Stereoselective pH-responsive peptide dendrimers for nucleic acid transfection

Abstract

The present invention relates to a compound of formula (D 3 )8-(B 3 -D 2 )4-(B 2 -D 1 )2-B 1 -Z(1), where Z is a hydrophobic core and Z is -XY 1 (1a), -XY 2 XY 2 (1b), -XY 3 The present invention relates to a stereoselective peptide dendrimer for nucleic acid transfection, comprising: -Cys (1c) or -HP (1d), X is selected from Lys or Glu, each Y is a fatty acid or an amino acid, HP is a hydrophobic peptide consisting of 3 to 5 hydrophobic amino acids, B is Lys, and each D, independently of other Ds, is a dipeptide consisting of one hydrophobic amino acid and one cationic amino acid (HC or CH), a dipeptide consisting of two cationic amino acids (CC), or a dipeptide consisting of two hydrophobic amino acids (HH). The present invention further includes a method for transfecting cells (ex vivo) using the above dendrimers.

Description

[Technical field]

[0001] Transfection of cells with nucleic acids allows the targeted expression of foreign genes or the knockdown of selected proteins and is an essential procedure in biological and biomedical research that is also envisaged for gene therapy applications. [Background technology]

[0002] Due to safety concerns regarding biological transfection agents such as viruses, a wide variety of non-viral transfection reagents have been developed. These reagents are amphiphilic polycations consisting of liposome-forming lipids (including Lipofectamine L2000, one of the best performing reagents used herein as a reference), polymers, modified poly(amidoamines), carbosilane or bora amphiphilic dendrimers, nanoparticles, protein-dendrimer conjugates, protein cages, or peptides. They complex nucleic acids to form aggregates with sizes of 50-150 nm, which enter cells by endocytosis, escape endosomes, and deliver their cargo to the cytosol (in the case of siRNA) or nucleus (in the case of DNA).

[0003] Unfortunately, almost all of the above reagents have undisclosed or heterogeneous structures or require complex multi-step syntheses using non-standard reagents, limiting the reproducibility and transferability of the results. Known peptide dendrimers have been used in conjunction with lipofectin as a helper lipid, but these dendrimers were inactive when used alone. Summary of the Invention

[0004] Terms and Definitions The term "amino acid" relates to naturally occurring amino acids, particularly proteinogenic amino acids, and to non-naturally occurring amino acids. The amino acids may be in the D- or L-configuration.

[0005] Based on the chemical properties of their side chains, amino acids are classified, for example, as hydrophobic or cationic.

[0006] A hydrophobic amino acid in the context of this specification is any α-amino carboxylic acid with a side chain that does not contain a hydrogen bond donor or acceptor. Hydrophobic amino acids include, but are not limited to, leucine (Leu), phenylalanine (Phe), tryptophan (Trp), cysteine ​​(Cys), norleucine (Nle) and aminooctanoic acid (Aoc). Cysteine ​​is classified as hydrophobic according to Nagano N, Ota M, Nishikawa K (1999) "Strong hydrophobic nature of cysteine ​​residues in proteins". FEBS Lett. 458(1):69-71.

[0007] In the context of this specification, an amino acid containing a cationic side chain is an α-amino carboxylic acid having a side chain that contains a chemical functional group that exists as a cation under physiological pH. Cationic amino acids include, but are not limited to, arginine (Arg), histidine (His) and lysine (Lys).

[0008] The amino acid sequences are presented from the N-terminus to the C-terminus. The terminal carboxy groups of the peptide dendrimers referred to herein are carboxylic acids or carboxylates (-COO - The C-terminus can be a carboxamide (-CONH 2 ) group.

[0009] Amino acids are expressed in three-letter code (Stryer, Biochemistry, 3 rd ed. p.21) or one-letter symbols. Unless otherwise indicated, amino acids may be in the D or L configuration. Particularly in the experimental section, capitalized one-letter symbols refer to L-amino acid enantiomers or diastereomers, and lowercase one-letter symbols refer to D-amino acid enantiomers or diastereomers.

[0010] explanation The present invention relates to a peptide dendrimer of formula 1, (D 3 ) 8 -(B 3 -D 2 ) 4 -(B 2 -D 1 ) 2 -B 1 -Z(1) Z is -X(Y 1 )(1a), -X(Y 2 )X(Y 2 )(1b), -X(Y 3 ) Cys(1c) or -HP(1d), wherein -X is selected from Lys, Orn, DAB (2,4 diaminobutyric acid), DAP (2,3 diaminopropionic acid), Glu or Asp, which is linked via its N-terminus to B 1 and is connected to Y through its side chain, -Y 1 When X is Lys, Orn, DAB or DAP, -C(=O)-C v -Alkyl, -C(=O)-C v -alkenyl, v is selected from - when X is Lys, 17 to 27, in particular 19 to 27, more particularly 21 to 25; - when X is Orn, 18 to 28, in particular 20 to 28, more particularly 22 to 26; - when X is DAB, 19 to 29, in particular 21 to 29, more particularly 23 to 27; - when X is DAP, it is 20 to 30, in particular 22 to 30, more particularly 24 to 28; or Y 1 When X is Glu or Asp, -(NH)-C a -Alkyl, -(NH)-C a -Alkenyl, -(N)-(C d -alkyl) 2 Or -(N)-(C d -alkenyl) 2 is selected from the group consisting of - the sum of d is 19 to 29, particularly 21 to 29, more particularly 23 to 27, when X is Glu; - the sum of d is 20 to 30, particularly 22 to 30, more particularly 24 to 28, when X is Asp; -a is 19 to 29, particularly 21 to 29, more particularly 23 to 27, when X is Glu; or -a is 20 to 30, particularly 22 to 30, more particularly 24 to 28 when X is Asp; -Each Y 2 When X is Lys, Orn, DAB or DAP, -C(=O)-C w -Alkyl and -C(=O)-C w -alkenyl, and the sum of w is independently selected from - when X is Lys, 18 to 36, in particular 22 to 36, more particularly 28 to 36; - when X is Orn, 20 to 38, in particular 24 to 38, more particularly 30 to 38; - 22 to 40, in particular 26 to 40, more particularly 32 to 40, when X is DAB; - when X is DAP, it is 24 to 42, in particular 28 to 42, more particularly 34 to 42; or Y 2 When X is Glu or Asp, -(NH)-C b -Alkyl, -(NH)-C b -Alkenyl, -(N)-(C e -alkyl) 2 Or -(N)-(C e -alkenyl) 2 is selected from the group consisting of the sum of e and b is 22 to 40, particularly 26 to 40, more particularly 32 to 40, when X is Glu; the sum of e and b is 24 to 42, particularly 28 to 42, more particularly 34 to 42, when X is Asp; -Y 3 When X is Lys, Orn, DAB or DAP, -C(=O)-C x -Alkyl and -C(=O)-C x-alkenyl, x is selected from - When X is Lys, 15-21, especially 15-17, - 16-22, especially 16-18, when X is Orn; - 17-23, especially 17-19, when X is DAB; - if X is DAP, it is 18-24, especially 18-20, and Y 3 When X is Glu or Asp, -(NH)-C c -Alkyl, -(NH)-C c -Alkenyl, -(N)-(C f -alkyl) 2 Or -(N)-(C f -alkenyl) 2 is selected from the group consisting of - the sum of f is 17 to 23, especially 17 to 19, when X is Glu; - the sum of f is 18 to 24, in particular 18 to 20, when X is Asp; -c is 17 to 23, in particular 17 to 19, when X is Glu; - c is 18 to 24, in particular 18 to 20, when X is Asp; -HP is a hydrophobic peptide consisting of 3 to 5 hydrophobic amino acids, - each B is Lys, - each D, independently of the other D, is a dipeptide consisting of one hydrophobic and one cationic amino acid (HC or CH), a dipeptide consisting of two cationic amino acids (CC), or a dipeptide consisting of two hydrophobic amino acids (HH); This relates to peptide dendrimers.

[0011] In certain embodiments, the present invention provides a peptide dendrimer of formula 1, (D 3 ) 8 -(B 3 -D 2 ) 4 -(B 2 -D 1 ) 2 -B 1 -Z(1) Z is -X(Y 1 )(1a), -X(Y 2 )X(Y 2 )(1b), -X(Y 3 )Cys(1c), -HP(1d), or -X(Y 4 ) Ala, and in particular Z is -X(Y 1 )(1a), -X(Y 2 )X(Y 2 )(1b), -X(Y 3 ) Cys(1c) or -HP(1d), During the ceremony, -X is selected from Lys, Orn, DAB, DAP, Glu or Asp, which is linked via its N-terminus to B 1 and is connected to Y through its side chain, -Y 1 When X is Lys, Orn, DAB or DAP, -C(=O)-C v -Alkyl, -C(=O)-C v -alkenyl, v is selected from - when X is Lys, 15 to 27, in particular 17 to 27, more particularly 19 to 27, even more particularly 21 to 25; - when X is Orn, 18 to 28, in particular 20 to 28, more particularly 22 to 26; - when X is DAB, 19 to 29, in particular 21 to 29, more particularly 23 to 27; - when X is DAP, it is 20 to 30, in particular 22 to 30, more particularly 24 to 28; or Y 1 When X is Glu or Asp, -(NH)-C a -Alkyl, -(NH)-C a -Alkenyl, -(N)-(C d -alkyl) 2 Or -(N)-(C d -alkenyl) 2 is selected from the group consisting of - the sum of d is 19 to 29, particularly 21 to 29, more particularly 23 to 27, when X is Glu; - the sum of d is 20 to 30, particularly 22 to 30, more particularly 24 to 28, when X is Asp; -a is 19 to 29, particularly 21 to 29, more particularly 23 to 27, when X is Glu; or -a is 20 to 30, particularly 22 to 30, more particularly 24 to 28 when X is Asp; -Each Y 2 When X is Lys, Orn, DAB or DAP, -C(=O)-C w -Alkyl and -C(=O)-C w -alkenyl, and the sum of w is independently selected from - when X is Lys, 18 to 36, in particular 22 to 36, more particularly 28 to 36; - when X is Orn, 20 to 38, in particular 24 to 38, more particularly 30 to 38; - 22 to 40, in particular 26 to 40, more particularly 32 to 40, when X is DAB; - when X is DAP, it is 24 to 42, in particular 28 to 42, more particularly 34 to 42; or Y 2 When X is Glu or Asp, -(NH)-C b -Alkyl, -(NH)-C b -Alkenyl, -(N)-(C e -alkyl) 2 Or -(N)-(C e -alkenyl) 2 is selected from the group consisting of the sum of e and b is 22 to 40, particularly 26 to 40, more particularly 32 to 40, when X is Glu; the sum of e and b is 24 to 42, particularly 28 to 42, more particularly 34 to 42, when X is Asp; -Y 3 and Y 4 When X is Lys, Orn, DAB or DAP, -C(=O)-C x -Alkyl and -C(=O)-C x -alkenyl, and x is independently selected from - When X is Lys, 15-21, especially 15-17, - 16-22, especially 16-18, when X is Orn; - 17-23, especially 17-19, when X is DAB; - if X is DAP, it is 18-24, especially 18-20, and Y 3 When X is Glu or Asp, -(NH)-C c -Alkyl, -(NH)-C c -Alkenyl, -(N)-(C f -alkyl) 2 Or -(N)-(C f -alkenyl) 2 is selected from the group consisting of - the sum of f is 17 to 23, especially 17 to 19, when X is Glu; - the sum of f is 18 to 24, in particular 18 to 20, when X is Asp; -c is 17 to 23, in particular 17 to 19, when X is Glu; - c is 18 to 24, in particular 18 to 20, when X is Asp; -HP is a hydrophobic peptide consisting of 3 to 5 hydrophobic amino acids, - each B is Lys, - each D, independently of the other D, is a dipeptide consisting of one hydrophobic and one cationic amino acid (HC or CH), a dipeptide consisting of two cationic amino acids (CC), or a dipeptide consisting of two hydrophobic amino acids (HH); This relates to peptide dendrimers.

[0012] In certain embodiments, the present invention provides a peptide dendrimer of formula 1, (D 3 ) 8 -(B 3 -D 2 ) 4 -(B 2 -D 1 ) 2 -B 1 -Z(1) Z is -X(Y 1)(1a), -X(Y 2 )X(Y 2 )(1b), -X(Y 3 ) Cys(1c) or -HP(1d), During the ceremony, -X is selected from Lys, Orn, DAB, DAP, Glu or Asp, which is linked via its N-terminus to B 1 and is connected to Y through its side chain, -Y 1 When X is Lys, Orn, DAB or DAP, -C(=O)-C v -Alkyl, -C(=O)-C v -alkenyl, v is selected from - when X is Lys, 15 to 27, in particular 17 to 27, more particularly 19 to 27, even more particularly 21 to 25; - when X is Orn, 18 to 28, in particular 20 to 28, more particularly 22 to 26; - when X is DAB, 19 to 29, in particular 21 to 29, more particularly 23 to 27; - when X is DAP, it is 20 to 30, in particular 22 to 30, more particularly 24 to 28; or Y 1 When X is Glu or Asp, -(NH)-C a -Alkyl, -(NH)-C a -Alkenyl, -(N)-(C d -alkyl) 2 Or -(N)-(C d -alkenyl) 2 is selected from the group consisting of - the sum of d is 19 to 29, particularly 21 to 29, more particularly 23 to 27, when X is Glu; - the sum of d is 20 to 30, particularly 22 to 30, more particularly 24 to 28, when X is Asp; -a is 19 to 29, particularly 21 to 29, more particularly 23 to 27, when X is Glu; or -a is 20 to 30, particularly 22 to 30, more particularly 24 to 28 when X is Asp; -Each Y 2 When X is Lys, Orn, DAB or DAP, -C(=O)-C w -Alkyl and -C(=O)-C w -alkenyl, and the sum of w is independently selected from - when X is Lys, 18 to 36, in particular 22 to 36, more particularly 28 to 36; - when X is Orn, 20 to 38, in particular 24 to 38, more particularly 30 to 38; - 22 to 40, in particular 26 to 40, more particularly 32 to 40, when X is DAB; - when X is DAP, it is 24 to 42, in particular 28 to 42, more particularly 34 to 42; or Y 2 When X is Glu or Asp, -(NH)-C b -Alkyl, -(NH)-C b -Alkenyl, -(N)-(C e -alkyl) 2 Or -(N)-(C e -alkenyl) 2 is selected from the group consisting of the sum of e and b is 22 to 40, particularly 26 to 40, more particularly 32 to 40, when X is Glu; the sum of e and b is 24 to 42, particularly 28 to 42, more particularly 34 to 42, when X is Asp; -Y 3 and Y 4 When X is Lys, Orn, DAB or DAP, -C(=O)-C x -Alkyl and -C(=O)-C x -alkenyl, and x is independently selected from - When X is Lys, 15-21, especially 15-17, - 16-22, especially 16-18, when X is Orn; - 17-23, especially 17-19, when X is DAB; - if X is DAP, it is 18-24, especially 18-20, and Y 3 When X is Glu or Asp, -(NH)-Cc -Alkyl, -(NH)-C c -Alkenyl, -(N)-(C f -alkyl) 2 Or -(N)-(C f -alkenyl) 2 is selected from the group consisting of - the sum of f is 17 to 23, especially 17 to 19, when X is Glu; - the sum of f is 18 to 24, in particular 18 to 20, when X is Asp; -c is 17 to 23, in particular 17 to 19, when X is Glu; - c is 18 to 24, in particular 18 to 20, when X is Asp; -HP is a hydrophobic peptide consisting of 3 to 5 hydrophobic amino acids, - each B is Lys, - each D, independently of the other D, is a dipeptide consisting of one hydrophobic and one cationic amino acid (HC or CH), a dipeptide consisting of two cationic amino acids (CC), or a dipeptide consisting of two hydrophobic amino acids (HH); This relates to peptide dendrimers.

[0013] Peptide dendrimers are suitable for transfection of cells, particularly mammalian cells.

[0014] Peptide dendrimers are easily obtained as pure products in a single solid-phase peptide synthesis (SPPS) operation followed by purification by HPLC. Peptide dendrimers are an excellent alternative to linear peptides as well-defined and easily transferable transfection reagents.

[0015] Dendrimer self-aggregation through intermolecular β-sheet formation at neutral pH allows complexation of nucleic acids to form nanoparticles that enter cells by endocytosis. Acidification of the endosome induces protonation of the amino termini and rearrangement to an α-helical conformation, forming smaller dendrimer / nucleic acid nanoparticles that escape the endosome and release their nucleic acid cargo in the cytosol of cells.

[0016] Beta-sheet formation is promoted by a hydrophobic core. Thus, Z is either a hydrophobic moiety comprising an acylated amino acid, or a hydrophobic peptide consisting of at least three or more hydrophobic amino acids.

[0017] While the core is important for self-aggregation, the branched structure of the peptide dendrimer is important for complexation of nucleic acid molecules. Several nucleic acid molecules are complexed by several self-aggregates forming nanoparticles.

[0018] The branched structure of the peptide dendrimer is connected to Z via its C-terminus and to one dipeptide D via its N-terminus. 1 and via its side chain, another dipeptide D 1 A first trifunctional branching unit B bonded to 1 , starting with lysine. Each dipeptide D 1 is a further branching unit B 2 The core unit Z is linked to G 0 It may also be called generation. Dipeptide D 1 The first branching unit B bound to 1 is the first generation G 1 forming dipeptide D 2 A second branching unit B bound to 2 is the second generation G 2 forming dipeptide D 3 A third branching unit B bound to 3 is the third generation G 3 Form.

[0019] The core unit Z is a hydrophobic moiety comprising the formulae 1a-1d. The moiety X is lysine or glutamic acid and is linked via its N-terminus to B 1 and via its C-terminus to a further moiety, X or cysteine. The side chain of X is acylated, for example by reaction of a carboxylate group of a fatty acid with an amino group of the side chain, or by reaction of a fatty amine with a carboxylate group of the side chain.

