficiency (PrimeCap T7, TaKaRa) (FIG. 37). The usage of PrimeCap T7 RNA polymerase resulted in more efficient incorporation of rn7GpppBn6AmpUpG using template T5 (<t>6.5-TGG) (compare lanes 8-10 in FIG 37A with lanes 13, and lane 15 in FIG 35) and increased IVT yield for m7GpppBn6AmpUmpG (compare lanes 22-24 in FIG 37B with lanes 3 and 4 in FIG 33 and lanes 7 and 8 in FIG 34), providing optimized conditions for obtaining Bn6-modified cap 1 and cap 2 structures of the mRNAs initiated withBn6AU sequence.

[0238] Significantly increased yield of mRNA in vitro transcribed using WT T7 RNA polymerase and initiated with heavily modifiedBn6A*U* sequence was obtained for an alternative tetranucleotide cap primer design. Particularly, m7GpppBn6A1*pU2*pA3tetranucleotide formed an initiation complex with the template T13 (<t>6.5-TAG) or T14 (<t>6.5-TAG with alternative 5' UTR), providing alignment of Bn6-modified adenosine A1* with the position -2 within the core promoter region of the non-coding template strand (FIG 38 and Table 5). Analogously to the previously described experiments (Example 10 and 17) when mRNA was in vitro transcribed without cap analog and using templates containing pyrimidine deoxyribonucleotide within coding strand at position following core promoter region (template T13 (<t>6.5-TAG) or T14 (<t>6.5-TAG with alternative 5' UTR)), generated RNA product was 1 nt shorter than the expected length (compare lanes 7 and 8 on FIG 12B vs. lanes 2 and 10 on FIG 39B with uncapped RNAs indicated with arrows on the bottom gels).

[0239] For optimization of the IVT conditions providing efficient incorporation of m7GpppBn6A1*pU2*pA3tetranucleotide primers, mRNAs were synthesized based on the procedure described in Example 4 and modifications introduced in Example 17, i.e.:

[0240] IVT buffer x1 containing 2 mM spermidine and 40 mM Bis-Tris pH 6.5 or 40 mM Tris pH 8.0;

[0241] 4, 5, or 10 mM tetranucleotide primer rn7GpppBn6AmpUpA or m7GpppBn6AmpUmpA and or 4 mM trinucleotide rn7GpppBn6AmpU or unmodified trinucleotide rn7GpppAmpU;

[0242] - 5 mM each of ATP, UTP, CTP and GTP;

[0243] 16.5-33 mM MgOAc and 25 or 30 mM MgCh;

[0244] 20 ng / pL of PCR-generated DNA template;

[0245] 75 U / pL T7 RNA polymerase (Roche) or 10 U / pL PrimeCap T7 RNA polymerase (TaKaRa) (details of the usage of particular conditions are denoted on respective figures). mRNAs were purified by affinity chromatography (described in Example 6) or using silica- based approach (Monarch® RNA Cleanup Kit, New England Biolabs) according to the manufacturer’s instruction. Purified mRNAs were cleaved at the 5' ends using a ribozyme, and separated on a denaturing polyacrylamide gel with 7 M urea followed by densitometric analysis to determine the capping efficiencies (Example 8), as described previously. [https: / / pubmed.ncbi. nlm.nih.gov / 35214060 / ] IVT efficiencies were determined spectrophotometrically after mRNA purification by measuring the volume of the eluate and the absorbance at 260 nm corresponding to mRNA concentration (Example 7).

[0246] Results of the performed experiments indicate that according to the invention described herein both m7GpppBn6A1*pU2*pA3analogs (i.e. cap 1 and cap 2) are efficiently incorporated into the 5' end of mRNA only if the template used for the IVT provides alignment ofBn6A1* with the position -2 of the non-coding strand of the template with TIS initiated with pyrimidine deoxyribonucleotide (template T13 (<t>6.5-TAG), FIG 38 and FIG 39). Moreover, using template T16 (<t>6.5-ATA) resulted in almost no mRNA produced during IVT, proving that the alignment of m7GpppBn6A1*pU2*pA3with positions +1 , +2 and +3 of the non-coding DNA strand highly affects initiation of IVT (compare lanes 6-8 and lanes 2-4 in FIG 39). This observation is also valid for Bn6-modified trinucleotides when applying conditions optimal for rn7GpppAmpU (lane 12 vs. lane 10, FIG 39A). The replacement of the WT T7 RNA polymerase (Roche) with PrimeCap T7 (TaKaRa) further improved capping efficiency. However, WT T7 enzyme also provides satisfactory IVT parameters, especially in terms of IVT yield (FIG 40 and FIG 41), and the capping efficiency can be increased when WT T7 RNA polymerase is combined with the template T14 (<t>6.5-TAG with alternative 5' UTR) (lanes 4-6 vs. lanes 1-3 in FIG 40 and lanes 18- 20 vs. lanes 15-17 in FIG 41).

