Method for the preparation of highly branched polypeptides
Description
The present invention relates to a method for the preparation of branched polypeptides, to the polypeptides obtainable with said method and the use of the polypeptides as therapeutic or diagnostic agents.
Branched or dendritic oligomers and polymers of L-lysine have attracted attention for the construction of peptide antigens111 and DNA delivery systems.121 Functionalization of the peripheral a- and e-amino groups of small dendritic oligo(L-lysine) scaffolds with one or different epitopes results in so-called multiple antigen peptides (MAPs), which are of interest for the development of vaccines and as diagnostic tools.111 Linear-dendritic di- and triblock copolymers of poly(ethylene glycol) and L-lysine dendrons have been reported, which can form water-soluble complexes with plasmid DNA.[21 These block copolymers show promise as carriers for gene delivery, since they combine the DNA complexing properties of cationic polymers with the water-solubility, non-immunogenicity and biocompatibility of poly(ethylene)glycol.
So far, dendritic polypeptides have been generally prepared in a stepwise fashion, either in homogeneous solution or via solid-phase synthesis, from appropriately N", Ne-diprotected building blocks.""31 Provided the synthesis and purification are performed carefully, this strategy will afford well- defined and perfectly monodisperse dendrimers/dendrons. Synthesis and purification, however, are usually laborious, and a large number of steps is required to obtain high molecular weight material.
Some earlier publications have reported the use of poly(L-Iysine) or copolymers of L-lysine and other σ-amino acids as macroinitiators to
prepare graft polymers, which were termed multichain polypeptides.171 However, in all of these reports only a single grafting step was described. Further, dendritic graft poly(ethylene imine)151, poly(ethylene imine)-poly(2- ethyl-2-oxazoline) copolymers,151 poly(styrene),161 poly(butadiene)181 and poly(styrene)-poly(n-butyl methacrylate) copolymers[91have beendescribed.
Therefore, it was an object of the invention to provide a method for the simple preparation of branched polypeptides, by means of which, in particular, high-molecular compounds can be obtained in a few reaction steps.
According to the invention this object is achieved by a method for the preparation of branched polypeptides, in particular, highly branched or dendritic graft polypeptides, comprising the steps: (i) providing an cr-amino acid N-carboxyanhydride (NCA) having a protected primary amino group, which is different from the -NH2 substituent, (ii) contacting the NCA with an initiator, thereby initiating a ring- opening polymerization of the NCA, (iii) removing the protective group,
(iv) contacting the polypeptide obtained with an α-amino acid N- carboxyanhydride (NCA) having a protected primary amino group, thereby performing a ring-opening polymerization of the NCA, and (v) repeating steps (iii) and (iv) at least once.
The method of the invention allows to repeat the deprotection/ring-opening polymerization cycle several times and to prepare highly branched or dendritic graft polypeptides. These polypeptides can be obtained in only a few reaction steps and are therefore an interesting alternative for the perfect polypeptide dendrimers that are prepared according to the state of the art in a time-consuming step-by-step fashion1171.
The principles of the inventive method are explained in detail in the following using the amino acid lysine as an example. However, it is to be noted that the inventive method can be performed using any amino acid having the characteristics as outlined in the claims.
The strategy, which is also outlined in Scheme 1 , is based on the ring- opening (co)polymerization of an σ-amino acid N-carboxyanhydride (NCA) having a protected primary amino group or of a mixture of NCAs in which at least one of the NCAs contains a protected primary amino group. If necessary, the side chains of the other NCA comonomers are protected with suitable orthogonal protective groups. Preferred is the use of two orthogonally N-protected σ-amino N-carboxyanhydrides (NCAs), e.g. two orthogonally Ne-protected L-lysine N-carboxyanhydrides. In this case one of the monomers contains a primary amine group masked with a temporary protective group (TPG) which can be removed under relatively mild conditions. Using e.g. lysine NCAs, the TPG group is fixed at the e-NH2 group. Functional groups present at the side chains of the other NCA comonomers, which could interfere with the polymerization, are protected with a permanent protective group (PPG) that is stable and is not removed under the conditions applied for the deprotection of the TPG. Using e.g. lysine NCAs, the e-NH2 group of the other L-lysine monomer is masked with a permanent protective group (PPG) which is stable and not removable under the conditions applied for the removal of the TPG.