[0020] The fatty acids or fatty amines may be saturated or unsaturated and straight-chain or branched, especially saturated and straight-chain.

[0021] Alternatively, the core unit Z may be formed by three or more hydrophobic amino acids (hydrophobic peptides). The hydrophobic peptides may contain different hydrophobic amino acids, for example Leu-Phe-Nle, or identical amino acids, for example Leu-Leu-Leu.

[0022] In certain embodiments, X is Lys, which is a peptide dendrimer of formula 1, (D 3 ) 8 -(B 3 -D 2 ) 4 -(B 2 -D 1 ) 2 -B 1 -Z(1) Z is -X(Y 1 )(1a), -X(Y 2 )X(Y 2 )(1b), -X(Y 3 )Cys(1c), -HP(1d), or -X(Y 4 ) Ala, and in particular Z is -X(Y 1 )(1a), -X(Y 2 )X(Y 2 )(1b), -X(Y 3 ) Cys(1c) or -HP(1d), wherein -X is Lys and is linked via its N-terminus to B 1 and is connected to Y through its side chain, -Y 1is -C(=O)-C v -Alkyl, -C(=O)-C v -alkenyl, where v is 15 to 27, in particular 17 to 27, more particularly 19 to 27, even more particularly 21 to 25; -Each Y 2 is -C(=O)-C w -Alkyl and -C(=O)-C w -alkenyl, and the sum of w is 18 to 36, particularly 22 to 36, more particularly 28 to 36; -Y 3 and Y 4 is -C(=O)-C x -Alkyl and -C(=O)-C x -alkenyl, where x is 15 to 21, in particular 15 to 17; -HP is a hydrophobic peptide consisting of 3 to 5 hydrophobic amino acids, - each B is Lys, - each D, independently of the other D, is a dipeptide consisting of one hydrophobic and one cationic amino acid (HC or CH), a dipeptide consisting of two cationic amino acids (CC), or a dipeptide consisting of two hydrophobic amino acids (HH); Peptide dendrimers are provided.

[0023] In certain embodiments, X is Lys, which is a peptide dendrimer of formula 1, (D 3 ) 8 -(B 3 -D 2 ) 4 -(B 2 -D 1 ) 2 -B 1 -Z(1) Z is -X(Y 1 )(1a), -X(Y 2 )X(Y 2 )(1b), -X(Y 3 ) Cys(1c) or -HP(1d), wherein -X is Lys and is linked via its N-terminus to B1 and is connected to Y through its side chain, -Y 1 is -C(=O)-C v -Alkyl, -C(=O)-C v -alkenyl, where v is 17 to 27, particularly 19 to 27, more particularly 21 to 25; -Each Y 2 is -C(=O)-C w -Alkyl and -C(=O)-C w -alkenyl, and the sum of w is 18 to 36, particularly 22 to 36, more particularly 28 to 36; -Y 3 is -C(=O)-C x -Alkyl and -C(=O)-C x -alkenyl, where x is 15 to 21, in particular 15 to 17; -HP is a hydrophobic peptide consisting of 3 to 5 hydrophobic amino acids, - each B is Lys, - each D, independently of the other D, is a dipeptide consisting of one hydrophobic and one cationic amino acid (HC or CH), a dipeptide consisting of two cationic amino acids (CC), or a dipeptide consisting of two hydrophobic amino acids (HH); Peptide dendrimers are provided.

[0024] High transfection efficiency can be achieved when using peptide dendrimers bearing lipid attachment moieties Z, i.e. peptide dendrimers in which Z comprises one or two amino acids X linked via their side chains to a fatty acid or fatty amine Y. In particular, the C-terminus of the C-terminal amino acid X is a carboxamide (-C(=O)-NH 2 ).

[0025] In certain embodiments, the peptide dendrimer has the formula 1a (D 3 ) 8 -(B 3 -D 2 ) 4 -(B2 -D 1 ) 2 -B 1 -X(Y 1 )(1a), where, -X is selected from Lys or Glu, in particular X is Lys, which is linked via its N-terminus to B 1 and is connected to Y through its side chain, - If X is Lys, then Y 1 is -C(=O)-C v -Alkyl, -C(=O)-C v -alkenyl; when X is Glu, Y 1 -(NH)-C a -Alkyl, -(NH)-C a -Alkenyl, -(N)-(C d -alkyl) 2 Or -(N)-(C d -alkyl) 2 wherein v is 15 to 27, a is 19 to 29, in particular 21 to 29, more particularly 23 to 27, and in particular Y 1 is -C(=O)-C v -Alkyl, -C(=O)-C v -alkenyl, where v is 17 to 27, particularly 19 to 27, more particularly 21 to 25; - each B is Lys, - each D, independently of the other D, is a dipeptide consisting of one hydrophobic and one cationic amino acid (HC or CH), a dipeptide consisting of two cationic amino acids (CC), or a dipeptide consisting of two hydrophobic amino acids (HH); It is a peptide dendrimer.

[0026] In certain embodiments, the peptide dendrimer has the formula 1a (D 3 ) 8 -(B 3 -D 2 ) 4 -(B 2 -D 1 ) 2 -B 1 -X(Y1 )(1a), where, -X is selected from Lys or Glu, in particular X is Lys, which is linked via its N-terminus to B 1 and is connected to Y through its side chain, - If X is Lys, then Y 1 is -C(=O)-C v -Alkyl, -C(=O)-C v -alkenyl; when X is Glu, Y 1 -(NH)-C a -Alkyl, -(NH)-C a -Alkenyl, -(N)-(C d -alkyl) 2 Or -(N)-(C d -alkyl) 2 wherein v is 17 to 27, in particular 19 to 27, more particularly 21 to 25; the sum of d is 19 to 29, in particular 21 to 29, more particularly 23 to 27; a is 19 to 29, in particular 21 to 29, more particularly 23 to 27; and especially Y 1 is -C(=O)-C v -Alkyl, -C(=O)-C v -alkenyl, where v is 17 to 27, particularly 19 to 27, more particularly 21 to 25; - each B is Lys, - each D, independently of the other D, is a dipeptide consisting of one hydrophobic and one cationic amino acid (HC or CH), a dipeptide consisting of two cationic amino acids (CC), or a dipeptide consisting of two hydrophobic amino acids (HH); It is a peptide dendrimer.

[0027] In certain embodiments, the peptide dendrimer has the formula 1b (D 3 ) 8 -(B 3 -D 2 ) 4 -(B 2 -D 1 ) 2 -B 1 -X(Y2 )X(Y 2 )(1b), where, -X is selected from Lys or Glu, in particular X is Lys, which is linked via its N-terminus to B 1 and is connected to Y through its side chain, - if X is Lys, then each Y 2 is -C(=O)-C w -Alkyl and -C(=O)-C w -alkenyl; and when X is Glu, Y 2 -(NH)-C b -Alkyl, -(NH)-C b -Alkenyl, -(N)-(C e -alkyl) 2 Or -(N)-(C e -alkyl) 2 wherein the sum of w is from 18 to 36, particularly from 22 to 36, more particularly from 28 to 36; the sum of e is from 22 to 40, particularly from 26 to 40, more particularly from 32 to 40; the sum of b is from 22 to 40, particularly from 26 to 40, more particularly from 32 to 40; and in particular each Y 2 is -C(=O)-C w -Alkyl and -C(=O)-C w -alkenyl, where the sum of w is 18 to 36, particularly 22 to 36, more particularly 28 to 36; - each B is Lys, - each D, independently of the other D, is a dipeptide consisting of one hydrophobic and one cationic amino acid (HC or CH), a dipeptide consisting of two cationic amino acids (CC), or a dipeptide consisting of two hydrophobic amino acids (HH); It is a peptide dendrimer.

[0028] In certain embodiments, one moiety Y 2 The number of carbon atoms in the other part Y 2 is equal to the number of carbon atoms in

[0029] In particular, DNA transfection efficiency can be further increased by using peptide dendrimers containing a cysteine ​​in the hydrophobic moiety Z. For example, Z can consist of an amino acid X (e.g. Lys or Glu) linked via its side chain to a fatty acid or fatty amine Y and linked via its C-terminus to a cysteine. The C-terminus of the cysteine ​​can in particular be a carboxamide (-C(=O)-NH 2 ).

[0030] In certain embodiments, the peptide dendrimer has the formula 1c (D 3 ) 8 -(B 3 -D 2 ) 4 -(B 2 -D 1 ) 2 -B 1 -X(Y 3 )Cys(1c), where: -X is selected from Lys or Glu, in particular X is Lys, which is linked via its N-terminus to B 1 and is connected to Y through its side chain, - If X is Lys, then Y 3 is -C(=O)-C x -Alkyl and -C(=O)-C x -alkenyl; when X is Glu, Y 3 -(NH)-C c -Alkyl, -(NH)-C c -Alkenyl, -(N)-(C f -alkyl) 2 Or -(N)-(C f -alkyl) 2 wherein x is 15 to 21, in particular 15 to 17; the sum of f is 17 to 23, in particular 17 to 19; c is 17 to 23, in particular 17 to 19; and especially Y 3 is -C(=O)-C x -Alkyl and -C(=O)-C x -alkenyl, where x is 15 to 21, in particular 15 to 17; - each B is Lys, - each D, independently of the other D, is a dipeptide consisting of one hydrophobic and one cationic amino acid (HC or CH), a dipeptide consisting of two cationic amino acids (CC), or a dipeptide consisting of two hydrophobic amino acids (HH); It is a peptide dendrimer.

[0031] Peptide dendrimers containing hydrophobic peptides as moieties Z allow for straightforward synthesis since they require only amino acids as building blocks. Moreover, such peptide dendrimers are more biocompatible than peptide dendrimers containing lipidated moieties Z.

[0032] In certain embodiments, the peptide dendrimer has formula 1d: (D 3 ) 8 -(B 3 -D 2 ) 4 -(B 2 -D 1 ) 2 -B 1 -HP(1d), where: -HP is a hydrophobic peptide consisting of 3 to 5 hydrophobic amino acids, - each B is Lys, - each D, independently of the other D, is a dipeptide consisting of one hydrophobic and one cationic amino acid (HC or CH), a dipeptide consisting of two cationic amino acids (CC), or a dipeptide consisting of two hydrophobic amino acids (HH); It is a peptide dendrimer.

[0033] In certain embodiments, the hydrophobic peptide consists of hydrophobic amino acids independently selected from Cys, Leu, Trp, Phe, Nle (norleucine), and Aoc (aminooctanoic acid).

[0034] In certain embodiments, the hydrophobic peptide consists of hydrophobic amino acids independently selected from Leu, Phe, Trp, Nle, and Aoc.

[0035] In certain embodiments, the hydrophobic peptides consist of identical hydrophobic amino acids.

[0036] The C-terminus of the peptide dendrimer may be a carboxamide instead of a carboxylate. The carboxamide further increases the hydrophobicity of the moiety Z, while the negatively charged carboxylate decreases the hydrophobicity of the core. As mentioned above, transfection requires self-assembly of the peptide dendrimer, which is mainly mediated by the hydrophobic core Z. The C-terminal carboxamide can be obtained by using Rink amide resin during the synthesis of the peptide dendrimer.

[0037] In certain embodiments, the C-terminus of the peptide dendrimer is a carboxamide.

[0038] As mentioned above, the branching of the dendrimer mediates nanoparticle formation with other self-aggregating peptide dendrimers and nucleic acid molecules. For this, the ratio of cationic to hydrophobic amino acids is important.

[0039] In certain embodiments, each D, independently of the other D, is selected from a dipeptide consisting of one hydrophobic and one cationic amino acid (HC or CH), a dipeptide consisting of two cationic amino acids (CC), or a dipeptide consisting of two hydrophobic amino acids (HH).

[0040] The branched portions of peptide dendrimers release nucleic acids in a pH-dependent manner, and His in particular may contribute to a proton sponge effect that aids in the release of nucleic acids.

[0041] In a particular embodiment, the dipeptide HH is selected from Leu and Nle, in particular Leu.

[0042] In a particular embodiment, the dipeptide HC or CH, in particular CH, is selected from Leu and Nle, in particular Leu, and H and C are selected from Lys, Arg and His, in particular Lys.

[0043] In certain embodiments, the dipeptide CC is selected from Lys, Arg and His, with at least one C being His.

[0044] In certain embodiments, D 1 is a dipeptide HH, where H is selected from Leu and Nle, in particular Leu.

[0045] In certain embodiments, D 1 is a dipeptide HC or CH, in particular CH, H is selected from Leu and Nle, in particular Leu, and C is Lys.

[0046] In certain embodiments, D 2 is the dipeptide HC or CH, in particular CH, H is Leu and C is selected from Lys, Arg and His, in particular Lys or Arg, more particularly Lys.

[0047] In certain embodiments, D 2 is a dipeptide CC, where C is selected from Lys, Arg and His, in particular Lys, and at least one C is His.

[0048] In certain embodiments, D 3 is the dipeptide HC or CH, in particular CH, H is Leu and C is selected from Lys, Arg and His, in particular Lys or Arg, more particularly Lys.

[0049] In certain embodiments, D 3 is a dipeptide CC, where C is selected from Lys, Arg and His, in particular Lys, and at least one C is His.

[0050] In certain embodiments, each D, independently of the other D, is selected from dipeptides consisting of one hydrophobic amino acid and one cationic amino acid (HC or CH, particularly CH), and dipeptides consisting of two hydrophobic amino acids (HH).

[0051] In certain embodiments, D 1 is a dipeptide (HH) consisting of two hydrophobic amino acids, and D 2 and D. 3 are dipeptides (CH) consisting of one cationic and one hydrophobic amino acid, respectively. These compounds are particularly suitable for siRNA transfection.

[0052] In certain embodiments, each D 1 , D 2 and D. 3 are dipeptides (CH) consisting of one cationic and one hydrophobic amino acid. These compounds are particularly suitable for DNA transfection.

[0053] In certain embodiments, the amino acids are independently selected from (L)-amino acids and (D)-amino acids.

[0054] In certain embodiments, the peptide is homochiral.

[0055] The present invention also provides a method for transfecting a cell (ex vivo), comprising the steps of: a. providing a transfection mixture comprising a peptide dendrimer according to any one of the preceding claims and a nucleic acid, b. contacting the cells with the transfection mixture to obtain transfected cells. The present invention relates to a method comprising the steps of:

[0056] In a particular embodiment, the transfection mixture is prepared by mixing a first solution comprising a peptide dendrimer according to any one of the preceding claims and a second solution comprising a nucleic acid.

[0057] In particular embodiments, the nucleic acid is or RNA, in particular circular DNA (plasmid / vector), linear DNA (cDNA), linear RNA (siRNA, saRNA, miRNA, mRNA, long RNA).

[0058] In certain embodiments, the solvent of the first and / or second solution is water or cell culture medium.

[0059] In a particular embodiment, the final concentration of the nucleic acid in the transfection mixture is between 10 nM and 200 nM, in particular between 20 and 100 nM, more particularly between 40 and 80 nM.

[0060] In a particular embodiment, the final concentration of the peptide dendrimer in the transfection mixture is between 1.5 μg / ml and 30 μg / ml, in particular between 3 μg / ml and 15 μg / ml, more particularly between 6 μg / ml and 12 μg / ml.

[0061] In certain embodiments, the N / P ratio (amine groups of the peptide dendrimer -NH 2 The ratio of the phosphate group of the nucleic acid to the phosphate group of the nucleic acid is 3 to 100, particularly 3 to 50, and more particularly 3 to 10.

[0062] In a particular embodiment, the transfection mixture is incubated, in particular for 10 to 60 minutes, more particularly for 20 to 40 minutes, and even more particularly for 25 to 35 minutes, after mixing the first and second solutions.

[0063] In a particular embodiment, the transfection mixture is incubated at a temperature between 18°C ​​and 30°C, in particular between 20°C and 25°C.

[0064] In a particular embodiment, the final concentration of the nucleic acid in the transfection mixture is between 10 nM and 200 nM, in particular between 20 nM and 100 nM, more particularly between 40 and 80 nM.

[0065] In a particular embodiment, the final concentration of the peptide dendrimer in the transfection mixture is between 1.5 μg / ml and 30 μg / ml, in particular between 3 and 15 μg / mL, more particularly between 6 and 12 μg / mL.

[0066] In a particular embodiment, the transfection mixture comprises fetal bovine serum, in particular up to 10% (v / v) fetal bovine serum.

[0067] In certain embodiments, the cell is a eukaryotic cell, in particular a mammalian cell.

[0068] In certain embodiments, the cells are contacted with the transfection mixture for between 4 and 72 hours.

