A General Strategy toward Synthesis of Well-Defined Polypeptides with Complex Chain Topologies

Open AccessCCS ChemistryRESEARCH ARTICLE7 Dec 2022A General Strategy toward Synthesis of Well-Defined Polypeptides with Complex Chain Topologies Lei Li, Jie Cen, Wenjin Li, Wenhao Pan, Yuben Zhang, Hao Yin, Jinming Hu and Shiyong Liu Lei Li CAS Key Laboratory of Soft Matter Chemistry, Department of Polymer Science and Engineering, School of Chemistry and Materials Science, University of Science and Technology of China, Hefei, Anhui 230026 Google Scholar More articles by this author , Jie Cen CAS Key Laboratory of Soft Matter Chemistry, Department of Polymer Science and Engineering, School of Chemistry and Materials Science, University of Science and Technology of China, Hefei, Anhui 230026 Google Scholar More articles by this author , Wenjin Li CAS Key Laboratory of Soft Matter Chemistry, Department of Polymer Science and Engineering, School of Chemistry and Materials Science, University of Science and Technology of China, Hefei, Anhui 230026 Google Scholar More articles by this author , Wenhao Pan CAS Key Laboratory of Soft Matter Chemistry, Department of Polymer Science and Engineering, School of Chemistry and Materials Science, University of Science and Technology of China, Hefei, Anhui 230026 Google Scholar More articles by this author , Yuben Zhang CAS Key Laboratory of Soft Matter Chemistry, Department of Polymer Science and Engineering, School of Chemistry and Materials Science, University of Science and Technology of China, Hefei, Anhui 230026 Google Scholar More articles by this author , Hao Yin Mass Spectrometry Lab, University of Science and Technology of China, Hefei, Anhui 230026 Google Scholar More articles by this author , Jinming Hu *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] CAS Key Laboratory of Soft Matter Chemistry, Department of Polymer Science and Engineering, School of Chemistry and Materials Science, University of Science and Technology of China, Hefei, Anhui 230026 Google Scholar More articles by this author and Shiyong Liu *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] CAS Key Laboratory of Soft Matter Chemistry, Department of Polymer Science and Engineering, School of Chemistry and Materials Science, University of Science and Technology of China, Hefei, Anhui 230026 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.022.202101704 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail The precise synthesis of polypeptides with varying chain topologies has attracted significant interest due to their diverse functional applications. However, conventional polymerization of environment-susceptible N-carboxylic anhydride monomers using primary amine initiators suffers from inevitable side reactions, which greatly compromise chain-end fidelity, thereby disabling access to polypeptides with high-order architectures. Herein, we have developed a general and robust strategy based on normal amine mechanism for the synthesis of various topological polypeptides via polymerization of moisture-insensitive and air-tolerant N-phenoxycarbonyl-functionalized α-amino acid precursors using primary amine hydrochloride as initiators. This strategy enabled the synthesis of a block, star, star-block, brush-type, and multiblock (co)polypeptides with desired sequences, predictable molecular weights, low polydispersity, and high-fidelity chain ends even under open-air conditions. Remarkably, the robustness of this approach has been exemplified by the precise synthesis of well-defined photoresponsive decablock copolypeptides, exhibiting a spontaneous morphological evolution from polymersomes to lamellae nanostructures upon light irradiation. This work provides a general and reliable tool for synthesizing polypeptides with varying topologies, shedding light on the development of synthetic polypeptide-based materials. Download figure Download PowerPoint Introduction Natural proteins implement many biological functions, taking advantage of their precise chain sequences and