Thus, for the realization of the synthesis depicted in Scheme 1 , preferably two orthogonally protected NCAs, in particular, Ne-protected L-lysine NCAs are elected. To explore the feasibility of the concept exemplarily Ne-(tert- butoxycarbonyI)-L-lysine NCA(BOC-lys NCA) and Ne-(benzyloxycarbonyl)-L- lysine NCA (Z-lys NCA) were used. Both, BOC-lys NCA[10] and Z-lys NCA11 11 are easily prepared following published procedures. Whereas the BOC groups are smoothly removed under the action of CF3COOH, Z-lys requires HBr/AcOH or hydrogenation for effective deprotection.1121 Thus, BOC-lys
NCA appears to be a suitable temporary protected monomer which, after deprotection, can act as a branching point and initiate the ring-opening copolymerization of a successive series of grafts.
In the first synthetic step preferably a primary amine, e.g. n-hexylamine, is used to initiate the ring-opening copolymerization of the two NCA's to prepare the core of the targeted polypeptide. Removal of the TPG generates a number of primary amine groups which can act as an initiator to graft a first generation of peptide arms onto the core. Repetition of this ring-opening polymerization/deprotection cycle yields highly branched, or dendritic graft polypeptides. Following commonly accepted conventions, a dendritic graft polymer of generation n - 1 (G = n - 1 ) is obtained after n deprotection/grafting cycles. [5,6]
The PPG can be removed in the very last step of the synthesis. In a preferred embodiment the PPG is removed to yield a polypeptide having free amino groups. In sum, the method according to the invention is a new synthetic route which yields highly branched polypeptides with high molecular weights, e.g. > 10 kDa in a relatively small number of reaction steps.
Thus, the invention relates to a new synthetic strategy for the preparation of highly branched or dendritic graft polypeptides. The method involves a repetitive sequence of NCA ring-opening polymerization and deprotection steps. Preferably, appropriately Ne-protected L-lysine derivatives are used as branching points. In this way dendritic graft polypeptides, e.g.- poly(L- lysine)s having a molecular weight of up to ~ 50 kDa were prepared in only 8 steps. Such highly branched polypeptides are of interest as carriers for gene delivery and for the development of synthetic vaccines.
Generally, the method of the invention allows to prepare branched polypeptides, especially, highly branched or dendritic graft polypeptides.
As starting material, an σ-amino acid N-carboxyanhydride (NCA = α-amino acid N-carboxyanhydride) having a protected primary amino group is used. The NCA used according to the invention is derived from an α-amino acid having an α-annino group forming the cyclic N-carboxyanhydride group and at least one further amino group which is/are different from the α-NH2 substituent and which is/are protected by a protective group. The NCA can be used alone or as a mixture, e.g. a mixture of NCAs in which at least one of the NCAs contains a protected primary amino group. If desired, the side chains of the other NCA comonomers are protected with suitable orthogonal protective groups. According to the present invention a mixture of two or more α-amino acid N-carboxyanhydrides (NCAs) can be used. In this case functional groups at the side chains of the different monomers can be blocked with orthogonal protective groups. This allows to deprotect the functional group, in particular, a NH2-group, that acts as the branching point without affecting the other protective groups. Preferably, NCAs are used having orthogonally protected primary amino groups which are blocked by protective groups which are removable under distinct different conditions. In another preferred embodiment a mixture of NCAs is used, wherein comonomers are employed having side chains which do not interfere with or take part in a NCA ring-opening polymerization, such comonomers are e.g. Ala-NCA or Leu-NCA, having hydrocarbon side chains. By selecting the type and amount of copolymer the chain length of each generation can be controlled.