[0069] In certain embodiments, the cells are incubated in a 5% CO 2 The cells are contacted with the transfection mixture at 37° C. in a humidified atmosphere containing 0.1% CO. [Brief description of the drawings]

[0070] [Figure 1] Figure 1 shows the solid phase peptide synthesis (SPPS) of peptide dendrimers. a. Coupling: 3 eq / coupling site Fmoc amino acid in NMP, 3 eq / coupling site PyBOP and 5 eq / coupling site DIEA, 60 min ON (manual synthesis) or 5 eq / coupling site Fmoc amino acid, 5 eq / coupling site Oxyma and 5 eq / coupling site DIC in NMP or DMF, 2-30 min (microwave synthesis). b. Fmoc deprotection: piperidine / DMF (1:4, v / v), 20 min. c. Cleavage: TFA (95%), TIS (2.5%), HO (2.5%) (non-Cys containing peptides) or TFA (94%), TIS (1%), HO (2.5%), EDT (2.5%) (Cys containing peptides), 300 min. [Diagram 2]Figure 2 shows the solid phase peptide synthesis (SPPS) of peptide dendrimers starting from generation 0 Alloc-protected lysine. a. Alloc deprotection: P(PPh3)4 (0.25 eq / Alloc group), PhSiH3 or (CH3)2NH·BH3 (25 eq / Alloc group), 2×60 min. b. Coupling: fatty acid (5 eq / coupling site), DIC (5 eq / coupling site), Oxyma (5 eq / coupling site) in NMP or DMF, 60 min ON. c. Fmoc deprotection: piperidine / DMF (1:4, v / v), 20 min. d. Cleavage: TFA (95%), TIS (2.5%), HO (2.5%) (non-Cys containing peptide) or TFA (94%), TIS (1%), HO (2.5%), EDT (2.5%) (Cys containing peptide), 300 min. [Diagram 3] Structural formulas and amino acid sequences of siRNA transfection dendrimers DMH13 and DMH18 (single-letter amino acids, branched lysines in italics). [Figure 4]Synthesis of transfection peptide dendrimers. (a) SPPS using orthogonal protection for lysine side chain acylation for the example of peptide dendrimer MH13. (b) High temperature SPPS in a microwave synthesizer using preacylated lysines in the dendrimer core. Conditions: 1a) i) 3 eq / coupling site Fmoc amino acid, 3 eq / coupling site PyBOP, 5 eq / coupling site i-Pr2NEt in NMP, 60 min overnight; ii) piperidine / DMF (1:4, v / v), 20 min. 1b) i) 5 eq / coupling site Fmoc amino acid, 5 eq / coupling site Oxyma, 5 eq / coupling site DIC in DMF, 1-5 min, 75-90 °C; ii) piperidine / DMF (1:4, v / v), 2 min, 75-90 °C. 2) P(PPh3)4 (0.25 equiv / alloc group), (CH3)2NH·BH3 (25 equiv / alloc group), 2 × 60 min. 3) Palmitic acid (5 equiv / coupling site), DIC (5 equiv / coupling site), Oxyma (5 equiv / coupling site) in NMP, 60 min overnight. 4a) Piperidine / DMF (1:4, v / v), 20 min. 4b) Piperidine / DMF (1:4, v / v), 2 min, 75–90 °C. 5) CF3CO2H (95%), i-Pr3SiH (2.5%), H2O (2.5%), 5 h, 25 °C. 6) Reverse phase C18 preparative HPLC, 0–70% CH3CN / H2O / 0.1% TFA gradient in 45 min. [Diagram 5]Knockdown efficiency of peptide dendrimer / siRNA complexes under different conditions in HeLa cells. (a) GAPDH activity in HeLa cells after 4 h transfection with siRNA (20–100 nM) and peptide dendrimer (N / P 10, 0.42–2.1 μM, 3–15 μg / mL) or L2000 (2:1, 0.532–2.66 μg / mL) followed by 48 h incubation in DMEM supplemented with 10% FCS. (b) GAPDH mRNA levels in HeLa cells by RT-PCR after 4 h transfection with siRNA (100 nM) and peptide dendrimer (N / P 10, 2.1 μM, 15 μg / mL) or L2000 (2:1, 2.66 μg / mL) completed by incubation in DMEM supplemented with 10% FCS for up to 24, 36 or 48 h. Results were normalized to the levels of untreated cells and each measurement was normalized to the levels of 18S. (c) GAPDH activity in HeLa cells after 4-72 h transfection with siRNA (100 nM) and peptide dendrimers (N / P 10, 2.1 μM, 15 μg / mL) or L2000 (2:1, 2.66 μg / mL) in the presence of 10% serum, completed by incubation in DMEM supplemented with 10% FCS for up to 72 h. (d) GADPH activity in HeLa cells after 4 h transfection with siRNA (100 nM) and peptide dendrimers (N / P 3-10, 0.63-2.1 μM, 4.5-15 μg / mL) or L2000 (2:1, 2.66 μg / mL) followed by 48 h incubation in DMEM supplemented with 10% FCS. [Figure 6]Knockdown efficiency of peptide dendrimer / siRNA complexes in different cell lines. GADPH activity in (a) HT-1080, (b) PC-3, (c) MCF-7, (d) SH-SY5Y and (e) Caco-2 cell lines after 4 h transfection in OptiMEM with siRNA (20–100 nM) and peptide dendrimer (N / P 10, 0.42–2.1 μM, 3–15 μg / mL) or L2000 (2:1, 0.532–2.66 μg / mL) followed by 48 h incubation in DMEM supplemented with 10% FCS. Results were normalized to parallel transfection with siNC (negative control). All experiments were performed in triplicate in three independent experiments. [Figure 7]Biological mechanisms of dendrimer-facilitated siRNA transfection. (a) Live-cell confocal microscopy of FAM-siRNA (green) in HeLa cells. HeLa cells were incubated with FAM-siRNA (100 nM) and peptide dendrimer (N / P 10, 2.1 μM, 15 μg / mL) or L2000 (2:1, 2.66 μg / mL) in OptiMEM for 4 h. Plasma membranes were stained with Cell Mask Deep Red. White scale bar represents 50 μm (lens x40 / 1.3). (b) Internalization of dendrimer / FAM-siRNA nanoparticles in HeLa cells measured by flow cytometry after 4 h transfection at 4 °C or prior to 1 h incubation with inhibitors. siRNA (100 nM) was combined with peptide dendrimer (N / P 10, 2.1 μM, 15 μg / mL) or L2000 (2:1, 2.66 μg / mL). Error bars represent standard deviation of two independent experiments, normalized to internalization under normal conditions. (c) GAPDH activity in HeLa cells after 4 h transfection with siRNA (100 nM) and peptide dendrimer (N / P 10, 2.1 μM, 15 μg / mL) or L2000 (2:1, 2.66 μg / mL) with 1 h pretreatment with 200 nM bafilomycin A1 (t = 0–1), 24 h incubation after transfection (t = 5–24), or both (t = 0–1 and 5–29), followed by 24 × 48 h incubation in DMEM supplemented with 10% FCS. [Figure 8]Formation of dendrimer / siRNA complexes. (a) Free siRNA assay by intercalation of Quant-it™ microRNA into complexes of siRNA (20 nM) and peptide dendrimer (N / P 1-10, 42-420 nM, 0.3-3 μg / mL) as a function of N / P ratio in Quant-it™ microRNA buffer at pH 7.4. (b) Same as (a) at pH 5 in Quant-it™ microRNA buffer acidified with 10 mM acetate buffer. (c) Fluorescence polarization of FAM-siRNA (20 nM) as a function of N / P ratio of peptide dendrimer (N / P 1-10, 42-420 nM, 0.3-3 μg / mL) at pH 7.4 in DPBS. (d) Same as (c) at pH 5 in DPBS. (e) Displacement of siRNA from complexes formed of siRNA (20 nM) and peptide dendrimer (N / P 10,420 nM, 3 μg / mL) or L2000 (2:1,532 ng / mL) by addition of heparin (0-4 μg / mL) measured by tracking of free siRNA with the Quant-it™ microRNA assay in Quant-it™ microRNA buffer at pH 7.4. Fluorescence normalized to siRNA alone was set as 100%. (f) Same as (e) with heparin concentrations (0-4 μg / mL) measured by FP in DPBS at pH 7.4. All experiments were performed in triplicate. [Figure 9] Nanoparticle formation observed by DLS and TEM. (a,c) Dynamic light scattering and (b,d) Zeta potential of complexes of siRNA (800nM) and peptide dendrimer (N / P 10, 16.8μM, 120μg / mL) or L2000 (2:1, 21.3μg / mL) formed in PB at pH 7.4 and 5. All data shown have a polydispersity index (PDI) <0.5. (e) Transmission electron microscopy of complexes of siRNA (800nM) and peptide dendrimer (N / P 10, 16.8μM, 120μg / mL) or L2000 (2:1, 21.3μg / mL) formed in Milli-Q water deposited on glow-discharged 400-mesh copper grids, dried, and stained with uranyl acetate. White scale bar represents 100nm. [Figure 10] Evidence for nanoparticle formation and aggregation by transfected peptide dendrimers. (a) Equilibrium between dendrimer / Cy3-siRNA nanoparticles and their constituents. (b) Fluorescence of Cy3-siRNA (100 nM) after dialysis through a 100 kDa cut-off membrane in the presence of peptide dendrimers (N / P 1–10, 0.21–2.1 μM, 1.5 × 15 μg / mL) or L2000 (2:1, 2.66 μg / mL) in OptiMEM pH 7.4 or pH 5.0. Fluorescence intensity is shown relative to untreated filtered Cy3-siRNA. (c) Fluorescein leakage assay from phosphatidylcholine lipid vesicles suspended in buffer (31 μM EYPC, 625 μM CF, 10 mM TRIS, 107 mM NaCl, pH 7.4). After 30 s, peptide dendrimer (1–15 μg / mL), L2000 (1–15 μg / mL), or complexes formed with siRNA (100 nM) and peptide dendrimer (N / P 10, 2.1 μM, 15 μg / mL) or L2000 (2:1, 2.66 μg / mL) were added to the lipid vesicle solution. After 270 s, Triton X-100 (0.07%) was added to completely release the fluorescein. It = fluorescence intensity ratio. [Figure 11] CMC and pKa determinations with peptide dendrimers. (a) Critical aggregation concentration measurements of peptide dendrimers and L2000 in PB pH 7.4 performed by serial dilution of compounds starting from 1 mg / mL or 0.1 mg / mL, respectively, added to dry Nile red (final concentration 0.2 μM). (b) Same as (a) in PB pH 5. (c) Peptide dendrimers or L2000 (0.1 mg / mL) in PB at pH 5–7.4 added to dry Nile red (final concentration 0.2 μM). Fluorescence was measured at λex=540 nm and λem=615 nm. RFU=relative fluorescence units. (d) pH titration of 0.7–1 μmol (5–7 mg in 7–10 ml Milli-Q water, 100 μM) peptide dendrimer with 2 μL steps of 0.1 M NaOH. [Figure 12]Conformation of transfected peptide dendrimers in solutions at pH 7.4 and pH 5.0. Circular dichroism spectra of MH13, DMH13, MH18, DMH18, MH46 and MH47 (200 μg / mL) in 8 mM phosphate buffer (PB) at pH 7.4 (a, b, c) or pH 5 (g, h, i) with various concentrations of trifluoroethanol. Percentage of α-helix and β-sheet as a function of the concentration of trifluoroethanol at pH 7.4 (d, e, f) and pH 5 (j, k, l) processed by Dichroweb using the CONTIN analysis program and criteria set 3. [Figure 13] Molecular dynamics study at pH 5 in the presence of 20% TFE. (a) Radius of gyration over the course of the simulation. (b) Ramachandran number R analysis of the internal secondary structure. Residues with α-helical conformation are shown in red (R ≈ 0.34), those with β-sheet conformation are shown in blue (R ≈ 0.52), and those with loop conformation are shown in cyan (R ≈ 0.62). The position of each residue according to its evolution is shown using the color code from the previous figure. Arrows represent the contiguous β-peptide moieties of the dendrimer. (c) Total backbone hydrogen bonds over the course of the simulation. Representative equilibrium structures obtained by clustering the last 100 ns of MD trajectories of peptide dendrimers (d) MH18 and (e) MH13, colored by residue type (blue = lysine, brown = leucine, grey = branched lysine, yellow = lipid chain).

[0071] General Procedure Solid-phase synthesis of peptide dendrimers Fmoc-SPPS is a widely used method for the synthesis of any kind of peptides with building blocks based on either natural, unnatural, L- or D-amino acids (these latter are denoted by upper and lower case letters, respectively). Peptide dendrimer synthesis differs in the use of branch points consisting of diamino acids such as lysine with amide bond linkages throughout the dendrimer structure (Figure 1). The amino acids are coupled under standard peptide coupling conditions and the Fmoc protecting groups orthogonal to the acid-labile side chain protecting groups are removed under mildly basic conditions. Chemical synthesis of peptides starts at the C-terminus and the peptide chain grows towards the N-terminus, whereas biosynthesis occurs in the opposite direction.

[0072] After completion of these sequences, the peptide dendrimers are cleaved from the resin with a TFA-based cleavage cocktail that also removes all acid-labile protecting groups from the side chains, precipitated into ether, purified by preparative RP-HPLC and characterized by MS, analytical LCMS and amino acid analysis.

[0073] Manual Solid-Phase Synthesis of Peptide Dendrimers Peptide dendrimers were synthesized by placing 300 mg of Tentagel S RAM resin (0.22-0.25 mmol / g) in a 10 mL polypropylene syringe equipped as described above. Agitation of the reaction mixture at a given step described below was performed by attaching the closed syringe to a rotating shaft. The resin was allowed to swell in DCM for 60 min. The following conditions were then used:

[0074] Removal of Fmoc protecting group: For each step, the Fmoc protecting group was removed with 8 mL of piperidine / DMF (1:4, v / v) for 2×10 min. After filtration, the resin was washed with NMP (3×6 mL), MeOH (3×6 mL) and DCM (3×6 mL).

[0075] Coupling of Fmoc-protected amino acids: 3 equivalents of Fmoc-protected amino acids per reactive site, 3 equivalents of PyBOP (benzotriazol-1-yl-oxytripyrrolidinophosphonium hexafluorophosphate) and 5 equivalents of DIPEA (N,N-diisopropylethylamine) in 8 mL of NMP / DCM (80:20, v / v) were added to the resin and the reaction was stirred for 60 min. The reaction was carried out depending on the dendrimer generation: 1 h for generation 0, 2 h for generation 1, 3 h for generation 2 and 4 h for generation 3. The resin was then washed with NMP (3 x 6 mL), MeOH (3 x 6 mL) and DCM (3 x 6 mL).

[0076] Solid-phase synthesis of peptide dendrimers using Biotage Initiator+Alstra Peptide dendrimers were also synthesized by Biotage Initiator Alstra using 300 mg of Tentagel S RAM resin (0.22-0.25 mmol / g). Agitation of the reaction mixture at a given step described below was performed by vortexing the vial. The resin was allowed to swell in DMF for 60 min at RT. The following conditions were then used:

[0077] Removal of Fmoc protecting group: At each step, the Fmoc protecting group was removed with 6 mL of piperidine / DMF (1:4, v / v) for 5 min at 75° C. After filtration, the resin was washed with DMF for 8 min.

[0078] Coupling of Fmoc-protected amino acids: 5 equivalents of Fmoc-protected amino acids, 5 equivalents of Oxyma and 5 equivalents of DIC, all at a concentration of 0.5M, were used as coupling reagents in 5 mL of DMF. The reaction was stirred at 75° C. for 8 min. The resin was then washed with DMF for 8 min. The coupling was repeated according to the generation: once for generation 0, twice for generation 1, four times for generation 2 and seven times for generation 3.

[0079] Solid-Phase Synthesis of Peptide Dendrimers with CEM Liberty Blue Peptide dendrimers were synthesized by CEM Liberty Blue using 300 mg of Tentagel S RAM resin (0.22–0.25 mmol / g) (scale 0.10 mmol). Stirring of the reaction mixtures in the given steps described below was performed by adding N 2 This was done by venting. The resin was swelled in DMF / DCM 50:50 for 15 min at RT. The following conditions were then used:

[0080] Removal of Fmoc protecting group: At each step, the Fmoc protecting group was removed with 5 mL of piperidine / DMF (1:4, v / v) for 2 min at 75° C. After filtration, the resin was washed five times with 5 mL of DMF.

[0081] Coupling of Fmoc-protected amino acids: 5 equivalents of Fmoc-protected amino acids, 5 equivalents of Oxyma and 5 equivalents of DIC, all at a concentration of 0.2M, were used as coupling reagents in 4 mL of DMF. The reaction was stirred at 75° C. for 5 min. The resin was then washed 4 times with 4 mL of DMF. The coupling was repeated according to the generation: once for generation 0, twice for generation 1, 4 times for generation 2 and 7 times for generation 3.

[0082] Deprotection and coupling of Lys(Alloc): The resin was dried under reduced pressure and bubbled with nitrogen twice for 5 min in dry DCM (8 ml). A solution of Pd(PPh3)4 (0.1 equiv, 10 mg) in dry DCM (3 mL) and (CH 3 ) 2 NH·BH 3 A solution of (25 equiv., 100 mg) was added to the resin and nitrogen was bubbled through for 1 h. The resin was washed with dry DCM (3×8 mL) and the reaction was repeated once for 2 h. The resin was washed with sodium diethyldithiocarbamate (0.02 M in DMF, 10 ml) for 20 min, as well as with NMP, MeOH and DCM (2×10 ml each). The carboxylic acid was then coupled according to the manual procedure.

[0083] Final Fmoc deprotection: After coupling of the last amino acid by three different methods and optionally after coupling of the carboxylic acid, Fmoc deprotection was carried out with 8 mL piperidine / DMF (1:4, v / v) for 20 min. After filtration, the resin was washed with NMP (3×6 mL), MeOH (3×6 mL) and DCM (3×6 mL).

[0084] Cleavage and purification: Cleavage was performed by centrifugation of the resin with 7 mL of TFA / DODT / TIS / H 2 The peptide solution was precipitated with 40 mL of TBME, centrifuged at 3500 rpm for 10 min (twice), evaporated, and dried under high vacuum for 60 min. The crude product was then purified by elution with 0.1% TFA in H2O (94:2.5:2.5:1, v / v / v / v) for 5 h. 2 O / CH 3 The compounds were dissolved in CN mixtures and purified by preparative RP-HPLC, adding a few drops of MeOH as needed. The crude product fractions were then lyophilized. The yields are given as SPPS total yields. In all cases, the yields are calculated for the corresponding TFA salt.

[0085] A manual SPPS procedure for the synthesis of dendrimer MH13 is shown (Figure 4a), and high-temperature SPPS in a microwave synthesizer using preacylated lysine building blocks for the dendrimer core, which resulted in higher preparative yields, is shown for the D-enantiomer dendrimer DMH13 (Figure 4b).