hierarchical structures.1–3 Synthetic polypeptides mimic proteins and exhibit unique advantages such as biocompatibility, biodegradability, and stimuli-responsiveness, which have found broad applications in targeted drug delivery,4,5 gene therapy,6 catalysis,7,8 antibacterial agents,9 and so on. For example, poly(ethylene glycol)-b-poly(l-glutamic acid) (PEG-b-PLGA) block copolymer with 7-ethyl-10-hydroxy-camptothecin conjugates (NK012) and micellar nanoparticles of cisplatin/PEG-b-PLGA (NC-6004) have been evaluated in clinical trials.10 In addition, Glatiramer acetate (Copaxone), a random copolypeptide consisting of four amino acids (glutamic acid, lysine, alanine, and tyrosine), has been approved for treating multiple sclerosis.4 In addition to linear polypeptides, synthetic polypeptides with various chain topologies have drawn increasing attention due to their specific three-dimensional structures that allow the fabrication of hierarchically assembled nanomaterials.11,12 Amphiphilic PEG-b-polypeptide,13 polypeptides brushes,14,15 star (block) polypeptides,16,17 and multiblock copolypeptides14 are typical topological polypeptides widely designed for biochemical applications.9 Especially multiblock copolypeptides potentially present more versatile self-assembled architectures due to tunable secondary structures, adapting protein-like biological functions.11 However, controlled synthesis of polypeptides with complex chain topologies remains a considerable challenge due to the presence of multiple side reactions of conventional α-amino acid N-carboxylic anhydride (NCA) polymerizations.18,19 Notably, although the polymerization of NCA monomers has been used extensively to prepare polypeptides, the labile nature of NCA derivatives to moisture and heat has been the major obstacle for its scalable production and practical utilization.18,20 On the other hand, the activated monomer mechanism (AMM) also contributes to side reactions due to amine basicity (pKa ∼10–12), generating impurities that remain in the final product, leading to uncertainties in terminal functionalities. Concurrent normal amine mechanism (NAM)/AMM are associated with primary amine initiators; thus, pose severe challenges toward the synthesis of copolypeptides with varying topologies such as star, star block, brush-type, and multiblock chain architectures, in which high-fidelity terminal functionality needs to be maintained throughout the entire intermediate stages.21,22 To improve the controllability of NCA polymerizations, numerous initiators, including primary ammonium salts,23–25 transition metal22,26,27 and rare earth complexes,19,28 trimethylsilyl amine and sulfide derivatives,29,30 and hexamethyldisilazane (HMDS)31 have been developed. Meanwhile, much effort has been devoted to improving operational procedures (e.g., high-vacuum technique-assisted ultrapure NCA polymerizations)32–34 and optimized reaction conditions (N2 flowing,35 low-temperature,36–38 and optimized pressure38). Hadjichristidis and co-workers developed accelerated amine mechanism via monomer activation (AAMMA)39,40 and hydrogen bonding-assisted organocatalysis41 strategies to achieve fast and controlled ring-opening polymerization (ROP) of NCAs under mild conditions. Later on, they reported cocatalyst-free controlled organocatalytic ROP of NCAs facilitated by fluorinated alcohol.42 Recently, hydroxyl compounds were used to initiate NCA polymerization in the presence of catalysts such as 1,1,3,3-tetramethylguanidine (TMG).43,44 However, TMG catalysis could also lead to the AMM reaction pathway and unwanted side reactions.45 Liu et al.46,47 utilized Li/Na/KHMDS and tetraalkylammonium carboxylate48 to initiate superfast NCA polymerizations, facilely conducted under open air. Quite recently, Cheng and co-workers reported superfast NCA polymerizations by taking advantage of the local cooperative milieu from neighboring α-helices of growing peptide chains11,49–51 and crown ether catalyzed NCA accelerated polymerization.52 Note that this strategy is selective to