It is also possible to use a mixture of two or more, preferably three or more NCAs, wherein at least two different NCAs derived from the same or different α-amino acid are employed having different TPGs. In this case sequential grafting to different, predefined sites of the polypeptide can be achieved.
In a preferred embodiment a first NCA with a first protective group and a second NCA with a second protective group being different from the first
protective group are provided, wherein the first and the second NCAs are derived from the same amino acid. In this case the same amino group is blocked by different protective groups in each of the two NCAs. However, it is also possible to use different NCAs derived from different amino acids.
Suitable amino acids, from which the NCAs can be derived and which can act as branching points include, but are not limited to lysine or ornithine.
According to the inventive method the NCA or the mixture of NCAs is contacted with an initiator to start a ring-opening polymerization of the NCAs. To this end, any initiator which starts the NCA ring-opening polymerization can be used. Such initiators are well-known and available. Compounds having one or more amino groups are preferred. Suitable initiators include alkoxide anions, pyridines, C, to C30 amines, e.g. tertiary, secondary or primary amines, in particular, C., to C12 alkylamines, C, to C12 alkyl diamines, amino acids, amino acids having at least two amino groups and dendrimers having free primary amino groups. In addition, any polymer having one or more primary amine groups, e.g. substituted poly(styrene), poly(ethy!ene oxide), saccharides etc. can also be used as initiator and result in hybrid materials. In an especially preferred embodiment the initiator used is an amino acid, from which an NCA according step (i) is derived, e.g. lysine or ornithine. However, it is also possible to initiate the ring-opening polymerization by another amino acid. Suitable dendrimers having free primary amino groups comprise, for example, polypropyleneimine (PPI) and polyamidoamine (PAMAM) dendrimers.
Basically all groups which, under the ring-opening polymerization conditions of the NCAs, prevent the further amino group present in the amino acids employed from participating in said polymerization reaction can be used as protective group. Thus, the protective groups are stable under the polymerization conditions employed and are not split off. Suitable protective groups, for example, are trifluoroacetyl, t-butoxycarbonyl,
benzyloxycarbonyl, 9-fluorenylmethoxycarbonyl and 1 -methyI-1 -(3,5- dimethoxyphenyl) ethoxycarbonyl. In case the primary amine substituted NCAs that act as branching point are copolymerized with NCAs containing functional groups that could interfere with the polymerization, it is essential according to the invention that protected NCAs are used, in particular, at least two NCAs which contain different protective groups each. Particularly preferred is the use of t-butoxycarbonyl (BOC) as a temporary protective group (TPG) and benzyloxycarbonyl (Z) as a permanent protective group (PPG) . When using said pair of orthogonally protective groups it is possible to split off the BOC protective group under conditions where the Z protective group is stable and not removed. In this way a particular type of protective group can be specifically split off in a partial deprotection step, while the protective group orthogonal thereto remains in the intermediate polypeptide formed. Preferably, a PPG is removed only after the final polymerization cycle.
The molar ratio of the first NCA having a temporary protected primary amino group to second NCAs having a permanent protected primary amino group or containing substituents that do not interfere with the NCA polymerization preferably ranges from 1 :20 to 5: 1 , more preferably from 1 : 10 to 1 : 1 and most preferably from 1 :5 to 1 :2.
The initiator, in particular, free amino groups of the initiator cause a ring opening of the α-amino acid N-carboxyanhydrides (NCAs), thereby forming a peptide bond and generating again a free amino group, namely the α- amino group derived from the NCA. Thus, a ring-opening polymerization of the NCAs proceeds. After this polymerization according to the invention protective groups are removed. The protective group removed in this step is termed temporary protective group herein. Protective groups orthogonal to this temporary protective group are not removed in step (iii), thus resulting in a polypeptide having free amino groups as well as side chain functionalities that do not interfere with the ring-opening polymerization
process or which are protected with suitable permanent protective groups, e.g. permanent protected amino groups. The free amino groups are derived from amino groups initially being protected by a temporary protective group. By contacting the thus obtained polypeptide again with an NCA having a temporary protected primary amino group or mixtures thereof with suitable orthogonally protected NCA comonomers, if necessary, the ring- opening polymerization of the NCAs is started again. In this second reaction step the polypeptide obtained in step (iii) functions as an initiator for the ring-opening polymerization of the N-carboxyanhydrides (NCAs), since it contains free amino groups.