[0086] All products were purified by preparative reverse-phase HPLC and the purest fractions were isolated as trifluoroacetate salts after lyophilization and adjusted as 1-5 mg / mL stock solutions in water for further use.

[0087] Transfection and engineering procedures siRNA Cell culture, transfection reagents and siRNA: HeLa, HEK-293, CHO, PC-3, HT-1080, SH-SY5Y, Caco-2 and MCF-7 cells (ATCC, Manassas, USA) were maintained in DMEM (Thermo Fisher Scientific) supplemented with 10% fetal calf serum (FCS, Thermo Fisher Scientific, Reinach, CH) at 37°C in a humidified atmosphere of 5% carbon dioxide. The sequence of siRNA targeting GAPDH (AM 4631) for the sense strand is GGUCAUCCAUGACAACUUUdTdT and for the antisense strand is AAAGUUGUCAUGGAUGACCdTdT. For FAM-siRNA (AM4650) or Cy3-siRNA (AM4649) targeting GAPDH, the sequence is identical to 5-carboxyfluorescein or cyanine 3 attached to the 5' end of the sense strand. The sequence of the validated negative control siRNA (siNC, AM4635) for the sense strand is AGUACUGCUUACGAUACGGdTdT, and for the antisense strand is CCGUAUCGUAAGCAGUACUdTdT. The siRNA consists of phosphodiester bonds and contains two nucleotide overhangs (dTdT) at the 3' end (Thermo Fisher Scientific). Lipofectamine® 2000 (L2000) was obtained from Thermo Fisher Scientific and used as a positive control with the transfection protocol according to the manufacturer's instructions.

[0088] siRNA transfection and protein quantification: HeLa, CHO, PC3, SH-SY5Y, Caco-2 and MCF-7 cells were plated at 5 × 10 per well in TPP 96-well plates (Faust Laborbedarf AG, Schaffhausen) the day before transfection. 3Cells were seeded at 5 × 10 per well for HEK-293 and HT-1080 the day before transfection after coating 96-well plates with poly-L-lysine (Sigma Aldrich, Buchs, CH) for 1 h and drying under airflow in a fume hood at RT. 3Cells were always used at passages 3–20 in exponential phase for transfection. siRNA transfection complexes were formed in OptiMEM by mixing siRNA (1–20 pmol, 0.02–0.4 μL from a 50 μM Milli-Q aqueous solution in 6.25 μL OptiMEM) with peptide dendrimer (N / P ratio 1–100, 21–420 pmol, 0.1–3.5 μL from a 1 mg / mL Milli-Q aqueous solution in 6.25 μL OptiMEM) or L2000 (w / w ratio 0.5:1–2:1, 0.0665–0.266 μg, 0.0665–0.266 μL from a 1 mg / mL commercial solution in 6.25 μL OptiMEM) for 30 min at room temperature (12.5 μL, concentration 80–1600 nM). siRNA and 1.68–33.6 μM, i.e., 12–240 μg / mL peptide dendrimer or 5.32–21.28 μg / mL L2000). The complexes were then diluted in OptiMEM or OptiMEM+10% FCS to a final volume of 100 μL per well (final concentrations 10–200 nM siRNA and 0.210–4.2 μM, i.e., 1.5–30 μg / mL peptide dendrimer or 0.665–2.66 μg / mL L2000). After removing the complete medium from the cells, 100 μL of transfection was added to each well. The plates were then incubated at 37 °C in a humidified atmosphere of 5% carbon dioxide for 4–72 h. The transfection complexes were replaced with complete growth medium after transfection. siGAPDH gene knockdown assays were performed 48 or 72 h after transfection by measuring the levels of GAPDH protein according to the manufacturer's instructions with the KDalert GAPDH Assay Kit (Thermo Fisher Scientific, Reinach, CH). Briefly, the medium was removed, cells were lysed with 200 μL of cold lysis buffer, and 2–10 μL of lysis buffer from each well was taken into a new 96-well plate containing 90 μL of master mix and read at λ = 1.0 μl using a Tecan Infinite M1000 Pro plate reader. ex = 560 nm and λ em Fluorescence was measured at 590 nm.

[0089] siRNA transfection in the presence of bafilomycin: HeLa cells were used untreated or treated with bafilomycin A1 (200 nM, Alfa Aesar, Karlsruhe, DE) in DMEM supplemented with 10% FCS for 1 h before transfection in 96-well TPP plates as described above. After transfection, the complexes were removed and replaced with DMEM supplemented with 10% FCS and incubated for 48 h, or replaced with DMEM supplemented with 10% FCS containing 200 nM bafilomycin A1 and incubated for 24 h, then replaced with DMEM supplemented with 10% FCS and incubated for an additional 24 h. HeLa cells were then treated with bafilomycin A1 1 h before transfection, 24 h after transfection, or both. GAPDH protein levels were assessed after a total of 48 h as described above.

[0090] Cellular uptake by flow cytometry: HeLa, CHO and HEK-293 cells were plated at 1 × 10 per well in 24-well TPP plates (Faust Laborbedarf AG, Schaffhausen) 24 h prior to transfection. 5Cells were seeded. Complexes of labeled FAM-siRNA (Thermo Fisher Scientific, Reinach, CH) (12–40 pmol, 0.24–0.8 μL from a 50 μM Milli-Q aqueous solution in 25 μL OptiMEM) and peptide dendrimer (N / P ratio 10, 252–840 pmol, 2.1–7 μL from a 1 mg / mL Milli-Q aqueous solution in 25 μL OptiMEM) or L2000 (w / w ratio 2:1, 0.319–1.064 μg, 0.319–1.064 μL from a 1 mg / mL commercial solution in 25 μL OptiMEM) were formed and incubated at room temperature for 30 min (50 μL, concentrations 240–800 nM siRNA and 5.04–16.8 μM, i.e., 36–120 μg / mL peptide dendrimer or 6.384–21.28 μg / mL L2000). The complexes were then diluted in OptiMEM to a final volume of 400 μL per well (final concentrations of 30–100 nM FAM-siRNA and 0.63–2.1 μM, i.e., 4.5–15 μg / mL peptide dendrimer or 0.798–2.66 μg / mL L2000). After removing the medium, the complexes were added to each well and incubated for 4 h at 37 °C in a humidified atmosphere of 5% carbon dioxide. The medium was then removed, the cells were washed with heparin in OptiMEM (2 mg / ml, 0.5 ml, three times), washed twice with PBS, and the cell membranes were labeled with CellMask Deep Red plasma membrane stain (Thermo Fisher Scientific, Reinach, CH) 0.5X in complete DMEM growth medium (0.25 μL / well in 0.5 mL) for 10 min at 37 °C. Cells were washed with PBS (1.0 mL / well, 3 times) and detached from the wells with 250.0 μL of 0.25% trypsin / EDTA solution at 37° C. Cells were harvested with 750.0 μL of complete DMEM growth medium, collected in a 1.5 mL Eppendorf tube, and centrifuged at 200 RCF for 5 min. Trypsin solution was discarded, the pellet was shaken, and cells were resuspended in 50.0 μL of PBS+1% FCS. The fluorescence intensity of each cell sample was analyzed by a combined flow cytometer-microscope ImageStream X Mark II (Merck Millipore) and processed with IDEAS® software.

[0091] Cellular uptake in the presence of inhibitors and at low temperature: Cells were treated with cytochalasin D (25 μg / mL), nystatin (25 μg / mL) or chlorpromazine (15 μg / mL) in DMEM supplemented with 10% FCS for 1 h, washed with PBS and then transfected in 24-well TPP plates as above. For temperature-dependent experiments, cells were transfected for 4 h under the same conditions as above, but at 4°C. Cells were washed and treated as above, and the fluorescence intensity of each cell sample was analyzed by a combined flow cytometer-microscope ImageStream X Mark II (Merck Millipore) and processed with IDEAS® software.

[0092] Quantitative RT-PCR: Cells were transfected in TPP 96-well plates as described above. After transfection, cells were washed with 50 μl cold PBS and the Cells-to-Ct kit (Thermo Fisher Scientific, Reinach, CH) was used. Briefly, cells were lysed with 49.5 μL lysis buffer + 0.5 μL DNase for 5 min, 5 μL stop solution was added and incubated for 2 min. Then, 2 μl of cell lysate was mixed with 18 μl of Master Mix containing 1 μL TaqMan GAPDH endogenous control (FAM / MGB, non-primer limited, 433764F, Thermo Fisher Scientific), 1 μL TaqMan 18S endogenous control (VIC / MGB, primer limited, 4319413E, Thermo Fisher Scientific), 5 μL TaqMan 1-Step qRT-PCR Mix and 11 μL water. RT-PCR was performed in double strand using GAPDH and 18S primers in the green and yellow channels, respectively, on a Corbett Rotorgene 6000 (Qiagen). Cycles: RT: 50°C / 5 min, RT inactivation: 95°C / 20 sec, amplification: 95°C / 15 sec and 60°C / 1 min for 40 cycles. mRNA levels were calculated by normalizing the Ct value of GAPDH to the Ct value of 18S and expressed as 2-ΔΔCt The results were quantified by the method.

[0093] Cell viability by AlamarBlue® assay: Cells were transfected in TPP 96-well plates as previously described. After transfection, the medium was removed and replaced with 10% AlamarBlue® (Thermo Fisher Scientific, Reinach, CH) in DMEM supplemented with 10% FCS. Cells were incubated for 4-24 hours at 37°C in a humidified atmosphere of 5% carbon dioxide. Plates were then read at λ = 1.0 μm using a Tecan Infinite M1000 Pro plate reader. ex = 560 nm and λ em = 590 nm and values ​​were normalized to those of untreated cells.

[0094] Confocal microscopy: Nunc Lab-Tek II 8-well chambered cover glass plates (Faust Laborbedarf AG, Schaffhausen) were treated with poly-L-lysine (Sigma Aldrich, Buchs, CH) for 1 h, dried at RT and cells were plated at 3 × 10 per well the day before transfection. 4Cells were plated with complexes of labeled FAM-siRNA (12–40 pmol, 0.24–0.8 μL from a 50 μM Milli-Q aqueous solution in 25 μL OptiMEM) and peptide dendrimer (N / P ratio 10, 252–840 pmol, 2.1–7 μL from a 1 mg / mL Milli-Q aqueous solution in 25 μL OptiMEM) or L2000 (w / w ratio 2:1, 0.319–1.064 μg, 0.319–1.064 μL from a 1 mg / mL commercial solution in 25 μL OptiMEM) and incubated at room temperature for 30 min (50 μL, concentrations 240–800 nM siRNA and 5.04–16.8 μM, i.e. 36–15 μg / mL peptide dendrimer or 6.384–21.28 μg / mL L2000). The complexes were then diluted in OptiMEM to a final volume of 400 μL per well (final concentrations of 30–100 nM FAM-siRNA and 0.63–2.1 μM, i.e., 4.5–15 μg / mL peptide dendrimer or 0.798–2.66 μg / mL L2000). The complexes were added to each well and incubated for 4 h at 37 °C in a humidified atmosphere of 5% carbon dioxide after removing the complete growth medium. The medium was then removed, the cells were washed with heparin in OptiMEM (2 mg / ml, 0.5 ml, three times), washed twice with PBS, and the cell membranes were labeled with CellMask Deep Red plasma membrane stain (Thermo Fisher Scientific, Reinach, CH) 0.5X in complete DMEM growth medium (0.25 μL / well in 0.5 mL) for 10 min at 37 °C. Cells were washed with PBS (1.0 mL / well, 3 times), FluoroBrite DMEM (Thermo Fisher Scientific, Reinach, CH) was added, and images were taken with a Zeiss LSM 880 confocal microscope equipped with a ×40 / 1.3 lens.

[0095] Free siRNA assay with Quant-iT™ microRNA: siRNA (10 pmol, 0.2 μL from a 50 μM Milli-Q aqueous solution in 6.25 μL OptiMEM pH 5 or pH 7.4) was complexed in OptiMEM (12.5 μL, concentration 800 nM) by mixing with peptide dendrimer (N / P ratio 1-10, 42-420 pmol, 0.15-1.5 μL from a 1 mg / mL Milli-Q aqueous solution in 6.25 μL OptiMEM pH 5 or pH 7.4) or L2000 (w / w ratio 0.2:1-8:1, 0.0266-1.064 μL from a 1 mg / mL commercial solution in 6.25 μL OptiMEM pH 7.4) for 30 min at room temperature. siRNA and 1.68-16.8 μM, i.e., 12-120 μg / mL peptide dendrimer or 0.133-5.32 μg / mL L2000). The Quant-iT™ MicroRNA Assay Kit (Thermo Fisher Scientific, Reinach, CH) was then used according to the manufacturer's protocol. Briefly, 1 μl of the reagent was diluted in 200 μl of buffer (pH 5 by acidification with 10 mM acetate buffer or standard pH 7.4) and 195 μL was added to a well of a TPP 96-well plate. 5 μl of the complex was then added to the well (200 μL, final concentration of 20 nM siRNA and 42-420 nM, i.e., 0.3-3 μg / mL peptide dendrimer or 0.0532-2.128 μg / mL L2000) and after 10 min, the λ was measured using a Tecan Infinite M1000 Pro plate reader. ex = 500 nm and λ em Fluorescence was measured at 525 nm. Quant-iT™ microRNA signals from the complexes were normalized to the "siRNA only" control to obtain a percentage of detected signal.

[0096] Competitive assay with Quant-iT™ microRNA: Complexes were formed similarly to the free siRNA assay, and 5 μL of complexes were diluted in 190 μL of assay buffer as described in the above section. Then, 5 μL of heparin at different concentrations was added to the wells of a TPP 96-well plate and incubated at room temperature for 30 min (200 μL, final concentrations 0-4 μg / mL heparin, 20 nM siRNA and 420 nM, i.e. 3 μg / mL peptide dendrimer or 532 ng / mL L2000). λ ex = 500 nm and λ em Fluorescence measured at = 525 nm was performed on a Tecan Infinite M1000 Pro plate reader. Quant-iT™ microRNA signals from complexes were normalized to the "siRNA only" control to obtain a percentage of detected signal.

[0097] Free FAM-siRNA assay by fluorescence polarization: FAM-siRNA (2 pmol, 0.04 μL from 50 μM Milli-Q aqueous solution in 6.25 μL DPBS pH 5 or pH 7.4) was complexed in DPBS (Thermofisher scientific, Reinach, CH) by mixing with peptide dendrimer (N / P ratio 1–10, 4.2–42 pmol, 0.03–0.3 μL from 1 mg / mL Milli-Q aqueous solution in 6.25 μL DPBS pH 5 or pH 7.4) or L2000 (w / w ratio 0.2:1–16:1, 0.00532–0.4256 μg, 0.00532–0.4256 μL from 1 mg / mL commercial solution in 6.25 μL DPBS pH 5 or pH 7.4) for 30 min at room temperature (12.5 μL, concentration 160 nM). The complexes were then diluted to 100 μL per well (final concentrations of 20 nM FAM-siRNA and 42–420 nM, i.e., 0.3–3 μg / mL peptide dendrimer or 0.0532–4.256 μg / mL L2000) in DPBS pH 5 or pH 7.4, added to Cellstar black, μClear, Greiner Bio one 96-well plates (Huberlab, Aesh, CH) and measured at λ = 1.0 nm with a Tecan Infinite M1000 Pro plate reader. ex = 470 nm and λ em Fluorescence anisotropy was measured at 520 nm = 520 nm. For each measurement, the G-factor was calibrated with a 1 nM fluorescein solution in 10 mM NaOH set at 20 mP, and for studies at pH 5, the polarization of FAM-siRNA alone was set at 55 mP, which is comparable to pH 7.4.

[0098] Competitive assay by fluorescence polarization: Complexes were formed as in the free FAM-siRNA in DPBS assay above and 50 μL was added to 50 μL of serial dilutions of 0-20 μg / mL heparin in DPBS in Cellstar black, μClear, Greiner bio one 96-well plates (100 μL, final concentrations 0-4 μg / mL heparin, 20 nM FAM-siRNA and 420 nM, i.e. 3 μg / mL peptide dendrimer or 532 ng / mL L2000). Plates were incubated for 2 h at RT and λ was measured using a Tecan Infinite M1000 Pro plate reader. ex = 470 nm and λ em Fluorescence anisotropy was measured at 520 nm = 520 nm. For each measurement, the G factor was calibrated using a 1 nM fluorescein solution in 10 mM NaOH set at 20 mP.

[0099] Dynamic Light Scattering (DLS): siRNA (80 pmol, 1.6 µL from a 50 µM Milli-Q aqueous solution in 50 µL PB pH 5 or pH 7.4) and peptide dendrimer (N / P ratio 10, 1680 pmol, 10-14 µL from a 1 mg / mL Milli-Q aqueous solution in 50 µL PB pH 5 or pH 7.4) or L2000 (w / w ratio 2:1, 2.128 µg, 2.128 µL from a 1 mg / mL commercial solution in 50 µL PB pH 5 or pH 7.4) were complexed in phosphate buffer (PB) at pH 5 or pH 7.4 and incubated for 30 min at room temperature (100 µL, final concentrations 800 nM siRNA and 16.8 µM, i.e. 120 µg / mL peptide dendrimer or 21.28 µg / mL L2000). 50 μL was then transferred to a low volume Univette (Sigma aldrich, Buchs, CH). Dynamic light scattering and zeta potential were then measured on an Anton Paar Litesizer 500 (Buchs, CH) and data were processed with the software provided by the manufacturer (Kalliope) using the "particle number" parameter.