solvents with low dielectric constants (e.g., CH2Cl2) rather than polar solvents such as dimethylformamide (DMF) or dimethylacetamide (DMAc), which likely limits the types of NCA monomers with specific pendant functional groups.53 Despite considerable progress, controlled NCA polymerization is still hampered by the unstable nature of NCA monomers and intrinsically concurrent side reactions.18,20 To address these drawbacks, moisture-insensitive activated urethane (i.e., N-aryloxycarbonyl) derivatives of α-amino acids have been utilized for polypeptide synthesis.54–56 However, the polymerization process is still associated with multiple side reactions (e.g., AMM, solvent-mediated initiation and termination, and cyclization) when amine initiators are used.54,55,57 Recently, we developed a novel strategy to synthesize well-defined polypeptides under NCA monomer-starved conditions using N-phenoxycarbonyl-functionalized α-amino acid (NPCA) as the monomer instead of unstable NCA and primary amine hydrochloride as the initiator, enabling controlled synthesis of polypeptides with predetermined molecular weights (MWs), narrow polydispersity (Đ), and high-fidelity terminal functionalities without noticeable side reactions.57 Herein, encouraged by the excellent control over NPCA polymerizations, we further extended the scope of the NCA monomer-starved strategy to synthesize topological polypeptides with varying (block, star, star-block, brush-type, and multiblock) chain architectures (Figure 1a). Using primary amine hydrochloride initiators, the resulting polypeptides had predetermined MWs, low polydispersity (Đ ∼1.1), and high-fidelity terminal functionality even under open-vessel conditions (Figure 1). Impressively, it was possible to synthesize photoresponsive multiblock (up to decablock) copolypeptides via sequential NPCA monomer addition under open air. Moreover, this sequence-defined decablock copolypeptides exhibited unique fluorescence emission and morphological transitions from polymeric vesicles to lamella upon light irradiation. This study extends the NCA monomer-starved strategy to precisely synthesize polypeptides with different chain topologies and paves the avenue for functional applications of polypeptide materials. Figure 1 | Facile synthesis of well-defined polypeptides with different chain topologies. (a) Schematics of controlled synthesis of polypeptides with well-defined chain topologies using moisture-stable NPCA precursors as monomers and protonated primary amines as initiators. This strategy renders the chain propagations more favorable over side reactions, thereby minimizing undesired contaminants and producing well-defined copolypeptides with varying chain topologies (e.g., block, star, star-block, brush-type, and multiblock), predetermined MWs, low polydispersity, and high-fidelity terminal functionality. (b) The chemical structures of NPCA precursors used for the synthesis of polypeptides with varying chain topologies in this work. Download figure Download PowerPoint Experimental Methods Synthesis of decablock copolypeptide via one-pot sequential polymerization using PEG45-NH3+Cl− as initiator For the synthesis of PEG45-b-PNBK5-b-PPhe5-b-PTrp5-b-PNBK5-b-PPhe5-b-PTrp5-b-PNBK5-b-PPhe5-b-PTrp5 (PEG-KFWKFWKFW) decablock copolypeptide, Nε-(o-nitrobenzyloxycarbonyl)-l-lysine (NBK) precursor (50.0 mg, 0.112 mmol) was placed in a glass vial and dissolved in DMAc (449 μL) to reach an [M]0 of 0.25 M. PEG45-NH3+Cl− (49.3 mg, 0.0224 mmol) was then added. After completion of polymerization for the first block, Phe precursor (24.9 mg, 0.087 mmol) in DMAc (0.25 M) was charged into the reaction vial for the second block With continuous stirring at 70 °C until complete polymerization of the newly charged Phe precursor had been attained. Following similar procedures, consecutive monomer feedings (in DMAc, 0.25 M) were charged into the reaction vial in the order described below: Trp precursor (24.2 mg, 0.075 mmol), NBK precursor (29.6 mg, 0.066 mmol), Phe precursor (17.2 mg, 0.060 mmol), Trp precursor (17.9 mg, 0.055 mmol), NBK precursor (22.7 mg, 0.051 mmol), Phe precursor (13.5 mg, 0.047 mmol), and Trp precursor (14.3 mg, 0.044 mmol) were consecutively charged into the reaction vial upon completion of the polymerization for the previous block (>99% NPCA conversion was obtained and validated by 1H NMR). Upon each chain extension, an aliquot of the reaction mixture was sampled out for further characterization by gel permeation chromatography (GPC) analysis. The final decablock copolypeptide of PEG-KFWKFWKFW was obtained as a pale-yellow solid by precipitation into an excess of diethyl ether and drying in a vacuum oven overnight. The detailed experimental methods are available in the Supporting Information. Results and Discussion Synthesis of PEG-NH3+Cl− initiator with high-fidelity primary ammonium functionality and polymerizations of NPCA precursors PEG-polypeptide amphiphilic block copolymers (BCPs) have been synthesized conventionally via ROP of NCA monomers using PEG-NH2 as the initiator.36 The ROP process was postulated to follow the NAM pathway, though side reactions associated with AMM; amine-relevant nucleophilicity might also occur.18,20 Dated back to 5 years ago, we attempted to synthesize amphiphilic PEG-polypeptide BCPs via NCA polymerization using PEG-NH2 as an initiator. To our surprise, we found that all commercially available PEG-NH2 starting materials are of unreliable quality ( Supporting Information Figure S1).58 GPC analysis typically gave bimodal or multimodal peaks; matrix-assisted laser desorption ionization-time of flight mass spectroscopy (MALDI-TOF MS) analysis revealed the presence of impurities not corresponding to the specified chemical structure. Moreover, we found that even when stored at −20 °C in a glovebox, the quality of PEG-NH2 was persistently worsened. This storage instability could be ascribed to the incompatibility between native amine and PEG backbones due to amine basicity and nucleophilicity.58,59 However, the use of a high-purity PEG-NH2 initiator is crucial for the synthesis of well-defined amphiphilic copolypeptides, especially in an instance when the fidelity of terminal functionality is a concerning issue. We hypothesized that the preparation of PEG derivatives with terminal ammonium salt should be more stable because both amine basicity and nucleophilicity were suppressed upon protonation. To this end, protonated PEG-NH2 (PEG-NH3+Cl−) was synthesized ( Supporting Information Scheme S1 and Figure 2a). We intentionally chose carbamate instead of ester linkage to install ammonium functionality due to the following considerations: (1) carbamate linkage is more stable than ester bond; (2) esterification is typically an equilibrium reaction, and ester linkage is incompatible with amine moieties.58 Specifically, PEG45-OH was converted into PEG45-Boc via reaction with Boc-containing isocyanate derivative in situ generated from acyl azide moieties upon heating (Figure 2 and Supporting Information Scheme S1 and Figures S2 and S3).60 The degree of functionalization of PEG45-Boc was quantitative (>99.9%), as evidenced by 1H NMR and MALDI-TOF MS data (Figures 2b and 2d and Supporting Information Figure S4). After deprotection with HCl in MeOH solution, PEG-NH3+Cl− bearing primary ammonium functionality was obtained, as confirmed by 1H NMR, GPC, electrospray ionization-mass spectrometry (ESI-MS), and electrospray ionization-ion mobility spectrometry-mass spectrometry (ESI-IMS-MS; Figures 2b and 2c and Supporting Information Figure S5). Figure 2 | Characterization data of synthetic intermediates of air-stable PEG45-NH3+Cl− initiator and as-synthesized amphiphilic PEG-polypeptides via NPCA polymerizations. (a) Synthetic routes employed for the preparation of PEG-NH3+Cl− and PEG45-b-polypeptides block copolypeptides. (b) 1H NMR spectra, recorded for PEG45-OH, PEG45-Boc, PEG45-NH3+Cl−, and PEG45-b-PNBK7 in DMSO-d6. (c) GPC elution traces, recorded for PEG45-OH, PEG45-Boc, PEG45-NH3+Cl−, and PEG45-b-PNBK7 using DMF as the eluent. The inset in (c) shows the