[0247] Example 18: In vitro translational properties of the mRNA co-transcriptionally capped with tetranucleotide primers m7GpppBn6A*pU*pG or m7GpppBn6A*pU*pA; see FIG 42 to 47.

[0248] A set of HPLC-purified transcripts was prepared fortesting the translational effects of the adaptations implemented for efficient production of mRNAs initiated withBn6A*U* sequence (see Example 17). Additionally, mRNAs capped with m7GpppBn6A*pA*pG primers and unmodified trinucleotides (rn7GpppAmpU, rn7GpppAmpG, and rn7GpppAmpA) were used to directly compare biological effects related to differently initiated transcripts, i.e. withBn6AmU,Bn6AmUm,Bn6AmA,Bn6AmAm, AmU, AmA or AmG sequence at the 5' end. mRNAs were prepared based on various IVT conditions summarized in Table 12 and being the effect of the optimization for particular cap primer::DNA template complexes (see also Example 17). In short, most of the mRNAs encoding codon-optimized firefly luciferase (construct FFLuc2) were in vitro transcribed using 40 mM Bis-Tris buffer pH 6.5 and 4 mM (m7GpppBn6AmpUpA and m7GpppBn6AmpUmpA) or 5 mM (m7GpppBn6AmpUpG, m7GpppBn6AmpUmpG, rn7GpppBn6AmpApG, and m7GpppBn6AmpAmpG) tetranucleotide primers for co-transcriptional capping. The reference mRNAs were in vitro transcribed using 40 mM Tris buffer pH 8.0 and 4 mM of unmodified trinucleotide rn7GpppAmpU and rn7GpppAmpG or 40 mM Bis-Tris buffer pH 6.5 and 10 mM rn7GpppAmpA. PCR-generated DNA templates T5 ( 6.5-TGG), T12 ( 6.5-TGA), T13 ( 6.5-TAG), T14 (<t>6.5-TAG with alternative 5' UTR), and T15 (<t>6.5-AGG) (see Table 5) were prepared as described in Example 2. The synthesized mRNAs were initially purified by affinity chromatography (described in Example 6) with oligo(dT)25 resin (POROS™ Oligo (dT)25 Affinity Resin, ThermoFisher Scientific), and the capping efficiencies were determined densitometrically for the products of the 5' terminal cleavage with ribozyme (described in Example 8, and based on Vlatkovic et al., Pharmaceutics (2022) 14(2):328; doi:

[0249] 10.3390 / pharmaceuticsl 4020328) (FIG 42). IVT yields were determined spectrophotometrically after mRNA purification by measuring the volume of the eluate and the absorbance at 260 nm corresponding to mRNA concentration (described in Example 7). To reduce the influence of dsRNA or low molecular weight contaminants, such as short RNA species, transcripts were further purified by RP-HPLC prior in vitro transfection. Each sample was applied on the RNASep™ Semi-Prep column (ADS Biotec) combined with Agilent Infinity 1260 II HPLC system, and eluted with the linear gradient (10-14.5%) of acetonitrile in 0.1 M TEAA, pH 7.0 at 55°C. Fractions containing mRNA of the highest purity confirmed by electrophoresis (using 1 xTBE 1 .2% agarose gel) were combined, desalted (by ultrafiltration using Amicon Ultra-15 50K, Millipore), precipitated overnight at -20°C (with 0.3 M NaOAc pH 5.2, one volume of isopropanol) and washed with 80% ethanol. Prior quality control and in vitro experiment mRNA pellets were resuspended in RNase-free water to obtain concentration of >1 pg / pL.