[0100] Transmission electron microscopy (TEM): siRNA (10 pmol, 0.2 µL from a 50 µM Milli-Q aqueous solution in 6.25 µL Milli-Q water) and peptide dendrimer (N / P ratio 10, 210 pmol, 1.25-1.75 µL from a Milli-Q aqueous solution of 1 mg / mL in 6.25 µL Milli-Q water) or L2000 (w / w ratio 2:1, 21.28 µg / mL, 0.266 µL from a commercial solution of 1 mg / mL in 6.25 µL Milli-Q water) were complexed as described previously in Milli-Q water for 30 min at room temperature (12.5 µL, final concentration 800 nM siRNA and 16.8 µM, i.e. 120 µg / mL peptide dendrimer or 21.28 µg / mL L2000). Carbon-coated copper TEM grids (400 mesh, Gloor Instruments AG, Kloten) were glow discharged in a CTA 010 Balzers Union. 5 μL of the complex was then dropped onto the grid, incubated at RT for 1 min, and then dried on filter paper. The grid was washed with water and dried on filter paper three times. The grid was then stained twice for 10 s with uranyl acetate (2% in 50% alcohol solution, 6 μl, Gloor Instruments AG, Kloten). After stock / air drying for about 2 h, imaging was performed on an FEI Tecnai spirit transmission electron microscope equipped with two digital cameras (Olympus-SIS Veleta CCD camera, FEI Eagle CCD camera).

[0101] Dialysis: Complexes were formed with labeled Cy3-siRNA (80 pmol, 1.6 μL from a 50 μM Milli-Q aqueous solution in 50 μL OptiMEM pH 5 or pH 7.4) and peptide dendrimer (N / P ratio 1-10, 168-1680 pmol, 1-14 μL from a 1 mg / mL Milli-Q aqueous solution in 50 μL OptiMEM pH 5 or pH 7.4) or L2000 (w / w ratio 2:1, 0.218-2.128 μg, 0.218-2.128 μL from a 1 mg / mL commercial solution in 50 μL OptiMEM pH 5 or pH 7.4) in OptiMEM and incubated for 30 min at room temperature (100 μL, concentration 80-800 nM Cy3-siRNA and 1.68-16.8 μM, i.e. 12-120 μg / mL peptide dendrimer or 2.66-21.28 μg / mL L2000). The complexes were then diluted to a final volume of 800 μL with OptiMEM pH 5 or pH 7.4 (final concentrations 100 nM Cy3-siRNA and 0.210-2.1 μM, i.e. 1.5-15 μg / mL peptide dendrimer or 0.266-2.66 μg / mL L2000). 500 μL of the complexes were added to an Amicon Ultra 0.5 mL dialysis eppendorf with a 100 KDa cutoff (Sigma aldrich, Buchs, CH) and centrifuged at 14000 g for 10 min. For the second experiment, after centrifugation, the lower compartment was replaced and 500 μL of OptiMEM pH 5 or pH 7.4 was added to the upper compartment of an Amicon Ultra 0.5 mL dialysis Eppendorf and centrifuged again for 10 min at 14000 g. Then, 100 μL of the dialysate before dialysis, after dialysis and after the additional wash were transferred to a TPP 96-well plate (Faust Laborbedarf AG, Schaffhausen) and analyzed by λ spectroscopy using a Tecan Infinite M1000 Pro plate reader. ex = 547 nm and λ em Fluorescence was measured at 563 nm.

[0102] Vesicle leakage: 4 mL MeOH / CHCl on a rotary evaporator at room temperature 3Egg yolk phosphatidylcholine (EYPC, Avanti Polar Lipids, Alablaster, USA) lipid thin layers were prepared by evaporating a 100 mg solution in 1:1 ethanol followed by evaporation under reduced pressure overnight. The resulting film was hydrated for 30 min with 4 mL of buffer A (50 mM 5(6)-carboxyfluorescein (CF, Sigma Aldrich, Buchs, CH), 10 mM TRIS, 10 mM NaCl, pH 7.4) and subjected to freeze-thaw cycles (7 times) and extrusion through polycarbonate membranes (pore size 100 nm) (15 times). Excess vesicular components were removed by gel filtration (Sephadex G-50) with buffer B (10 mM TRIS, 107 mM NaCl, pH 7.4). Final concentration: approximately 2.5mM EYPC; internal: 50mM CF, 10mM TRIS, 10mM NaCl, pH 7.4 buffer; external: 10mM TRIS, 107mM NaCl, pH 7.4 buffer.

[0103] EYPC stock solution (10 µL) was diluted to 700–792 µL with buffer B (10 mM TRIS, 107 mM NaCl, pH 7.4) and placed in a fluorescence cuvette, thermostated (25 °C) and gently stirred (final concentrations 31 µM EYPC, 625 µM CF, 10 mM TRIS and 107 mM NaCl). Complexes were formed with siRNA (80 pmol, 1.6 μL from a 50 μM Milli-Q aqueous solution in 50 μL buffer) and peptide dendrimer (N / P ratio 10, 1680 pmol, 12 μL from a 1 mg / mL Milli-Q aqueous solution in 50 μL buffer B) or L2000 (w / w ratio 2:1, 2.128 μg, 2.128 μL from a 1 mg / mL commercial solution in 50 μL buffer B) in buffer B and incubated for 30 min at room temperature (100 μL, concentration 800 nM siRNA and 16.8 μM, i.e. 120 μg / mL peptide dendrimer or 21.28 μg / mL L2000). CF effluent was measured at λ = 1.0 μM on an Agilent Cary Eclipse fluorescence spectrophotometer. ex = 492 nm and λ emThe fluorescence was monitored at t = 517 nm as a function of time for 360 s. At t = 30 s, peptide dendrimer (0.8-12 μL from a 1 mg / mL Milli-Q water stock solution), L2000 (0.8-12 μL from a 1 mg / mL commercial solution) or complex (100 μL from the complex solution above) was added to the cuvette (final concentrations of 0-100 nM siRNA and 0.14-2.1 μM / 1-15 μg / mL peptide dendrimer or 1-15 μg / mL L2000). At t = 270 s, 1.2% Triton X-100 was added to the cuvette (50 μL, final concentration 0.07%). Fluorescence intensity was normalized to the specific emission intensity I(t) using I(t) = (It-I0) / (I∞-I0), where I0 is the intensity before addition of peptide dendrimer and I∞ is the intensity at saturation after lysis with Triton X-100.

[0104] Critical micelle concentration (CMC): Nile red is known to have higher fluorescence when surrounded by a hydrophobic environment, and this experiment is derived from a known procedure but applied to a 96-well plate. To confirm whether the CMC measurement now works in our assay, we first performed measurements with dodecylphosphocholine and observed a value of 0.3125 mg / mL corresponding to 0.9 mM, which matches values ​​of 0.9-1.1 mM from the literature (data not shown). Briefly, Nile red (Sigma aldrich, Buchs, CH) was diluted to a concentration of 2 μM in methanol and 5 μL was added to each well of a TPP 96-well plate (Faust Laborbedarf AG, Schaffhausen) and dried for 1 h at room temperature under fume hood air flow. Serial dilutions of peptide dendrimers, L2000 and dodecylphosphocholine (Avanti polar lipids, Alablaster, USA) were performed in 10 mM phosphate buffer (pH 5 or pH 7.4) starting from 1 mg / mL to 0.5 μg / mL, and 50 μL were added to the plate containing dried Nile Red fluorescent dye (final concentration 0.2 μM). For the second experiment, a solution of peptide dendrimers at 1 mg / mL in Milli-Q water was diluted to 0.1 mg / mL in 100 μL PB with pH ranging from 7.4 to 4 and added to the dried Nile Red. The plate was incubated for 2 h before measuring λ in a Tecan Infinite M1000 Pro plate reader. ex = 540 nm and λ em Fluorescence was measured at = 615 nm. The CMC value can be determined at the inflection point of the curve.

[0105] Diffusion NMR (DOSY) measurements: Diffusion NMR experiments are 2 The analysis was carried out using a Bruker DRX500 with a solution of dendrimer (15 mg / mL) in 20O (pH 5 or 7.4 at 303 K). -4 Tcm -1 The gradient with the maximum intensity of D 2The HOD proton signal in O was used for calibration (99.997%). The diffusion time was 125 ms and the gradient duration was 6 ms. Data analysis was performed using Bruker Simfit software to calculate the diffusion coefficient D [m 2 s -1 The hydrodynamic radius is D at 303K. 2 For O, the Boltzmann constant k = 1.380 × 10 -23 J.K. -1 , calculated from the diffusion coefficient D using the Stokes-Einstein equation Rh=kT / 6πηD at temperature T (K) and viscosity η=1.089 mPa.s.

[0106] Titration with NaOH: Peptide dendrimers (0.8–1 μmol, 5–7 mg) were diluted in 7–10 mL of Milli-Q water (final concentration 100 μM) and acidified with 1 M HCl to pH 3. 0.1 M NaOH was then added to the solution in 2 μL steps using a Dosimat plus (Metrohm, Zofingen, Switzerland) and the pH was measured with a 692 pH / ion meter (Metrohm).

[0107] Circular dichroism (CD) spectroscopy measurements: CD spectra were recorded using a Jasco J-715 spectrometer equipped with a PFD-350S temperature controller and a PS-150J power supply. All experiments were measured using a Hellma Suprasil R 100-QS 0.1 cm cuvette. Stock solutions of peptide dendrimers (1 mg / mL) were freshly prepared in Milli-Q water. For measurements, peptides were diluted to a final concentration of 200 μg / mL in PBS buffer (pH = 7.4, final concentration 10 mM). For solvent-dependence studies, peptides were diluted to 200 μg / mL in phosphate buffer (pH = 5 or 7.4, final concentration 8 mM) and TFE or ACN (0, 5, 10, 15, 20 or 40%). The measurement range was 185-260 nm, the scan speed was 10 nm / min or 20 nm / min, the pitch was 0.5 nm, the response was 16 s and the band was 1.0 nm. Nitrogen flow was maintained above 8 L / min. The baseline (solvent) was recorded under the same conditions and manually subtracted. Each sample was subjected to two accumulations. Before each measurement, the cuvette was filled with 1 M HCl, mQ deionized H 2 O and washed with PBS or PB buffer.

[0108] Molecular Dynamics: A dendrimer model was constructed by processing the GROMACS topology of a linear peptide of the same sequence using in-house software. An initial starting conformation was generated by setting the (Φ, Ψ) angle pairs of all residues in a helical conformation using PyMol (Molecular Graphics System, version 1.8 (Schrodinger, LLC)).

[0109] Molecular dynamics (MD) simulations were performed using GROMACS software version 2016.1 and the Gomos53a6 force field. A dodecahedral box was created around the dendrimer 1.0 nm from the edge of the dendrimer and filled with extended simple point-charge water molecules. Sodium and chloride ions were added to produce an electroneutral solution with a final concentration of 0.15 M NaCl.

[0110] Before subjecting the system to a two-phase positionally constrained MD equilibration procedure, the energy was minimized using the steepest gradient method to remove close contacts. The system was first generated for 100 ps in the canonical NVT (N is the number of particles, V is the system volume, and T is the temperature) ensemble at 300 K and equilibrated for another 100 ps in the NPT (P is the system pressure) ensemble at 1.0 bar before pressure coupling was turned on.

[0111] All bond lengths were constrained to their equilibrium values ​​by using the LINCS algorithm. The neighbor lists for the calculation of nonbonded interactions were updated every five time steps using a cutoff of 1.0 nm with a step size of 2 fs. A twin-range cutoff of 1.0 nm was used for both Coulomb and Lennard-Jones interactions. The system was divided into two groups, "protein" and "non-protein", which were separately coupled to a temperature bath using the V-rescale algorithm with a time constant of 0.1 ps, while pressure coupling was performed using an isotropic Parrinello-Rahman barostat with a time constant of 2.0 ps.

[0112] The stability of helical structures in peptide dendrimers under different conditions was evaluated by determining the unfolding kinetics as evidenced by the radius of gyration and RMSD values. After pre-equilibration of the system (see above), the structures were subjected to MD at 300 K for 1000 ns in water with 0.15 M NaCl with or without 20% v / v TFE. After unfolding the main α-peptide chain helix, its main chain RMSD and the total number of i → i + 4 H-bonds were calculated. The overall stability of the internal structures was evaluated using the total number of main chain H-bonds over time. The spontaneous emergence of β-sheet and random coil secondary structures was evidenced by Ramachandran number analysis. The last 100 ns of each 1 ms MD run (10001 structures) were clustered using the GROMACS method and a cutoff of 0.3 nm, and the central structure of the main cluster was used for the analysis. Ramachandran numbers were calculated using the python package as implemented by Mannige R. The same structure was used as a representative structure of the equilibrated dendrimer at each condition, which was simulated using PyMol software to build a 3D model.

[0113] DNA Cell lines, transfection reagents and plasmids: Cells were incubated at 4 °C for 2 h in 5% CO 2 The cells were maintained in DMEM medium containing 10% (v / v) FCS and 1% (v / v) P / S in a humidified atmosphere at 37 °C. Plasmid CRISPR-Cas9-GFP was purchased from DNA2.0. Lipofectamine 2000 (L2000) was obtained from Sigma-Aldrich. L2000 was used as a positive control transfection agent according to the manufacturer's instructions.

[0114] Transfection procedure: 24 h prior to transfection, cells were seeded in 96-well plates to reach 70–90% confluency. Peptide dendrimer / DNA complexes were formed by mixing dendrimer (6.25 μL, 5 μg–6.4 μg at an N / P ratio of 5) with plasmid DNA (250 ng; 6.25 μL). These mixtures were incubated in OptiMEM for 30 min at 25 °C. Transfection control complex L2000 was mixed with plasmid DNA (250 ng; 6.25 μL) at the respective manufacturer's recommended concentration. Before layering the DNA complexes onto the cells, OptiMEM was added to dilute the complexes so that each complex contained 250 ng of DNA in a total volume of 100 μL in one well of a 96-well plate. After removing the complete medium from the cells, the complexes were added to the plates. The plates were incubated at 37 °C for 4 h. The transfection solution was then replaced with complete growth medium for 40 h before assaying transfection efficiency.

[0115] Transgene expression assay: Cells were washed twice with PBS and incubated with trypsin for 20 min at 37° C. Then, 100 μL of PBS was added to each well and transfection efficiency was assessed by FACS analysis (BD LSRFortessa). [Table 1]

[0116] Two D-enantiomeric G3 peptide dendrimers, DMH13 and DMH18, act with remarkable efficiency as single-component reagents for helper lipid-free siRNA transfection (Figure 3). These dendrimers combine the polycationic branches of our previously optimized siRNA cotransfection peptide dendrimers, now used in D-enantiomeric form, with a hydrophobic core composed of either hydrophobic amino acids or fatty acids as an alternative to lipofectin to promote endosomal escape. Both dendrimers transfect siRNA as efficiently as or better than the gold standard Lipofectamine L2000 in a variety of cell lines and conditions, without significant toxicity to cells.

[0117] Quantification of GADPH knockdown by measuring the residual enzyme activity in an NADH reporter assay in dendrimer / siRNA versus dendrimer / siNC treated cells showed that several dendrimers induced knockdown as potently as the commercial reagent L2000 (33±10% residual activity for siNC treated cells, Table 1). Cells were not significantly affected by treatment with dendrimer / siNC complexes, which did not induce GADPH knockdown or a decrease in cell viability compared to untreated cells. With regard to dendrimer structure, increasing G0 hydrophobicity increased the transfection efficiency in the three dendrimer series up to an optimal hydrophobicity, which was achieved with one C26 fatty acid for dendrimers with one side-chain acylated lysine in G0 (MH06, 30±6%), two C16 fatty acids for dendrimers with two side-chain acylated lysines in G0 (MH13, 32±5%), and four leucines (MH18, 28±9%), three phenylalanines (MH22, 32±3%), and four tryptophans (MH25, 34±3%) for dendrimers with linear peptides in G0. Investigating further variations of G0 showed that transfection was also particularly efficient with a hydrophobic core consisting of four norleucines (MH40, 23±1%).

[0118] To further investigate the effect of dendrimer structure on transfection efficiency, we varied the composition of the G1-G3 branches around the tetraleucine G0 core of MH18, yielding analogs with different types or numbers of cationic and hydrophobic residues (Table 1). Substitution of all cationic lysines in the branches of MH18 with histidines weakened siRNA binding and cellular uptake and abolished transfection. On the other hand, comprehensive substitution with arginines (MH44) led to very strong siRNA binding (unbound siRNA: 1.6 ± 0.3%) and cellular uptake (GMFI: 594 ± 92%), while transfection efficiency remained stable (residual GADPH: 34 ± 13%).

[0119] Changes in the ratio and distribution of leucine and lysine in the G1-G3 branch also abolished transfection, for example by global deletion of leucine or by KL exchange in G1, G2 or G3, whereas Lys-Leu in G2 still showed strong siRNA binding and internalization into cells.

[0120] Considering that both our peptide dendrimers and siRNAs are homochiral, we further investigated whether changes in amino acid chirality could affect transfection. For MH13 (2xC16 core), transfection efficiency was maintained upon switching to D-residues either at G3 only (MH13D1) or inclusively (DMH13). In contrast, for MH18 (L4 core), switching to D-amino acids at either G3 (MH18D1), G0 (MH18D2) or branched lysine (MH18D3) partially or completely abolished transfection. Surprisingly, the diastereoisomer MH18D3 also lost siRNA binding and cellular uptake compared to MH18. Nevertheless, siRNA transfection was preserved for the enantiomeric DMH18, indicating that homochirality, rather than absolute configuration, is important for siRNA transfection in this dendrimer.