macroscopic image of the PEG45-NH3+Cl− initiator. (d) MALDI-TOF MS spectra, recorded for PEG45-OH, PEG45-Boc, and PEG45-NH3+Cl−. (e) MALDI-TOF MS spectrum of amphiphilic PEG45-b-PNBK7, synthesized via polymerization of NBK precursor ([M]0 = 0.25 M, DMAc, 70 °C) using PEG45-NH3+Cl− as the initiator. (f) GPC traces recorded for amphiphilic PEG-polypeptide BCPs (PEG45-b-PBocDab, PEG45-b-PBocO, and PEG45-b-PBocK) initiated by PEG45-NH3+Cl−. (g) GPC traces recorded for amphiphilic PEG45-b-PNBOn synthesized using PEG45-NH3+Cl− initiator at varying [M]0/[I]0 molar ratios. (h) GPC traces of amphiphilic PEG45-b-P(NBOx-co-Phe1-x)n BCPs synthesized via random copolymerization of two types of NPCA precursors ([NBO]0/[Phe]0 = 5/5 and 7/3; target DPs of polypeptide block: 80 and 115) using PEG45-NH3+Cl− as the initiator. All polymerizations were conducted at [M]0 = 0.25 M in DMAc and 70 °C unless otherwise noted. Download figure Download PowerPoint GPC elution traces of PEG45-OH, PEG45-Boc, and PEG45-NH3+Cl− shown in Figure 2c are symmetric and monomodal without any discernible shoulder peaks. In sharp contrast, commercially available PEG45-NH2, PEG45-NH3+Cl− exhibited only one single set of peaks in MALDI-TOF MS spectra, corresponding to [M+Na]+; the interval between each adjacent peak was ∼44.026 Da, in agreement with the –CH2CH2O– repeating units (Figure 2d). In addition, ESI-MS and two-dimensional (2D) ESI-IMMS spectra further consolidated the high purity of PEG45-NH3+Cl− (>99%) ( Supporting Information Figures S5c and S5d). Unexpectedly, PEG45-NH3+Cl− was highly stable, as revealed by the unchanged ESI-MS data for the sample after 1-year storage under ambient conditions ( Supporting Information Figure S6). In addition, PEG23-NH3+Cl− and PEG113-NH3+Cl− were synthesized and characterized ( Supporting Information Scheme S1). GPC results showed that the target products displayed monomodal distribution with narrow Đ (<1.08), high purity, and high-fidelity terminal ammonium functionality ( Supporting Information Figures S7–S9). With high-fidelity PEG45-NH3+Cl− in hand, we could, in principle, use its base-neutralized deprotonated counterpart as the initiator for either NCA or NPCA polymerizations. However, the introduction of organic bases would inevitably implicate the AMM pathway into the polymerization process,18,36 whereas the use of inorganic bases might lead to heterogeneity in reaction media. Inspired by the work of Schlaad and co-workers,23–25 who used a primary ammonium salt to achieve improved control of NCA polymerizations, we attempted to use PEG45-NH3+Cl− directly as an initiator to polymerize NPCA precursors. At elevated temperatures, the protonation-deprotonation equilibrium of amine residues would shift to the deprotonated state (i.e., initiation and chain-growth), and NPCA precursors in situ could transform into their corresponding NCA monomers.54,55,61–63 The combination of primary ammonium initiators and NPCA precursors might also solve their stability issues and moisture/heat-sensitive limitations of NCA monomers. In the trial polymerizations, NBK was chosen as a model NPCA precursor. The reaction was conducted at 70 °C using PEG45-NH3+Cl− as the initiator and DMAc as the solvent (entry 13, Table 1). The degree of polymerization (DP) of the poly(Nε-o-nitrobenzyloxycarbonyl)-l-lysine) (PNBK) block was calculated to be ∼7 by 1H NMR analysis, in line with the initial feed ratio (Figure 2b). GPC elution trace revealed an apparent number average MW (Mn) of ∼5.1 kDa with Đ of ∼1.04 (Figure 2c), whereas the PEG45-NH3+Cl− macroinitiator had an Mn of ∼1.9 kDa, indicating the formation of well-defined PEG45-b-PNBK7 BCPs. MALDI-TOF MS of PEG45-b-PNBK7 revealed relatively narrow polydispersity, and the corresponding peak assignments supported the NAM polymerization pathway (Figure 2e). Specifically, the theoretical m/z value ([M+Na]+, 4335.66 Da) of PEG45-b-PNBK7 was quite close to the experimental result (4335.89 Da), while the peak intervals concurred well with the NBK repeating unit (307.12 Da). More importantly, the MALDI-TOF MS pattern was unexpectedly clean, and no unassignable MS shoulder peaks corresponding to impurities and side reactions could be discerned. Table 1 | Controlled Synthesis of Amphiphilic PEG-Polypeptide BCPs from NPCA Precursors Initiated by PEG45 -NH3+Cl− Entry Monomer [M]0/[I]0 Temp.