[0250] To assess the translational activity of mRNAs 5'-capped with rn7GpppBn6AmpUpG, m7GpppBn6AmpUmpG, m7GpppBn6AmpUpA, m7GpppBn6AmpUmpA, rn7GpppBn6AmpApG, rn7GpppBn6AmpAmpG, and rn7GpppAmpU, rn7GpppAmpG or rn7GpppAmpA we have transfected selected cell types: HEK293T and murine or human primary macrophages (also marked as mM<t> and hM<t>, respectively) and measured luminescence at several timepoints. 50x103murine and human primary macrophages and 10x103HEK293T were seeded on a 96 well plate. The next day, cells were transfected with 20 or 100 ng mRNA using Lipofectamine MessengerMax. Medium was collected 24 h after transfection. Luciferase activity was measured with BrightGlo at 4, 24, 48, and in some cases also 72 h. Viability of the cells was measured with CellTiter Blue 24 (each of tested cell types) and 48 h (HEK293T and human primary macrophages) after transfection. At all tested time points in HEK293T cells transfected with mRNA initiated with AUGG sequence obtained using T5 template (i.e. <t>6.5-TGG) and rn7GpppBn6AmpUpG, m7GpppBn6AmpUmpG or rn7GpppAmpU primers yielded the highest overall luminescence in comparison to other transcripts (FIG 43 and 44). In some cases also mRNAs in vitro transcribed using template T15 (<t>6.5-AGG) and rn7GpppBn6AmpApG, rn7GpppBn6AmpAmpG primers provided equally good translatability. These results are consistent with previous observations for the HEK293T cells (see Example 12, and FIG 18 and 19). Notably, Bn6modification increased mRNA translatability in primary cells. In human macrophages (FIG 45 and 46) this effect was the most dominant for mRNAs possessing Bn6-modified cap 2 structures, especially when the transcripts were initiated with AUGG or AAGG sequence, i.e., for mRNAs co-transcriptionally capped with rn7GpppBn6AmpUmpG or m7GpppBn6AmpAmpG tetranucleotides using template T5 (<t>6.5- TGG) and T15 (<t>6.5-AGG), respectively. <t>hM transfected with 100 ng of mRNA initiated with AUAG sequence (i.e. for transcripts obtained using m7GpppBn6AmpUmpA tetranucleotide and the template T13 (<t>6.5-TAG)) also provided relatively high luminescence readouts. On the other hand, in murine macrophages transfected with mRNAs capped with rn7GpppBn6AmpUpG or rn7GpppBn6AmpApG tetranucleotides (using template T5 (<t>6.5-TGG) and T15 (<t>6.5-AGG), respectively) the luminescence was significantly higher in comparison to the corresponding transcripts with unmodified 5' cap structures. However, unlike hM<t>, in murine macrophages Bn6-modified mRNA with cap 1 structure improves translatability more than Bn6-modified cap 2 (compare FIG 47 with FIG 45 and 46). Overall, these results show again that the highly modified initiation complexes according to the present invention produce mRNAs with highly modified 5’ ends characterized by superior translational properties, and this effect is observed regardless of the 5’-terminal RNA sequence.

[0251] In vitro experiments highlighted that mRNAs with Bn6-modified 5' end starting with m7GpppBn6A*pU*pG or m7GpppBn6A*pA*pG structure provide the best translatability of all tested transcripts. Introduction of such modifications, including cap 2 structure, was possible because of co-transcriptional positioning of heavily modified nucleotide N1at position -2 of the template T5 (<t>6.5-TGG, see FIG 13) and position -1 of the template T15 (<t>6.5-AGG, which is analogous to the template T1 (see FIG 6), and differs only with regard to the, coding sequence, i.e. standard (template T1) or codon-optimized (template T15)), respectively. It was shown above that trinucleotide primer rn7GpppBn6AmU can be efficiently incorporated into 5' end of in vitro transcribed mRNA (lane 7 in FIG 14A and 14B, lane 17 and 18 in FIG 34 and lane 7 in FIG 48). However, as it was proven by in vitro experiment using Bn6-modified tetranucleotide with the second ribose being also 2'- O methylated (i.e. m7GpppBn6AmpUmpG) could increase translatability of mRNAs possessing such modification of the 5' end, and the cap 2 structure is directly incorporable only for m7GpppBn6AmpUmpG primer combined with template T5 (<t>6.5-TGG) (see results depicted in FIG 45 and 46). Alternative design of the cap primer: :DNA template complex for obtaining heavily modified 5' end of RNA using m7GpppBn6A*pU*pA tetranucleotide and the templates with TAG as a TIS (i.e. template T13 (<t>6.5-TAG) and T14 (<t>6.5-TAG with alternative 5' UTR), see FIG 38) is also useful, and in particular comparisons giving improvement over analogous, but unmodified transcripts.