In step (iv) the same NCA or NCA mixture as employed in step (i) can be used. However, it is also possible to use a NCA or NCA mixture different from that of step (i), e.g. a mixture having a different ratio of constituents or different types of constituents. In step (iv) new polypeptide chains are grafted onto to the polypeptide formed in the steps before by reacting the amino groups being protected in the first reaction steps by a temporary protective group with further α-amino acid N-carboxyanhydride (NCA) after removal of the temporary protective groups.
The chain length of linear segments can be easily adjusted by the molar ratio of α-amino acid N-carboxyanhydride (NCA) employed to free primary amine groups. This molar ratio preferably ranges from 50: 1 to 5: 1 , more preferably from 30: 1 to 10: 1 in steps (ii) or/and (iv). During synthesis the molar ratio of NCA to free primary amine groups does not necessarily have to be the same during each grafting cycle but can be varied resulting in dendritic graft polypeptides with arm lengths that are different for each generation. Deprotection/grafting can be repeated as often as desired. Preferably, steps (iii) and (iv) are repeated one to ten times, more preferably three to eight times. However, it is also possible to repeat these steps twenty or more times depending on the desired size of the polypeptide to be formed. With the method of the invention it is easily
possible to prepare high-molecular branched polypeptides having a molecular weight of greater than 10 kDa, in particular, greater than 30 kDa in only a few reaction steps. By using more repetitions of the deprotection and polymerization/grafting steps it is, however, also possible to prepare polypeptides having a molecular weight of greater than 100 kDa or even greater than 1000 kDa.
The branching/grafting density of the dendritic graft polypeptides depends on the molar ratio between the temporary protected primary amine substituted NCA that acts as branching point and the other NCA comonomers. The highest branching density, one graft per amino acid residue, is obtained when the temporary protected primary amine substituted NCA is homopolymerized. By copolymerization of this monomer with (mixtures of) other NCAs that are protected with permanent protective groups or carry substituents that do not interfere with the NCA polymerization, the branching density can be adjusted by the molar ratio of the monomers. Typically, the temporary protected primary amine substituted NCA that acts as the branching point constitutes between 1 0- 60 mol% of the total monomer mixture. The branching density does not need to be fixed throughout the synthesis, but can be varied from generation to generation.
With the method of the invention fully protected, partially protected or totally deprotected polypeptides can be obtained. If the last step performed is a step (iv), fully protected polypeptides are obtained which contain at least one kind of protective group. If the polypeptide contains functional groups that could have interferred with the NCA ring-opening polymerization, preferably at least two types of orthogonal protective groups will be present. Partially protected polypeptides are obtained, if the last step performed is a step (iii), i.e. specifically temporary protective groups are removed and permanent protective groups that are eventually present remain in the molecule. It is especially preferred, however, to
remove all protective groups, i.e. both temporary protective groups and permanent protective groups that are eventually present, by treating the polypeptide under respective conditions after completion of the polymerization steps, i.e. when the polypeptide has the desired size. In this way a totally deprotected polypeptide is obtained which has free amino groups.
The free amino groups of the dendritic graft polypeptide can be further functionalized to provide the molecules with a desired functionality. For example, the amino groups can be functionalized with one or more epitopes, resulting in multiple antigen peptides (MAPs) . The free amino groups can also be used to attach synthetic polymers that possess a suitable end-group, e.g. a carboxylic acid end-functionalized poly(styrene) or poly(ethylene oxide), or can be modified with functional groups that can initiate radical or ring-opening polymerization reactions of vinyl-type and cyclic monomers, respectively. In this way, hybrid structures composed of a highly branched peptide core and a shell of a synthetic polymer become accessible. Such molecules are soluble in organic solvents and are of interest for storage, transport and release of small molecules and catalysis.
