Three component lipid nanoparticles

Abstract

The disclosure relates to a lipid nanoparticle (LNP) comprising a plurality of lipids. The plurality of lipids comprises (i) a helper lipid, (ii) a sterol and (iii) a cationic lipid and / or an ionisable lipid. The plurality of lipids does not comprise a PEGylated lipid. The disclosure also provides pharmaceutical compositions and vaccines comprising the LNP, and medical uses thereof. The disclosure also extends to a method of calculating a hydrogen bond potential for a lipid nanoparticle (LNP) formulation.

Description

[0001] Three Component Lipid Nanoparticles The invention provides a three-component lipid nanoparticle (LNP). The invention extends to formulations, pharmaceutical compositions and vaccines comprising the LNP and to methods of identifying stable LNP formulations. The first approved messenger RNA (mRNA) vaccines were developed in response to the SARS-CoV-2 global pandemic. Lipid nanoparticles (LNPs) were found to be a vital component of the vaccines. In particular, the US Food and Drug Administration (FDA) approved Pfizer-BioNTech BNT162b2 and Moderna mRNA-1273 SARS-CoV-2 vaccines both comprise LNPs. The LNPs in both vaccines are made of four components, namely cholesterol, a helper lipid, an ionisable lipid and a polyethylene glycol (PEG) lipid. The PEG lipids have been shown to be a critical component of LNPs. In particular, they have been found to prevent aggregation, increase particle stability, prolong circulation time and improve delivery in vivo. Thus, PEG containing LNPs have been at the forefront of research and development for a variety of therapeutic modalities. However, in rare cases immunoglobulin (Ig)E-mediated anaphylactic reactions can be caused by medications, bowel preparations and / or laxatives containing PEG lipids. Furthermore, a general population sample in the US has revealed that 5% to 9% of patients had anti-PEG IgG and 0.1% had anti-PEG IgE. Current RNA vaccines and therapeutics are not recommended for those with a history of anaphylaxis to PEG or its derivatives. The incidence of anti-PEG sensitisation is only likely to become bigger following the wide-spread use of mRNA vaccines. The present invention arose from the inventors’ work in attempting to address the problems associated with the prior art. In accordance with a first aspect of the invention, there is provided a lipid nanoparticle (LNP) comprising a plurality of lipids, wherein the plurality of lipids comprises (i) a helper lipid, (ii) a sterol and (iii) a cationic lipid and / or an ionisable lipid, and the plurality of lipids does not comprise a PEGylated lipid. The inventors have shown that they can provide a stable LNP formulation which does not comprise a PEGylated lipid. These LNP formulations have been shown to have similar encapsulation efficiency and transfection efficiency to LNP formulations comprising a PEGylated lipid. Omission of a PEGylated lipid from the LNP is advantageous as enables the production of medicaments comprising LNPs without the risk of anaphylaxis to PEG or its derivatives. In some embodiments, the plurality of lipids consists of (i) the helper lipid, (ii) the sterol and (iii) the cationic lipid and / or the ionisable lipid. The helper lipid may be a phospholipid. The phospholipid may be di-oleoyl- phosphatidylethanolamine (DOPE), di-oleoyl-phosphatidylcholine (DOPC), di-oleoyl- phosphatidylserine (DOPS), 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC), 1,2- distearoyl-sn-glycero-3-phosphoethanolamine (DSPE), a naturally occurring or synthetically derived phospholipid, or a combination thereof. In some embodiments, the phospholipid is DSPC. The plurality of lipids may comprise at least 2 mol%, at least 4 mol%, at least 6 mol% or at least 8 mol% helper lipid. In some embodiments, the plurality of lipids comprises at least 10 mol%, at least 15 mol%, at least 20 mol% or at least 22 mol% helper lipid. In some embodiments, the plurality of lipids comprises at least 24 mol%, at least 26 mol%, at least 28 mol%, at least 30 mol% or at least 32 mol% helper lipid. The plurality of lipids may comprise less than 80 mol%, less than 70 mol%, less than 60 mol% or less than 55 mol% helper lipid. In some embodiments, the plurality of lipids comprises less than 50 mol%, less than 45 mol%, less than 40 mol% or less than 38 mol% helper lipid. In some embodiments, the plurality of lipids comprises less than 36 mol%, less than 34 mol%, less than 32 mol%, less than 30 mol% or less than 28 mol% helper lipid. The plurality of lipids may comprise between 2 and 80 mol%, between 4 and 70 mol%, between 6 and 60 mol%, between 8 and 55 mol%, between 10 and 50 mol%, between 15 and 45 mol%, between 20 and 40 mol% or between 22 and 38 mol% helper lipid. In some embodiments, the plurality of lipids comprises between 24 and 50 mol%, between 26 and 45 mol%, between 28 and 42 mol%, between 30 and 40 mol% or between 32 and 38 mol% helper lipid. In some embodiments, the plurality of lipids comprises between 14 and 36 mol%, between 16 and 34 mol%, between 18 and 32 mol%, between 20 and 30 mol% or between 22 and 28 mol% helper lipid. The sterol may be cholesterol, a cholesterol derivative or a combination thereof. The cholesterol derivative may be beta-sitosterol, Vitamin D2, Vitamin D3, Calcipotriol, Stigmasterol, Campesterol, Fucosterol, Brassicasterol, Ergosterol, 9,11- dehydroergosterol, Daucosterol, beta-Sitosterol-Acetate, Betutin, Lupeol, Ursotic acid, Oleanotic acid, an Oxysterol, a natural sterol, a cholesterol ester, a glycosylated sterol or an oxidised sterol. The cholesterol derivative (e.g. the oxysterol) may be substituted. For instance, the cholesterol derivative may be fluorinated, sulfonated and / or phosphorylated. The cholesterol derivative may be A-ring substituted, B-ring substituted, D-ring substituted, side chain substituted and / or double substituted. The cholesterol derivative may be polyunsaturated. The cholesterol derivative may be cholestanoic acid, cholesterol (ovine), cholesterol sulphate, desmosterol, stigma sterol, lanosterol, 7-dehydrocholesterol, dihydrolanosterol, zymosterol, lathosterol, T- MAS, 8(9)-dehydrocholesterol, 8(14)-dehydrocholesterol, FF-MAS, dysgenic, DHEA sulphate, DHEA, sitosterol, lanosterol-95, Dihydro FF-MAS-d6, Dihydro T-MAS-d6, sitosterol, zymostenol, sitostanol, campestanol, campersterol, 7-dehydrodesmosterol, pregnenolone, Dihydro T-MAS, Delta 5-avenasterol, Brassicasterol, Dihydro FF-MAS, 24 methylene cholesterol, 17:10 cholesterol ester, 18:1 chol ester, sitoindosine I, glucose stigma sterol, glucose sitisterol, sitoindoside II, 18:1 stigmasteryl glucose, 16:0 stigmasteryl glucose, galactosyl cholesterol, BbGL-1, ox-18:2 cholesterol or Cholesterol-Amino-Phosphate (CAP) CAP2. In some embodiments, the sterol is cholesterol. The plurality of lipids may comprise at least 5 mol%, at least 10 mol%, at least 15 mol% or at least 18 mol% sterol. In some embodiments, the plurality of lipids comprises at least 20 mol%, at least 22 mol% or at least 24 mol% sterol. In some embodiments, the plurality of lipids comprises at least 26 mol%, at least 28 mol% or at least 30 mol% sterol. The plurality of lipids may comprise less than 70 mol%, less than 60 mol%, less than 55 mol% or less than 50 mol% sterol. In some embodiments, the plurality of lipids comprises less than 45 mol%, less than 40 mol%, less than 38 mol% or less than 36 mol% sterol. In some embodiments, the plurality of lipids comprises less than 34 mol%, less than 32 mol%, less than 30 mol% or less than 28 mol% sterol. The plurality of lipids may comprise between 5 and 70 mol%, between 10 and 60 mol%, between 14 and 55 mol%, between 16 and 50 mol%, between 18 and 45 mol%, between 20 and 40 mol%, between 22 and 38 mol% or between 24 and 36 mol% sterol. In some embodiments, the plurality of lipids comprises between 26 and 40 mol%, between 28 and 36 mol% or between 30 and 34 mol% sterol. In some embodiments, the plurality of lipids comprises between 14 and 36 mol%, between 16 and 34 mol%, between 18 and 32 mol%, between 20 and 30 mol% between 22 and 28 mol%, between 24 and 26 mol% or between 25 and 27 mol% sterol. The cationic lipid may be understood to have a permanent cationic charge. The cationic lipid may be monovalent or multivalent. The cationic lipid may be DOTAP, DC-cholesterol Hal, DC-cholesterol-d7, DOBAQ, 16:0 DPCB, 16:0 DPSB, 18:0 DAP, 16:0 DAP, 14:0 DAP, 18:0 DAP, DODMA, ALC-0315, DORI, DC-6-14, 12:0 EPC (Cl salt), 14:0 EPC (Cl salt), 16:0 EPC (Cl salt), 18:0 EPC (Cl salt), 14:1 EPC (Tf salt), 18:0 DDAB, 14:0 TAP, 16:0 TAP, 18:0 TAP, 18:1 TAP (DOTAP), 18:1 TAP (DOTAP, MS salt), DOTMA, DODMA, N-TETAMINE-pLys40, MVL5, DOSPA or GL67. In some embodiments, the cationic lipid is DOTAP. Conversely, the ionisable lipid may be capable of forming a charge. The ionisable lipid may be an ionisable cationic lipid. The ionisable cationic lipid may be capable of forming a cationic charge. The ionisable cationic lipid may be neutral at physiological pH. The ionisable cationic lipid may be protonated at low pH. Low pH may be understood to be a pH of less than 7, less than 6 or less than 5. The pH may be understood to be the pH at 20°C. The ionisable lipid may be a linker-degradable ionisable lipid (LDIL), an multi-tail ionizable phospholipid, a polymeric lipid and / or a zwitterionic amino lipid. The ionisable lipid may be DLin-MC3-DMA, C12-200, 306-O12B, 306Oi10, C24, cKK- E12, LP-01, PPZ-A10, 4A3-SC8, 4A3-SCC-PH, 4A3-SCC-10, lipid 5, 9A1P9, 4A3-SCC- 10, Lipid 319, 306-N16B, OC2-K3-E10, 93-O17S, 93-O17O, FTT5, TT3, 11-A-M, Lipid 10, Lipid M, SM-102 (Lipid H), ATX100, YSK05, OF-02, ALC-0315, ssPalmE-P4C2, ssPalmE, CL4H6, YSH-12, MIC2, Lipid A6, MIC1 / Lipid C2, Lipid OA2, 1-A-N I, 80-O18, AA-T3A-C12, 4-O10b1, DOG-IM4, A18-Iso5-2DC18, 9A19P, 98N(12)-5, 304O(13), Go- C14, 7C1, ZA3-Ep10, SM102 or a combination thereof. In some embodiments, the cationic and / or ionisable lipid is an ionisable lipid. Accordingly, in some embodiments, the plurality of lipids consists of (i) the helper lipid, (ii) the sterol and (iii) the ionisable lipid. The plurality of lipids may comprise at least 5 mol%, at least 10 mol%, at least 15 mol% or at least 20 mol% cationic lipid and / or ionisable lipid. In some embodiments, the plurality of lipids comprises at least 22 mol%, at least 24 mol%, at least 26 mol% or at least 28 mol% cationic lipid and / or ionisable lipid. In some embodiments, the plurality of lipids comprises at least 30 mol%, at least 32 mol%, at least 34 mol%, at least 36 mol% or at least 38 mol% cationic lipid and / or ionisable lipid. The plurality of lipids may comprise less than 80 mol%, less than 70 mol%, less than 60 mol% or less than 55 mol% cationic lipid and / or ionisable lipid. In some embodiments, the plurality of lipids comprises less than 50 mol%, less than 48 mol%, less than 46 mol% or less than 45 mol% cationic lipid and / or ionisable lipid. In some embodiments, the plurality of lipids comprises less than 44 mol%, less than 42 mol%, less than 40 mol%, less than 38 mol%, less than 36 mol%, less than 34 mol% or less than 32 mol% cationic lipid and / or ionisable lipid. The plurality of lipids may comprise between 5 and 80 mol%, between 10 and 70 mol%, between 15 and 60 mol%, between 20 and 55 mol%, between 22 and 52 mol%, between 24 and 50 mol%, between 26 and 48 mol%, between 28 and 46 mol% or between 29 and 45 mol% cationic lipid and / or ionisable lipid. In some embodiments, the plurality of lipids comprises between 30 and 50 mol%, between 32 and 48 mol%, between 34 and 46 mol%, between 36 and 44 mol%, between 38 and 42 mol% or between 39 and 41 mol% cationic lipid and / or ionisable lipid. In some embodiments, the plurality of lipids comprises between 30 and 52 mol%, between 35 and 50 mol%, between 40 and 48 mol%, between 42 and 46 mol% or between 43 and 45 mol% cationic lipid and / or ionisable lipid. In some embodiments, the plurality of lipids comprises between 23 and 44 mol%, between 24 and 40 mol%, between 25 and 38 mol%, between 26 and 36 mol% between 27 and 34 mol%, between 28 and 32 mol% or between 29 and 31 mol% cationic lipid and / or ionisable lipid. It may be appreciated that each of the lipids in the plurality of lipids (e.g. the helper lipid, the sterol and the cationic lipid and / or the ionisable lipid) may be viewed as comprising a head group and one or more tail groups. The helper lipid, the sterol and the cationic lipid and / or the ionisable lipid may comprise one or more linker groups, wherein each linker group is disposed between the head group and a tail group. The head group may be polar. The head group may contain one or more hydrogen bond donors (HBDs) and / or one or more hydrogen bond acceptors (HBAs). Each HBD may independently be an NH or OH group. Each HBA may independently be a nitrogen, oxygen or sulfur atom. In some embodiments, each HBA is a nitrogen atom or an oxygen atom. The nitrogen or oxygen may have a neutral charge. It may be appreciated that a positively charged nitrogen would not be a HBA. Accordingly, it may be appreciated that an NH or OH group could be both a HBD and a HBA. The head group may be defined as a continuous portion of the structure of a lipid which contains all of the heteroatoms therein. The heteroatoms may be nitrogen, oxygen and / or sulfur. Accordingly, the head group may contain all of the HBDs and HBAs which are in the lipid. One or more of the heteroatoms may be charged. For instance, if a cationic lipid comprises an N+group, this may be part of the head group. The term “continuous” may be understood to mean that the head group contains any atoms required to connect the heteroatoms. The head group may be defined as the smallest continuous portion of the structure of a lipid which contains all of the heteroatoms therein, and may further comprise a chain extending between 1 and 5 atoms from the smallest continuous portion at each point where the smallest continuous portion connects to a remainder of the lipid structure. In embodiments where the lipid comprises only one heteroatom, the smallest continuous portion of the structure of a lipid which contains all of the heteroatoms therein may be understood to the heteroatom and the carbon atom to which it is bonded. If the atom at the point where the smallest continuous portion connects to a remainder of the lipid structure is a carbon atom, and the carbon atom is bonded to two adjacent carbon atoms, each of which are part of the remainder of the lipid structure, then the chain extending from the smallest continuous portion may only comprise one of these adjacent carbon atoms. Any hydrogen atoms bonded to atoms within the smallest continuous portion or bonded to the chain extending from the smallest continuous portion may also be understood to be part of the head group. In an embodiment, the head group is defined as consisting of: the smallest continuous portion of the structure of a lipid which contains all of the heteroatoms therein, the first adjacent atom to the smallest continuous portion of the structure at each point where the smallest continuous portion connects to a remainder of the lipid structure, and optionally any hydrogen atoms bonded to atoms within the smallest continuous portion or bonded to the first adjacent atom to the smallest continuous portion. If the atom at the point where the smallest continuous portion connects to a remainder of the lipid structure is a carbon atom, and it is bonded to two adjacent carbon atoms, each of which are part of the remainder of the lipid structure, then the head group may comprise only one of the adjacent carbon atoms. The remainder of the lipid structure may be understood to comprise the one or more tail groups. The remainder of the lipid structure may also comprise one or more linker groups. In some embodiments, the remainder of the lipid structure may be understood to consist of the one or more tail groups. The helper lipid, the sterol and the cationic lipid and / or the ionisable lipid may each comprise at least one HBD and / or at least one HBA. In an embodiment, the helper lipid, the sterol and the cationic lipid and / or the ionisable lipid each comprise at least one HBA. The helper lipid may comprise between 1 and 20, between 2 and 15 or between 5 and 10 HBAs. In some embodiments, the help lipid comprises 8 HBAs. The helper lipid may comprise between 0 and 10 or between 0 and 5 HBDs. In some embodiments, the helper lipid comprises no HBDs. The sterol may comprise between 1 and 20, between 1 and 15, between 1 and 10, between 1 and 5 or between 1 and 3 HBAs. In some embodiments, the sterol comprises 1 HBA. The sterol may comprise between 1 and 20, between 1 and 15, between 1 and 10, between 1 and 5 or between 1 and 3 HBDs. In some embodiments, the sterol comprises 1 HBD. In some embodiments, the sterol comprises one OH group. The cationic lipid and / or the ionisable lipid may comprise between 1 and 40, between 2 and 30, between 3 and 25 or between 3 and 20 HBAs. In some embodiments, the cationic lipid and / or the ionisable lipid, may comprise 3, 4, 7, 8, 10, 11 or 19 HBAs. The cationic lipid and / or the ionisable lipid may comprise between 0 and 15, between 0 and 10 or between 0 and 7 HBDs. In some embodiments, the lipid and / or the ionisable lipid comprises no HBDs. In alternative embodiments, the lipid and / or the ionisable lipid comprises between 1 and 6 HBDs. In some embodiments, the cationic lipid and / or the ionisable lipid, may comprise 2, 5 or 6 HBDs. The number of HBDs and HBAs defined above may be for the lipid at a neutral pH (i.e. a pH of 7 at 20°C). Accordingly, the number of HBAs and / or HBDs in the ionisable lipid may be understood to be the number of HBAs and / or HBDs in a neutral (i.e. not ionised) form thereof. The head group may have a cross-sectional area of between 1 and 300 Å2, between 10 and 200 Å2or between 20 and 100 Å2. In some embodiments, the head group of the sterol may have a cross-sectional area of between 1 and 100 Å2, between 