[0252] Table 12. Summarized IVT conditions for tetrameric and trimeric capped oligonucleotide primers used for co-transcriptional capping of mRNA for the in vitro experiment with HEK293T, and murine or human primary macrophages (see Example 18). 1 TIS -trinucleotide initiating sequence of the dsDNA template. Indicated sequence corresponds to the nucleotides of the coding strand following <P6.5 promoter. First transcribed nucleotide which corresponds to the position +1 of the template strand is underlined. Bold font indicates nucleotide which is omitted while in vitro transcription is performed without cap analog added to the IVT mix.

[0253] 2T7 RNA polymerase used for in vitro transcription: WT - wild type enzyme manufactured by Roche; PrimeCap T7 - low dsRNA variant of T7 RNA polymerase manufactured by TaKaRa.

[0254] 19: m7GpppBn6AmpApU or m7GpppBn6AmpAmpU tetranucleotide cap primers are with DNA templates initiated with ATG sequence (<t>6.5-ATG).

[0255] Under some circumstances, modification of TIS ATG in the template could be undesired; therefore using cap analogs compatible exclusively with <t>6.5-ATG template might be obligatory. Co-transcriptional introduction of several cap structures, including combination of cap 2 with benzyl group, requires tetranucleotide primer. However, for <t>6.5-ATG template (e.g. template T4) IVT involving co-transcriptional capping with primer rn7GpppBn6AmpUmpG is inefficient and mRNA production in the presence of this analog is almost completely inhibited (lanes 1 and 3 in FIG 14A and lanes 1-4 and 9-12 in FIG 34). Therefore, using tetranucleotide m7GpppBn6AmpUmpG requires modification of the template <t>6.5-ATG sequence. To overcome this obstacle with the template modification, we have designed another complex according to the present invention, comprising a rn7GpppBn6AmpAmpU primer and DNA template T4 (<t>6.5-ATG).

[0256] To determine IVT and capping efficiency of mRNAs co-transcriptionally capped with tetranucleotides m7GpppBn6A*pA*pU we used PCR-generated template T4 encoding FFLuc2codon-optimized firefly luciferase (construct FFLuc2) and carrying <t>6.5 promoter followed by ATG as the trinucleotide initiating sequence (see Table 5). The mRNAs were synthesized as described in Example 4, including following modifications:

[0257] IVT buffer x1 containing 2 mM spermidine and 40 mM Bis-Tris pH 6.5 or 40 mM Tris pH 8.0;

[0258] 2, 5, or 8 mM tetranucleotide rn7GpppBn6AmpApU or m7GpppBn6AmpAmpU and 4 mM rn7GpppAmpU trinucleotide or 8 mM trinucleotide rn7GpppBn6AmpU;

[0259] - 5 mM each of ATP, UTP, CTP and GTP;

[0260] 27 and 30 mM MgOAc;

[0261] 20 ng / pL of PCR-generated DNA template;

[0262] 75 U / pL T7 RNA polymerase (Roche)

[0263] (details of the usage of particular conditions are denoted on respective figures).

[0264] The mRNAs were purified by affinity chromatography (described in Example 6) using oligo(dT)25 resin (POROS™ Oligo (dT)25 Affinity Resin, ThermoFisher Scientific), and the capping efficiencies were determined densitometrically for the products of the 5' terminal cleavage with ribozyme (described in Example 8, and based on Vlatkovic et al., Pharmaceutics (2022) 14(2):328; doi: 10.3390 / pharmaceuticsl 4020328) (FIG 48). IVT efficiencies were determined spectrophotometrically after mRNA purification by measuring the volume of the eluate and the absorbance at 260 nm corresponding to mRNA concentration (Example 7).

