Smart metallopoly(L-glutamic acid) polymers: reversible helix-to-coil transition at neutral pH

Abstract

Among the smart polymers, smart polypeptides have a unique polymeric scaffold made of amino acids whose structuring can be controlled by an external stimuli. Herein we present how coordination chemistry to Zn species can be reversibly used to control the helix-to-coil transition of synthetic poly(L-glutamic acid) (PGA) polymers at a neutral pH of 7 in aqueous solutions.

---

Synthetic polypeptide polymers are made of natural building blocks (amino acids) and they have provided an important breakthrough in material science applications including those requiring smart polymers, i.e. polymers able to respond to an external stimulus.¹ Smart polypeptides are ideal candidates to mimic adaptive biological systems such as natural proteins, by undergoing structural or conformational changes in response to biologically relevant external stimuli including stimuli from other (macro)molecules or from the environment (temperature, pH, redox…).² Synthetic polypeptide polymers can reproduce secondary structures from natural proteins including α-helices or β-strands structures.³ Compared to polymers presenting a coil structure, structured polypeptides exhibit intriguing physico-chemical properties either in bulk, at the surface or in solution.⁴ Interestingly, synthetic polypeptides can easily undergo secondary structure transitions that can be easily implemented and tuned by tailoring amino-acid side chains.^3,5 For instance, helix-to-coil transitions can be controlled by means of pH changes for poly(L-glutamic acid)⁶ and poly(L-lysine).⁷ It is known that secondary structures of polypeptides can be induced by using metal salts.⁸ In fact, metals interaction with synthetic polypeptides polymers has been an important topic in the 80's, and in particular with synthetic anionic polypeptides.⁹ Helix induction has been studied with poly(L-glutamic acid) when coordinated to Cu^2+ 10 but also with other transition metals.^9a,11 Since these pioneer works, the use of coordination chemistry as a way to tune the physico-chemistry of synthetic polypeptides has been poorly explored, even if coordination chemistry represents nowadays an important tool to design advanced materials with enhanced properties.¹² Herein, we propose the use of coordination chemistry to develop a smart polymeric platform from poly(L-glutamic acid) that undergoes reversible helix-to-coil transitions at a neutral pH of 7 in aqueous conditions.

The first step of our smart polymer preparation involved the synthesis of the polypeptide backbone. A small library of poly(L-glutamic acid) spanning a wide range of molecular weights were obtained by using a two-step synthetic methodology. In a first step, controlled ring-opening polymerization of γ-benzyl-L-glutamate-N-carboxyanhydride was used to prepare a set of 4 poly(γ-benzyl-L-glutamate) (PBLG) polymers. Analysis of the polymers by size-exclusion chromatography (SEC) and ¹H NMR spectroscopy confirmed a low polydispersity index (PDI) of ∼1.2 and good agreement of the polymer composition with the monomer feed ratio (see Fig. S1 and Table S1†). In a second step, deprotection of each PBLG was achieved in smooth acidic conditions¹³ to afford poly(L-glutamic acid) PGA polymers, for which the polymerization degrees were determined by both SEC and ¹H NMR (see Fig. S2 and Table S1†). PGA 1 presented a weight-average molecular weight (M_n) of 4200 g mol⁻¹, PGA 2 a molecular weight of 9600 g mol⁻¹, PGA 3 a molecular weight of 15 200 g mol⁻¹ and PGA 4 a molecular weight of 27 200 g mol⁻¹. All these M_n values were associated with low polydispersities index of ∼1.1. Circular dichroism of the different PGA polymer revealed (1) an alpha helix structure at an acidic pH of 4 presenting two minima both at around λ = 210 and 224 nm;^6a and (2) an extended structure at a neutral pH of 7 (coil structure) presenting a maximum at λ = 218 nm (Fig. S3 and S4†).