2 and 50 Å2, between 5 and 25 Å2or between 8 and 15 Å2. In some embodiments, the head group of the helper lipid may have a cross-sectional area of between 1 and 200 Å2, between 5 and 100 Å2, between 10 and 50 Å2, between 20 and 40 Å2or between 25 and 35 Å2. In some embodiments, the head group of the ionisable lipid and / or the cationic lipid may have a cross-sectional area of between 10 and 200 Å2, between 15 to 100 Å2or between 20 and 80 Å2. In some embodiment, the head group of the cationic lipid and / or the ionisable lipid may have a cross-sectional area of between 30 and 70 Å2or between 40 and 60 Å2. The or each tail group may be hydrophobic. The one or more tail groups may be defined as being any portion of the structure of the helper lipid, the sterol and the cationic lipid and / or the ionisable lipid which is not the head group. The or each tail group may be an alkyl, an alkenyl or an alkynyl. The or each tail group may be a C1-50 alkyl, a C2-50 alkenyl or a C2-50 alkynyl. The alkyl, alkenyl and / or alkynyl may be straight or branched. The LNP may have a diameter of less than 300 nm, less than 250 nm or less than 200 nm. In some embodiments, the LNP has a diameter of less than 180 nm, less than 160 nm, less than 150 nm, less than 140 nm or less than 120 nm. In some embodiments, the LNP has a diameter of less than 110 nm or less than 100 nm. The diameter of the LNP may be measured as defined in the examples. The diameter of the LNP may be understood to be the Z-average size or Z-average mean. The diameter of the LNP may be calculated as defined in ISO 22412. The LNP may have a diameter of between 60 and 200 nm, between 70 and 180 nm, between 75 and 160 nm, between 80 and 140 nm, between 85 and 120 nm, between 90 and 110 nm or between 95 and 105 nm. In some embodiments, the LNP may comprise a payload. The payload may be a biomolecule and / or an active pharmaceutical ingredient (API). The API may be a hydrophobic or hydrophilic API. The API may be a macromolecule or a small molecule. It may be appreciated that a small molecule could be considered to be a molecule with a molecular weight of less than 900 daltons. In some embodiments, a small molecule may have a molecular weight of less than 800 daltons, less than 700 daltons, less than 600 daltons, less than 500 daltons or less than 400 daltons. Similarly, a macromolecule may be considered to be a molecule with a molecular weight of at least 900 daltons. In an embodiment, the payload is a biomolecule. For instance, the biomolecule may be or comprise an amino acid, a peptide, an affimer, a protein, a glycoprotein, a lipopolysaccharide, an antibody or a fragment thereof, or a nucleic acid. The nucleic acid may be DNA, RNA or a DNA / RNA hybrid sequence. In an embodiment, the nucleic acid is DNA or RNA. In an embodiment, the nucleic acid is RNA. The RNA may be single stranded or double stranded. The RNA may be selected from the group consisting of: messenger RNA (mRNA); self-amplifying RNA (saRNA); antisense RNA (asRNA); RNA aptamers; interference RNA; micro RNA (miRNA); short interfering RNA (siRNA); short hairpin RNA (shRNA); and small RNA. In an embodiment, the RNA is self-amplifying RNA (saRNA) or messenger RNA (mRNA). The skilled person would appreciate that self-amplifying RNAs may contain the basic elements of mRNA (a cap, 5’ UTR, 3’UTR, and poly(A) tail of variable length), but may be considerably longer (for example 9-12 kb). The nucleic acid sequence, preferably RNA, may be at least 10 bases in length, at least 20 bases in length, at least 50 bases in length, at least 100 bases in length, at least 200 bases in length, at least 300 bases in length, at least 400 bases in length, at least 500 bases in length, at least 600 bases in length at least 700 bases in length, at least 800 bases in length or at least 900 bases in length. In one preferred embodiment, the RNA is saRNA or mRNA. The nucleic acid sequence, preferably RNA, and most preferably saRNA or mRNA, may be at least 1000 bases in length, at least 2000 bases in length, at least 3000 bases in length, at least 4000 bases in length, at least 5000 bases in length, at least 6000 bases in length, at least 7000 bases in length, at least 8000 bases in length, at least 9000 bases in length at least 10000 bases in length, at least 11000 bases in length or at least 12000 bases in length. In one embodiment, the nucleic acid sequence is at least 6000 bases in length. In one embodiment, the RNA is at least 6000 bases in length. In a preferred embodiment, the saRNA is at least 6000 bases in length. In an alternative embodiment, the nucleic acid sequence is at least 900 bases in length. In one embodiment, the RNA is at least 900 bases in length. In a preferred embodiment, the mRNA is at least 900 bases in length. The nucleic acid sequence, preferably RNA, and most preferably saRNA, may be between 5000 and 20000 bases in length, between 5000 and 15000 bases in length, between 5000 and 14000 bases in length, between 5000 and 13000 bases in length, between 5000 and 12000 bases in length, between 5000 and 11000 bases in length, between 5000 and 10000 bases in length, between 6000 and 20000 bases in length, between 6000 and 15000 bases in length, between 6000 and 14000 bases in length, between 6000 and 13000 bases in length, between 6000 and 12000 bases in length between, between 6000 and 11000 bases in length, between 6000 and 10000 bases in length, between 7000 and 20000 bases in length, between 7000 and 15000 bases in length, between 7000 and 14000 bases in length, between 7000 and 13000 bases in length, between 7000 and 12000 bases in length, between 7000 and 11000 bases in length, between 7000 and 10000 bases in length, between 8000 and 20000 bases in length, between 8000 and 15000 bases in length, between 8000 and 14000 bases in length, between 8000 and 13000 bases in length, between 8000 and 12000 bases in length, between 8000 and 11000 bases in length, between 8000 and 10000 bases in length, between 9000 and 20000 bases in length, between 9000 and 15000 bases in length, between 9000 and 14000 bases in length, between 9000 and 13000 bases in length, between 9000 and 12000 bases in length, between 9000 and 11000 bases in length or between 9000 and 10000 bases in length. Alternatively, the nucleic acid sequence, preferably RNA, and most preferably mRNA, may be between 50 and 10000 bases in length, between 100 and 9000 bases in length, between 200 and 8000 bases in length, between 300 and 7000 bases in length, between 400 and 6000 bases in length, between 500 and 6000 bases in length, between 600 and 5000 bases in length, between 700 and 4000 bases in length, between 800 and 3000 bases in length or between 900 and 2000 bases in length. In one embodiment, the nucleic acid sequence is between 6000 and 15000 bases in length. The nucleic acid sequence may be between 8000 and 12000 bases in length. The RNA may be between 6000 and 15000 bases in length. The RNA may be between 8000 and 12000 bases in length. Preferably, the saRNA is between 6000 and 15000 bases in length. Preferably the saRNA is between 8000 and 12000 bases in length. In an alternative embodiment, the nucleic acid sequence is between 400 and 14000, between 500 and 10000, between 600 and 7500, between 700 and 5000, between 800 and 4000 or between 900 and 2000 bases in length. The RNA may between 400 and 14000, between 500 and 10000, between 600 and 7500, between 700 and 5000, between 800 and 4000 or between 900 and 2000 bases in length. Preferably, the mRNA is between 400 and 14000, between 500 and 10000, between 600 and 7500, between 700 and 5000, between 800 and 4000 or between 900 and 2000 bases in length. The skilled person would appreciate that when the nucleic acid is double stranded, for example double stranded RNA, “bases in length” will refer to the length of base pairs. The nucleic acid may encode at least a portion of a virus. The virus may be the SARS- CoV-2 virus or an influenza virus. The nucleic acid may encode a SARS-CoV-2 spike protein, more preferably a pre-fusion stabilized SARS-CoV-2 spike protein. Alternatively, the nucleic acid may encode as the H1 hemagglutinin of the influenza virus. In some embodiments, the nucleic acid is RNA. In some embodiments, the nucleic acid is saRNA or mRNA. The LNP may have an encapsulation efficiency of at least 60%, at least 70% or at least 80%. In some embodiments, the LNP has an encapsulation efficiency of at least 82%, at least 84%, at least 86% or at least 88%. In some embodiments, the LNP has an encapsulation efficiency of at least 92%, at least 94% or at least 95%. The molar ratio of the cationic lipid and / or an ionisable lipid to the payload may be between 1:50 and 50:1, between 1:20 and 20:1, between 1:10 and 10:1, between 1:5 and 5:1, between 1:3 and 3:1, between 1:2 and 2:1 or between 1:1.5 and 1.5:1. The weight ratio of the cationic lipid and / or an ionisable lipid to the payload may be between 1:1 to 1: 200, between 1:10 and 1:150 or between 1:20 and 1:100. In some embodiments, the weight ratio of the cationic lipid and / or an ionisable lipid to the payload may be between 1:30 and 1:90 or between 1:45 and 1:78. The N / P ratio of the LNP may be at least 14, at least 15, at least 20 or at least 30. In some embodiments, the N / P ratio is at least 40, at least 50, at least 55 at least 60 or at least 65. In some embodiments, the N / P ratio is between 14 and 250, between 15 and 200, between 20 and 150, between 25 and 120, between 30 and 100, between 35 and 80 or between 40 and 75. In some embodiments, the N / P ratio is between 50 and 90, between 60 and 80 or between 65 and 75. In some embodiments, the N / P ratio is between 15 and 80, between 20 and 70, between 30 and 60, between 35 and 50 or between 40 and 45. It may be appreciated that the N / P ratio is the molar ratio of the ionized nitrogen and / or ionizable nitrogen in the cationic lipid and / or the ionisable lipid to the phosphate in the nucleic acid. The inventors believe that LNPs with a high N / P ratio are novel and inventive per se. Accordingly, in a second aspect of the invention, there is provided a lipid nanoparticle (LNP) with an N / P ratio of at least 15. The N / P ratio of the LNP may be at least 20 or at least 30. In some embodiments, the N / P ratio is at least 40, at least 50, at least 55 at least 60 or at least 65. In some embodiments, the N / P ratio is between 14 and 250, between 15 and 200, between 20 and 150, between 25 and 120, between 30 and 100, between 35 and 80 or between 40 and 75. In some embodiments, the N / P ratio is between 50 and 90, between 60 and 80 or between 65 and 75. In some embodiments, the N / P ratio is between 15 and 80, between 20 and 70, between 30 and 60, between 35 and 50 or between 40 and 45. The LNP may be understood to comprise a plurality of lipids. The plurality of lipids may comprise (i) a helper lipid, (ii) a sterol, (iii) a cationic lipid and / or an ionisable lipid, and / or (iv) a PEGylated lipid. In some embodiments, the plurality of lipids comprises or consists of (i) a helper lipid, (ii) a sterol, (iii) a cationic lipid and / or an ionisable lipid, and (iv) a PEGylated lipid. The helper lipid, the sterol, the cationic lipid and / or the ionisable lipid may be as defined in relation to the first aspect. The PEGylated lipid may be DMG-PEG 2000, DSPE-PEG2000, DSG-PEG2000 or c-DMG- PEG 2000. The plurality of lipids may comprise at least 0.1 mol%, at least 0.5 mol%, at least 1 mol%, at least 1.5 mol%, at least 2 mol%, at least 2.2 mol% or at least 2.4 mol% PEGylated lipid. The plurality of lipids may comprise less than 10 mol%, less than 7.5 mol%, less than 5 mol%, less than 4 mol%, 3 mol%, less than 2.8 mol% or less than 2.6 mol% PEGylated lipid. The plurality of lipids may comprise between 0.1 and 10 mol%, between 0.5 and 7.5 mol%, between 1 and 5 mol%, between 1.5 and 4 mol%, between 2 and 3 mol%, between 2.2 and 2.8 mol% or between 2.4 and 2.6 mol% PEGylated lipid. The diameter of the LNP the second aspect may be as defined in relation to the first aspect. In some embodiments, the LNP may comprise a payload. The payload may be as defined in relation to the first aspect. In accordance with a third aspect, there is provided a formulation comprising a plurality of LNPs of the first aspect or the second aspect. The plurality of LNPs may have a polydispersity index (PDI) of less than 0.4, less than 0.35 or less than 0.3. In some embodiments, the plurality of LNPs may have a PDI or less than 0.28, less than 0.26, less than 0.24, less than 0.22 or less than 0.2. In some embodiments, the plurality of LNPs may have a PDI or less than 0.18, less than 0.16, less than 0.14 or less than 0.12. The formulation may comprise a solvent. The solvent may be or comprise water and / or an alcohol. The alcohol may be ethanol. The formulation may comprise at least one stabilizing molecule. The formulation may comprise the at least one stabilizing molecule in a concentration between 5 and 60 % m / m, between 10 and 50 % m / m, between 15 and 40 % m / m or between 20 and 30 % m / m. % m / m may be understood to be percentage mass. The or each stabilizing molecule may be a carbohydrate and / or an amino acid. In some embodiments, the formulation may comprise a carbohydrate and an amino acid. The carbohydrate may be a sugar. The carbohydrate may be a monosaccharide, a disaccharide, a trisaccharide or a polysaccharide. The monosaccharide may be selected from a group consisting of: glucose; galactose; fructose; mannose; and xylose or a pharmaceutically acceptable complex, salt, solvate, tautomeric form, stereoisomer or polymorphic form thereof. The disaccharide may be selected from a group consisting of: trehalose; sucrose; lactose; maltose; isomaltose; lactitol; lactulose; mannobiose; and isomalt or a pharmaceutically acceptable complex, salt, solvate, tautomeric form, stereoisomer or polymorphic form thereof. The trisaccharide may be selected from a group consisting of: nigerotriose; maltotriose; melezitose; maltotriulose; raffinose; and kestose or a pharmaceutically acceptable complex, salt, solvate, tautomeric form, stereoisomer or polymorphic form thereof. The polysaccharide may be selected from the group consisting of: dextran; amylose; amylopectin; glycogen; galactogen; inulin; callose; cellulose; chitosan; and chitin or a pharmaceutically acceptable complex, salt, solvate, tautomeric form, stereoisomer or polymorphic form thereof. In some embodiments, the carbohydrate is a disaccharide. In some embodiments, the carbohydrate is sucrose or trehalose, or a pharmaceutically acceptable complex, salt, solvate, tautomeric form, stereoisomer or polymorphic form thereof. In some embodiments, the carbohydrate is trehalose, or a pharmaceutically acceptable complex, salt, solvate, tautomeric form, stereoisomer or polymorphic form thereof. The trehalose may be synthetic trehalose or natural trehalose. The formulation may comprise the carbohydrate in a concentration between 5 and 60 % m / m, between 10 and 50 % m / m, between 12.5 and 40 % m / m, between 15 and 35 % m / m, between 17.5 and 30 % m / m or between 20 and 25 % m / m. The amino acid may be any natural amino acid. The amino acid may be histidine, arginine or a pharmaceutically acceptable complex, salt, solvate, tautomeric form, stereoisomer or polymorphic form thereof. The formulation may comprise the amino acid in a concentration between 0.1 and 20 % m / m, between 0.25 and 15 % m / m, between 0.5 and 12.5 % m / m, between 0.75 and 10 % m / m or between 1 and 8 % m / m. In some embodiments, the formulation comprises the amino acid in a concentration between 1 and 20 % m / m, between 2 and 15 % m / m, between 4 and 12.5 % m / m, between 5 and 10 % m / m or between 6 and 8 % m / m. In alternative embodiments, the formulation comprises the amino acid in a concentration between 0.1 and 5 % m / m, between 0.25 and 3 % m / m, between 0.5 and 2 % m / m, between 0.75 and 1.75 % m / m or between 1 and 1.5 % m / m. The formulation may comprise a buffer. The buffer may be configured to maintain the formulation at a pH of less than 9, less than 8, less than 7 or less than 6.5. The buffer may be configured to maintain the formulation at a pH of more than 3, more than 4, more than 5 or more than 5.5. The buffer may be configured to maintain the formulation at a pH of between 3 and 9, between 4 and 8, between 5 and 7 or between 5.5 and 6.5. The pH may be understood to be the pH when measured at 20°C. The buffer may be phosphate-buffered saline (PBS) or Tris. The formulation may comprise a surfactant. The surfactant may be an ionic or a non- ionic surfactant. In some embodiments, the surfactant is a non-ionic surfactant. The surfactant may be a polysorbate. In some embodiments, the polysorbate is polysorbate 20. The formulation may comprise the surfactant in an amount which is between 0.0001 and 1 % m / m, between 0.0005 and 0.5 % m / m, between 0.001 and 0.2 % m / m, between 0.005 and 0.15 % m / m, between 0.01 and 0.1 % m / m, between 0.02 and 0.08 % m / m or between 0.04 and 0.06 % m / m. In a fourth aspect, there is provided a pharmaceutical composition comprising the LNP of the first or second aspect or the formulation of the third aspect and a pharmaceutically acceptable vehicle. In a fifth aspect, there is provided a method of preparing the pharmaceutical composition according to the fourth aspect, the method comprising contacting the LNP of the first or second aspect or the formulation of the third aspect with a pharmaceutically acceptable vehicle. In a sixth aspect, there is provided the LNP of the first or second aspect, the formulation of the third aspect or the pharmaceutical composition of the fourth aspect, for use as a medicament. In a seventh aspect, there is provided a method of treatment, the method comprising administering, or having administered, to a subject in need thereof, a therapeutic amount of the LNP of the first or second aspect, the formulation of the third aspect or the pharmaceutical composition