(°C) ReactionTime (h) Mn,NMR(kDa)a Conv.(%)a DPa Mn,GPC(kDa)b Đc 1 BocO 10 70 24 4.3 >99 10 4.9 1.03 2 BocO 13 70 36 5.2 >99 14 5.3 1.03 3 BocO 30 70 48 8.4 97.7 29 7.7 1.06 4 BocK 10 70 24 4.4 >99 10 5.4 1.03 5 BocK 30 70 48 9.0 >99 30 8.3 1.06 6 BocK 100 70 72 24.1 97.3 96 23.4 1.17 7 BocDab 10 70 24 4.2 >99 10 4.5 1.03 8 NBO 15 70 36 6.3 93.3 14 5.4 1.03 9 NBO 45 70 48 15.6 >99 46 13.7 1.08 10 NBO 75 70 72 25.0 >99 78 23.1 1.09 11 NBO 100 70 72 32.9 >99 105 29.8 1.14 12 NBO 200 70 96 62.8 >99 207 60.1 1.15 13 NBK 7 70 24 4.3 >99 7 5.1 1.04 14 NBO/Phe 80 (7/3) 70 72 22.2 >99 56/25 20.3 1.09 15 NBO/Phe 80 (5/5) 70 72 19.9 >99 40/41 18.7 1.07 16 NBO/Phe 115 (7/3) 70 80 29.6 >99 75/37 28.7 1.13 17 NBO/Phe 115 (5/5) 70 80 27.7 >99 58/60 26.9 1.14 aCalculated from 1H NMR spectra. bDetermined by GPC using refractive index (RI) detector (eluent: DMF, 10 mM LiBr; 1 mL/min). cPolydispersity index (Mw/Mn) determined by GPC. Synthesis of amphiphilic polypeptide BCPs using PEG45-NH3+Cl− initiator The synthesis and self-assembly of amphiphilic polypeptides have gained increasing attention due to their potential applications in various fields such as drug delivery, tissue engineering, and gene vector,13 in which the PEG segment is typically used as the hydrophilic components.64–66 As demonstrated in the trial polymerizations (Figures 2b–2e), NPCA polymerization using PEG45-NH3+Cl− as the macroinitiator afforded amphiphilic copolypeptides. However, due to the presence of side reactions, multiple polymerization mechanisms, and the poor quality of commercial PEG-amine starting materials, conventional synthesis of PEG-polypeptide BCPs with relatively short polypeptide chain length and high high-fidelity terminal functionalities needed further optimization.57 Thus, we fabricated an array of amphiphilic polypeptide BCPs via controlled polymerization of NPCA monomers using PEG45-NH3+Cl− as the initiator. Three Boc-protected NPCA precursors, BocO, BocDab, and BocK, were polymerized at a fixed [M]0/[I]0 of 10. The final DPs agreed quite well with the initial feed ratios in all cases, and GPC results showed monomodal elution peaks with narrow Đ (entries 1, 4, and 7 in Table 1; Figure 2f and Supporting Information Figures S10–S14). By varying the initial feed ratios, PEG45-b-PNBOn diblock copolypeptides could be readily synthesized (entries 8–12 in Table 1; Figure 2g and Supporting Information Figures S15–S19). GPC elution traces revealed a continuous shift to high MWs upon increasing NPCA feed ratios, while keeping relatively narrow Đ values (1.03–1.14) (Figure 2g). To further demonstrate the versatility of our approach, we synthesized NPCAs based on γ-benzyl-l-glutamate, l-tyrosine (LT), and S-(3-hydroxypropyl)- l-cysteine (HLC), which were fully characterized ( Supporting Information Scheme S2 and Figures S20–S22).57 These NPCA precursors were further polymerized using a PEG45-NH3+Cl− initiator. The as-synthesized polypeptides had narrow dispersity (1.08–1.15), and the MWs were in good agreement with the feed ratios ( Supporting Information Table S1 and Figures S23–S26). Notably, the polymerization of NPCAs containing unprotected phenolic or hydroxyl groups (LT and HLC) was still feasible. In addition to homopolymerization, the copolymerization of NPCA precursors was also examined, enabling the synthesis of copolypeptides.4 Taking the copolymerization of NBO and Phe precursors as an example, we achieved PEG45-b-P(NBO-co-Phe)n BCPs (NBO = Nδ-(o-nitrobenzyloxycarbonyl)-l-ornithine) successfully with varying compositions and chain lengths, which could be finely tuned by adjusting the initial feed ratios of two NPCA precursors. Remarkably, all the synthesized copolymers exhibited monomodal GPC traces and