A further object of the present invention are branched polypeptides, in particular, dendritic graft polypeptides obtainable by the method described above. Such polypeptides preferably contain units derived from α-amino acids having one or more primary amino groups in addition to the α-amino groups, wherein the additional amino groups are partially free and partially constitute branching points. By the above-described manufacturing method polypeptides can be prepared, wherein some of the amino groups serve as branching points (those provided with temporary protective groups) and other functional groups, in particular, amino groups can be obtained in free form after completion of the polymerizations (those, to which permanent protective groups were bound) . The ratio of free amino groups to amino
groups constituting branching points preferably ranges from 20: 1 to 1 :5, more preferably from 10: 1 to 1 : 1 .
These polypeptides are of interest as vaccines and as diagnostic tools. Further, they can be used as carriers for gene delivery, since they have DNA-compIexing properties. The invention, therefore, also relates to a therapeutic or/and diagnostic agent containing a polypeptide as described above.
The invention is further illustrated by the following Example and the Figures, wherein
Scheme 1 shows the preparation of dendritic graft polypeptides via a repetitive sequence of NCA ring-opening polymerization and deprotection steps. (TPG = temporary protective group, PPG
= permanent protective group) .
Fig.1 is part of the 700 MHz 1H-NMR spectrum of the different dendritic graft poly(L-lysine)s recorded in D2O, and
Fig.2 shows GPC traces of the different dendritic graft poly(Ne
(benzyloxycarbonyl)-L-lysine)s.
Example 1
The preparation of the dendritic graft poly(L-lysine)s was carried out in N,N-dimethylformamide (DMF), using a monomer mixture containing 20 mole% BOC-lys NCA and an initial molar ratio of NCA to primary amine groups of 20 throughout all polymerization steps."31 The deprotection/grafting cycle was repeated three times to ultimately afford a 2nd generation (G = 2) dendritic graft poly(Z-lys) . Treatment of this dendritic graft poly(Z-lys) with HBr/AcOH to remove the PPGs yielded the
corresponding water-soluble polypeptide in quantitative yield. The polypeptides were analyzed with 1H-NMR spectroscopy and gel permeation chromatography (GPC). Figure 1 shows an expansion of the methine region of the 1H-NMR spectrum of the different dendritic graft poly(L-lysine)s.[14'15i Comparison of the integral of the signal at 0.9 ppm, which is due to the methyl group of the n-hexylamine initiator, with the sum of the integrals of the resonances representing the different methine protons yields the total number average degree of polymerization of the polypeptides. The integral of the peak at ~ 4.30 ppm, which can be assigned to the methine proton of the terminal α-amino acid of a branch, represents the number of grafting sites of that particular generation, or the number of arms of the next higher generation. Combination of these different integrals provides information about the number average degree of polymerization and the BOC-lys content of the arms that have been introduced in a particular generation. The results of the 1H-NMR experiments are summarized in Table 1 .
able 1 : Isolated yields and the results of the H-NMR characterization of the dendritic graft poly(L-lysine)s.
olymer Yield'1' 1 H-NMR integrals12' DPn (3) M DPn/arm(b) Total number of BOC-lys
(%) (-) (-) (g/mol) (-) branching content'6' points (-) (mol %) (1 ) Isolated
A B D (A + B)/C yi e ld , afte r ore 74 18 0 1 1 18 20 2670 20 6 25 precipitation
10 60 84 6 6 n.d. 15 96 12400 13 14 10 a nd freeze- i1 62 204 17 13.5 n.d. 16 235 30200 10 25 8 drying,
!2 54 300 22.5 24.5 n.d. 13 347 44600 5 n.d. n.d. (2) I nteg rals were etermined relative to that of the triplet of the methyl group of the initiator moiety at 0.8 ppm, which was set equal to 3.
3) Total number average degree of polymerization of the polypeptide.
4) Total number average molecular weight of the polypeptide.