of the fourth aspect. The LNP of the formulation may be used to treat an infectious disease, cancer and / or an autoimmune disorder. The infectious disease may be caused by a protozoa, a bacterium, a virus or a fungus. In an eighth aspect, there is provided a vaccine composition comprising the LNP of the first or second aspect, the formulation of the third aspect or the pharmaceutical composition of the fourth aspect. The vaccine may comprise a suitable adjuvant. The vaccine may be a vaccine for COVID-19. The vaccine may be a vaccine for influenza virus. In a ninth aspect, there is provided the LNP of the first or second aspect, the formulation of the third aspect, the pharmaceutical composition of the fourth aspect or the vaccine of the eighth aspect, for use in stimulating an immune response in a subject. The immune response may be stimulated against a protozoa, bacterium, virus, fungus or cancer. The virus may be COVID-19. The virus may be influenza virus. In a tenth aspect of the invention, there is provided a method of vaccinating a subject, the method comprising administering, or having administered, to a subject in need thereof, a therapeutic amount of the first or second aspect, the formulation of the third aspect, the pharmaceutical composition of the fourth aspect or the vaccine of the eighth aspect. The LNP, the formulation, the pharmaceutical composition or the vaccine of the invention may be combined in compositions having a number of different forms depending, in particular, on the manner in which the composition is to be used. Thus, for example, the composition may be in the form of a powder, tablet, capsule, liquid, ointment, cream, gel, hydrogel, aerosol, spray, micellar solution, transdermal patch, liposome suspension or any other suitable form that may be administered to a person or animal in need of treatment. It will be appreciated that the vehicle of medicaments according to the invention should be one which is well-tolerated by the subject to whom it is given. The LNP, the formulation, the pharmaceutical composition or the vaccine of the invention may also be incorporated within a slow- or delayed-release device. Such devices may, for example, be inserted on or under the skin, and the medicament may be released over weeks or even months. The device may be located at least adjacent the treatment site. In a preferred embodiment, however, medicaments according to the invention may be administered to a subject by injection into the blood stream, muscle, skin or directly into a site requiring treatment. Injections may be intravenous (bolus or infusion), subcutaneous (bolus or infusion), intradermal (bolus or infusion), intramuscular (bolus or infusion), intrathecal (bolus or infusion), epidural (bolus or infusion) or intraperitoneal (bolus or infusion). It will be appreciated that the amount of the LNP, the formulation, the pharmaceutical composition or the vaccine that is required is determined by its biological activity and bioavailability, which in turn depends on the mode of administration, the physiochemical properties of the LNP, the formulation, the pharmaceutical composition or the vaccine and whether it is being used as a monotherapy or in a combined therapy. The frequency of administration will also be influenced by the half-life of the active agent within the subject being treated. Optimal dosages to be administered may be determined by those skilled in the art, and will vary with the LNP, the formulation, the pharmaceutical composition or the vaccine in use, the strength of the pharmaceutical composition, the mode of administration, and the type of treatment. Additional factors depending on the particular subject being treated will result in a need to adjust dosages, including subject age, weight, gender, diet, and time of administration. The required dose may depend upon a number of factors including, but not limited to, the active agent being administered, the disease being treated and / or vaccinated against, the subject being treated, etc. Generally, a dose of between 0.001 µg / kg of body weight and 10 mg / kg of body weight, or between 0.01 mg / kg of body weight and 1 mg / kg of body weight, of the LNP, the formulation, the pharmaceutical composition or the vaccine of the invention may be used, depending upon the active agent used. A dose may be understood to relate to the quantity of the payload molecule which is delivered. Doses may be given as a single administration (e.g., a single injection). Alternatively, the LNP, the formulation, the pharmaceutical composition or the vaccine may require more than one administration. As an example, the LNP, the formulation, the pharmaceutical composition or the vaccine may be administered as two or more doses of between 0.07 mg and 700 mg (i.e., assuming a body weight of 70 kg). Alternatively, a slow-release device may be used to provide optimal doses of the LNP, the formulation, the pharmaceutical composition or the vaccine according to the invention to a patient without the need to administer repeated doses. Routes of administration may incorporate intravenous, intradermal subcutaneous, intramuscular, intrathecal, epidural or intraperitoneal routes of injection. Known procedures, such as those conventionally employed by the pharmaceutical industry (e.g., in vivo experimentation, clinical trials, etc.), may be used to form specific formulations of the LNP, the formulation, the pharmaceutical composition or the vaccine according to the invention and precise therapeutic regimes (such as doses of the agents and the frequency of administration). A “subject” may be a vertebrate, mammal, or domestic animal. Hence, compositions and medicaments according to the invention may be used to treat any mammal, for example livestock (e.g., a horse), pets, or may be used in other veterinary applications. Most preferably, however, the subject is a human being. A “therapeutically effective amount” of the LNP, the formulation, the pharmaceutical composition or the vaccine is any amount which, when administered to a subject, is the amount of the aforementioned that is needed to produce a therapeutic effect. For example, a therapeutically effective amount of the LNP, the formulation, the pharmaceutical composition or the vaccine of the invention may comprise from about 0.001 mg to about 800 mg of the payload molecule, and preferably from about 0.01 mg to about 500 mg of the payload molecule. A “pharmaceutically acceptable vehicle” as referred to herein, is any known compound or combination of known compounds that are known to those skilled in the art to be useful in formulating pharmaceutical compositions. In one embodiment, the pharmaceutically acceptable vehicle may be a solid, and the composition may be in the form of a powder, a capsule or tablet. A solid pharmaceutically acceptable vehicle may include one or more substances which may also act as flavouring agents, lubricants, solubilisers, suspending agents, dyes, fillers, glidants, compression aids, inert binders, sweeteners, preservatives, dyes, coatings, or tablet-disintegrating agents. The vehicle may also be an encapsulating material. In powders, the vehicle is a finely divided solid that is in admixture with the finely divided active agents according to the invention. In tablets, the active agent (e.g., the LNP of the invention) may be mixed with a vehicle having the necessary compression properties in suitable proportions and compacted in the shape and size desired. The pharmaceutical vehicle may be a gel and the composition may be in the form of a cream or the like. Alternatively, the pharmaceutical vehicle may be a liquid, and the pharmaceutical composition is in the form of a solution. Liquid vehicles are used in preparing solutions, suspensions, emulsions, syrups, elixirs and pressurized compositions. The LNP according to the invention may be dissolved or suspended in a pharmaceutically acceptable liquid vehicle such as water, an organic solvent, a mixture of both or pharmaceutically acceptable oils or fats. The liquid vehicle can contain other suitable pharmaceutical additives such as solubilisers, emulsifiers, buffers, preservatives, sweeteners, flavouring agents, suspending agents, thickening agents, colours, viscosity regulators, stabilizers or osmo-regulators. Suitable examples of liquid vehicles for oral and parenteral administration include water (partially containing additives as above, e.g., cellulose derivatives, preferably sodium carboxymethyl cellulose solution), alcohols (including monohydric alcohols and polyhydric alcohols, e.g., glycols) and their derivatives, and oils (e.g., fractionated coconut oil and arachis oil). For parenteral administration, the vehicle can also be an oily ester such as ethyl oleate and isopropyl myristate. Sterile liquid vehicles are useful in sterile liquid form compositions for parenteral administration. The liquid vehicle for pressurized compositions can be a halogenated hydrocarbon or other pharmaceutically acceptable propellant. Liquid pharmaceutical compositions, which are sterile solutions or suspensions, can be utilized by, for example, intramuscular, intrathecal, epidural, intraperitoneal, intravenous and subcutaneous injection. The LNP, the formulation, the pharmaceutical composition or the vaccine of the invention may be prepared as any appropriate sterile injectable medium. The LNP, the formulation, the pharmaceutical composition or the vaccine may be administered by inhalation. For instance, the LNP, the formulation, the pharmaceutical composition or the vaccine may be provided in the form of an aerosol. The LNP, the formulation, the pharmaceutical composition or the vaccine of the invention may be administered orally in the form of a sterile solution or suspension containing other solutes or suspending agents (for example, enough saline or glucose to make the solution isotonic), bile salts, acacia, gelatin, sorbitan monoleate, polysorbate 80 (oleate esters of sorbitol and its anhydrides copolymerized with ethylene oxide) and the like. The LNP, the formulation, the pharmaceutical composition or the vaccine according to the invention can also be administered orally either in liquid or solid composition form. Compositions suitable for oral administration include solid forms, such as pills, capsules, granules, tablets, and powders, and liquid forms, such as solutions, syrups, elixirs, and suspensions. Forms useful for parenteral administration include sterile solutions, emulsions, and suspensions. In an eleventh aspect, there is provided a method of calculating a hydrogen bond potential for a lipid nanoparticle (LNP) formulation, the method comprising: a) selecting three or more different lipids; b) identifying all hydrogen bond acceptors (HBAs) and hydrogen bond donors (HBDs) in each of the different lipids; c) selecting a ratio of the three or more different lipids; d) modelling a physical distribution of the three or more different lipids at the ratio; and e) calculating a hydrogen bond energy based upon interactions between neighbouring molecules in the physical distribution to thereby determine a hydrogen bond potential for the LNP formulation. Advantageously, the method of the eleventh aspect allows the design and identification of LNPs maximising the formation of hydrogen bonds. This facilitates strong intermolecular interactions, which can provide LNPs with structural and thermodynamic stability. The method enables the design of LNPs of enhanced stability, increased RNA encapsulation, prolonged circulation and improved delivery in vivo. The method of the eleventh aspect may be a computer implemented method. The three or more different lipids may comprise (i) a helper lipid, (ii) a sterol and (iii) a cationic lipid and / or an ionisable lipid. The helper lipid, the sterol and the cationic lipid and / or the ionisable lipid may be as defined in relation to the first aspect. In some embodiments, the three or more different lipids may not comprise a PEGylated lipid. Accordingly, the method may design an LNP in accordance with the first aspect. In alternative embodiments, the three or more different lipids may further comprise a PEGylated lipid. The PEGylated lipid may be as defined in relation to the second aspect. The ratio may be a molar ratio. Selecting a ratio of the three or more different lipids may comprise selecting a ratio which comprises the sterol at least 5 mol%, at least 10 mol%, at least 15 mol% or at least 20 mol%. Identifying all HBAs and HBDs in each of the different lipids may comprise identifying all HBAs and HBDs in each of the different lipids at a pH of 7 at 20°C. If the three or more different lipids comprise an ionisable lipid, the method may comprise identifying the HBAs and HBDs in the ionisable lipid in a neutral (i.e. not ionised) form thereof. If the three or more different lipids comprise a PEGylated lipid, the PEGylated groups in the PEGylated lipid may be considered to not include any HBAs and / or HBDs. The HBAs and HBDs may be identified: - based upon the chemical groups in each of the different lipids; - based upon an electron density map for each of the different lipids; - based upon an atomic charge analysis for each of the different lipids; and / or - based upon experimental data. Prior to identifying all HBAs and HBDs in each of the different lipids, the method may comprise generating a three-dimensional conformation of at least a portion of each lipid. The three-dimensional conformation may influence the HBAs and HBDs which are identified. If the HBAs and HBDs are identified based upon the chemical groups in each of the different lipids, then each HBD may independently be an NH or OH group. Alternatively, or additionally, each HBA may independently be a nitrogen atom, an oxygen atom or a sulfur atom. In some embodiments, each HBA may independently be a nitrogen atom or an oxygen atom. The nitrogen or oxygen may have a neutral charge. It may be appreciated that a positively charged nitrogen would not be a HBA. Modelling a physical distribution of the three or more different lipids may comprise: d-i) generating a three-dimensional conformation of at least a portion of each lipid. The at least a portion of each lipid may be or comprise the head group of each lipid. The head group may be as defined in relation to the first aspect. In embodiments where the three or more different lipids comprise a PEGylated lipid the PEGylated group(s) in the PEGylated lipid may not form part of the head group. Accordingly, the head group may be defined as a continuous portion of the structure of a lipid which contains all of the heteroatoms therein except for any heteroatoms which are present in a PEGylated group. The head group may otherwise be as defined in relation to the first aspect. Modelling a physical distribution of the three or more different lipids may comprise: d-ii) generating a lipid mixture, wherein the lipid mixture comprises a plurality of three-dimensional conformations of each of the three or more different lipids, such that the plurality of three-dimensional conformations is representative of the selected ratio. The lipid mixture may be generated in a two-dimensional plane. The method may further comprise: d-iii) optionally randomising the distribution of the plurality of three-dimensional conformations. The generated lipid mixture may be a static lipid mixture. Advantageously, a static lipid mixture requires less computational power to generate and less a priori knowledge of the system being modelled than a dynamic lipid mixture. In this embodiment, the distribution of the plurality of three-dimensional conformations is preferably randomised. Alternatively, the generated lipid mixture may be a dynamic lipid mixture. In this embodiment, the distribution of the plurality of three-dimensional conformations could be non-random. Modelling a physical distribution of the three or more different lipids may further comprise: d-iv) rotating and / or translating the three-dimensional conformations to maximise cross-sectional area thereof in a plane, minimise distance between neighbouring conformations, avoiding overlap between neighbouring conformations, maximise a calculated potential energy in the system and / or optimise interactions between the three-dimensional conformations. It may be appreciated that molecular dynamics simulations could comprise rotating and / or translating the three-dimensional conformations to maximise a calculated potential energy in the system. The calculated potential energy may be based on a force field. In this method, hydrogen bonds may be assessed using geometric criteria, such as specific distances and angles. The geometric criteria may be derived from the force field, not from static translocation methods. It may be appreciated that docking algorithms may optimise interactions between the three-dimensional conformations. The interactions may be optimised to achieve the best fit. The best fit may be determined by considering interactions between the three-dimensional conformations. These interactions could include hydrogen bond interactions, Van-der-Waals interactions and / or hydrophilic and hydrophobic interactions. In an embodiment, the method comprises rotating and / or translating the three- dimensional conformations to maximise cross-sectional area thereof in a plane, minimise distance between neighbouring conformations and / or avoiding overlap between neighbouring conformations. In an embodiment, the method comprises rotating and translating the three-dimensional conformations to maximise cross- sectional area thereof in a plane, minimise distance between neighbouring conformations and avoiding overlap between neighbouring conformations. The method may further comprise optimising the three-dimensional conformations. The three-dimensional conformations may be optimised based upon intramolecular interactions within each lipid. Accordingly, this would ensure intramolecular stability. Accordingly, the three-dimensional conformations