[0121] Transfection of HeLa cells with decreasing amounts of siRNA at constant N / P=10 showed that the two D-enantiomer dendrimers DMH13 and DMH18 produced the most sustained GADPH knockdown after 48 h with as little as 40 nM siRNA (Figure 5a). In both cases, GADPH knockdown was essentially complete at 24, 36 and 48 h as measured by the amount of remaining mRNA (Figure 5b). Notably, all dendrimers except the arginine-containing MH44 were also effective in the presence of 10% serum for 8-72 h of incubation, suggesting potential use in complex environments (Figure 5c). Similar results were obtained with CHO and HEK-293 cells. When different N / P ratios were investigated at constant siRNA (100 nM), it was observed that the most active dendrimers, i.e., lipidated dendrimers MH06 (C26) and MH13 / DMH13 (2 × C16) as well as the D-enantiomer leucine dendrimer DMH18, tolerated a wide range of N / P ratios (from 3 to 10) (Figure 5d). On the other hand, lowering the siRNA concentration to 10 and 20 nM at higher N / P ratios (100 and 50, respectively) only yielded efficient transfection with dendrimers MH25 (Trp), MH40 (Nle), DMH13 (D-, 2 × C16) and DMH18 (D-leu).

[0122] Transfection experiments were performed in five additional cell lines using the most potent transfection dendrimers MH06, MH13, MH18, MH40, DMH13 and DMH18 in comparison to L2000. Quantification of GADPH knockdown showed that DMH13 and DMH18, as well as some other dendrimers, effectively transfected all five cell lines, including SH-SY5Y and Caco-2 cells, which are notoriously difficult to transfect (Figure 6). Cell viability was only slightly affected in all of these transfection experiments, with most cytotoxicity observed for L2000.

[0123] Taken together, these experiments showed that the D-enantiomer dendrimers DMH13 and DMH18 performed best across all conditions and cell lines tested. The better siRNA transfection of D-peptide dendrimers versus L-peptide dendrimers contrasts with reports of better performance of linear L-peptides versus linear D-peptides for DNA transfection.

[0124] Biological mechanisms of cellular uptake To understand how our dendrimers enabled siRNA transfection, we investigated cellular uptake of siRNA using the nine transfection dendrimers discussed above, along with L2000 as a positive control and, as negative controls, two close analogs of MH18 that showed low (MH46, LK in G1) or high (MH47, KL in G2) internalization of siRNA but no significant GADPH knockdown.

[0125] Live cell confocal microscopy images of fluorescently labeled siRNA showed that FAM-siRNA was mostly localized to intracellular compartments upon internalization in HeLa cells in all cases (Figure 7a). Consistent with the expected role of lipid components in promoting endosomal escape, fluorescence diffusion in cells was observed for lipidated dendrimers MH06, MH13 and DMH13 as well as L2000, indicating endosomal release of siRNA. Nevertheless, the endosomal escape observed in these live cell confocal images did not correlate with transfection efficiency, as exemplified by two of the best transfected dendrimers, MH18 and DMH18, lacking significant siRNA fluorescence in the cytosol. Indeed, a portion of the siRNA remained trapped in intracellular compartments even after 24 h for all dendrimers investigated and for L2000, independent of their transfection efficiency. The effect was most pronounced when transfecting the arginine dendrimer MH44 and the untransfected control dendrimer MH47, both of which showed much stronger cellular uptake than the other dendrimers. Similar effects were observed in CHO and HEK-293 cells.

[0126] Cellular uptake of siRNA was inhibited in the presence of various cellular internalization inhibitors, as evidenced by flow cytometry (Figure 7b). Cellular uptake was completely blocked at low temperature (+4 °C). Uptake was also strongly inhibited by cytochalasin D, which blocks actin filaments indicative of macropinocytosis, and by chlorpromazine, indicative of clathrin-mediated endocytosis. Furthermore, nystatin inhibited cellular uptake by the dendrimer MH18 / DMH18, its arginine analogue MH44, and the inactive control dendrimer MH46, suggesting caveolae-mediated entry in these cases. Other dendrimers were only partially affected by nystatin (MH06, MH13, MH22, MH47) or not affected at all (DMH13, MH25, MH40).

[0127] To investigate whether transfection requires endosomal acidification as assumed for most transfection reagents, we measured the overall transfection efficiency in the presence of bafilomycin A1, which inhibits vacuolar ATPase and endosomal acidification. Treatment with bafilomycin 1 h before transfection had no effect, but transfection was completely inhibited when bafilomycin was present throughout the experiment, and was strongly reduced even when the compound was added t = 4 h after endocytosis had occurred, indicating that dendrimer-mediated transfection requires endosomal acidification. In contrast, bafilomycin treatment had no effect on L2000-mediated transfection (Figure 7c).

[0128] Taken together, the above imaging and inhibition experiments were consistent with findings reported for various nonviral transfection reagents, which occurred via energy-dependent endocytosis followed by acidification-dependent endosomal escape, and suggested that peptide dendrimer-mediated siRNA delivery should be accompanied by the formation of dendrimer / siRNA nanoparticles.

[0129] Nanoparticle formation Measuring the extent of siRNA binding by quantifying free siRNA as a function of N / P ratio using the Quant-it™ assay showed that all eleven selected dendrimers bound siRNA completely independent of their transfection efficiency at N / P>2, both at pH 7.4 corresponding to the extracellular medium and at pH 5.0 corresponding to acidified endosomes (Figure 8a / b). To test whether siRNA is incorporated into nanoparticles upon dendrimer complexation, FAM-siRNA binding was followed using fluorescence polarization (FP). Indeed, strong FP signals were recorded upon addition of the various dendrimers, suggesting that FAM-siRNA was incorporated into higher molecular weight aggregates presumably nanoparticles. At pH 7.4, the FP signal intensity reached a maximum above N / P=5, suggesting that complete complexation of siRNA at N / P=2 was followed by further aggregation into larger nanoparticles by incorporating additional dendrimer molecules (Figure 8c). In contrast, at pH 5.0, the FP signal reached saturation at N / P = 2 and remained stable upon addition of more dendrimers, suggesting that the formed dendrimer / siRNA nanoparticles did not incorporate additional dendrimers at pH 5.0 (Figure 8 d).

[0130] Measurements of the displacement of FAM-siRNA from dendrimer complexes formed at N / P=10 upon addition of polyanionic heparin, tracking either free siRNA (Figure 8e) or the decrease in FP signal intensity (Figure 8f), in both cases required a threshold concentration of heparin before the signal began to change, likely reflecting the displacement of excess dendrimer from the nanoparticles until a stoichiometry corresponding to N / P=2 was reached. Above that threshold, the extent of displacement varied greatly between the different dendrimers. FAM-siRNA was bound most strongly by the polyarginine dendrimer MH44 and the non-transfected control MH47, both of which showed very high cellular uptake, suggesting that these dendrimers formed very tight siRNA complexes and were unable to release their siRNA cargo. Heparin displacement was also incomplete for amino acid-only dendrimers (MH18, MH22, MH25, MH40, MH46, DMH18) and was most facile for lipidated dendrimers, suggesting that these formed significantly weaker complexes with siRNA (MH06, MH13, DMH13). Note that the commercially available reagent Lipofectamine L2000 released siRNA immediately at the lowest heparin concentration used, indicating much weaker complex formation compared to our dendrimers.

[0131] Nanoparticle formation was further confirmed using dynamic light scattering (DLS), which showed that dendrimer / siRNA complexes formed nanoparticles in the size range of 80 × 200 nm at pH 7.4 (Figure 9a). The nanoparticles were significantly smaller at pH 5.0, confirming the effect seen by FP (Figure 9c). The largest particles were formed by dendrimer MH47, which strongly bound and internalized but did not transfect siRNA. All dendrimer / siRNA nanoparticles had positive zeta potential values ​​ranging from +10.4 mV (MH46) to +17.7 mV (DMH13), indicating an overall positive charge, consistent with FP measurements indicating further aggregation of dendrimers above N / P=2 (Figure 9b). The zeta potential values ​​were higher and more spread out at pH 5.0, ranging from +13.3 mV (MH46) to +40.3 mV (DMH13), possibly indicating increased nanoparticle stability at that pH (Figure 9d). 55 Note that our dendrimers behaved very differently from L2000, which formed larger aggregates with comparable sizes at both pH values, but the zeta potential values ​​changed from strongly negative (-50 mV) at pH 7.4 to slightly positive (+10.7 mV) at pH 5.0. Transmission electron microscopy (TEM) further confirmed the formation of nanoparticles (Figure 9e). TEM images obtained by combining the dendrimers as trifluoroacetate salts with siRNA in pure water, resulting in approximately pH = 5, showed a wide range of sizes of dendrimer / siRNA complexes, consistent with the size distribution observed by DLS at pH 5.0.

[0132] Two further experiments were carried out showing that both siRNA and dendrimer were strongly bound within the nanoparticles. In the first experiment, the complex formed between fluorescently labeled Cy3-siRNA and our dendrimers was extensively dialyzed using a dialysis membrane with a molecular weight cutoff of 100 kDa, which should separate uncomplexed Cy3-siRNA from the dendrimer / Cy3-siRNA nanoparticles (Figure 9a). Strong fluorescence was detected in the dialysis filtrate in the absence of dendrimer, indicating that free Cy3-siRNA was able to diffuse through the membrane, but decreased to undetectable levels upon increasing the dendrimer concentration up to N / P=10, indicating that Cy3-siRNA was fully incorporated into aggregates larger than 100 kDa. siRNA was also strongly bound by our dendrimers at pH=5.0, whereas L2000 bound strongly to siRNA only at pH 5.0 and only weakly at pH 7.4 (Figure 9b).

[0133] In a second experiment, we measured the leakage of fluorescein from vesicles composed of phosphatidylcholine, a zwitterionic lipid that forms neutral membranes associated with the membranes of eukaryotic cells. In this experiment, the fluorescence is self-quenched under high fluorescein concentrations within the vesicles, but is restored by dilution if fluorescein is released from the vesicles, which occurs upon addition of a membrane-disrupting compound. Indeed, a strong fluorescence increase was observed upon addition of 15 μg / mL of free dendrimer (red curve, Figure 10c). The effect was significant for all transfected dendrimers, even at 1 μg / mL (yellow curve, Figure 10c), as could be expected from their cationic / hydrophobic composition, but was much weaker for the two negative controls MH46 and MH47, which have a higher and lower ratio of cationic / hydrophobic side chains, respectively, compared to the transfected dendrimers. Remarkably, the addition of siRNA to 15 μg / mL of dendrimers (cyan curves, FIG. 10c) either completely blocked (MH06, MH22) or very strongly reduced (e.g. MH13 / DMH13, MH40) fluorescein release from vesicles, or at least to the level observed with 1 μg / mL of free dendrimer alone (e.g. MH18, MH44), indicating that the dendrimer was almost completely incorporated into the siRNA complex and had no significant membrane disruptive activity by itself. This effect is striking, since the peptide dendrimers at N / P=10 are in large excess compared to the siRNA. In contrast to our dendrimers, L2000 did not cause measurable vesicle leakage, with or without siRNA (black curves, FIG. 10c).

[0134] pH-Dependent Aggregation of Peptide Dendrimers The amphiphilic topology of our dendrimers and the effect of hydrophobic core size and composition on siRNA binding and transfection suggested that dendrimer-siRNA interactions may be mediated by self-aggregation of the dendrimers prior to siRNA binding. Such self-aggregation in the absence of siRNA involves intermolecular interactions that potentially explain the much larger difference in siRNA binding observed between diastereoisomers (MH18 / MH18D3) than between enantiomers (MH18 / DMH18).

[0135] Measuring aggregation using the Nile Red assay57 showed that all peptide dendrimers with strong siRNA binding properties formed aggregates at pH 7.4, a critical micelle concentration (CMC) value below the concentration range used in our transfection experiments (Figure 11a). In contrast, dendrimers that bound weakly to siRNA, such as MH18D3 and MH46, did not show measurable aggregation at that pH across the entire concentration range tested. At pH=5, there was a sharp decrease in aggregation indicated by higher or undetectable CMCs across all dendrimers except for the negative control dendrimer MH47, which bound but did not release siRNA and whose CMC was the same at pH=7.4 and pH=5 (Figure 11b). CMC and siRNA binding correlated with dendrimer hydrophobicity when comparing the optimally transfected dendrimer MH18 with the negative controls MH46 (double LK exchange in G1 of MH18 decreasing hydrophobicity, no self-aggregation, no siRNA binding) and MH47 (four-fold KL exchange in G2 of MH18 increasing hydrophobicity, strong self-aggregation at both pH and strong siRNA binding without transfection). This effect was also seen when examining aggregation as a function of pH, showing that the lipidated dendrimer MH06 / MH13 / DMH13 aggregated at lower pH than the amino acid-only dendrimer (Figure 11).

[0136] Dendrimer aggregation was further evidenced by hydrodynamic radius (Rh) measurements by DOSY NMR at pH 7.4 and 5.0 in the case of MH13, DMH13, MH18, DMH18, MH18D3, MH46 and MH47 (Table 3). MH18D3 and MH46, despite the relatively high concentration measured by NMR (15 mg / mL), gave comparable values ​​of Rh=2.42 / 2.52 nm and 1.98 / 2.06 nm, compatible with a monomeric state at both pH values, whereas MH18 and DMH18 were in the monomeric state at pH 5.0 (Rh=2.52 nm and 2.42 nm) but aggregated at pH 7.4 (Rh=3.21 nm and 3.11 nm), whereas MH13 and DMH13 aggregated at both pH values ​​(Rh=4.66 / 4.97 nm and Rh=4.14 / 4.95 nm). [Table 2]

[0137] The pH dependence of aggregation correlated with the acid-based properties of the dendrimers. Titration of dendrimers MH13, DMH13, MH18, DMH18, MH18D3, MH46 and MH47 showed two apparent pKa values ​​with relative stoichiometries consistent with ionization of the amino terminus at acidic pH (pKa1 = 6.5) and ionization of the lysine side chain at higher pH (pKa2 = 8.5-10) (Figure 11d). Protonation of the amino terminus at pKa1 = 6.5 led to an increase of eight positive charges in the dendrimer, easily explaining the switch in aggregation between pH 7.4 and pH 5. On the other hand, the pKa2 values ​​of the lysine side chain protonation decreased in the series MH18D3 / MH46>MH18 / DMH18>MH13 / DMH13, following the order of increasing self-aggregation tendency, indicating that aggregation is driven by decreasing basicity of the lysine side chain. MH47 precipitated during acid-base titration and its pKa2 value could not be determined.

[0138] Taken together, CMC, DOSY-NMR and acid-base titration experiments demonstrated a consistent picture of pH-dependent self-aggregation of our dendrimers enabling siRNA complexation, including the surprising diastereoselective complexation of siRNA by self-aggregating MH18 / DMH18 compared to the diastereoisomer MH18D3, which did not bind to siRNA or aggregate. However, these effects were mainly according to the hydrophobicity of the whole dendrimer, which could not explain the difference between MH18 and the diastereoisomer MH18D3.

[0139] Circular dichroism Since stereochemistry should influence conformation, the secondary structure content of our dendrimers was measured by circular dichroism (CD) in aqueous media containing up to 20% v / v trifluoroethanol (TFE) as an inducer of secondary structure (Figure 12).

[0140] CD spectra of lipidated dendrimers MH13 / DMH13 at pH=7.4 showed significant β-sheet content, which increased from ∼35% in water to ∼40% upon addition of 20% v / v TFE at the expense of α-helix content, despite the fact that trifluoroethanol (TFE) normally induces α-helices (Figure 12a / d). β-sheet content also increased with transfected dendrimers MH18 / DMH18 up to 10% v / v TFE at pH 7.4 (25%–33% β-sheet), but decreased again at 20% v / v TFE (22% β-sheet) in favor of α-helices (30% α-helix), probably reflecting the α-helical propensity of the linear tetraleucine core (Figure 12b / e). An increase in β-sheet content without induction of α-helices in transfected dendrimers was similarly observed upon addition of acetonitrile as a co-solvent. The β-sheet content also increased as a function of hydrophobic core size in dendrimers with G0 containing one (MH01–MH07) or two (MH08–MH15) side-chain acylated lysines, or zero to five leucines (MH30, MH16–MH20), a trend that followed their aggregation ability as well as their siRNA binding and cellular internalization. On the other hand, dendrimer MH47, which aggregated and bound strongly to siRNA but did not transfect, maintained an overall stable α-helix and β-sheet content when 20% v / v TFE (11%–14% α-helix, 36%–33% β-sheet) was added (Figure 12c / f). Moreover, at pH 7.4, the β-sheet content of dendrimer MH46, which did not aggregate or bind to siRNA, decreased due to its α-helix content (16%–26% α-helix) from pure water to 20% v / v TFE (28%–20% β-sheet).

[0141] CD spectra recorded at pH 5.0 showed a predominantly α-helical conformation with enhanced α-helical content upon addition of TFE for all dendrimers under conditions where the dendrimers were not aggregated or were only partially aggregated, with the exception of the strongly aggregated and siRNA-bound nontransfected dendrimer MH47, which retained a predominantly β-sheet conformation.

[0142] The β-sheet propensity of our transfection dendrimers was surprising to us because previous conformational studies with structurally related antimicrobial peptide dendrimers (AMPDs) showed a strong preference for α-helical conformations in the CD spectra. These AMPDs were significantly less hydrophobic and did not form aggregates in water, suggesting that the β-sheet conformation observed in our transfection dendrimers may reflect intermolecular interactions in the aggregated state.

[0143] molecular dynamics To test whether our transfected dendrimers could contain a β-sheet conformation in the monomeric state, molecular dynamics (MD) simulations were performed with transfected dendrimers MH13 and MH18 and non-transfected controls MH18D3, MH46 and MH47 in explicit water with or without 20% v / v TFE as an inducer of secondary structure. As starting conformations, 3D models were constructed using PyMol (version 1.8, Schrodinger, LLC) with all dendrimer branches and fully protonated dendrimers in prefolded α-helical conformations to simulate pH 5.0, and dendrimers with protonated side chains and amino termini as free bases to simulate pH 7.4. Simulations of the monomeric species were run for 1 microsecond using GROMACS.