5) Number average degree of polymerization of the arms added in a particular generation.
6) BOC lys content of the arms added in a particular generation.
The 1H-NMR spectra do not only give information about the number average degree of polymerization and the composition of the polypeptides, but also provide evidence for their branched topology. This becomes evident when comparing the integral of the methine protons of the terminal α-amino acid residues (labelled "C" in Fig.1 ) with that of the signals of the other methine protons (labelled "A" and "B" in Fig.1 ) . The ratio (A + B)/C would continuously increase with increasing degree of polymerization for a linear polypeptide. Table 1 , however, shows that for the polypeptides prepared according to the route outlined in Scheme 1 , this number is relatively small and slowly decreases from 1 8 to 1 3. This observation is consistent with a highly branched topology and a gradual decrease in the number average degree of polymerization per arm with increasing generation.
The data presented in Table 1 and Fig.1 clearly prove the feasibility of the synthetic strategy outlined in Scheme 1 and show that dendritic graft poly(L-lysine)s with number average molecular weights up to ~ 50 kDa can be prepared in only 8 steps. The 1 H-NMR data are supported by the results of the GPC analysis,1161 which are shown in Fig.2. All GPC traces possess a monomodal character and the average molecular weights (see fable 2), which can be calculated from the chromatograms rapidly increase with increasing generation. However, since the elution times were converted to molecular weights using a calibration curve constructed with linear poly(ethylene oxide) standards, the GPC experiments can only provide qualitative support.
Table 2: The results of the GPC analysis of the different poly(Z-lys) dendritic graft polymers.
Polymer Mn (g/mol) W Mw (g/mol)( ) Mw/Mn (-)
.Core 2550 4600 1.8 GO 18800 31900 1.7 G1 60700 102200 1.7 G2 212900 409900 1.9
(1) Number-average molecular weight.
(2) \Λ/eight~average molecular weight.
Table 1 , however, also indicates that the synthesis is not completely free of side reactions. This is illustrated by the continuous decrease both in the number average degree of polymerization of the branches and in the BOC- lys content in these arms with increasing generation number. Although a 20-fold excess of NCA with respect to the number of grafting sites was used throughout the synthesis, the number average degree of polymerization per branch was found to decrease from ~ 20 for the core to ~ 5 for the G2 dendritic graft poly (L-lysine). This decrease in the DPn is accompanied by a decrease in the isolated yields. These findings might be interpreted in terms of a reduced accessibility of the grafting sites for the NCA monomers with increasing generation number.
References and Notes
[1] See e.g.: (a) D.N.Posnett, H.McGrath, J.P.Tam, J.Biol.Chem.1988,
263, 1719-1725. (b) J.P.Tam, Proc.Natl.Acad.Sci. USA 1988, 85, 5409-5413. (c) J.P.Tam, J. Immunol. Methods 1996, 196, 17-32.
[2] (a) T.M. Chapman, G.L.Hillyer, E.J.Mahan, K.A.Schaffer,
J.Am.Chem.Soc. 1994, 116, 11195-11196. (b) J.S.Choi, E.J.Lee,
Y.H.Choi, Y.J.Yeong, J.S.Park, Bioconjugate Chem. 1999, 10, 62-
65. (c) J.S.Choi, D.K.Joo, C.H.Kim, K.Kim, J.S.Park, J.Am.Chem.Soc.2000, 122, 474-480.
[3] The first polypeptide dendrimers were prepared by stepwise coupling and deprotection of Nσ,Nε-di-(tert-butoxycarbonyl)-L-lysine: (a)
R.G.Denkewalter, J.F.Kolc, W.J.Lukasavage (Allied Corp.), U.S. US
4,410,688, 1983 [Chem.Abstr. 1984, 100 103907p]. (b) R.G.Denkewalter, J.F.Kolc, W.J.Lukasavage (Allied Corp.), U.S. US
4,289,872, 1981 [Chem.Abstr. 1985, 102 79324q]. (c)
S.M.Aharoni, N.S.Murthy, Polym.Commun. 1983, 24, 132-136.