may be optimised prior to generating the lipid mixture. The three-dimensional confirmations may be optimised using the Universal Force Field (UFF). In an alternative embodiment, the three-dimensional conformations may be optimised based upon intramolecular interactions. Accordingly, the three-dimensional conformations may be optimised after generating the lipid mixture. The three- dimensional conformations may be optimised using a docking algorithm. The three- dimensional conformations may be optimised based upon hydrogen bond interactions, Van-der-Waals interactions and / or hydrophilic and hydrophobic interactions. Accordingly, in this embodiment, the three-dimensional conformations may be optimised at the same time as rotating and / or translating the three-dimensional conformations. In embodiments where the method comprises generating a dynamic lipid mixture, the method may not comprise optimising the three-dimensional conformations. Calculating the hydrogen bond energy based upon interactions between neighbouring molecules in the physical distribution may comprise: e-i) determining the number of hydrogen bonds between neighbouring molecules in the modelled physical distribution. Determining the number of hydrogen bonds between neighbouring molecules in the modelled physical distribution may comprise: e-i-a) optionally assigning HBDs and HBAs on neighbouring molecules in the physical distribution into HBD-HBA pairs; and e-i-b) optionally counting a HBD-HBA pair as defining a hydrogen bond if the HBD- HBA pair are less than a predetermined distance apart and / or the HBD-HBA pair define an angle within a predetermined range. The method may comprise ensuring each HBD is assigned as being in only one HBD- HBA pair. The method may comprise ensuring each HBA is assigned as being in only one HBD-HBA pair. If a HBD may be paired with two or more HBAs, or vice versa, the method may comprise assigning the HBA and HBD which are the closest to each other into a HBD-HBA pair. In some embodiments, the method comprises counting a HBD-HBA pair as defining a hydrogen bond if the HBD-HBA pair are less than a predetermined distance apart. The predetermined distance may be between 1 and 10 Å, between 2 and 7.5 Å, between 2.5 and 5 Å or between 3 and 4 Å. In some embodiments, the predetermined distance is 3.5 Å. The method may comprise: e-ii) calculating a hydrogen bond energy based upon the number of hydrogen bonds between neighbouring molecules in the physical distribution to thereby determine a hydrogen bond potential for the LNP formulation. Calculating hydrogen bond energy may comprise calculating the hydrogen bond energy using any method known in the art. The hydrogen bond energy may be calculated for each HBD-HBA pair which defines a hydrogen bond, using molecular dynamics (MD), alternative quantum mechanics and / or a molecular docking algorithm. In an embodiment, the hydrogen bond energy is calculated for each HBD-HBA pair which defines a hydrogen bond. The energy may be calculated using the following formula: E = -C / r2where E is the energy for a hydrogen bond between a HBA and a HBD, C is a constant and r is the distance between the HBA and the HBD. It may be appreciated that the exact value of C is not important. Any constant may be used to calculate an energy, and this may then give a value which may be used to compare different formulations. In an embodiment, C may be 33 kJ.Å2. mol-1. Calculating hydrogen bond energy may comprise determining the total hydrogen bond energy for all of the HBD-HBA pairs which define a hydrogen bond. Accordingly, the hydrogen bond energy calculated for each HBD-HBA pair may be added together to determine total hydrogen bond energy for all of the HBD-HBA pairs. The total hydrogen bond energy may be normalised. The total hydrogen bond energy may be normalised by the number of three-dimensional conformations in the physical distribution. Alternatively, calculating the hydrogen bond energy based upon interactions between neighbouring molecules in the physical distribution may be calculated using a quantum mechanics model. Accordingly, the calculating the hydrogen bond energy based upon interactions between neighbouring molecules in the physical distribution may comprise: e-i) calculating the energy of the physical distribution and calculating the sum of the energies for all of the lipids which form the physical distribution in isolation; and e-ii) calculating the difference between the energy of the physical distribution and the sum of the energies for all of the lipids which form the physical distribution in isolation. It may be appreciated that the above method does not rely solely on geometric criteria but rather computes the electronic interactions that contribute to HB interaction. In embodiments where the method comprises generating a dynamic lipid mixture, the method may comprise calculating the hydrogen bond energy for the dynamic mixture. It may be appreciated that the calculated the hydrogen bond energy may change over time. The method may comprise monitoring how the calculated hydrogen bond energy changes over time. The method may comprise recording the average calculated hydrogen bond energy for the system and / or the maximum hydrogen bond energy for the system. In embodiments where the method comprises generating a static lipid mixture, the method may comprise calculating the hydrogen bond energy for the static mixture. The method may comprise: f) optionally repeating steps (d-ii) to (e). Steps (d-ii) to (e) may be repeated at least once, at least twice, at least 3 times, at least 4 times, at least 5 times, at least 6 times, at least 7 times or at least 8 times. The method may comprise repeating steps (d-ii) to (e) between 1 and 50 times, between 2 and 40 times, between 4 and 20 times, between 6 and 15 times or between 8 and 10 times. The method may comprise: g) optionally repeating steps (d) to (e) or steps (d) to (f). In particular, the method may comprise optionally repeating steps (d-i) to (e) or steps (d) to (f). Steps (d) to (e) or steps (d) to (f) may be repeated at least once, at least twice, at least 3 times or at least 4 times. The method may comprise repeating steps (d) to (e) or steps (d) to (f) between 1 and 30 times, between 1 and 20 times, between 2 and 10 times, between 3 and 8 times or between 4 and 5 times. The method may comprise identifying the highest or mean hydrogen bond energy calculated as the potential hydrogen bond energy for the formulation at the selected ratio. Advantageously, repeating steps (d) to (e) or steps (d) to (f) may optimise the three-dimensional conformations and the physical distribution for a high hydrogen bond potential. The method may comprise: h) selecting a new ratio of the three or more different lipids and then repeating steps (d) to (e), steps (d) to (f) or steps (d) to (g) for the new ratio. Step (h) may be repeated to provide the hydrogen bond potential for a plurality of ratios. The method may comprise identifying specific ratios as being potentially stable formulations based upon the calculated hydrogen bond potential. It may be appreciated that formulations with a higher hydrogen bond potential may be predicted to be more stable than formulations with a lower hydrogen bond potential. Stability can then be verified experimentally. All features described herein (including any accompanying claims, abstract and drawings), and / or all of the steps of any method or process so disclosed, may be combined with any of the above aspects in any combination, except combinations where at least some of such features and / or steps are mutually exclusive. For a better understanding of the invention, and to show how embodiments of the same may be carried into effect, reference will now be made, by way of example, to the accompanying Figures, in which:- Figure 1 displays the simulated conformation states of lipid component head groups projected in the lipid-bilayer plane. The displayed conformers provide optimized intramolecular stability and maximum intramolecular hydrogen bond forces. PPZ- AA10: 2 O and 6 N act as Hydrogen Bond Acceptor (HBA), 2 N act as Hydrogen Bond Donor (HBD). 4a3-sc8: 12 O and 2 N act as HBA. 306-O12B: 8 O and 3 N act as HBA. 306-Oi10: 8 O and 3N act as HBA, 2 N act as HBD. cKK-E12: 2 O and 4 OH act as HBA and HBD, 4 N act as HBA, 2 N act as HBD. LP-01: 9 O and 1 N act as HBA. Dlin-MC3: 2 O and 1 N act as HBA. C12-200: 5 OH act as HBD and HBA, 5 N act as HBA. C24: 4 O and 3 N act as HBA. DSPC: 2 O act as HBA; Figure 2 shows the maximum and average hydrogen bond forces (in kCal / mol) in three component formulation designs composed of C12-200, 1,2-Distearoyl-sn- glycero-3-phosphocholine (DSPC) and cholesterol at different molar ratios (% mol). The square with the white border is where previously reported 50 / 10 / 38.5 / 1.5 ionizable lipid / DSPC / cholesterol / PEGylated lipid formulation would fall; Figure 3 shows the evaluation of the size a subset of the library of the three component LNPs for mRNA delivery. Bars represent means ± standard deviations for n = 3; Figure 4 shows the evaluation of the polydispersity index (PDI) for a subset of the library of the three component LNPs for mRNA delivery. Bars represent means ± standard deviations for n = 3; Figure 5 shows the evaluation of the Encapsulation Efficiency (EE%) for a subset of the library of the three component LNPs for mRNA delivery. Figure 6 shows the evaluation of the size of a subset of the library of the three component LNPs for saRNA delivery. Bars represent means ± standard deviations for n = 3; Figure 7 shows the evaluation of the PDI of a subset of the library of the three component LNPs for saRNA delivery. Bars represent means ± standard deviations for n = 3; Figure 8 shows the evaluation of the Encapsulation Efficiency (EE%) for a subset of the library of the three component LNPs for saRNA delivery. Figure 9 displays the evaluation of a subset of the library of the three component LNPs for ability to deliver RNA to HEK cells. Relative luciferase activity values for mRNA. Bars represent means ± standard deviations for n = 3; Figure 10 displays the evaluation of a subset of the library of three component LNPs for ability to deliver RNA to HEK cells. Relative luciferase activity values for saRNA. Bars represent means ± standard deviations for n = 3; Figure 11 displays the evaluation of a subset of the library of hydrogen-bond stabilized four component LNPs for ability to deliver RNA to HEK cells. Relative luciferase activity values for saRNA. Bars represent means ± standard deviations for n = 3; Figure 12 shows the comparison between PEGylated LNPs and a subset of the three component LNP library delivering saRNA to human skin explant (C12-200 LNP Alternative 1 and 2). Imaging performed 5 days after administration; Figure 13 shows the comparison between PEGylated LNPs and a subset of the three component LNP library delivering saRNA to human skin explant. Bars represent means ± standard deviations for n = 3. Imaging was conducted at 24 hours, 3, 5 and 7 days and show the firefly luciferase activity (expressed as photon emission per second); Figure 14 displays the evaluation of a subset of the library of three components LNPs for ability to deliver RNA in vivo; and Figure 15 shows the in vivo prolonged saRNA delivery ability of a three component LNP compared to the PEGylated LNP at 18 hrs, 48 hrs, 7 days and 14 days. Bars represent means ± standard deviations for n = 5. Figure 16 shows the in vivo prolonged mRNA delivery ability of a three component LNP compared to the PEGylated LNP at 18 hrs, 48 hrs, 3 days and 6 days. Bars represent means ± standard deviations for n = 5. Figure 17 shows the intravenous in vivo mRNA delivery ability of a three component LNP compared to the PEGylated LNP at 6 hrs and 24 hrs. Bars represent means ± standard deviations for n = 4. EXAMPLE 1: Design and optimisation of three component LNPs The inventors looked at optimising hydrogen bonding in three lipid component formulations. A hydrogen bond simulation and optimization algorithm were implemented in Python 3.12.1. Rdkit 2023.09.4 and Scipy 1.13.1 packages were used for lipid conformer optimization, translation and visualisation. Numpy 1.26.3 was used to generate random sequences of lipids for simulation. Energy bond interaction calculations were calculated using in-house code. The following algorithm was implemented to design and optimize hydrogen bond- stabilised three lipid component formulations: 1. Identify canonical Simplified Molecular Input Line Entry System (SMILES) decomposition of the lipid head groups. The head group is considered to be the smallest continuous portion of the lipid which comprises all of the heteroatoms present in the lipid and the first adjacent atom to the smallest continuous portion of the structure at each point where the smallest continuous portion connects to a remainder of the lipid structure. For PEGylated lipids, the PEGylated portion is not considered to be part of the head group. In embodiments where the lipid comprises only one heteroatom (e.g. cholesterol), the smallest continuous portion of the structure of a lipid which contains all of the heteroatoms therein may be understood to the heteroatom and the carbon atom to which it is bonded (e.g. the smallest continuous portion of the structure for cholesterol will be understood to be CH-OH). If the atom at the point where the smallest continuous portion connects to a remainder of the lipid structure is a carbon atom, and it is bonded to two adjacent carbon atoms, each of which are part of the remainder of the lipid structure, then the head group comprises only one of the adjacent carbon atoms (e.g. the head group for cholesterol is CH2-CH- OH, as shown below). Ionizable lipids are considered in their neutral form. 2. Convert SMILES to 3D conformations. The conformations were generated based on experimental torsional-angle and ring geometry preferences (ETKDG) algorithm (Wang et al., J Chem Inf Model. 2020 Apr 27; 60(4): 2044-2058. However, it is noted that alternative approaches could be used. a. Generate multiple 3D conformations. b. Optimise conformations using the Universal Force Field (UFF) algorithm to ensure intramolecular stability. It is noted that alternative algorithms could be used to optimise the conformations. c. Return and display optimised conformational states for each molecule. d. Identify hydrogen donor and acceptor atoms for each conformational state. 3. Rotate and translate molecules to maximise cross-sectional area in the lipid-bilayer plane (a projection in a 2D grid) based on the convex hull algorithm. 4. Initialise lipid mixture grid for simulation. a. Create initial lipid sequence based on given lipid ratios. c. Shuffle lipid sequence to randomise the distribution of lipids based on Mersenne Twister Algorithm. d. Reshape lipid sequence into the 2D grid representing the lipid-bilayer plane. e. Define a grid space based on the cross-sectional area of each lipid. 5. Translate and rotate molecules to minimise distance between neighbouring molecules and avoid overlap. 6. Establish hydrogen bond counts according to the distance between acceptor and donor atoms between neighbouring molecules, e.g. the distance between a nitrogen or oxygen atom in one molecule and a hydrogen atom in a neighbouring molecule. Herein, the distance threshold to establish hydrogen bond interactions was set at 3.5 Å, i.e. hydrogen bonds are only counted for donors and acceptors which are separated by a distance of equal to or less than 3.5 Å. 7. Calculate the hydrogen bond energy between neighbouring molecules based on distance, electronegativity, and functional groups. a. Ensure each acceptor bonds with only one donor and vice versa by selecting the closest pairs. b. Return total hydrogen bond count and energy for each final pair based on the Gerber Method: E = -C / r2Where C is equal to 33 kCal.Å2.mol-1, and r is the distance between hydrogen acceptor and donor atoms in Å. The result is normalised by the number of molecules. 8. Simulate (repeat steps 3 to 7) to optimise the 2D grid for the given 3D conformations. 9. Simulate (repeat steps 2 to 8) to optimise conformers and lipid sequences for high hydrogen bond potential. Both the mean value and the highest value obtained are assessed and considered relevant. 10. Simulate (repeat steps 2 to 9) for different lipid ratios to identify formulation designs with high hydrogen bond potential and ensuring at least 20% cholesterol (mol ratio) to prevent lipids precipitation. 11. Perform space-filling design-of-experiments (DoE) within the area defined in step 10 to fine-tune formulation design for stability and efficient RNA delivery. Herein, the present algorithm has been implemented to identify formulation designs of maximum hydrogen-bond interactions for C12-200, 1,2-distearoyl-sn-glycero-3- phosphocholine (DSPC) and cholesterol lipids. It will be appreciated that more than one conformational state is possible / feasible. Accordingly, for each formulation component, 10 stable conformational states have been screened to select conformers with the highest intermolecular hydrogen bond potential, i.e. step 8 comprises repeating steps 3 to 7 9 times. For each formulation design and lipid conformational state, 5 mixture grids have been randomly generated and used for hydrogen bond calculation, i.e. step 9 repeats steps 2 to 84 times. Details of the head groups for the lipids used in this study are provided in Table 1. Additionally, FIG 1 displays optimized headgroup 3D conformers projected in the lipid- bilayer 2D plane. FIG. 2 displays the output from step 11 in the above method. Table 1: Structure of the head group, cross-sectional-area of the head group for the optimised conformer and number of hydrogen bond donors (HBDs) and hydrogen bond acceptors (HBAs) for the lipids used in this study Cross- sectional No. No. area of Lipid Head group of of head HBDs HBAs group / Ų Dlin-MC3 26.80 0 3 C12-200 47.14 5 10