[0265] Initially, we found that m7GpppA1mpA2mpU3tetranucleotide is highly compatible with template T4 (<t>6.5-ATG) (lane 3 on FIG 12 and lane 5 on FIG 14A), and such primer::DNA complex guarantees complete alignment of nucleotides N1, N2and N3with the template at positions -1 , +1 and +2 of the non-coding strand (see FIG 13). An analogous primer with nucleotide N1substituted with the benzyl group was further examined in broad range of cap concentration, i.e. 2, 5 and 8 mM, providing extraordinary IVT yield and capping efficiency at all tested conditions (FIG 48). 5 mM m7GpppBn6AmpApU and m7GpppBn6AmpAmpU are the best choice to reach yields comparable to unmodified trinucleotide rn7GpppAmpU (lanes 2 and 5 vs. lane 8 on FIG 48). However, decreasing the concentration of both m7GpppBn6A*pA*pU variants to even 2 mM did not affect capping efficiency, allowing for a reduction in the consumption of the capping reagent.

[0266] Example 20: In vitro translational properties of the mRNAco-transcriptionally capped with tetranucleotide primers m7GpppBn6A*pA*pU or m7GpppBn6A*pA*pA; see FIG 49-54.

[0267] Set of HPLC-purified transcripts was prepared to validate translatability of mRNA capped with m7GpppBn6A*pA*pU tetranucleotide primers and directly compare them with other analogs introducing bulky substitutions within 5' end, i.e.: a) rn7GpppBn6AmpUpG and m7GpppBn6AmpUmpG tetranucleotides forming cap primer::DNA template complex with templates initiated with TGG sequence (template T5 (<t>6.5-TGG)). Bn6-modified rn7GpppBn6AmpU and unmodified rn7GpppAmpU were used as the relevant trinucleotide controls; b) rn7GpppBn6AmpApG and m7GpppBn6AmpAmpG tetranucleotides forming cap primer::DNA template complex with templates initiated with AGG sequence (template T15 (<t>6.5-AGG)); Bn6-modified rn7GpppBn6AmpG and rn7GpppBn6AmpA and unmodified rn7GpppAmpG and rn7GpppAmpA were used as the relevant trinucleotide controls; c) rn7GpppBn6AmpApA and m7GpppBn6AmpAmpA tetranucleotides forming cap primer::DNA template complex with templates initiated with AGG sequence (template T15 (<t>6.5-AGG)); Bn6-modified m7GpppBn6AmpA and unmodified rn7GpppAmpA were used as the relevant trinucleotide controls. mRNAs were prepared based on various IVT conditions summarized in Table 13 and being the effect of the optimization for particular cap primer::DNA template complexes (see also Example 17 and 19). In short, the mRNAs encoding codon-optimized firefly luciferase (construct FFLuc2) were in vitro transcribed using 40 mM Bis-Tris buffer pH 6.5 and co-transcriptionally capped with tetranucleotide primers (m7GpppBn6AmpApU, m7GpppBn6AmpAmpU, m7GpppBn6AmpUpG, m7GpppBn6AmpUmpG, rn7GpppBn6AmpApG, rn7GpppBn6AmpAmpG, rn7GpppBn6AmpApA and m7GpppBn6AmpAmpA). The reference mRNAs capped with unmodified trinucleotides were in vitro transcribed using 40 mM Tris buffer pH 8.0 and 4 mM of rn7GpppAmpU and rn7GpppAmpG or 10 mM of rn7GpppAmpA. PCR-generated DNA templates T4 ( 6.5-ATG), T5 (<t>6.5-TGG) and T15 (<t>6.5-AGG) (see Table 5) were prepared as described in Example 2. The synthesized mRNAs were initially purified by affinity chromatography (described in Example 6) with oligo(dT)25 resin (POROS™ Oligo (dT)25 Affinity Resin, ThermoFisher Scientific). IVT yields were determined spectrophotometrically after mRNA purification by measuring the volume of the eluate and the absorbance at 260 nm corresponding to mRNA concentration (described in Example 7). To reduce the influence of dsRNA or low molecular weight contaminants, such as short RNA species, transcripts were further purified by RP-HPLC prior in vitro transfection. Each sample was applied on the RNASep™ Semi-Prep column (ADS Biotec) combined with Agilent Infinity 1260 II HPLC system, and eluted with the linear gradient (10-14.5%) of acetonitrile in 0.1 M TEAA, pH 7.0 at 55°C. Fractions containing mRNA of the highest purity confirmed by electrophoresis (using 1 xTBE 1.2% agarose gel) were combined, desalted (by ultrafiltration using Amicon Ultra-15 50K, Millipore), precipitated overnight at -20°C (with 0.3 M NaOAc pH 5.2, one volume of isopropanol) and washed with 80% ethanol. Prior quality control and in vitro experiment mRNA pellets were resuspended in RNase-free waterto obtain concentration of >1 pg / pL. Cap primer: :DNA template complex formation was determined based on electrophoretic profile obtained for the products of mRNAs 5' terminal cleavage with ribozyme (described in Example s, and based on Vlatkovic et al., Pharmaceutics, 2022, 14(2):328; doi: 10.3390 / pharmaceuticsl 4020328) (FIG 49).