[4] For reviews on the synthesis and polymerization of α-amino acid N- carboxyanhydrides, see e.g.: (a) H.R.Kricheldorf, α-Amino acid N- carboxyanhydrides and related heterocycles, Springer-Verlag: Berlin
1987. (b) T.J.Deming, Adv. Mater. 1997, 9, 299-311. (c)
T.J.Deming, J.Polym.Sci. Part A: Polym.Chem. 2000, 38, 3011-
3018.
[5] D.A.Tomalia, D.M.Hedstrand, M.S.Ferritto, Macromolecules 1991, 24, 1435-1438.
[6] M.Gauthier, M.Moller, Macromolecules 1991, 24, 4548-4553.
[7] See eg.: (a) M.Sela, S.Fuchs, R.Arnon, Biochem. J. 1962, 85, 223-
235. (b) A.Yaron, A.Berger, Biochim.Biophys.Acta 1965, 107, 307-
332. (c.) M.Sakamoto,
J.Po/ym.Sci.:Po/ym. Chem. Ed. 1978, 16, 1107-1122. (d) M.Sakamoto, Y.Kuroyanagi, R.Sakamoto,
J.Polym.Sci.: Polym.Chem. Ed. 1978, 16, 2001-2017.
[8] M.A.Hempenius, W.Michelberger, M.Mδller, Macromolecules 1997, 30, 5602-5605.
[9] R.B.Grubbs, C.J. Hawker, J.Dao, J . M . J . Frechet, Angew.Chem.lnt.Ed.Engl. 1997, 36, 270-272. [10] R.Wilder, S.Mobashery, .Orβf.CΛetr?. 1992, 57, 2755-2756.
[11] D.S.Poche, M.J.Moore, J.L.Bowles, Synth. Commun. 1999, 29, 843-854.
[12] T.W.Greene, P.G.M.Wuts, Protective Groups in Organic Synthesis, 3rd edition, John Wiley & Sons, Inc., New York 1999. [13] Polymerizations were conducted in dry DMF by adding the initiator (either neat, in case of n-hexylamine, or dissolved in DMF, in case of the macroinitiators) to a DMF solution containing the appropriate amount of monomer (~ 0.2 g/mL). After 5 days, the polymers were isolated by precipitation in Et2O. BOC groups were removed by treating the protected precursor polypeptide with CF3COOH for 15-
30 min at room temperature, followed by precipitation in NaHCO3 (aq) and freeze-drying.
[14] 1H-NMR experiments were performed at room temperature on a Bruker WS700 spectrometer. D2O was used as the solvent, and the residual proton signal was taken as internal standard.
[15] 1H-NMR resonances were assigned according to earlier published data on L-lysine oligomers: W.N.E. van Dijk-Wolthuis, L.van de
Water, P. van de Wetering, M.J.van Steenbergen, J.J.Kettenes-van den Bosch, W.J.W.Schuyl, W.E.Hennink, Macromol.Chem.Phys. 1997, 198, 3893-3906.
[16] GPC experiments were performed at 60°C with a setup consisting of a Waters 510 pump and a series of three PSS SDV columns with pore sizes of 500, 104 and 106 A. A 0.1 M solution of LiBr in DMF was used as the mobile phase and sample elution was monitored with simultaneous UV-Vis and refractive index detection. Elution times were converted to molecular weights using a calibration curve
constructed from narrow polydispersity poly(ethylene oxide) standards. [17] An alternative approach for the preparation of highly branched polypeptides involves end-capping the NCA polymerization with the active ester of an Nσ,Nf-diprotected lysine derivative. Removal of the protective groups generates two primary amine groups at the peptide's N-terminus that can be used to initiate the polymerization of a next generation of arms, see: A.C.Birchall, M. North, Chem. Commun. 1 998, 1 335. This synthetic approach, however, is fundamentally different from the present method of invention and is also much less flexible, e.g. in varying the grafting density of the polymer.