[0002] Cross- sectional No. No. area of Lipid Head group of of head HBDs HBAs group / Ų 6-O12B 63.60 0 11

[0003] Cross- sectional No. No. area of Lipid Head group of of head HBDs HBAs group / Ų 06-Oi10 53.61 0 11 cKK-E12 49.68 6 10

[0004] Cross- sectional No. No. area of Lipid Head group of of head HBDs HBAs group / Ų LP-01 50.91 0 10 PPZ-A10 47.78 8 2

[0005] Cross- sectional No. No. area of Lipid Head group of of head HBDs HBAs group / Ų A3-SC8 70.78 0 19 C24 46.09 0 7 Cross- sectional No. No. area of Lipid Head group of of head HBDs HBAs group / Ų DOTAP 26.52 0 4 DSPC 31.12 0 8 Cholesterol 10.12 1 1 The inventors conducted the algorithm described in example 1 to identify three component formulations which they believed would be stable and efficiently deliver RNA. Each formulation comprised 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC), cholesterol and a given ionisable lipid. The ionisable lipids were DLin-MC3- DMA, C12-200, 306-O12B, 306Oi10, C24, cKK-E12, LP-01, PPZ-A10, 4A3-SC8 and DOTAP. Compositions of LNP formulations predicted to be stable are shown in Tables 2 and 3 below. PEGylated formulations were selected based on literature and previously reported designs. Table 2: Formulation composition for LNPs and calculated hydrogen bond energy (HB) using the method described in example 1. IL= Ionizable Lipid. The percentages given are mol %. The weight ratio of mRNA and saRNA to ionizable lipid is 0.013 and 0.022. Corresponding N to P ratio is reported. F Ho D C r S h P mIL(N / (sN(kC B(k / H P o ( Em%leG P a R P aaC a B R vCIuI l l ( / / R R MDLNs e%N(lam m%(a atA)are%Aat t tx)o oii)irg)o o ° o)o)l len) )lC12-200 LNP1 35 16 46.5 2.5 40 70 -9.80 -12.79 C12-200 LNP2 43 22 35 0 40 70 -11.54 -15.82 C12-200 LNP3 45 21 33 0 40 70 -13.94 -15.14 C12-200 LNP4 45 24 31 0 40 70 -13.74 -18.75 C12-200 LNP5 45 23 32 0 40 70 -14.99 -15.69 C12-200 LNP6 44 23 33 0 40 70 -14.16 -16.02 C12-200 LNP7 44 25 31 0 40 70 -13.29 -13.69 C12-200 LNP8 43 25 32 0 40 70 -12.63 -14.26 C12-200 LNP9 43 24 33 0 40 70 -11.87 -15.42 C12-200 LNP10 44 24 32 0 40 70 -13.97 -15.84 C12-200 LNP11 44 26 30 0 40 70 -14.50 -16.56 C12-200 LNP12 42 26 32 0 40 70 -12.99 -15.16 C12-200 LNP13 50 26 32 0 40 70 -14.87 -16.16 C12-200 LNP14 44 24 29.5 2.5 40 70 C12-200 LNP15 44 24 30.5 1.5 40 70 C12-200 LNP16 44 24 31.5 0.5 40 70 CKK-E12 LNP1 35 16 46.5 2.5 32 56 -9.62 -13.20 CKK-E12 LNP2 44 26 30 0 32 56 -17.14 -17.70 CKK-E12 LNP3 40 30 30 0 32 56 -13.79 -15.59 CKK-E12 LNP4 40 20 40 0 32 56 -12.52 -15.45 CKK-E12 LNP5 44 24 32 0 32 56 -16.44 -18.67 CKK-E12 LNP6 31 22 32 0 32 56 -10.65 -11.87 306-O12B LNP1 50 10 38.5 1.5 24 42 -2.09 -2.18 306-O12B LNP2 30 40 30 0 24 42 -3.70 -4.01 306-O12B LNP3 40 40 20 0 24 42 -2.90 -3.56 306-O12B LNP4 40 30 30 0 24 42 -2.55 -2.76 306-O12B LNP5 30 30 40 0 24 42 -3.94 -4.29 306-O12B LNP6 40 24 36 0 24 42 -2.59 -3.07 F o D C h P(N(N(kH s B(rkmILS P o Em / a / C C H a B LI (%leG R P R P a I u v a Dla(%C s(NRNRl / el / M(%) te %A a t A a m tram a io) ro)io)°tioo g o x n))l l)el)06-O12B LNP7 40 20 40 0 24 42 -2.96 -3.3206-O12B LNP8 31 22 47 0 24 42 -3.54 -4.2606-O12B LNP9 30 20 50 0 24 42 -4.20 -4.6306-O12B LNP10 30 37 33 0 24 42 -4.40 -5.0306-O12B LNP11 33 37 30 0 24 42 -3.66 -4.0606-O12B LNP12 44 26 30 0 24 42 -2.35 -3.2206-O12B LNP13 30 38 32 0 24 42 -4.49 -6.0506-O12B LNP14 30 36 34 0 24 42 -3.85 -4.0706-O12B LNP15 30 40 30 0 24 42 -3.45 -3.9306-O12B LNP16 44 24 32 0 24 42 -3.55 -4.10O6-O12B LNP17 50 20 30 0 24 42 -2.38 -2.62 C24 LNP1 50 10 38.5 1.5 24 42 -2.56 -2.70 C24 LNP2 30 40 30 0 24 42 -3.97 -4.90 C24 LNP3 40 40 20 0 24 42 -3.50 -4.15 C24 LNP4 40 30 30 0 24 42 -2.96 -3.31 C24 LNP5 39 33 28 0 24 42 -2.80 -3.83 C24 LNP6 41 32 27 0 24 42 -2.85 -3.70 C24 LNP7 40 34 26 0 24 42 -2.63 -3.15 C24 LNP8 50 10 40 0 24 42 -2.21 -2.74 C24 LNP9 37 30 33 0 24 42 -2.14 -2.68 C24 LNP10 40 33 27 0 24 42 -3.06 -3.48 4a3sc8 LNP1 50 10 38.5 1.5 24 42 -1.98 -2.35 4a3sc8 LNP2 40 24 32 0 24 42 -2.69 -3.46 LP01 LNP1 50 10 38.5 1.5 8 14 -1.51 -1.85 LP01 LNP2 40 30 30 0 8 14 -3.00 -3.80 LP01 LNP3 40 35 25 0 8 14 -3.17 -3.52 LP01 LNP4 40 40 20 0 8 14 -3.30 -4.12 LP01 LNP5 45 35 20 0 8 14 -2.54 -2.80 LP01 LNP6 35 40 25 0 8 14 -3.43 -3.99 LP01 LNP7 38 37 25 0 8 14 -3.75 -4.31 LP01 LNP8 36 37 27 0 8 14 -3.11 -4.31 LP01 LNP9 35 35 30 0 8 14 -3.93 -3.45 F o D C h P(N(N(kH s B(rkmILS P o Em / a / C H PaC P B IIu C(%leG ( R R a / v a LDla %s(NRNRl% a m el / Mma( tA a) re%Aat t tx)o oii i)rg)o °o oo))l len) )l306-Oi10 LNP1 50 10 38.5 1.5 24 42 -2.46 -2.83 306-Oi10 LNP2 30 40 30 0 24 42 -4.16 -5.17 306-Oi10 LNP3 50 10 40 0 24 42 -2.44 -2.96 DOTAP LNP1 50 10 38.5 1.5 8 14 -1.81 -2.46 DOTAP LNP2 35 16 49 0 8 14 -3.36 -3.64 Dlin-MC3- LNP1 50 10 38.5 1.5 8 14 -2.27 -2.99 DMA Dlin-MC3- LNP2 31 22 47 0 8 14 -3.87 -4.44 DMA Dlin-MC3- LNP3 33 18 49 0 8 14 -3.42 -3.48 DMA Dlin-MC3- LNP4 23 22 55 0 8 14 -3.92 -4.44 DMA Dlin-MC3- LNP5 25 18 57 0 8 14 -4.43 -5.14 DMA Table 3: Formulation composition for LNPs using alternative mRNA and saRNA to ionizable lipid ratio. IL= Ionizable Lipid. The percentages given are mol %. Corresponding N to P ratio is reported. Ionizable Formulation Ionizable DSPC Cholesterol PEG N / P Lipid ID Lipid % % % % Ratio (saRNA) C12-200 LNP17 35 16 46.5 2.5 13.3 C12-200 LNP18 35 16 46.5 2.5 23.3 C12-200 LNP19 35 16 46.5 2.5 40 C12-200 LNP14 to C12-200 LNP 19 would display similar HB potential to C12-200 LNP1, as the formulation design is similar. EXAMPLE 3: Physicochemical characterisation of LNPs The inventors then produced the LNP formulations identified in example 2 with saRNA and mRNA. The inventors then characterised the physicochemical properties of the formulations. The results are shown in Tables 4 and 5 below and Figures 3 to 8. Table 4: Size, PDI and encapsulation efficiency data for the 3-component mRNA-LNPs compared to PEGylated LNPs. Standard deviation (SD) is based on measurement triplicates. Formulation ID Size SD PDI SD Encapsulation (nm) efficiency (%) C12-200 LNP1 59.812 0.443 0.209 0.053 98.80 C12-200 LNP2 110.826 5.18 0.215 0.071 94.80 C12-200 LNP3 120.633 21.529 0.163 0.04 92.90 C12-200 LNP4 133.668 9.527 0.126 0.019 92 C12-200 LNP5 110.024 7.94 0.265 0.053 90.80 C12-200 LNP6 101.284 3.891 0.19 0.024 90 C12-200 LNP7 154.608 3.618 0.365 0.019 91.80 C12-200 LNP8 119.006 11.923 0.257 0.066 92.20 C12-200 LNP9 219.613 14.433 0.332 0.038 90 C12-200 LNP10 103.3 5.807 0.109 0.019 96.10 cKK-E12 LNP1 140.205 10.079 0.184 0.06 94 cKK-E12 LNP2 158.024 10.47 0.26 0.088 81.90 cKK-E12 LNP3 112.117 18.178 0.178 0.021 83.50 cKK-E12 LNP4 120.253 8.245 0.115 0.018 85.40 306-O12B LNP1 88.042 5.56 0.103 0.004 93.10 306-O12B LNP2 128.831 27.283 0.067 0.007 91 306-O12B LNP3 147.422 25.713 0.243 0.019 83.60 306-O12B LNP4 106.403 10.879 0.243 0.078 96.80 306-O12B LNP5 238.778 29.498 0.208 0.051 80.50 306-O12B LNP6 77.767 3.456 0.097 0.02 97.60 306-O12B LNP7 98.072 8.053 0.268 0.074 96 306-O12B LNP8 87.009 7.522 0.196 0.038 94.40 306-O12B LNP9 258.185 118.264 0.405 0.086 81.90 306-O12B LNP14 184.923 45.885 0.283 0.036 67.80 C24 LNP1 89.52 5.582 0.126 0.012 98.50 C24 LNP2 117.986 21.858 0.152 0.006 95 C24 LNP3 331.808 67.472 0.426 0.065 97.60 C24 LNP4 113.936 8.504 0.248 0.057 99.40 C24 LNP8 236.439 20.499 0.189 0.042 99.30 Dlin-MC3 LNP1 69.666 2.623 0.211 0.039 85.90 Formulation ID Size SD PDI SD Encapsulation (nm) efficiency (%) Dlin-MC3 LNP2 124.590 21.586 0.109 0.015 93.50 Table 5: Size, PDI, Zeta Potential and Encapsulation Efficiency data for the 3- component saRNA-LNPs compared to PEGylated LNPs. Standard deviation (SD) is based on measurement triplicates Formulation Size SD PDI SD Zeta Encapsulation ID (nm) Potential efficiency (mV) (%) C12-200 80.63 11.11 0.18 0.02 -10.65 96.62 LNP1 C12-200 97.74 7.2 0.18 0.04 1.45 96.32 LNP11 C12-200 88.02 9.18 0.1 0.02 -2.33 95.98 LNP10 C12-200 143.4 17.52 0.09 0.02 -2.85 88.11 LNP12 C12-200 152.971 9.173 0.312 0.06 -2.452 98.30 LNP13 cKK-E12 127.103 19.27 0.283 0.037 0.403 81.20 LNP5 cKK-E12 113.295 5.324 0.16 0.032 0.327 43.50 LNP6 306-O12B 102.9 12.38 0.314 0.039 -3.37 95.2 LNP1 306-O12B 177 12.96 0.21 0.01 -3.8 83.19 LNP10 306-O12B 226.7 14.45 0.3 0.05 -4.68 75.02 LNP11 306-O12B 271.5 27.05 0.288 0.023 -3.1 90.74 LNP12 306-O12B 152.413 31.186 0.203 0.008 -2.67 91.30 LNP13 306-O12B 126.936 1.211 0.118 0.013 -5.55 86.20 LNP14 Formulation Size SD PDI SD Zeta Encapsulation ID (nm) Potential efficiency (mV) (%) 306-O12B 82.553 15.147 0.258 0.009 -5.768 93.30 LNP4 306-O12B 164.90 4.14 0.19 0.00 79.76 LNP15 306-O12B 176.90 6.85 0.03 0.02 93.08 LNP16 C24 LNP1 108.9 9.17 0.136 0.032 0.741 97.8 C24 LNP9 251.10 1.50 0.17 0.02 0.71 97.32 C24 LNP10 135.20 7.33 0.10 0.01 0.64 94.78 C24 LNP2 202.70 26.72 0.322 0.039 0.031 92.96 C24 LNP5 158.50 15.88 0.247 0.059 0.359 85.79 C24 LNP6 206.50 16.70 0.302 0.045 0.117 92.17 C24 LNP7 164.20 1.77 0.17 0.01 92.92 4a3sc8 LNP1 195.178 20.341 0.224 0.001 -6.316 62.10 4a3sc8 LNP2 219.455 33.357 0.184 0.038 -4.447 65.50 LP01 LNP1 148.756 3.46 0.239 0.017 -8.826 85.60 LP01 LNP2 204.8 31.87 0.073 0.011 -5.08 66.1 LP01 LNP3 186.1 20.61 0.21 0.01 -5.77 77.99 LP01 LNP4 196.2 28.84 0.21 0.01 -6.32 66.47 LP01 LNP5 183.5 29.17 0.22 0.04 -5.52 83.89 LP01 LNP6 160.3 12.36 0.21 0.01 -4.3 78.72 306-Oi10 124.8 9.46 0.33 0.04 -0.15 34.02 LNP1 306-Oi10 195.3 3.95 0.26 0.00 -3.77 67.31 LNP2 306-Oi10 129.8 6 0.08 0.04 -5.11 98.58 LNP3 DOTAP LNP1 135.60 4.50 0.29 0.04 -29.58 100 DOTAP LNP2 271.10 1.40 0.13 0.02 -27.81 96.28 Dlin-MC3 62.225 3.817 0.105 0.013 95.90 LNP1 Dlin-MC3 109.381 4.335 0.063 0.002 91.70 LNP2 The majority of the formulations developed have an LNP size below 200 nm and / or comparable to the relevant PEGylated LNPs, see Figures 3 and 6. The majority of the formulations developed also showed PDI on average below 0.3 and / or comparable to the relevant PEGylated LNPs, see Figures 4 and 7. The majority of formulations developed also had an EE% on average above 80% and / or comparable to the relevant PEGylated LNPs, see Figures 5 and 8. The zeta potential of the formulations developed with ionizable lipids showed zeta potential below 15mV, below or comparable to the relevant PEGylated LNPs. EXAMPLE 4: In vitro transfection efficiency of LNPs The inventors then investigated the in vitro transfection efficiency of the LNPs. The results are provided in Tables 6 and 7 and Figures 9 and 10. Table 6: In vitro transfection efficiency for the three component LNP formulations for mRNA compared to PEGylated LNPs. Standard deviation (SD) based on triplicate measurements Formulation ID Transfection Efficiency SD (Fold increase in Luminescence) C12-200 LNP1 676094 92543 C12-200 LNP2 317610 21078 C12-200 LNP3 396692 42512 C12-200 LNP4 211794 40186 C12-200 LNP5 680488 25245 C12-200 LNP6 640880 47611 C12-200 LNP7 535679 25358 C12-200 LNP8 406870 40680 C12-200 LNP9 440693 89404 C12-200 LNP10 1335750 86938 cKK-E12 LNP1 126399 3086 cKK-E12 LNP2 1241500 103003 cKK-E12 LNP3 945265 83345 cKK-E12 LNP4 558996 43019 306-O12B LNP1 122680 70231 306-O12B LNP2 1392000 242987 306-O12B LNP3 1244000 217712 Formulation ID Transfection Efficiency SD (Fold increase in Luminescence) 306-O12B LNP4 1401000 169600 306-O12B LNP5 224660 29244 306-O12B LNP6 1306250 127463 306-O12B LNP7 598266 39273 306-O12B LNP8 464027 72478 306-O12B LNP9 258710 24466 306-O12B LNP14 456061 102543 C24 LNP1 221401 69408 C24 LNP2 405600 130112 C24 LNP3 592690 77453 C24 LNP4 696512 63602 Dlin-MC3 LNP1 993058 12086 Dlin-MC3 LNP2 290950 26823 Table 7: In vitro transfection efficiency for the three component LNP formulations for saRNA compared to PEGylated LNPs. Standard deviation (SD) based on triplicate measurements Formulation ID Transfection Efficiency SD (Fold increase in Luminescence) C12-200 LNP1 692072 39018 C12-200 LNP11 798016 81456 C12-200 LNP12 345183 39163 C12-200 LNP13 930082 24514 C12-200 LNP14 998237 23934 C12-200 LNP15 1198167 90885 306-O12B LNP1 1716667 123792 306-O12B LNP10 431345 60283 306-O12B LNP11 243162 34796 306-O12B LNP12 332297 24584 306-O12B LNP13 1312000 259675 306-O12B LNP14 1557667 45369 306-O12B LNP4 1183000 94016 Formulation ID Transfection Efficiency SD (Fold increase in Luminescence) 306-O12B LNP15 926098 56287 306-O12B LNP16 208940 15273 C24 LNP1 1495673 124012 C24 LNP2 776698 32226 C24 LNP5 346850 4745 C24 LNP6 375236 22281 C24 LNP7 1891963 83003 C24 LNP9 287819 198043 C24 LNP10 515971 362910 4a3sc8 LNP1 1081862 192123 4a3sc8 LNP2 814578 113505 LP01 LNP1 1716667 123792 LP01 LNP2 630598 156368 LP01 LNP3 1649126 104643 LP01 LNP4 1434442 164261 LP01 LNP5 219540 54268 LP01 LNP6 874335 94486 306-O10 LNP1 619600 117042 306-O10 LNP2 1288836 30470 306-O10 LNP3 1183956 5284 DOTAP LNP1 265344 38049 DOTAP LNP2 449509 27052 Dlin-MC3 LNP1 1220208 313275 Dlin-MC3 LNP2 1622250 144440 Figures 9 and 10 show that it is possible to obtain PEG-free LNP compositions with a higher transfection efficiency compared to the PEGylated LNP composition. As shown in Figures 9 and 10, it is possible to obtain compositions with a higher transfection efficiency compared to the PEGylated LNP composition. Figure 11 shows that hydrogen-bond optimized formulations also provide high transfection efficiency with the addition of PEG. EXAMPLE 5: Ex vivo transfection efficiency of LNPs in human skin explants The inventors then investigated the ex vivo transfection efficiency of LNPs in human skin explants. The three component formulations to be tested were chosen based on transfection efficacy and stability data, as discussed in examples 3 and 4, as well as their potential suitability for subcutaneous delivery. The results are provided in Table 8 below and Figures 12 and 13. Table 8: Ex vivo transfection efficiency for the three component LNP formulations for saRNA compared to PEGylated LNPs, expressed as photon emission per second 5 days following saRNA-LNP administration. Standard deviation (SD) based on triplicate measurements Formulation ID Emission SD (photon / second) C12-200 LNP1 70592402 34236789 C12-200 LNP17 7015050 2033242 C12-200 LNP18 16745580 7184276 C12-200 LNP19 21360716 11502464 C12-200 LNP10 86371585 25058033 C24 LNP1 33003867 24496206 C24 LNP7 32111295 2285874 306-O12b LNP1 106002246 15450246 306-O12b LNP14 48288104 17881701 cKK-E12 LNP1 144129 27179 cKK-E12 LNP5 340241 292238 Dlin-MC3 LNP1 14252669 2882214 Dlin-MC3 LNP2 47799244 15951447 As shown in Figure 13, the firefly luciferase activity (expressed as photon emission per second) of saRNA delivered via the three component LNPs was higher or comparable to the relevant PEGylated LNPs. EXAMPLE 6: In vivo transfection efficiency of LNPs following intramuscular administration The in vivo transfection efficiency of the LNPs was then investigated using a mouse model. The results are provided in Table 9 and 10 below and Figure 14, 15 and 16. Table 9: In vivo transfection efficiency for the three component LNP formulations for saRNA compared to PEGylated LNPs, expressed as photon emission following saRNA- LNP intramuscular administration. Standard deviation (SD) based on n=5. Emission Formulation ID SD Day (photon / second) 540800000 139730812.611668200000 66098124037506000000 23322907195C12-200 LNP1 119120000001191200000074466000000 160941915010580400000 30023957114135200000 67451464.031550600000 26088464932752000000 18237790445C12-200 LNP11 2232400000 816785651.271265200000 39554418210354800000 144366547.41466429197.2 45188661.031669000000 38722603232818000000 5004198245C24 LNP1 4332000000 160915195171548000000 5779013761096340000 33816756.2144416705.36 1232983.1011265644528 24288174432306000000 8370961715C24 LNP7 2198000000 982583329.87178060000 109920235105222000 3406885.67514180800000 41775590.9611281600000 55905885233347200000 285397077853O6-O12B LNP1 3662000000 299271949971505400000 83971173610365400000 463041358.814 Emission Formulation ID SD Day (photon / second) 58925847.2 24495373.4911104000000 39369023431784000000 69658452553O6-O12B LNP16 2094000000 934949196.572461800000 225723751510259660000 169045786.7141072000000 487257632.111905000000 964079872311058000000 55841669035Dlin-MC3 LNP1 14980000000 464833303579834000000 468394385110623400000 352723404.4141051600000 881135801.111395600000 82787003835713200000 41033365945Dlin-MC3 LNP2 3112800000 19059719837613000000 2644276841080420000 58721307.8914Table 10: In vivo transfection efficiency for the three component LNP formulations for mRNA compared to PEGylated LNPs, expressed as photon emission following mRNA- LNP intramuscular administration. Measurement is performed 16h post-injection on day 1. Standard deviation (SD) based on n=5. Emission Formulation ID SD Day (photon / second) 39658390 45772656137738078 307810392C12-200 LNP1 69334093 677536483121658538933997642727193 21871518112971515 131335902C12-200 LNP11 16001168 2236124234038661 21510676 Emission Formulation ID SD Day (photon / second) 282187196 1746125841207891171 1500524032C24 LNP1 149346146 137186725316921426 12606867622834790 11565512170766365 847977532C24 LNP7 9655921 858163536178662 5524799637149545 14673308146739841 2427504423O6-O12B LNP1 8956947 758019535012256 427587164256086 352698213O6-O12B LNP162142008 13150442821655 34956631224883 7413106134118492 145269925197387100 392714332Dlin-MC3 LNP1 17987229 6016474311070117 6375179615900597 919856315499742Dlin-MC3 LNP24 26018822067811 23663973931294 6466236The results provide in vitro data regarding the intracellular delivery of three component LNPs using various different ionisable lipids. In particular, the results show that the three component