[0268] To assess the translational activity of mRNAs 5'-capped with rn7GpppBn6AmpApU, m7GpppBn6AmpAmpU, m7GpppBn6AmpUpG, m7GpppBn6AmpUmpG, rn7GpppBn6AmpApG, m7GpppBn6AmpAmpG, m7GpppBn6AmpApA, m7GpppBn6AmpAmpA and rn7GpppBn6AmpU, rn7GpppAmpU, m7GpppBn6AmpG, rn7GpppAmpG, rn7GpppBn6AmpA or rn7GpppAmpA we have transfected selected cell types: HEK293T and murine (mM<t>) or human (hM<t>) primary macrophages and measured luminescence at several timepoints. 50x103murine and human primary macrophages and 10x103HEK293T were seeded on 96 well plate. The next day, cells were transfected with 20 or 100 ng mRNA using Lipofectamine MessengerMax. Medium was collected 24 h after transfection. Luciferase activity was measured with BrightGlo at 4, 24 and 48 h. Viability of the cells was measured with CellTiter Blue 24 (each of tested cell types) and 48 h (HEK293T and human primary macrophages) after transfection.

[0269] As observed in experiment described in Example 18 in HEK293T cells transfected with lower dose (20 ng / well) of mRNA initiated with AUGG sequence obtained using T5 template (i.e. <t>6.5-TGG) and rn7GpppBn6AmpUpG, m7GpppBn6AmpUmpG, rn7GpppBn6AmpU or rn7GpppAmpU primers yielded the highest overall luminescence in comparison to other transcripts (FIG 50 and 51). At these conditions comparable translatability was obtained for mRNAs in vitro transcribed using template T4 (<t>6.5-ATG) and T15 (<t>6.5-AGG). However, in HEK293T cells transfected with increased dose of mRNA (100 ng / well) luminescence reads for the transcripts obtained using template T15 (<t>6.5-AGG) and T5 template (<t>6.5-TGG) were comparable to each other, and as previously reported (see Example 12, and FIG 18 and 19 and Example 18, and FIG 43 and 44), obvious advantage of Bn6modification was not observed.