[0144] Simulations gave comparable results in pure water at both pH values, under which conditions all dendrimers rapidly unfolded and rearranged into compact conformations with nearly constant radii of gyration throughout the entire trajectory. Clustering of the final 100 ns produced a well-defined 3D model, which was analyzed both as an average structure as the 3D model and in terms of individual residue conformations throughout the dendrimer using Ramachandran numbers. In nearly all cases, residues along the central α-peptide chain adopted α-helical conformations from G0 to G2, whereas residues in the branches starting from the isopeptide bonds were either disordered or in β-sheet conformations. In the presence of TFE, the dendrimers unfolded and rearranged similarly, but were more open and flexible, exhibiting a slightly larger radius of gyration than in water, while the number of backbone H-bonds was similar, if not higher, in TFE compared to water (Figure 13a / c). This analysis showed a conserved α-helix in the central α-peptide chain at the G0 / G1 / G2 level (Figure 13b / d), as exemplified for the case of transfected dendrimer MH18 at pH 5. One notable exception was MH13 at pH 5, under which conditions the dendrimer was fully unfolded and had a β-sheet conformation throughout the structure (Figure 13b / e).

[0145] The present MD simulations described above reproduced the CD data very poorly with respect to α-helical and β-sheet content between different dendrimers, and variations in pH and solvent conditions. We consider this discrepancy as further evidence that our transfected dendrimers directly interact with each other in the aggregated state, presumably by forming intermolecular β-sheets. Although this situation cannot be easily modeled, we have previously observed β-sheet bridging between peptide dendrimers in the crystal structure of a lectin-binding glycopeptide dendrimer engaging in a trimeric aggregate.

[0146] Mechanistic model for dendrimer-enhanced siRNA transfection Considering the cellular uptake, nanoparticle formation, pH-dependent aggregation, CD and MD data presented above, we propose the following overall mechanistic model for dendrimer-mediated siRNA transfection. First, transfection dendrimers exhibiting a preferred arrangement of cationic and hydrophobic groups self-aggregate at pH 7.4 through intermolecular β-sheet crosslinks in a stereoselective process that is available for homochiral dendrimers such as MH18 / DMH18, but not for their diastereoisomer MH18D3. These dendrimer aggregates then complex siRNA to form stable nanoparticles, accumulating excess dendrimers on siRNA until N / P>6. In the third step, these nanoparticles enter cells by endocytosis and localize in endosomes. Acidification of the endosome then induces protonation of the dendrimer amino termini, which causes disaggregation, presumably by electrostatic repulsion.

[0147] Disaggregation leads to dissolution of the intermolecular β-sheet bridges between dendrimers and rearrangement into monomeric α-helical structures, which releases excess free monomeric dendrimers and leaves smaller dendrimer / siRNA nanoparticles with lower N / P ratios. Protonation of the amino termini probably also causes a proton sponge effect that allows endosomal escape, as has also been hypothesized for many transfection reagents. In agreement with studies of siRNA-transfected lipids showing that less than 4% of siRNA actually escapes from endosomes, this process is only very partial, as evidenced by confocal images showing endosomal localization of FAM-siRNA up to 24 hours in almost all cases. In the case of MH13 / DMH13, where a more extensive spread of FAM-siRNA into the cytosol is observed, we hypothesize that endosomal escape is further assisted by disruption of the endosomal membrane by free dendrimers present in an open conformation that probably exposes hydrophobic groups for membrane disruption, similar to the model obtained by MD for MH13 at pH 5 in the presence of TFE.

[0148] For the few dendrimer / siRNA nanoparticles that reach the cytosol after endosomal escape, equilibration to pH 7.4 is expected to partially liberate siRNA due to the lower N / P ratio of these nanoparticles. To model this process, diafiltration experiments were repeated at pH 5 to form rearranged dendrimer / siRNA nanoparticles, re-equilibrated to pH 7.4 to mimic entry into the cytosol, and a new diafiltration was performed at that pH value. In this experiment, there was very little release from the tightly binding transfected MH44 and non-transfected dendrimer MH47, whereas transfected dendrimers such as MH18 / DMH18, MH22, MH25, and MH40 released significant amounts of FAM-siRNA. Note that a shift to pH 5 was required to induce siRNA release. Non-transfected dendrimer MH46 also released its siRNA, but the dendrimer was not significantly internalized into the cells. Finally, lower amounts of siRNA were released from lipidated dendrimers MH06 and MH13 / DMH13, presumably compensated for by much more extensive endosomal escape of the nanoparticles in these cases.

[0149] Gene knockdown is most efficient with dendrimers that exhibit intermediate levels of siRNA binding. The most efficient peptide dendrimers are the D enantiomer peptide dendrimers DMH13 and DMH18.

[0150] Structure-activity relationships and mechanistic studies provide unprecedented insight into the transfection mechanism. Efficient transfection depends on the favorable ratio and arrangement of hydrophobic and cationic groups in the dendrimer, as well as its ability to aggregate in a stereoselective process involving intermolecular β-sheet cross-linking to allow the formation of dendrimer / siRNA nanoparticles. Stereoselectivity is highlighted by the lack of aggregation and siRNA binding with MH18D3, a diastereoisomer of the best transfection dendrimer DMH18. Protonation of the dendrimer amino terminus (pKa ∼6.5) upon endosomal acidification is also essential to allow nanoparticle rearrangement and endosomal escape, possibly via the proton sponge effect.

[0151] The described dendrimers are easily obtained as pure products from commercially available building blocks by SPPS and subsequent preparative HPLC purification, ensuring the reproducibility of the biological results and the transferability of the reagents to other laboratories. The possibility to fine-tune their activity by modifying their amino acid sequence and the nature of the lipidated core opens the door to further optimization for targeted applications.

[0152] DNA transfection results For example, a test to transfect plasmid DNA using a similar peptide dendrimer and a large plasmid DNA (about 9Kbp) encoding CRISPR-Cas9-GFP. The Cas9-2A-GFP vector allows for the co-expression of Cas9 and GFP from the same mRNA. Since the self-cleaving peptide 2A allows for the release of the GFP reporter in the cytoplasm after expression, the transfection efficiency was determined by tracking the expression of GFP (green fluorescent protein) in various cell lines upon exposure to the corresponding plasmid complex with the dendrimer at N / P=5. The transfection efficiency was tracked by fluorescence-activated cell sorting, and the percentage of successfully transfected cells that appeared fluorescent was counted. The results showed that DNA transfection was more efficient with the composition (KL) combined with a hydrophobic core according to the examples shown in the table below.8 (KKL) 4 (KKL) 2 It was shown that the dendrimer branching of K is required for DNA transfection. Most notably, dendrimer-mediated transfection is found to be much more efficient than that of the reference reagent L2000. We found that the presence of a cysteine ​​residue is favorable as one of the hydrophobic residues in the dendrimer core, and found that the corresponding disulfide-bridged dimeric dendrimers had similar transfection capabilities. However, cysteine-containing monomers were also transfected in the presence of reducing agents that prevented dimer formation, indicating that disulfide bridge formation is not necessary for DNA transfection. Mechanistic studies on aggregation and nanoparticle formation indicate that similar to siRNA-transfecting dendrimers, DNA-transfecting dendrimers also undergo self-aggregation and nanoparticle formation as a necessary step for DNA transfection. [Table 3]

[0153] Further examples of DNA transfection are given below. [Table 4]

[0154] example MH02((KL) 8 (KKL) 4 (KLL) 2 K.K.(C 18 ) was obtained from a CEM Liberty Blue synthesizer as a foamy colorless solid after preparative RP-HPLC (6.0 mg, 0.86 μmol, 2%). Analytical RP-HPLC:t R = 3.44 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 234 H 453 N 57 O 37 calc. / obs.4654.53 / 4654.56 [M+H +.

Chem.

[0155] MH03((KL) 8 (KKL) 4 (KLL) 2 KK(C 20 )) was obtained as a foamy colorless solid after manual synthesis and preparative RP-HPLC (4.1 mg, 0.59 μmol, 1%). RP-HPLC for analysis: t R = 4.01 min (from 100% A to 100% D in 7.5 min, λ = 214 nm). MS(ESI+): C 236 H 457 N 57 O 37 calc. / obs. 4682.56 / 4682.70 [M+H + .

Chem.

[0156] MH04((KL) 8 (KKL) 4 (KLL) 2 KK(C 22 )) was obtained as a foamy colorless solid after manual synthesis and preparative RP-HPLC (3.2 mg, 0.46 μmol, 1%). RP-HPLC for analysis: t R = 4.20 min (from 100% A to 100% D in 7.5 min, λ = 214 nm). MS(ESI+): C 238 H 461 N 57 O 37 calc. / obs. 4710.59 / 4710.73 [M+H + .

Chem.

[0157] MH05((KL) 8 (KKL) 4 (KLL) 2 KK(C 24) was obtained after manual synthesis as a foamy colorless solid after preparative RP-HPLC (4.5 mg, 0.6 μmol, 1%). Analytical RP-HPLC:t R = 4.40 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 240 H 465 N 57 O 37 calc. / obs.4738.63 / 4738.63[M+H + ] [ka]

[0158] MH06((KL) 8 (KKL) 4 (KLL) 2 K.K.(C 26 ) was obtained after manual synthesis as a foamy colorless solid after preparative RP-HPLC (3.1 mg, 0.44 μmol, 1%). Analytical RP-HPLC:t R = 4.67 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 242 H 469 N 57 O 37 calc. / obs.4766.66 / 4766.65[M+H + ]. [ka]

[0159] MH07((KL) 8 (KKL) 4 (KLL) 2 K.K.(C 28 ) was obtained after manual synthesis as a foamy colorless solid after preparative RP-HPLC (3.1 mg, 0.44 μmol, 1%). Analytical RP-HPLC:t R = 4.81 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 244 H 473 N 57 O 37calc. / obs.4794.69 / 4794.70[M+H + ]. [ka]

[0160] MH12((KL) 8 (KKL) 4 (KLL) 2 K.K.(C 14 )K(C 14 ) was obtained after manual synthesis as a foamy colorless solid after preparative RP-HPLC (2.3 mg, 0.32 μmol, 1%). Analytical RP-HPLC:t R = 4.27 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 250 H 483 N 59 O 39 calc. / obs.4936.76 / 4936.90[M+H + ]. [ka]

[0161] MH13((KL) 8 (KKL) 4 (KLL) 2 K.K.(C 16 )K(C 16 ) was obtained from a Biotage Initiator+Alstra synthesizer as a foamy colorless solid after preparative RP-HPLC (26.8 mg, 3.68 μmol, 4%). Analytical RP-HPLC:t R = 4.68 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 254 H 491 N 59 O 39 calc. / obs.4992.83 / 4992.82[M+H + ]. [ka]

[0162] MH14((KL) 8 (KKL) 4 (KLL) 2 KK(C 18 )K(C 18 )) was obtained as a foamy colorless solid after manual synthesis and preparative RP-HPLC (2.4 mg, 0.3 μmol, 1%). RP-HPLC for analysis: t R = 5.08 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS(ESI+): C 258 H 499 N 59 O 39 calc. / obs. 5048.89 / 5048.91 [M+H + .

Chemical formula

[0163] MH13D1((kl) 8 (KKL) 4 (KLL) 2 KK(C 16 )K(C 16 )) was obtained from a CEM Liberty Blue synthesizer as a foamy colorless solid after preparative RP-HPLC (16.2 mg, 2.23 μmol, 2%). RP-HPLC for analysis: t R = 3.81 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS(ESI+): C 254 H 491 N 59 O 39 calc. / obs. 4992.83 / 4992.85 [M+H + .

Chemical formula

[0164] DMH13((kl) 8 (kkl) 4 (kll) 2 kk(C 16 )k(C 16) was obtained from a CEM Liberty Blue synthesizer as a foamy colorless solid after preparative RP-HPLC (128.4 mg, 17.65 μmol, 7%). Analytical RP-HPLC:t R = 4.67 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 254 H 491 N 59 O 39 calc. / obs.4992.83 / 4992.84[M+H + ]. [ka]

[0165] MH18((KL) 8 (KKL) 4 (KLL) 2 KLLLL) was obtained from a CEM Liberty Blue synthesizer as a foamy colorless solid after preparative RP-HPLC (52.7 mg, 7.53 μmol, 8%). Analytical RP-HPLC: R = 3.29 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 234 H 451 N 59 O 39 calc. / obs.4712.51 / 4712.52[M+H + ]. [ka]

[0166] MH19((KL) 8 (KKL) 4 (KLL) 2 KLLLLL) was obtained after manual synthesis and then preparative RP-HPLC as a foamy colorless solid (4.7 mg, 0.66 μmol, 1%). Analytical RP-HPLC: R = 3.36 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 240 H 462 N 60 O 40calc. / obs.4825.60 / 4825.61[M+H + .

Chem.

[0167] MH22((KL) 8 (KKL) 4 (KLL) 2 KFFF) was obtained as a foamy colorless solid after manual synthesis and preparative RP-HPLC (13.7 mg, 1.96 μmol, 3%). RP-HPLC for analysis: t R = 3.30 min (from 100% A to 100% D in 7.5 min, λ = 214 nm). MS(ESI+): C 237 H 434 N 58 O 38 calc. / obs.4701.38 / 4701.40[M+H + .

Chem.

[0168] MH25((KL) 8 (KKL) 4 (KLL) 2 KWWWW) was obtained as a foamy colorless solid after manual synthesis and preparative RP-HPLC (4.5 mg, 0.6 μmol, 1%). RP-HPLC for analysis: t R = 3.27 min (from 100% A to 100% D in 7.5 min, λ = 214 nm). MS(ESI+): C 254 H 446 N 62 O 40 calc. / obs.5004.49 / 5004.50[M+H + .

Chem.

[0169] MH26((KL) 8 (KKL) 4 (KLL) 2KWWWWW) was obtained as a foamy colorless solid after manual synthesis and preparative RP-HPLC (1.7 mg, 0.23 μmol, 1%). Analytical RP-HPLC: tR=3.33 min (100% A to 100% D in 7.5 min, λ=214 nm). MS(ESI+): C 265 H 457 N 65 O 40 calc. / obs.5190.57 / 5190.63[M+H + ]. [ka]

[0170] MH40((KL) 8 (KKL) 4 (KLL) 2 KNleNleNleNle) was obtained after manual synthesis as a foamy colorless solid after preparative RP-HPLC (5.1 mg, 0.73 μmol, 1%). Analytical RP-HPLC:t R = 3.34 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 234 H 451 N 59 O 39 calc. / obs.4712.51 / 4712.53[M+H + ]. [ka]

[0171] MH46((KL) 8 (KKL) 4 (KKL) 2 KLLLL) was obtained after manual synthesis and then preparative RP-HPLC as a foamy colorless solid (8.5 mg, 1.17 μmol, 2%). Analytical RP-HPLC: R = 3.11 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 234 H 453 N 61 O 39 calc. / obs.4742.53 / 4742.55[M+H + ]. [ka]

[0172] MH47((KL) 8 (KLL) 4 (KLL) 2 KLLLL) was obtained after manual synthesis and then preparative RP-HPLC as a foamy colorless solid (10.1 mg, 1.56 μmol, 2%). Analytical RP-HPLC: R = 4.15 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 234 H 447 N 55 O 39 calc. / obs.4652.47 / 4652.49[M+H + ]. [ka]

[0173] DMH18((kl) 8 (kkl) 4 (kll) 2 kllll) was obtained from a Biotage Initiator+Alstra synthesizer as a foamy colorless solid after preparative RP-HPLC (73.6 mg, 10.5 μmol, 11%). Analytical RP-HPLC: R = 3.27 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 234 H 451 N 59 O 37 calc. / obs.4712.51 / 4712.52[M+H + ]. [ka]

[0174] kl3,2-fff((kl) 8 (kkl) 4 (kll) 2kfff) was obtained from a CEM Liberty Blue synthesizer as a foamy colorless solid after preparative RP-HPLC (22.7 mg, 3.25 μmol, 5%). Analytical RP-HPLC: R = 3.27 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 237 H 434 N 58 O 38 calc / found 4701.38 / 4701.41[M+H + ]. [ka]

[0175] D4((kl) 8 (kkl) 4 (kll) 2 klllc) was obtained from a CEM Liberty Blue synthesizer after preparative RP-HPLC as a foamy colorless solid (129.5 mg, 18.5 μmol, 21%). Analytical RP-HPLC: R = 2.97 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 231 H 445 N 59 O 39 S calc / found 4702.44 / 4702.48[M+H + ]. [ka]

[0176] kl3,2-nlenlenlenle ((kl) 8 (kkl) 4 (kll) 2 knlenlenlenle) was obtained from a CEM Liberty Blue synthesizer after preparative RP-HPLC as a foamy colorless solid (19.3 mg, 2.76 μmol, 4%). Analytical RP-HPLC: R = 3.31 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 234H 451 N 59 O 39 calc / found 4712.51 / 4712.53[M+H + ]. [ka]

[0177] Kl3,2-aocaocaoc((kl) 8 (kkl) 4 (kll) 2 kaocaocaoc) was obtained from a CEM Liberty Blue synthesizer as a foamy colorless solid after preparative RP-HPLC (8.4 mg, 1.21 μmol, 2%). Analytical RP-HPLC:t R = 3.48 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 234 H 452 N 58 O 38 calc / found 4683.52 / 4683.52[M+H + ]. [ka]

[0178] kl3,2-aocaocaocaoc((kl) 8 (kkl) 4 (kll) 2 Kaocaocaocaoc) was obtained from a CEM Liberty Blue synthesizer as a foamy colorless solid after preparative RP-HPLC (13.3 mg, 1.87 μmol, 3%). Analytical RP-HPLC: R = 3.56 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 242 H 467 N 59 O 39 calc / found 4824.64 / 4824.63[M+H + ]. [ka]