LNPs display better intracellular delivery in vitro compared to the control of the PEGylated LNPs. EXAMPLE 7: In vivo immunogenicity of LNPs The inventors wanted to demonstrate the capacity of LNP formulations to elicit antigen-specific B and T cells responses using a vaccine candidate. Animals were immunised with saRNA encoding the Rabies virus surface glycoprotein antigen and each group of mice was vaccinated with a different PEGylated and PEG-free LNP formulation, including the selected ionizable lipids C12-200, 306-O12B, C24 and MC3. Mice were injected with LNP formulated Rabies saRNA or Rabies mRNA in a total volume of 50 μL. After 6 weeks the mice were bleed, then received a boost vaccination with the same dose and LNP. The results provide evidence regarding in vivo ability of PEG-free LNPs to trigger immunogenicity (Table 11, 1213 and 14). These results indicate that the PEG free compositions show similar immunogenicity to the PEGylated compositions, i.e. removal of PEG does not appear to impact immunogenicity. Table 11: Murine serum antibody concentration (ng / mL) for the three component LNP formulations for saRNA compared to PEGylated LNPs. Standard deviation (SD) based on n=5. Antibody serum Formulation ID concentration SD Dose (ng / mL) PrimeC12-200 LNP1132070 197171254417 259555BoostC12-148857 22698Prime200 LNP11 860867 172740BoostPrimeC24 LNP167959 10142987983 245186Boost66883 14374PrimeC24 LNP7 372827 97054Boost150003 15464Prime3O6-O12B LNP1 1151417 239606BoostPr3O6-O12B LNP1692422 16276ime1000700 225556BoostD98277 15520Primelin-MC3 LNP1 1074583 117645Boost139987 29876PrimeDlin-MC3 LNP2 425890 79303Boost Table 12: Murine serum antibody concentration (ng / mL) for the three component LNP formulations for saRNA compared to PEGylated LNPs. Standard deviation (SD) based on n=5. Antibody serum Formulation ID concentration SD Dose (ng / mL) PrimeC12-200 LNP157112 179751614133 256438BoostC81787 8486Prime12-200 LNP11 1294007 311645BoostC24 LNP129321 6355Prime963527 157921Boost43415 8235PrimeC24 LNP7 459820 130050Boost36091 9746Prime3O6-O12B LNP1 1331507 427137Boost53404 221Prime3O6-O12B LNP16971185087 197326Boost34499 6377PrimeDlin-MC3 LNP1 908273 162601BoostDlin-MC3 LNP273209 36100Prime500173 152154BoostTable 13: Murine spleen T cell response (SFU / 10 splenocytes) following intramuscular injection of a 2 µg saRNA prime on day 1. SFU / 10 Formulation SD Dose splenocytes C12-200 LNP11692 775PrimeC12-200 LNP11806 372PrimeC24 LNP12878 1210PrimeC24 LNP71417 680Prime3O6-O12B LNP1 880 282Prime3O6-O12B LNP16628 222PrimeDlin-MC3 LNP11405 582PrimeDlin-MC3 LNP22519 789Prime Table 14: Murine spleen T cell response (SFU / 10 splenocytes) following intramuscular injection of a 2 µg saRNA prime on day 1. SFU / 10 Formulation SD Dose splenocytes C12-200 LNP12424 290PrimeC12-200 LNP112336 304PrimeC24 LNP11262 421PrimeC24 LNP7891 403Prime3O6-O12B LNP1 2131 753Prime3O6-O12B LNP163126 932PrimeDlin-MC3 LNP12030 609PrimeDlin-MC3 LNP21200 537PrimeEXAMPLE 8: In vivo transfection efficiency of LNPs following systemic administration The inventors evaluated the capability of PEG-free LNP formulation for systemic mRNA delivery. To assess systemic biodistribution, mRNA–LNP formulations encoding Green Lantern fLuc fusion constructs were intravenously into a murine model (Table 15). In comparison to PEGylated formulations, C12-200 PEG-free LNPs exhibited a modest increase in splenic expression and a log-fold reduction in hepatic expression. This suggests that PEG free compositions would be advantageous for indications where spleen expression is desirable, such as infectious disease prophylaxis, therapies, cancer immunotherapies or autoimmune disorder treatments. Table 15: In vivo transfection efficiency for the three component LNP formulations for mRNA compared to PEGylated LNPs, expressed as photon emission following mRNA- LNP intraveneous administration. Standard deviation (SD) based on n=4. Emission Time Formulation ID Organ SD (photon / second) (hours) Lungs 64662502042250Heart 726250291332Liver 6855000003874297106 Spleen 120195008999547C12-200 LNP1 Kidneys 24124751284301Lungs 3928000786925Heart 761150257541 24Liver 55787500088119289 Emission Time Formulation ID Organ SD (photon / second) (hours) Spleen 54335001099182Kidneys 1983000373466Lungs 34197501368577Heart 614950169624Liver 46487500269055296 Spleen 145760007223811145278C12-200 LNP11Kidneys 389500Lungs 1942500342513Heart 900125476769Liver 72190000904839924 Spleen 109695003465373Kidneys 506100104661EXAMPLE 9: Cytokine release and complement activation of LNPs The inventors wanted to assess the reactogenicity of LNP formulations. Some PEG-free and PEGylated LNP formulations were tested for cytokine release and complement activation. C3b levels were strongly inhibited in all PEG-free LNP formulations, regardless of RNA dose (Tables 16 and 17), demonstrating the additional safety of PEG-free formulations. While the 306-O12B PEG-free LNP is associated with higher cytokine release (including IL-1RA, IL-8, IL-6 and TNF-α), the other PEG-free LNPs show a more moderate cytokine induction, which would be preferable for reduced reactogenicity. Table 16: Cytokine profiles of PEGylated and PEG-free LNPs for RNA delivery, including on Formulation IL-1RA IL-8 TNF-α Dose SD SD SD ID (Log2FC) (Log2FC) (Log2FC) (μg / mL 2.33 0.82 3.41 3.08 5C12-2003.28 0.33LNP14.10 0.723.00 0.49 4.57 3.24 104.75 1.653.66 1.86 4.54 3.13 204.06 1.092.48 0.69 3.76 3.47 5C12-200 2.21 0.59 4.13 3.46 10LNP114.31 1.122.90 2.061.52 1.14 3.95 2.92 20 Formulation IL-1RA IL-8 TNF-α Dose SD SD SD ID (Log2FC) (Log2FC) (Log2FC) (μg / mL 1.08 1.450.00 0.33 0.93 1.86 53O6-O12B1.80 1.09 2.79 2.73 10LNP13.10 1.793.43 0.803.22 1.79 3.55 4.40 204.48 1.541.74 1.28 3.61 2.55 53O6-O12B 2.71 1.52 5.41 2.62 10LNP165.22 2.545.25 2.523.24 1.57 6.28 2.44 202.23 1.630.00 1.23 1.61 1.17 5C24 LNP14.40 1.970.22 1.37 1.82 1.22 103.49 0.800.00 0.71 2.13 1.55 202.67 1.410.20 1.82 0.63 0.74 5C24 LNP72.71 0.640.65 1.56 0.77 0.93 103.09 0.240.00 0.98 0.87 1.00 203.20 0.890.00 1.50 0.43 0.85 5Dlin-MC3 0.56 1.75 2.16 1.46 10LNP14.21 1.154.71 1.830.55 1.54 2.67 1.83 203.66 0.470.33 0.92 2.45 1.90 5Dlin-MC3 0.51 0.75 2.74 1.91 10LNP24.07 0.894.24 1.291.08 1.28 2.19 1.49 20Table 17: Cytokine and complement activation profiles of PEGylated and PEG-free LNPs for RNA delivery, including IL-1B and IL-6 expressed as log fold increase from the untreated sample, and iC3b concentration expressed as percentage from untreated control. Percentage increases from untreated samples was calculated as follows, ((Concentration iC3b test compound-Concentration iC3b iC3b (% Formulation IL-1B IL-6 Dose SD SD untreated SD ID (Log2FC) (Log2FC) (μg / mL control) 1.09 2.183.68 3.01 222 143 5C12-200 4.99 3.55 361 82 10LNP12.78 2.332.85 2.165.22 2.88 269 136 201.93 2.274.26 3.59 146 58 5C12-200 4.53 3.60 195 94 10LNP112.45 2.882.48 2.244.36 3.03 237 195 20 iC3b (% Formulation IL-1B IL-6 Dose SD SD untreated SD ID (Log2FC) (Log2FC) (μg / mL control) 0.00 0.001.36 2.73 260 168 53O6-O12B3.41 2.87 274 167 10LNP10.44 0.882.07 2.844.16 4.28 260 139 200.68 0.804.22 2.09 138 10 53O6-O12B 2.33 1.895.40 2.41 104 17 10LNP16 3.62 2.746.08 2.43 116 37 200.00 0.002.27 1.82 116 37 5C24 LNP10.00 0.002.38 1.92 222 139 100.00 0.002.49 1.99 306 212 200.00 0.001.58 1.22 133 21 5C24 LNP70.00 0.002.04 1.44 141 43 100.00 0.001.45 1.74 165 46 200.00 0.001.52 1.39 228 130 5Dlin-MC3 0.26 0.523.16 2.12 231 124 10LNP1 0.27 0.553.03 2.18 221 135 200.00 0.003.15 2.40 133 56 5Dlin-MC3 3.20 2.32LNP20.27 0.53171 58 100.76 0.973.18 2.29 185 75 20 The inventors wanted to investigate the stability of LNP formulations. Some of the saRNA C12-200 LNP formulations were selected as exemplary formulations. Buffers or PBS were added to the C12-200 LNP formulations at a 1:1 volume ratio, as described in the materials and methods section below. The size of the LNPs, the PDI and the transfection efficiency was determined when the buffered formulations were fresh and after 7 days at 4°C. The results are provided in Tables 18 to 26. Table 18: Size, PDI and in vitro transfection (expression in HEK293 cells) for saRNA formulation:buffer systems determined immediately after the formulation and buffer had been combined. Standard deviation (SD) based on triplicate measurements System Size SD PDI SD Transfection SD ID (nm) Efficiency (Fold increase in Luminescence) S1 114.0 25.14 0.32 0.037 100174 32987 S2 92.00 9.44 0.22 0.01 108492 5722 S3 82.47 2.04 0.18 0.02 222601 43611 S4 86.85 8.24 0.18 0.02 268510 62761 S5 90.17 6.91 0.23 0.01 175192 38654 S6 93.74 7.59 0.27 0.03 186457 16898 S7 154.4 16.07 0.24 0.04 188683 21332 S8 156.1 19.00 0.19 0.03 451003 28777 S9 170.7 7.67 0.16 0.02 656303 20383 S10 159.6 13.62 0.17 0.01 532119 19304 S11 158.4 5.90 0.14 0.03 597171 7576 S12 161.2 16.97 0.19 0.02 658143 29222 S13 156.1 19.00 0.19 0.03 451003 28777 S14 148.7 15.06 0.15 0.02 275001 23577 S15 136.10 10.33 0.16 0.04 259693 16455 S16 121.30 7.83 0.30 0.01 982395 24568 S17 86.66 0.39 0.09 0.06 206427 12708 S18 129.8 4.59 0.08 0.04 1183956 87055 S19 153.6 2.92 0.21 0.01 940230 49950 S20 143.9 1.99 0.04 0.03 764716 25242 S21 155.7 4.16 0.21 0.01 996475 75646 S22 129.80 4.59 0.08 0.04 1183956 87055 S23 87.44 0.72 0.09 0.01 189520 5617 S24 136.10 10.33 0.16 0.04 242629 12200 formulation:buffer systems stored at 4°C for seven days. Standard deviation (SD) based on triplicate measurements System Size (nm) SD PDI SD Transfection SD ID Efficiency (Fold increase in Luminescence) S1 114.29 7.027 0.249 0.001 183219 139752 S2 86.69 3.35 0.149 0.006 122448 2841 S3 88.00 10.85 0.216 0.044 112913 13806 S4 80.02 2.86 0.130 0.003 160900 5564 S5 79.89 2.53 0.122 0.004 133270 4072 S6 82.94 2.71 0.172 0.018 147947 12862 S7 146.3 6.65 0.204 0.024 151983 6329 S8 185.0 9.17 0.264 0.034 114994 8296 S9 169.8 4.66 0.211 0.014 586192 32483 S10 154.2 6.30 0.226 0.014 433492 10781 S11 161.4 5.06 0.160 0.012 715581 16102 S12 187.7 3.59 0.314 0.012 833148 20317 S13 185 9.17 0.264 0.034 114994 8296 S14 143.60 0.95 0.156 0.027 232246 18228 S15 125.80 2.50 0.143 0.028 176535 18299 S16 132.50 2.77 0.248 0.023 594325 31433 S17 94.53 2.20 0.110 0.017 136841 22163 S18 121.9 4.7 0.089 0.01 438930 17264 S19 144.7 5.93 0.199 0.018 462596 17080 S20 142.1 1.96 0.055 0.034 543327 7605 S21 157.8 3.67 0.213 0.002 480534 18253 S22 121.90 4.70 0.089 0.010 438930 17264 Table 20: Size, PDI and in vitro transfection (expression in HEK293 cells) for saRNA formulation:buffer systems stored at 4°C for fourteen days. Standard deviation (SD) based on triplicate measurements System Size SD PDI SD Transfection SD ID (nm) Efficiency (Fold increase in Luminescence) S17 101.30 4.72 0.182 0.044 178008 6150 S18 116.6 2.05 0.264 0.034 279320 93459 S19 153.7 5.47 0.031 0.006 817719 89276 S20 278.5 2.37 0.207 0.012 1184886 361243 S21 132.5 2.4 0.082 0.037 1392000 63663 S22 116.60 2.05 0.031 0.006 279320 93459 S23 97.56 3.94 0.159 0.037 - - Table 21: Size, PDI and in vitro transfection (expression in HEK293 cells) for saRNA formulation:buffer systems stored at -20°C for seven days. Standard deviation (SD) based on triplicate measurements System Size (nm) SD PDI SD Transfection SD ID Efficiency (Fold increase in Luminescence) S23 106.0 8.66 0.219 0.023 112962 12717 S17 101.70 3.86 0.208 0.018 206427 12708 S24 138.20 2.19 0.206 0.023 256858 4427 Table 22: Size, PDI and in vitro transfection (expression in HEK293 cells) for saRNA formulation:buffer systems stored at -20°C for fourteen days. Standard deviation (SD) based on triplicate measurements System Size SD PDI SD Transfection SD ID (nm) Efficiency (Fold increase in Luminescence) S1 100.5 1.67 0.223 0.006 126365 6240 S17 99.33 1.02 0.205 0.040 221215 7131 S24 139.9 2.80 0.250 0.023 419592 28532 Table 23: Size and PDI or mRNA formulation:buffer systems determined immediately after the formulation and buffer had been combined. Standard deviation (SD) based on triplicate measurements System Size SD PDI SD Zeta SD ID (nm) potential (mV) S1 59.81 0.443 0.209 0.053 -5.239 1.41 S25 96.32 6.51 0.115 0.0050 -0.163 0.516 S18 147.42 25.7 0.243 0.019 -2.79 1.54 S26 77.77 3.46 0.097 0.020 -5.15 0.394 S27 113.9 8.50 0.248 0.057 -5.49 1.76 S28 158.0 10.5 0.260 0.88 - - S2 96.322 5.51 0.115 0.00500 - - based on triplicate measurements System Size SD PDI SD ID (nm) S1 66.56 4.66 0.165 0.043 S2 129.4 10.06 0.213 0.037 Table 25: Size, PDI and in vitro transfection (expression in HEK293 cells) for mRNA formulation:buffer systems stored at 4°C for twelve days. Standard deviation (SD) based on triplicate measurements System Size SD PDI SD ID (nm) S18 148.56 5.15 0.258 0.055 S26 83.29 2.89 0.065 0.007 S27 118.6 17.67 0.263 0.071 Table 26: Size, PDI and in vitro transfection (expression in HEK293 cells) for mRNA formulation:buffer systems stored at 4°C for thirty-five days. Standard deviation (SD) based on triplicate measurements System Size SD PDI SD ID (nm) S25 122.9 8.93 0.147 0.020 System Size SD PDI SD ID (nm) S28 171.9 11.69 0.241 0.017 The three component LNP formulations show comparable storage stability to the PEGylated formulations. Buffer B1, B4 and B5 show the best stability performance. SUMMARY The inventors have devised a lipid-bilayer simulation algorithm which assesses hydrogen bonding forces potential between lipid components based on molecular conformation simulation and thermodynamic calculation. The algorithm can be used to identify LNP formulations which maximise hydrogen bonding interactions between lipid components. Due to this, the optimized formulations do not require PEGylation to ensure nanoparticle stability and RNA delivery. This enables the provision of three component LNP formulations. The inventors have shown that hydrogen-bond stabilized three component LNP formulations provide efficient RNA delivery in vitro, ex vivo and in vivo. The inventors have also shown that these formulations are stable upon storage. MATERIALS AND METHODS In vitro transcription (IVT) Self-amplifying RNA (saRNA) derived from the Venezuelan Equine Encephalitis Virus (VEEV) encoding firefly luciferase (fLuc) was prepared using in vitro transcription. pDNA was transformed in Escherichia coli and cultured in 50 mL of LB with 1 mg / mL carbenicillin (Sigma–Aldrich, U.K.) and isolated using a Plasmid Plus Maxiprep kit (QIAGEN, U.K.). pDNA concentration and purity was measured on a NanoDrop One (ThermoFisher, U.K.) and then linearized using MluI for 2 h at 37 °C and heat inactivated at 80 °C for 20 min. Uncapped in vitro RNA transcripts were synthesized using 1 μg of linearized DNA template in a MEGAScript reaction (Promega, U.K.), according to the manufacturer’s protocol. Transcripts were then purified by overnight LiCl precipitation at −20 °C, pelleted by centrifugation at 14000 rpm for 20 min, washed once with 70% EtOH, centrifuged at 14 000 rpm for 5 min, and then resuspended in UltraPure H2O. Purified transcripts were then capped using the ScriptCap m7G Capping System (CellScript, Madison, USA) and ScriptCap 2′-O- Methyltransferase Kit (CellScript, USA) simultaneously, according to the manufacturer’s protocol. Capped transcripts were then purified again by LiCl precipitation, resuspended in ultraPure H2O, and stored at −80 °C until use. Messenger RNA (mRNA) was obtained from Centillion Technology, UK Lipid nanoparticle (LNP) formulation of RNA The following lipid components were used to formulate saRNA / mRNA in lipid nanoparticle (LNP) for in vitro and in vivo studies: Ionisable lipid, helper lipid, and cholesterol. Each LNP consisted of 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC) (Avanti® Polar Lipids, US), cholesterol (Avanti® Polar Lipids), and a given ionisable lipid. In the case of PEGylated LNPs, DMG-PEG 2000 was also added. Compositions for certain LNPs are shown in Table 1 in example 2. The ionisable lipids used include DLin- MC3-DMA (Cambridge bioscience, UK), C12-200 (Corden Pharma, Switzerland), and 306-O12B, 306Oi10, C24, cKK-E12, LP-01, and PPZ-A10 (Echelon Biosciences, USA), and DOTAP (Avanti® Polar Lipids, US). Each lipid component was dissolved in absolute ethanol at different stock concentrations and the molar ratio of each lipid component. saRNA / mRNA solutions were prepared in 50 mM sodium acetate and 100 mM sodium chloride buffer at pH 5.5. The weight ratio of saRNA and mRNA to ionizable lipid was 0.013 and 0.022 unless specified otherwise. RNA-LNPs were formulated using the NanoAssemblr Ignite (Precision NanoSystems Incorporated, Canada), where the flow rate ratio of lipid to RNA was 1:3 and the total flow rate was 8 mL / min. The formulated RNA-LNPs were diluted in 1X DPBS five times the volume of formulated material and was then concentrated through a VivaSpin 6 filter tube (Sartorius, Germany) that has a 10 kDa molecular weight cut-off (MWCO) at 4000 rpm, 18°C until the desired concentration is achieved. LNP size and surface charge characterisation For characterisation formulation size and surface charge, 8.5 μL of LNPs was diluted into 841.5 μL PBS (Sigma, U.K.) and equilibrated at room temperature prior to analysis. The LNPs was characterized on a Zetasizer Nano ZS (Malvern Instruments, U.K.) with Zetasizer 7.1 software (Malvern, U.K.) using 850 μL of diluted particles in a 1 mL cuvette and the following settings: material refractive index, 1.529; absorbance, 0.010; dispersant viscosity, 0.8820 cP; refractive index, 1.330; and dielectric constant, 79. Each sample was