[0270] Analogously to the results described in Example 18 Bn6modification increased mRNA translatability in primary cells, i.e. mM<t> and hM<t>. In later timepoints (24 and 48 h) some mRNAs with Bn6-modified cap 2 structures (m7GpppBn6AmpAmpU, m7GpppBn6AmpUmpG and rn7GpppBn6AmpAmpG) provided higher luminescence in comparison to their cap 1- containing counterparts obtained using tetranucleotide (m7GpppBn6AmpApU, rn7GpppBn6AmpUpG and rn7GpppBn6AmpApG) or trinucleotide primers (rn7GpppBn6AmpA, rn7GpppBn6AmpU and rn7GpppBn6AmpG) (FIG 52 and 53). This effect was dominating for the transcripts initiated with AUGG or AAGG sequence, i.e. for mRNAs co- transcriptionally capped with m7GpppBn6AmpUmpG or m7GpppBn6AmpAmpG tetranucleotides using template T5 (<t>6.5-TGG) and T15 (<t>6.5-AGG), respectively. In murine macrophages the highest luminescence was obtained for mRNAs capped with Bn6-modified cap 1 trinucleotides: rn7GpppBn6AmpU (transcript initiated with AUGG sequence using template T5 (<t>6.5-TGG)), rn7GpppBn6AmpA and rn7GpppBn6AmpG (transcript initiated with AAGG sequence using template T15 (<t>6.5-AGG)) (see FIG 54). The luminescence readouts were the lowest in mM<t> transfected with transcripts obtained using template T4 (<t>6.5-ATG). However, heavily modified trinucleotides rn7GpppBn6AmpU and rn7GpppBn6AmpA provided better translatability than their unmodified counterparts (rn7GpppAmpU and rn7GpppAmpA, respectively). In vitro experiments described herein and in Example 18 emphasized that Bn6-modified cap primers provide increased translatability, especially in primary cells. In case of human primary macrophages representing immune cells often being targeted at therapeutic applications usage of exogenous mRNAs with cap 2 structure seems to be beneficial in terms of mRNA’s translatability. For incorporation of cap analog combining both modifications (benzyl group present on nucleotide N1and methyl group present on the ribose ring of nucleotide N2), it is mandatory to apply cap primer::DNA template complex which aligns Bn6-modified nucleotide at position -1 within <t>6.5 promoter. RNAs initiated with AUG sequence (e.g. saRNA) are efficiently in vitro transcribed only in the presence of cap analog appropriately aligned with two types of templates: <t>6.5-ATG and <t>6.5- TGG (see Fig 12). Results shown on FIG 45-47 and FIG 52-54 prove that various cap primers containing nucleotide N1substituted with benzyl group (rn7GpppBn6AmpUpG, m7GpppBn6AmpUmpG, m7GpppBn6AmpApU, m7GpppBn6AmpAmpU, rn7GpppBn6AmpU, rn7GpppBn6AmpA) deliver superior RNA translatability over transcripts capped with analogous primers without Bn6-modification. Additionally, in vitro results described in Example 20 reveal possibility of obtaining mRNAs with highly modified cap 2 structure present on the 5' end, by usage of appropriate tetranucleotide primers for co- transcriptional capping.

[0271] Table 13 Summarized IVT conditions for tetrameric and trimeric capped oligonucleotide primers used for co-transcriptional capping of mRNA for the in vitro experiment with HEK293T, and murine or human primary macrophages (see Example 20).

[0272] 1TIS — trinucleotide initiating sequence of the dsDNA template. Indicated sequence corresponds to the nucleotides of the coding strand following <P6.5 promoter. First transcribed nucleotide which corresponds to the position +1 of the template strand is underlined. Bold font indicates nucleotide which is omitted while in vitro transcription is performed without cap analog added to the IVT mix. 2 T7 RNA polymerase used for in vitro transcription: WT - wild type enzyme manufactured by

[0273] Roche; PrimeCap T7 - low dsRNA variant of T7 RNA polymerase manufactured by TaKaRa.

Claims

CLAIMS1 . A molecular complex suitable for in vitro transcription with a T7 RNA polymerase comprising(i) a Cap analog according to the general formula (I)wherein Base1, Base2and Base3are independently a natural, unnatural or modified base;R1, R2, R3and R4are independently OH, O-Me, H, F, Cl, O-alkyl, O-aryl, O-arylalkyl, O- acyl; wherein at least one from R1and R2, or both, is / are not OH; and X1, X2, X3are independently O, S, BH3, Se;Y1and Y2are independently O, CH2, CHCI, CHF, CF2, CCI2, NH;W1and W2are independently O, S;Z1, Z2are independently O', S', CH3, BHs' ;(ii) a DNA template comprising a T7 polymerase promoter and a trinucleotide initiating sequence;optionally: wherein nucleobase Base1hybridizes to the -1 position and / or preferably wherein nucleobase Base2hybridizes to the +1 position and Base3hybridizes to the +2 position of the trinucleotide initiating sequence; or optionally: provided the DNA template comprises a core promoter of SEQ ID NO: 3 or the promoter comprises a sequence of SEQ ID NOs: 12 or 13, 20 or 21 , 22 or 23, nucleobase Base1hybridizes to the -2 position; nucleobase Base2hybridizes to the -1 position and Base3hybridizes to the +1 position of the trinucleotide initiating sequence.