narrow Đ (entries 14–17 in Table 1; Figure 2h and Supporting Information Figure S27–S30). Collectively, PEG45-NH3+Cl− could be used successfully to initiate NPCA polymerizations at elevated temperatures, yielding PEG-peptide BCPs with well-defined chain architectures. Synthesis of polypeptide brush and star (block) copolypeptides via primary amine hydrochloride-initiated NPCA polymerizations In addition to linear chain architectures, polypeptides with nonlinear chain topologies have received increasing attention.14 For example, polypeptide brushes composed of densely grafted amino acids in the side chains are attractive platforms to generate biomimetic surfaces and interfaces.67 Encouraged by the excellent control over NPCA polymerizations using protonated amines as initiators, we further explored the possibility of the fabrication of copolypeptides with nonlinear chain topologies including brush, star, and star-block copolypeptides (Figure 3). For the synthesis of polypeptide brushes, PEG45-b-PBocK30 was synthesized at first via BocK polymerization using PEG45-NH3+Cl− as the initiator (entry 5, Table 1). The Boc protecting groups were then removed (Figures 3a and 3b and Supporting Information Figure S31), followed by the polymerization of NBO precursors using side-chain primary ammonium moieties as initiating sites, affording the target PEG-polypeptide brush. The formation of polypeptide brush was evidenced by the successive shift of GPC elution peaks from ∼16.9 min for PEG45-NH3+Cl− to ∼15.3 min for PEG45-b-PBocK30, and further to ∼12 min (Figure 3c). Notably, the GPC elution trace of the polypeptide brush was monomodal with Đ of 1.13, revealing an excellent control over the chain extension process even in the presence of multiple initiating sites along one backbone chain. The MW of copolypeptide brush was determined to be ∼116.6 kDa (average DP ∼11.7 for each poly(Nδ-(o-nitrobenzyloxycarbonyl)-l-ornithine) (PNBO) graft) by GPC equipped with a multiangle laser light scattering (MALLS) detector (Figure 3c), consistent with 1H NMR results (Figure 3d and Supporting Information Figure S32). Figure 3 | Synthesis of polypeptide brush and star (block) copolypeptides via NPCA polymerization initiated by primary amine hydrochloride. (a) Schematic illustration of the synthesis of polypeptide brush via a “grafting from” approach. (b) 1H NMR spectra recorded in DMSO-d6 for PEG45-b-PBocK30 and protonated PEG45-b-PLys30 macroinitiator upon deprotection; the dotted region indicates complete removal of Boc protecting groups. (c) GPC elution traces, recorded for PEG45-NH3+Cl− initiator, PEG45-b-PBocK30, and target polypeptide brush. (d) 1H NMR spectrum recorded for polypeptide brush in DMSO-d6. (e) Schematic illustration of the synthesis of four-arm star block copolypeptides via one-pot sequential monomer addition. (f) GPC elution traces, recorded for four-arm star block copolypeptides (from Star-1 to Star-5). (g) Mn,calcd, Mw,GPC-MALLS, and Đ recorded for four-arm star block copolypeptides (from Star-1 to Star-5); note that the monomer feed ratio varies for each extension of star blocks (see Supporting Information for details). (h) 1H NMR spectra recorded in DMSO-d6 for protonated four-arm star amine initiator and four-arm star block copolypeptide. All polymerizations were conducted at [M]0 = 0.25 M in DMAc and 70 °C. Download figure Download PowerPoint Next, we synthesized four-arm star block copolypeptides via one-pot sequential polymerization of NPCA precursors using four-arm star primary amine hydrochloride as the initiator (Figure 3e). The star-type initiator bearing four protonated primary amine moieties was synthesized via the reaction of acyl azide derivative with pentaerythritol upon heating, followed by removing the Boc protecting groups with HCl in methanol ( Supporting Information Scheme S3 and Figures S33 and S34). The protonated star-type amine initiator was then used to actuate the NPCA polymerization. NBO and NBK precursors were sequentially added upon completion of the previously added ba