analyzed for up to 100 runs until the measurement stabilized. Encapsulation efficiency Quant-iT RiboGreen assay (Thermo Fisher Scientific) was used according to the manufacturer’s instructions to quantify the encapsulation efficiency and the amount of saRNA / mRNA encapsulated inside the LNP. In brief, RNA-LNP samples were diluted in 1X TE buffer to 6 μg / mL and were then mixed with either 1X TE buffer or 2% Triton X- 100 at a 1:1 ratio in a black 96-well Costar® plate (Thermo Fisher Scientific). RNA standards were prepared according to the manufacturer's instructions and pipetted into the same 96-well Costar® plate. Both samples and standards were incubated at 37 °C for 10 min and subsequently, were incubated at RT for 5 min. RiboGreen reagent was diluted in 1X TE buffer at a 1:100 dilution and 100 μL of the diluted RiboGreen reagent was added into each well. The plate was analyzed on the FLUOstar Omega microplate reader (BMG Labtech, UK) at an excitation of 480 nm and emission of 520 nm. Based on the standard curve, the concentration of total RNA and non- encapsulated RNA were calculated. The concentration of encapsulated RNA was calculated by subtracting the non- encapsulated RNA concentration from the total NA concentration. A Zetasizer Nano ZS (Malvern Instruments, UK) was used to analyze the RNA-LNP Z- average size, polydispersity index (PDI) and zeta potential. RNA-LNP samples were diluted in 850 μL of 1X DPBS at a 1:100 dilution in RT and were transferred into a DTS1070 cuvettes before inserting into the Zetasizer Nano ZS. The parameters used were: material refractive index of 1.529, absorbance of 0.010, dispersant viscosity of 0.8820 cP, refractive index of 1.330 and dielectric constant of 79. Cell culture and in vitro transfection HEK293T.17 (ATCC, UK) and HELA (ATCC) cells were cultured in complete Dulbecco’s Modified Eagle’s Medium (DMEM) (Gibco, UK) containing 10% fetal bovine serum (FBS), 1% L-glutamine (L-glu) and 1% penicillin-streptomycin (Pen-strep) (Thermo Fisher Scientific, UK). THP1 cells (ATCC) were cultured in complete Roswell Park Memorial Institute (RPMI) 1640 medium with 10% FBS, 1% L-glu and 1% pen-strep. Cells were plated in a 96-well plate 24 h prior to transfection at a density of 5 x104cells per well for HEK293T.17 and HeLa cells; and 8x104cells per well THP-1 cells. Transfection of saRNA encoding fLuc and mRNA encoding nanofLuc were performed using Lipofectamine MessengerMAX (Thermo Fisher, UK) and according to the manufacturer’s instructions. At 24 h post transfection, half of the cell supernatants were discarded from each well and an equal volume of the ONE-GloTM Luciferase (Promega, UK) added to each well for saRNA samples and 100 μL of Nano-Glo™ Live substrate was added and mixed well for mRNA formulations then placed in the dark. Cells were incubated with the luciferase reagent for 5 min to ensure complete lysis of the cells. Cells were then pipetted up and down and transferred to a white 96-wellCostar® plate (Thermo Fisher Scientific). The luminescence was then measured on the FLUOstar Omega microplate reader (BMG Labtech) using gain 2500. Human Skin Explant Injection, Culture, and Imaging Surgically resected specimens of human skin tissue were collected at Charing Cross Hospital, Imperial College London, U.K. All tissues were collected after receiving signed informed consent from all patients, under protocols approved by the Local Research Ethics Committee. The tissue was obtained from patients undergoing elective abdominoplasty or mastectomy surgeries. Tissue was refrigerated until its arrival in the laboratory, where it was cut into 1 cm2section, and the subcutaneous layer of fat removed. Explants were incubated at 37 °C with 5% CO2 in Petri dishes with 10 mL of Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% FBS, 5 mg / mL l- glutamine, and 5 mg / mL penicillin / streptomycin (Thermo Fisher, U.K.). Media was replaced every 3 days, and explants were cultured for up to 7 days. Explants were injected intradermally using a Micro-Fine Demi 0.3 mL syringe (Becton Dickinson, U.K.) with 2 μg of RNA and 25 μL of LNPs in PBS, unless otherwise indicated. Samples were imaged on an AMI HTX (Spectral Instruments Imaging equipped with Aura Imaging Software) for 60 min. Signal from each tissue explant was analyzed using Molecular Imaging software and expressed as total flux (p / s). In vivo methods and assays for saRNA / mRNA All animals were handled, and procedures were performed in accordance with the UK Home Office Animals (Scientific Procedures) Act 1986 under protocol number 1 of the project license P4EE85DED and personal license I92512599, in which all studies performed were approved by the Animal Welfare and Ethical Review Body (AWERB). Food and water were provided ad libitum. Female BALB / c mice aged 6-8 weeks old were placed into group of n = 5. Mice were injected I.M. with 10 μg of LNP formulated VEEV-Fluc saRNA or 8 μg of LNP formulated mRNA in a total volume of 50 μL. On day 1-5 after injection, mice were injected intraperitoneally (I.P) with 150 μL of XenoLight RediJect D-Luciferin Substrate (Perkin Elmer, UK). Mice were left to rest for 10 min and during the 10 min, the mice legs that were injected I.M. were shaved. Mice were then anesthetized with isoflurane and were subsequently imaged on an AMI HTX (Spectral Instruments Imaging equipped with Aura Imaging Software) for 30 sec. Signal from each injection site was quantified as Total Flux (photons / second) using the Aura imaging software. Storage stability studies for saRNA and mRNA three component LNP formulations For storage stability studies, the excipients trehalose, trehalose dihydrate, sucrose, L- arginine HCl, histidine HCl (Sigma Aldrich, UK) at the desired amounts described in Table 27 were dissolved in deionised water and PBS for preparation of formulation storage buffers. For the storage buffers the pH of each solution was adjusted to 6 or 7.4 as specified by addition of NaOH and / or HCl using the pH electrode Mettler Toledo InLab Micro (UK). Then buffers were added to the produced formulations at a 1:1 volume ratio to provide the buffer:formulation systems shown in Table 28. Table 27: Buffer design composition for optimized LNP stability studies. % m / m is % mass. d h B(%isHihT(L S % u T utiism y dre-aw(m c%e N Bdf rurm ptho m m / / efi geradiaHanm m sfn / fCi / / rneei letn ome 2 Im ml) )i rne sD H(0%)e( / ) )e%C l B1 2.44 40 0 0 0 0 PBS 6 B2 2 20 20 0 0 0 PBS 6 B3 2 30 20 0 0 0 PBS 6 B4 2 45 0 0 0 0 PBS 6 B5 2 45 13 0 0 0 PBS 6 B6 2.44 40 0 0 0 0 PBS 6 B7 0 40 0 35 0 0 PBS 6 B8 0 0 0 10 0 0 PBS 7.4 B9 0 0 0 0 0.1 5 PBS 6 B10 0 0 0 10 0 0 Tris 7.4 B11 0 10 0 0 0 0 Tris 7.4 B12 2.44 40 0 0 0 0 Tris 6 B13 2 20 20 0 0 0 Tris 6 B14 2 30 20 0 0 0 Tris 6 B15 0 0 0.1 45 0.1 5 PBS 6 Table 28: Buffer:formulation systems for optimized LNP stability studies System ID Formulation Buffer S1 C12-200 LNP1 PBS S2 C12-200 LNP1 B1 S3 C12-200 LNP1 B2 System ID Formulation Buffer S4 C12-200 LNP1 B3 S5 C12-200 LNP1 B4 S6 C12-200 LNP1 B5 S7 C12-200 LNP11 PBS S8 C12-200 LNP11 B1 S9 C12-200 LNP11 B2 S10 C12-200 LNP11 B3 S11 C12-200 LNP11 B4 S12 C12-200 LNP11 B5 S13 C12-200 LNP11 B6 S14 C12-200 LNP11 B7 S15 C12-200 LNP11 B9 S16 C12-200 LNP11 B10 S17 C12-200 LNP11 B11 S18 306-Oi10 LNP3 PBS S19 306-Oi10 LNP3 B12 S20 306-Oi10 LNP3 B13 S21 306-Oi10 LNP3 B14 S22 306-Oi10 LNP1 PBS S23 C12-200 LNP10 B10 S24 C12-200 LNP11 B15 S25 C12-200 LNP10 B1 S26 306-O12B LNP6 PBS S27 C24 LNP4 PBS S28 cKK-E12 LNP2 B1 Immunogenicity Mice were injected I.M. with LNP formulated Rabies saRNA or Rabies mRNA in a total volume of 50 μL. After 6 weeks the mice were bleed, then received a boost vaccination with the same dose and LNP. Blood was again harvested and the end of the experiment and the rabies-specific antibody responses in the peripheral blood were measured using a semi-quantitative antigen-specific ELISA. Briefly, maxiSorp high binding ELISA plates (Nunc) were coated with 100 μL per well of 1 μg / mL recombinant Rabies protein in phosphate-buffered saline (PBS). For the standard IgG, three columns on each plate were coated with 1 in 1000 dilution each of goat anti-mouse κ- and λ-light chains (Southern Biotech). After overnight incubation at 4 °C, the plates were washed four times with PBS–Tween 200.05% (v / v) and blocked for 1 h at 37 °C with 200 μL per well blocking buffer (1% bovine serum albumin (w / v) in PBS–Tween-20 0.05%(v / v)). The plates were then washed and the diluted samples or a fivefold dilution series of the standard IgG added using 50 μL per well volume. Plates were incubated for 1 h at 37 °C, then washed and secondary antibody added at 1 in 2000 dilution in blocking buffer (100 μL per well), and incubated for 1 h at 37 °C. After incubation and washes, plates were developed using 50 μL per well SureBlue TMB (3,3′, 5,5′-tetramethylbenzidine) substrate and the reaction stopped after 5 min with 50 μL per well stop solution (Insight Biotechnologies). The absorbance was read on a Versamax Spectrophotometer at 450 nm (BioTek Industries). Statistical analyses were performed on log-transformed data. Spleens harvested at cull were physically dissociated, and single-cell suspension splenocytes were prepared by passing cells through 70 μM cell strainers and density gradient separation centrifugation using lymphoprep containing SepMate-15 columns (StemCell Technologies, UK). For analysis of IFNγ production by ELISpot, splenocytes were stimulated with 15-mer peptides (overlapping by 11) spanning the length of the rabies protein (final concentration of 1 µg / mL) on pre-coated Mouse IFN-gamma ELISpot Plus plates (MabTech, Sweden). After 18–20 h of stimulation at 37 °C, IFNγ spot forming cells were detected by staining membranes with anti-mouse IFNγ biotin (1 mg / mL) (R4-6A2) followed by streptavidin-ALP (1 mg / mL) and development with ALP conjugate substrate, as per the MabTech protocol. Spots were enumerated using an AID ELISpot reader and software (AID). In vivo Bioluminescence imaging: Group of four female BALB / c mice (8-10 weeks old) were housed under standard conditions (22 °C, 55% humidity, 12 h light / dark cycle) and provided with a standard diet ad libitum. Each mouse received a 50 µL (5 µg mRNA) of formulation intravenously (IV) via the lateral tail vein. Bioluminescence imaging was performed using an IVIS Spectrum (REVVITY (Wales, UK)), and data capture and analysis were conducted via Living Image software 4.8.2. mRNA expression was detected based on the bioluminescence emission of firefly luciferase (560 nm). Imaging parameters included medium binning, an f / stop of 2, and an acquisition time determined using auto-exposure settings. For conducting the bioluminescence imaging, the mice were given subcutaneously (SC) with 150 mg / kg D-luciferin solution 10 minutes prior to each imaging time point. Then mice were anaesthetised with 4% Isoflurane for induction, with anaesthesia maintained at 2.5% Isoflurane during imaging. The images were captured at 0.5 and 6 h for the first cohort, and at 0.5, 6 and 24 h for the second cohort. Then mice were sacrificed for each cohort at the last imaging time point using a Schedule 1 method, dissected and the following individual organs were imaged on a petri dish: the heart, lungs, liver, spleen, kidneys and pancreas, taking bioluminescence readings from each of these sites. and 24 h post-injection. In Vivo Imaging (CT and Bioluminescence Imaging): All CT images were acquired at the 6 h time point using the Quantum GX2 micro-CT imaging system for small animal X-ray imaging (REVVITY, Wales, UK). Mice were administered a subcutaneous (SC) injection of D-luciferin (150 mg / kg) 10 minutes prior to imaging. Anaesthesia was induced with 4% isoflurane and maintained at 2.5% during scanning. Mice were positioned using the Mouse Imaging Shuttle (MIS) to ensure stability throughout the imaging session. CT scanning parameters were as follows: tube voltage, 90 kV; tube current, 88 µA; exposure time, 8 seconds; field of view, 46 × 46 mm; and pixel resolution, 0.09 × 0.09 mm. Three-dimensional (3D) bioluminescence images were acquired for each mouse using the IVIS Spectrum system (REVVITY, Wales, UK) in 3D mode. Bioluminescence signals were then overlaid onto the corresponding CT-derived full- body skeletal images using Living Image Software (v4.8.2, REVVITY), enabling precise anatomical localization of expression. Ex vivo blood exposures for cytokine assays: Blood was collected fresh in tubes containing the anticoagulant Li-heparin from healthy volunteers using the PharmB ethics. Within 30 minutes of being drawn, the blood was diluted 1:4 with complete culture media (RPMI-1640 10% v / v FBS), 400 µL of diluted blood was then seeded into 48-well plates, and 100 µL of each test compound diluted in media was added. The final concentration of ionisable lipid (C12- 200 / 3O6-012B / C24 / MC3) in the formulations tested were: 5, 10 and 20 μg / mL. Plates were incubated for 24 hours at 37°C, 5% CO2. Samples were then centrifuged at 860xg for five minutes, and 100 µL aliquots of the supernatants were frozen at -80 °C until analysis. Supernatants were thawed, and cytokine analysis followed the Human Magnetic Luminex Assay protocol. Samples, standards and all reagents were allowed to equilibrate to 15-30 °C. The standards provided in the kit were reconstituted with Calibrator Diluent RD6-52 using the volumes specified on the certificate of analysis and allowed to stand for 15 minutes with gentle agitation. 100 µL of each was then combined with Calibrator Diluent RD6-52 to make it up to 1 mL to create standard 1. A 3-fold dilution series in Calibrator Diluent RD6-52 was performed to develop five further standards. The wash buffer was made by adding 20 mL of wash buffer concentrate to 480 mL of distilled water. Samples were centrifuged at 860xg for five minutes, and 50 µL of sample or standards were plated in their respective wells; all standards and samples were read in duplicate. The human magnetic microparticle cocktail was vortexed, and 500 µL was added to 5 mL of Diluent RD2-1 to create the diluted microparticle cocktail. 50 µL of microparticle cocktail was added to every well on the 96-well plate and incubated at room temperature on a plate shaker set at 800 RPM for two hours. The plate was washed thrice with the addition and aspiration of 100 µL of wash buffer on the Bio-Plex Pro II wash station. 500 µL Biotin-antibody cocktail was added to 5 mL of Diluent RD2-1 to create the diluted biotin-antibody cocktail. 50 µL of diluted Biotin antibody cocktail was added to each well and incubated at room temperature on a plate shaker set at 800 RPM for one hour. During this incubation, the Bio-Plex 200 Luminex and bead regions and standard values, the Bio-Plex 200 Luminex were calibrated, and bead regions and standard values were entered into the software. Streptavidin-PE concentrate was vortexed, and 220 µL was added to 5.35 mL of Wash buffer in a polypropylene test tube wrapped with aluminium foil to protect it from light. The plate wash was repeated, and 50 µL of diluted streptavidin-PE and 50 µL of diluted streptavidin-PE were added to each well and incubated at 15-30 °C on a plate shaker set at 800 RPM for 30 minutes. The plate wash was repeated, and 100 µL of wash buffer was added to each well and incubated for two minutes on a plate shaker set to 800 RPM. The plate was analysed using the Bio-Plex 200 Luminex, setting the sample volume at 50 µL, bead type as Bio-Plex MagPlex Beads, set double discriminator gates at 8000 and 23000, reporter gain settings set to low RP1 target value for CAL2 setting, 50 counts / region and collect median fluorescence intensity (MFI). Complement analysis in plasma: Blood was collected fresh in tubes containing the anticoagulant hirudin from healthy volunteers using the PharmB ethics. Within 30 minutes of being drawn, the blood was centrifuged at 2500xg for ten minutes. Individual Eppendorf’s were then set up for each donor with 100 μL of PBS, 100 μL of test compound diluted in PBS and 100 μL of fresh plasma. The final concentration of ionisable lipid (C12-200 / 3O6-012B / C24 / MC3) in the formulations tested were: 5, 10 and 20 μg / mL. Untreated control (PBS alone) and two positive controls were set up: Doxil at 200 / 16.42 µg / mL (total lipid content / Doxorubicin content) and Cobra venom factor (CVF) at 1035 units / mL. All the Eppendorf’s were then incubated at 37 °C for 30 minutes. Plasma samples were assayed for iC3b concentrations immediately following the 30 minute incubation. Samples were analysed in duplicate using the MicroVue™ Complement iC3b EIA following the manufacturer’s protocol. Samples were diluted either 1:40 or 1:20 in iC3b specimen diluent for the positive controls and all other samples respectively. Each of the standards and controls were reconstituted with 2 mL of hydrating reagent, left to sit for 15 minutes, and mixed gently. Wash buffer was made up by diluting the wash buffer concentrate with deionised water with a 1:20 dilution. 100 μL of each the blank, standards, controls, or samples was added to the corresponding wells of the 96-well plate in duplicate. OD was read at 405 nm and data analysed by removing the blank and plotting the Concentration (x-axis) against Absorbance (y-axis) and the test sample concentrations interpolated from the curve created using GraphPad Prism software.