2. The molecular complex according to claim 1 , wherein the DNA template is selected from a T1 to T8 template as detailed in FIG 4A to 4D, or from a T13, 14 and 16 template as detailed in FIG 38A or wherein the DNA template comprises a sequence as defined by any of SEQ ID NOs: 1 to 25, and / or the complement thereof, or a sequence having at least 99% identity to the respective sequence.

3. The molecular complex according to claims 1 and 2, wherein Base1and Base2of the Cap analog as defined in claim 1 (i) are independently selected from adenine or an analog of adenine, Base3is guanine or an analog of guanine, and wherein the DNA template as defined in claim 1 (ii) comprises a promoter with a core region with a sequence as defined by any one of SEQ ID NO: 1 to 3, or wherein the initiating sequence is selected from an AGG, an AAG, an ATG, a GGG, a AG, an AGA or a GAG.

4. The molecular complex according to claims 1 and 2, wherein Base1is independently adenine, N6-benzyladenine, N6-methyladenine, N6-(2- phenylethyl)adenine, and R1 and / or R2 are independently O-Me, and(i) Base2and Base3are both guanine, or(ii) Base2is adenine and Base3is guanine, or(iii) Base2and Base3are independently guanine or adenine, or(iv) Base2is adenine and Base3is uracil, or(v) Base2is uracil and Base3is guanine, or(vi) Base2is uracil and Base3is adenine, or(vii) Base2is adenine and Base3is adenine, in the Cap analog as defined in claim 1 .

5. The molecular complex according to any one the preceding claims, additionally comprising a T7 RNA polymerase and / or a homolog thereof and / or a modified variant thereof.

6. A vector, or more than one vector, for use in forming a molecular complex according to any of claims 1 to 5, wherein the at least one vector comprises the at least one DNA template as defined in claims 1 , 2 or 3.

7. A DNA molecule, or more than one than one DNA molecule, for use in forming a molecular complex according to any of claims 1 to 5, wherein the at least one DNA molecule comprises the at least one DNA template as defined in claims 1 , 2, or 3.

8. A use of the molecular complex according to any of claims 1 to 5 for in vitro transcription, wherein the capping efficiency is at least 80%, 81 %, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or up to 100%.

9. A method of producing at least one m7G-capped mRNA molecule, the method comprising:(i) providing a reaction mixture comprising the Cap analog as defined in claims 1 , 3 and 4, the DNA template as defined in claims 1 to 3 and / or the vector according to claim 6 and / or the DNA molecule according to claim 7, and at least one T7 RNA polymerase and / or the sequence encoding the same;(ii) allowing in vitro transcription from said DNA template and / or from said vector and obtaining at least one m7G-capped mRNA;(iii) optionally: purifying said at least one m7G-capped mRNA; and / or(iv) optionally: formulating said at least one m7G-capped mRNA molecule yielding a pharmaceutical composition.

10. The method of claim 9 further comprising a step of:(v) introducing or contacting said at least one m7G-capped mRNA molecule obtained in step (ii) into or with at least one target cell in an in vitro cellular system for transfecting said at least one target cell and / or for in vitro studying the functionality of said at least one m7G-capped mRNA.

11. A pharmaceutical composition for use in a method of preventing and / or treating a disease, the pharmaceutical composition comprising the at least one m7G-capped mRNA molecule obtained or obtainable by the method according to claim 9 or 10, additionally comprising at least one pharmaceutically acceptable carrier or excipient.

12. The pharmaceutical composition for use in a method of preventing and / or treating a disease according to claim 11 , the pharmaceutical composition additionally comprising at least one lipid nanoparticle, wherein the at least one lipid nanoparticle is selected from the group of at least one liposome (LPs), at least one liposome-like nanoparticle (LLP), at least one solid lipid nanoparticle (SLN), at least one nanostructured lipid carrier (NLC), at least one lipid-polymer hybrid nanoparticle (LPN), at least one lipoprotein particle (LPT), at least one nanoemulsion, at least one cationic nanoemulsion (CNE) and at least one exosome.

13. A kit comprising the Cap analog as defined in claims 1 , 3 or4, optionally comprising the DNA template as defined in claims 1 to 3, and / or the vector according to claim 6 and / or the DNA molecule according to claim 7, optionally wherein the kit comprises at least one further reagent and / or at least one aqueous solution, preferably a buffer.

Images