[0179] MH44((RL) 8 (KRL) 4 (KLL) 2 KLLLL) was obtained from a Biotage Initiator+Alstra synthesizer as a foamy colorless solid after preparative RP-HPLC (25.9 mg, 3.53 μmol, 5%). Analytical RP-HPLC: R = 3.40 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 234 H 450 N 82 O 40 calc. / obs.5048.59 / 5048.61[M+H + ]. [ka]

[0180] rl3-kl2-nlenlenlenle((rl) 8 (kkl) 4 (kll) 2 knlenlenlenle) was obtained from a CEM Liberty Blue synthesizer after preparative RP-HPLC as a foamy colorless solid (1.6 mg, 0.22 μmol, 1%). Analytical RP-HPLC: R = 3.40 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 234 H 451 N 75 O 39 calc / found 4936.56 / 4936.56[M+H + ]. [ka]

[0181] kl3-rl2-nlenlenlenle((kl) 8 (krl) 4 (kll) 2knlenlenlenle) was obtained from a CEM Liberty Blue synthesizer after preparative RP-HPLC as a foamy colorless solid (17.3 mg, 2.43 μmol, 5%). Analytical RP-HPLC: R = 3.35 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 234 H 451 N 67 O 39 calc / found 4824.53 / 4824.53[M+H + ]. [ka]

[0182] kl3,2-nlenle1-C 16 C 16 ((kl) 8 (kkl) 4 (Knlenle) 2 kk(C 16 )k(C 16 ) was obtained from a CEM Liberty Blue synthesizer as a foamy colorless solid after preparative RP-HPLC (12.7 mg, 1.69 μmol, 3%). Analytical RP-HPLC:t R = 4.47 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 254 H 491 N 59 O 39 calc / found 4992.8251 / 4992.8246[M+H + ]. [ka]

[0183] ((KL) 8 (KKL) 4 (KKL) 2 K.K.(C 18 )K(C 18 )) was obtained after preparative RP-HPLC as a foamy colorless solid (2.5 mg, 0.49 μmol, 1%). Analytical RP-HPLC:tR =5.06 points (7.5 points, 100% A, 100% D, λ = 214nm). MS(ESI+):C 258 H 501 N 61 O 39 calc / found 5078.91 / 5078.90[M+H + ].

change

[0184] ((KL) 8 (KKL) 4 (KKL) 2 KK (オレイン acid) C) was separated and obtained as a vesicular colorless solid after RP-HPLC (2.3 mg, 0.48 μmol, 2%). Analytical RP-HPLC:t R =3.70 points (7.5 points, 100% A, 100% D, λ = 214nm). MS(ESI+):C 237 H 458 N 60 O 38 S calc / found 4785.55 / 4785.56[M+H + ].

change

[0185] ((KL) 8 (KKL) 4 (KKL) 2 KK(C 18 )C), fractionate and obtain vesicular colorless solid after RP-HPLC (2.5 mg, 0.52 μmol, 2%). Analytical RP-HPLC:t R =3.83 points (7.5 points, 100% A, 100% D, λ = 214nm). MS(ESI+):C 237 H 460 N 60 O 38 S calc / found 4787.56 / 4787.57[M+H + ].

change

[0186] ((KH) 8 (KKH) 4 (KKL) 2 K(oleic acid) was obtained as a foamy colorless solid after preparative RP-HPLC (2.7 mg, 0.50 μmol, 3%). Analytical RP-HPLC: R = 4.48 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 258 H 449 N 85 O 39 calc / found 5362.58 / 5362.58[M+H + ]. [ka]

[0187] ((KL) 8 (KKL) 4 (KLL) 2 K.K.(C 22 )) was obtained after preparative RP-HPLC as a foamy colorless solid (11.4 mg, 0.24 μmol, 8%). Analytical RP-HPLC:t R = 4.18 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 238 H 463 N 59 O 37 calc / found 4740.62 / 4740.63[M+H + ]. [ka]

[0188] ((KL)8(KKL)4(KKL)2KK(C 16 )K(C 16 )) was obtained after preparative RP-HPLC as a foamy colorless solid (39.1 mg, 5.18 μmol, 8%). Analytical RP-HPLC:t R=4.35 points (7.5 points, 100% A, 100% D, λ = 214nm). MS(ESI+):C 254 H 493 N 61 O 39 calc / found 5022.85 / 5022.86[M+H + ].

change

[0189] ((kl)8(kkl)4(kkl)2kk(C 16 )k(C 16 )), fractionate and obtain vesicular colorless solid after RP-HPLC (24.9 mg, 3.3 μmol, 5%). Analytical RP-HPLC:t R =4.45 points (7.5 points, 100% A, 100% D, λ = 214nm). MS(ESI+):C 254 H 493 N 61 O 39 calc / found 5022.85 / 5022.88[M+H + ].

change

[0190] ((kl) 8 (kkl) 4 (kkl) 2 kk(C 16 )c), fractionate and obtain vesicular green solid after RP-HPLC (6.04 mg, 0.13 μmol, 3%). Analytical RP-HPLC:t R =3.65 points (7.5 points, 100% A, 100% D, λ = 214nm). MS(ESI+):C 235 H 456 N 60 O 38 S calc / found 4759.53 / 4759.54[M+H + ].

change

[0191] ((kl) 8 (kkl) 4 (kkl) 2 kk(C 18 )c) was obtained after preparative RP-HPLC as a foamy colorless solid (3.1 mg, 0.65 μmol, 3%). Analytical RP-HPLC:t R = 3.73 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 237 H 460 N 60 O 38 S calc / found 4787.56 / 4787.57[M+H + ]. [ka]

[0192] ((kl) 8 (kkl) 4 (kkl) 2 kk (oleic acid) c) was obtained as a foamy solid after preparative RP-HPLC (5.85 mg, 0.12 μmol, 4%). Analytical RP-HPLC: R = 3.41 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 237 H 458 N 60 O 38 S calc / found 4786.00 / 4785.55[M+H + ]. [ka]

[0193] ((kl) 8 (kkl) 4 (kkl) 2 kk(C18)a) was obtained as a foamy green solid after preparative RP-HPLC (114 mg, 24 μmol, 18%). Analytical RP-HPLC: R = 3.66 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 237 H460 N 60 O 38 calc / found 4755.59 / 4755.59.

change

[0194] ((kl) 8 (kkl) 4 (kkl) 2 kk(C18)), and after fractionation by RP-HPLC, a vesicular green solid was obtained (20 mg, 4.1 μmol, 23%). Analytical RP-HPLC:t R =3.76 points (7.5 points, 100% A, 100% D, λ = 214nm). MS(ESI+):C 234 H 455 N 59 O 37 calc / found 4684.55 / 4684.56.

change

[0195] ((kl) 8 (kkl) 4 (kkl) 2 kk(C16)). After fractionation and RP-HPLC, a vesicular green solid was obtained (22 mg, 4.6 μmol, 16%). Analytical RP-HPLC:t R =3.60 points (7.5 points, 100% A, 100% D, λ = 214nm). MS(ESI+):C 232 H 451 N 59 O 37 calc / found 4656.52 / 4656.52.

change

[0196] ((rl) 8 (krl) 4 (krl) 2kk(C18)) was obtained after preparative RP-HPLC as a foamy green solid (13 mg, 2.5 μmol, 15%). Analytical RP-HPLC: R = 3.90 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 234 H 455 N 87 O 37 calc / found 5076.64 / 5076.63. [ka]

[0197] ((rl) 8 (krl) 4 (krl) 2 kk(C16)) was obtained after preparative RP-HPLC as a foamy green solid (14 mg, 2.7 μmol, 11%). Analytical RP-HPLC: R = 3.69 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 232 H 451 N 87 O 37 calc / found 5048.61 / 5048.60. [ka]

[0198] ((rl) 8 (krl) 4 (krl) 2 kk(C 18 )c) was obtained after manual synthesis as a foamy colorless solid after preparative RP-HPLC (26.6 mg, 5.1 μmol, 14%). Analytical RP-HPLC:t R = 3.99 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 237 H 460 N 88 O 38 S calc. / obs.5179.65 / 5179.66. [ka]

[0199] ((rl) 8 (krl) 4 (krl) 2 kk(C 16 )c) was obtained after manual synthesis as a foamy colorless solid after preparative RP-HPLC (9 mg, 1.6 μmol, 4%). Analytical RP-HPLC:t R = 3.77 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 235 H 456 N 88 O 38 S calc. / obs.5151.62 / 5151.62 [ka]

[0200] ((rl) 8 (krl) 4 (krl) 2 kk(C18)a) was obtained after preparative RP-HPLC as a foamy green solid (25 mg, 4.9 μmol, 12%). Analytical RP-HPLC: R = 3.90 min (100% A to 100% D in 7.5 min, λ = 214 nm). MS (ESI+): C 237 H 460 N 88 O 38 calc / found 5147.68 / 5147.68. [ka] [Sequence List Free Text]

[0201] Sequence Listing 1 <223> siRNA targeting GAPDH (sense strand) Sequence Listing 2 <223> siRNA targeting GAPDH (antisense strand) Sequence Listing 3 <223> Control siRNA for the sense strand Sequence Listing 4 <223> Control siRNA for the sense strand

Claims

1. A peptide dendrimer of formula 1, (D ３ ) ８ -(B ３ -D ２ ) ４ -(B ２ -D １ ) ２ -B １ -Z(1) Z is -X(Y １ )(1a), -X(Y ２ ) X(Y ２ )(1b), -X(Y ３ )Cys(1c), -HP(1d), or -X(Y ４ ) Ala, During the ceremony, X is selected from Lys, Orn, DAB, DAP, Glu or Asp, which is linked via its N-terminus to B １ and is connected to Y through its side chain, Y １ When X is Lys, Orn, DAB or DAP, -C(=O)-C ｖ -alkyl, -C(=O)-C ｖ -alkenyl, v is selected from 15 to 27 when X is Lys; when X is Orn, 18 to 28; 19 to 29 when X is DAB; When X is DAP, it is 20 to 30; or Y １ When X is Glu or Asp, it is represented by -(NH)-C ａ -alkyl, -(NH)-C ａ -alkenyl, -(N)-(C ｄ -alkyl) ２ Or -(N)-(C ｄ -alkenyl) ２ is selected from the group consisting of the sum of d is 19 to 29 when X is Glu; The sum of d is 20 to 30 when X is Asp, a is 19 to 29 when X is GIu; or a is 20 to 30 when X is Asp; Each Y ２ When X is Lys, Orn, DAB or DAP, -C(=O)-C ｗ -Alkyl and -C(=O)-C ｗ -alkenyl, and the sum of w is When X is Lys, it is 18 to 36; When X is Orn, 20 to 38; 22 to 40 when X is DAB; When X is DAP, it is 24 to 42; or Y ２ When X is Glu or Asp, it is represented by -(NH)-C ｂ -alkyl, -(NH)-C ｂ -alkenyl, -(N)-(C ｅ -alkyl) ２ Or -(N)-(C ｅ -alkenyl) ２ is selected from the group consisting of the sum of e and b is 22 to 40 when X is Glu; the sum of e and b is 24 to 42 when X is Asp; Y ３ and Y ４ When X is Lys, Orn, DAB or DAP, -C(=O)-C ｘ -Alkyl and -C(=O)-C ｘ -alkenyl, x is independently selected from 15 to 21 when X is Lys, 16 to 22 when X is Orn, 17 to 23 when X is DAB; When X is DAP, it is 18 to 24; Y ３ When X is Glu or Asp, it is represented by -(NH)-C ｃ -alkyl, -(NH)-C ｃ -alkenyl, -(N)-(C ｆ -alkyl) ２ Or -(N)-(C ｆ -alkenyl) ２ is selected from the group consisting of the sum of f is 17 to 23 when X is Glu; the sum of f is 18 to 24 when X is Asp; c is 17 to 23 when X is Glu; c is 18 to 24 when X is Asp; HP is a hydrophobic peptide consisting of 3 to 5 hydrophobic amino acids, each B is Lys; Each D, independently of the other D, is a dipeptide consisting of one hydrophobic amino acid and one cationic amino acid (HC or CH), a dipeptide consisting of two cationic amino acids (CC), or a dipeptide consisting of two hydrophobic amino acids (HH). Peptide dendrimers.

2. Z is -X(Y 1 )(1a), -X(Y 2 )X(Y 2 )(1b), -X(Y 3 )Cys(1c) or -HP(1d); Y 1 is selected from -C(═O)-C v -alkyl, -C(═O)-C v -alkenyl when X is Lys, Orn, DAB or DAP, and v is When X is Lys, it is 17 to 27; when X is Orn, it is 20 to 28; when X is DAB, it is 21 to 29; When X is DAP, it is 22 to 30; or Y 1 when selected from -(NH)-C a -alkyl, -(NH)-C a -alkenyl, -(N)-(C d -alkyl) 2 or -(N)-(C d -alkenyl) 2 when X is Glu or Asp; the sum of d is 21 to 29 when X is Glu; the sum of d is 22 to 30 when X is Asp; a is 21 to 29 when X is GIu; or a is 22 to 30 when X is Asp; Each Y 2 is independently selected from -C(═O)-C w -alkyl and -C(═O)-C w -alkenyl when X is Lys, Orn, DAB or DAP, and the sum of w is: 22 to 36 when X is Lys; When X is Orn, 24 to 38; 26 to 40 when X is DAB; When X is DAP, it is 28 to 42; or Y 2 when selected from -(NH)-C b -alkyl, -(NH)-C b -alkenyl, -(N)-(C e -alkyl) 2 or -(N)-(C e -alkenyl) 2 when X is Glu or Asp; the sum of e and b is 26 to 40 when X is Glu; the sum of e and b is 28 to 42 when X is Asp; Y 3 and Y 4 are independently selected from -C(=O)-C x -alkyl and -C(=O)-C x -alkenyl, when X is Lys, Orn, DAB or DAP, where x is 15 to 17 when X is Lys, 16 to 18 when X is Orn, 17 to 19 when X is DAB, When X is DAP, it is 18 to 20; Y 3 when selected from -(NH)-C c -alkyl, -(NH)-C c -alkenyl, -(N)-(C f -alkyl) 2 or -(N)-(C f -alkenyl) 2 when X is Glu or Asp; the sum of f is 17 to 19 when X is Glu; the sum of f is 18 to 20 when X is Asp; c is 17 to 19 when X is GIu; c is 18 to 20 when X is Asp; The peptide dendrimer according to claim 1.

3. 3. The peptide dendrimer of claim 1 or 2, wherein X is Lys.

4. The peptide dendrimer according to any one of claims 1 to 3, wherein the hydrophobic peptide is composed of hydrophobic amino acids independently selected from Cys, Leu, Trp, Phe, Nle (norleucine), and Aoc (aminooctanoic acid).

5. The peptide dendrimer of claim 4, wherein the hydrophobic peptide consists of hydrophobic amino acids independently selected from Leu, Phe, Trp, Nle and Aoc.

6. The peptide dendrimer according to any one of claims 1 to 5, wherein the hydrophobic peptides are composed of identical hydrophobic amino acids.

7. The peptide dendrimer according to any one of claims 1 to 6, wherein the C-terminus is a carboxamide.

8. A peptide dendrimer according to any one of claims 1 to 7, comprising the formula: For HH, CH and HC, a.D １ The hydrophobic amino acids in are selected from Leu and Nle, １ The cationic amino acid in is Lys, b.D ２ The hydrophobic amino acid in is Leu, ２ the cationic amino acid in is selected from Lys, Arg and His; C.D ３ The hydrophobic amino acid in is Leu, ３ the cationic amino acid in is selected from Lys, Arg and His; And in the case of CC, D ２ Or D ３ the cationic amino acids in are selected from Lys, Arg and His, and at least one C is His; Peptide dendrimers.

9. A peptide dendrimer according to any one of claims 1 to 8, comprising the formula: D １ is a dipeptide (HH) consisting of two hydrophobic amino acids, and D ２ and D. ３ are dipeptides (CH) each consisting of one cationic amino acid and one hydrophobic amino acid, or Each D １ , D ２ and D. ３ is a dipeptide (CH) consisting of one cationic amino acid and one hydrophobic amino acid; Peptide dendrimers.

10. A peptide dendrimer according to any one of claims 1 to 9, wherein the amino acids are independently selected from (L)-amino acids and (D)-amino acids.

11. The peptide dendrimer according to any one of claims 1 to 10, wherein the peptide is homochiral.

12. 1. A method for transfecting a cell (ex vivo), comprising: a. Providing a transfection mix comprising a peptide dendrimer according to any one of claims 1 to 11 and a nucleic acid; b. Contacting the cells with the transfection mix to obtain transfected cells. The method includes:

13. 13. The method of claim 12, wherein the transfection mix is ​​prepared by mixing a first solution comprising a peptide dendrimer according to any one of claims 1 to 11 with a second solution comprising the nucleic acid.

14. The method according to any one of claims 12 to 13, wherein the nucleic acid is DNA or RNA.

15. The method of claim 14, wherein the nucleic acid is a circular DNA (plasmid / vector), a linear DNA (cDNA), or a linear RNA (siRNA, saRNA, miRNA, mRNA, long RNA).

16. The method according to any one of claims 12 to 15, wherein the solvent of the first and / or second solution is water or a cell culture medium.

17. The method according to any one of claims 12 to 16, wherein the final concentration of the nucleic acid in the transfection mix is ​​between 10 nM and 200 nM.

18. The method of claim 17, wherein the final concentration of the nucleic acid in the transfection mix is ​​20 to 100 nM.

19. The method according to any one of claims 12 to 18, wherein the final concentration of the peptide dendrimer in the transfection mix is ​​between 1.5 μg / ml and 30 μg / ml.

20. The method of claim 19, wherein the final concentration of the peptide dendrimer in the transfection mix is ​​between 3 μg / ml and 15 μg / ml.

Images