Claims

Claims 1. A lipid nanoparticle (LNP) comprising a plurality of lipids, wherein the plurality of lipids comprises (i) a helper lipid, (ii) a sterol and (iii) a cationic lipid and / or an ionisable lipid, and the plurality of lipids does not comprise a PEGylated lipid.

2. The LNP of claim 1, wherein the plurality of lipids consists of (i) the helper lipid, (ii) the sterol and (iii) the cationic lipid and / or the ionisable lipid.

3. The LNP of claim 1 or claim 2, wherein the helper lipid is a phospholipid, optionally wherein the phospholipid is di-oleoyl-phosphatidylethanolamine (DOPE), di- oleoyl-phosphatidylcholine (DOPC), di-oleoyl-phosphatidylserine (DOPS), 1,2- distearoyl-sn-glycero-3-phosphocholine (DSPC), 1,2-distearoyl-sn-glycero-3- phosphoethanolamine (DSPE), a naturally occurring or synthetically derived phospholipid, or a combination thereof.

4. The LNP of any one of the preceding claims, wherein the plurality of lipids comprises between 2 and 80 mol%, between 4 and 70 mol%, between 6 and 60 mol%, between 8 and 55 mol%, between 10 and 50 mol%, between 15 and 45 mol%, between 20 and 40 mol% or between 22 and 38 mol% helper lipid.

5. The LNP of any one of the preceding claims, wherein the sterol is cholesterol, a cholesterol derivative or a combination thereof 6. The LNP of any one of the preceding claims, wherein the plurality of lipids comprises between 5 and 70 mol%, between 10 and 60 mol%, between 14 and 55 mol%, between 16 and 50 mol%, between 18 and 45 mol%, between 20 and 40 mol%, between 22 and 38 mol% or between 24 and 36 mol% sterol.

7. The LNP of any one of the preceding claims, wherein the cationic lipid is DOTAP and / or the ionisable lipid is DLin-MC3-DMA, C12-200, 306-O12B, 306Oi10, C24, cKK- E12, LP-01, PPZ-A10, 4A3-SC8 or a combination thereof.

8. The LNP of any one of the preceding claims, wherein the plurality of lipids comprises between 5 and 80 mol%, between 10 and 70 mol%, between 15 and 60 mol%, between 20 and 55 mol%, between 22 and 52 mol%, between 24 and 50 mol%, between 26 and 48 mol%, between 28 and 46 mol% or between 29 and 45 mol% cationic lipid and / or ionisable lipid.

9. The LNP of any one of the preceding claims, wherein the LNP has a diameter of less than 300 nm, less than 250 nm or less than 200 nm or less than 150 nm, or less than 100 nm.

10. The LNP of any one of the preceding claims, wherein the LNP comprises a payload.

11. The LNP of claim 10, wherein the payload is a biomolecule and / or an active pharmaceutical ingredient (API).

12. The LNP of claim 11, wherein the payload is a biomolecule, and the biomolecule is or comprises an amino acid, a peptide, an affimer, a protein, a glycoprotein, a lipopolysaccharide, an antibody or a fragment thereof, or a nucleic acid.

13. The LNP of claim 12, wherein the biomolecule is a nucleic acid and the nucleic acid is DNA, RNA or a DNA / RNA hybrid sequence.

14. The LNP of any one of claims 10 to 13, wherein the molar ratio of the cationic lipid and / or an ionisable lipid to the payload is between 1:50 and 50:1, between 1:20 and 20:1, between 1:10 and 10:1, between 1:5 and 5:1, between 1:3 and 3:1, between 1:2 and 2:1 or between 1:1.5 and 1.5:

1.

15. The LNP of any one of claims 10 to 14, wherein the N / P ratio of the LNP is at least 14, at least 20 or at least 30.

16. A lipid nanoparticle (LNP) with an N / P ratio of at least 15.

17. A formulation comprising a plurality of LNPs as defined by any one of the preceding claims.

18. The formulation of claim 17, wherein the plurality of LNPs have a polydispersity index (PDI) of less than 0.4, less than 0.35 or less than 0.

3.

19. A pharmaceutical composition comprising the LNP of any one of claims 1 to 16 or the formulation of claim 17 or claim 18 and a pharmaceutically acceptable vehicle.

20. The LNP of any one of claims 1 to 16, the formulation of claim 17 or claim 18 or the pharmaceutical composition of claim 19, for use as a medicament.

21. A vaccine composition comprising the LNP of any one of claims 1 to 16, the formulation of claim 17 or claim 18 or the pharmaceutical composition of claim 19.

22. The LNP of any one of claims 1 to 16, the formulation of claim 17 or claim 18, the pharmaceutical composition of claim 19 or the vaccine of claim 21, for use in stimulating an immune response in a subject.

23. A method of calculating a hydrogen bond potential for a lipid nanoparticle (LNP) formulation, the method comprising: a) selecting three or more different lipids; b) identifying all hydrogen bond acceptors (HBAs) and hydrogen bond donors (HBDs) in each of the different lipids; c) selecting a ratio of the three or more different lipids; d) modelling a physical distribution of the three or more different lipids at the ratio; and f) calculating a hydrogen bond energy based upon interactions between neighbouring molecules in the physical distribution to thereby determine a hydrogen bond potential for the LNP formulation.

24. The method of claim 23, wherein the three or more different lipids comprise (i) a helper lipid, (ii) a sterol and (iii) a cationic lipid and / or an ionisable lipid.

25. The method of claim 23 or claim 24, wherein selecting a ratio of the three or more different lipids comprises selecting a ratio which comprises the sterol at a concentration of at least 5 mol%, at least 10 mol%, at least 15 mol% or at least 20 mol%.

26. The method of any one of claims 23 to 25, wherein modelling a physical distribution of the three or more different lipids comprises: d-i) generating a three-dimensional conformation of at least a portion of each lipid, and optionally optimising the three-dimensional conformation.

27. The method of claim 26, wherein the at least a portion of each lipid is or comprises the head group of each lipid.

28. The method of claim 26 or 27, wherein modelling a physical distribution of the three or more different lipids comprises: d-ii) generating a lipid mixture, wherein the lipid mixture comprises a plurality of three-dimensional conformations of each of the three or more different lipids, such that the plurality of three-dimensional conformations is representative of the selected ratio; and d-iii) randomising the distribution of the plurality of three-dimensional conformations.

29. The method of claim 28, wherein modelling a physical distribution of the three or more different lipids further comprises: d-iv) rotating and / or translating each of the three-dimensional conformations to (a) maximise cross-sectional area thereof in a plane, minimising distance between neighbouring conformations, and / or avoiding overlap between neighbouring conformations.

30. The method of any one of claims 23 to 29, wherein calculating the hydrogen bond energy based upon interactions between neighbouring molecules in the physical distribution comprises: e-i) determining the number of hydrogen bonds between neighbouring molecules in the modelled physical distribution.

31. The method according to claim 30, wherein determining the number of hydrogen bonds between neighbouring molecules in the modelled physical distribution comprises: e-i-a) assigning HBDs and HBAs on neighbouring molecules in the physical distribution into HBD-HBA pairs; and e-i-b) counting a HBD-HBA pair as defining a hydrogen bond if the HBD-HBA pair are less than a predetermined distance apart.

32. The method according to claim 30 or claim 31, wherein the method further comprises: e-ii) calculating a hydrogen bond energy based upon the number of hydrogen bonds between neighbouring molecules in the physical distribution to thereby determine a hydrogen bond potential for the LNP formulation.

33. The method of any one of claims 28 to 32, wherein the method further comprises:f) repeating steps (d-ii) to (e).

34. The method of any one of claims 23 to 33, wherein the method further comprises: g) repeating steps (d) to (e).

35. The method of any one of claims 23 to 34, wherein the method may further comprises: h) selecting a new ratio of the three or more different lipids and then repeating steps (d) to (e) for the new ratio.

Images