We further screened helix induction with various transition metals. It is important to note that coordination of an inorganic metal salt to an amino-acid ligand decreases the aqueous pH of a water solution, a parameter that was neglected in previous work.^9a In this study, we have carefully prepared our polymer solutions by adjusting the pH to 7 before CD analysis. In a first attempt, we mixed 1 with various metal salts including ZnSO₄, CuSO₄, NiSO₄, CoSO₄ and CaSO₄, each time in neutral pH aqueous solution (pH = 7) for which the polymer alone was in a coil structure (maximum at λ = 218 nm, Fig. S3†) and at a mixing ratio C_M/C_P = 1 (C_P the polymer concentration on the monomer basis and C_M the concentration of the metal salt). As depicted in Fig. 1, CD spectra between 200 nm and 250 nm of the different solutions were monitored at a concentration of 300 μM in monomer units. Neither Co²⁺ salts nor Ni²⁺ salts were able to induce significantly alpha helix structures, in marked contrast to previous results obtained with PGA.^9a For those transition metals as well as for Ca²⁺ salts, the CD spectra showed a maximum at λ = 218 nm that was indicative of a coil structure (Fig. 1 top).‡ Only the solutions containing Cu²⁺ or Zn²⁺ salts revealed significant changes in CD spectra with appearance of minimum values that were attributable to an alpha helix structure (Fig. 1 down).^6a,9a As compared to PGA alone at pH = 4, with Zn²⁺ salts, the CD signature of the helix at pH = 7 displayed a significant smaller 210 nm minimum (θ_obs value of −14.6 mdeg instead of −29.2 mdeg) and a slight red shifted n–π* band at 225 nm (Fig. 1 arrows). These changes resulted in a CD distortion that previous works have attributed to the occurrence of helix nanoaggregation.^9a,10d

Fig. 1 Influence of the presence of metal salts on the CD spectra of an aqueous solution of PGA 1 at a buffered pH of 7 at a mixing ratio C_M/C_P = 1. [PGA] = 300 μM in monomer units.

Overall, significant helix induction was observed for ZnSO₄ as compared to other salts. Considering that coordination of Zn²⁺ to PGA was also much less studied than Cu²⁺, we further focused on Zn²⁺ ability to induce helix secondary structure by comparing the coordinating ability of ZnSO₄ to the other PGA 2–4. For all the polypeptides, addition of Zn²⁺ resulted in CD spectra exhibiting helix signature (two minima at 210 and 224 or 225 nm respectively). As depicted in Fig. S3–S5† and in Fig. 2, influence of the molecular weight over helicity was similar at an acidic pH of 4 without metal or at a neutral pH of 7 after metal coordination. Overall, increased θ_obs values at 224 nm were obtained when the molecular weight was higher. For instance, at an acidic pH of 4 and at 300 μM in monomer units, a θ_obs value of −33.1 mdeg was measured for 1 and a much lower θ_obs value of −67.3 mdeg was measured for 4 (Fig. S3†). At the same concentration, with Zn²⁺ and at neutral pH, a θ_obs value of −23.2 mdeg was measured for 1 and a much lower θ_obs value of −53.2 mdeg was measured for 4 (Fig. 2). The fact that (1) Ca²⁺ did not induce helix structuring at C_M/C_P = 1 for all the PGA (Fig. S6†) and that (2) PGA structuring was the same for C_M/C_P = 0.5 once Zn²⁺ coordinated to 2, 3 and 4 (Fig. S7†) were both indicative of a bidentate coordination. Current studies are indeed dedicated to better determine this coordination mode.

Fig. 2 Zn²⁺ induced structuring of PGA 1–4 in aqueous solution at pH 7 at a mixing ratio C_M/C_P = 1. [PGA] = 300 μM in monomer units.

Pursuing the design of smart polypeptide systems, we then focused on helix destructuring once Zn²⁺ coordinated. In a first intention, we studied how coordination to Zn²⁺ influenced the temperature destructuring of the metallopolypeptide. In fact and as previously reported with PGA alone,¹⁴ increase of temperature triggers acidic PGA helix destructuring and the resulting helix-to-coil transition occurs in a very broad temperature range from 10 to 80 °C. Therefore, CD spectra of the different PGA at an acidic pH of 4 have been recorded in a temperature range from 10 °C to 80 °C to confirm such helix destructuring (Fig. S8†).

CD spectra obtained at a neutral pH of 7 with PGA 1–4, once coordinated to Zn²⁺, were further monitored. Molar ellipticities taken from these analyses, at 222 nm, have been plotted, as shown in Fig. 3. For PGA 1–4, once coordinated to Zn²⁺, partial helix-to-coil transition were observed when temperature was increased from 10 °C to 80 °C. For instance for 4, a [θ]₂₂₂ value of −16.4 deg cm² dmol⁻¹ was calculated at 20 °C and a much shallower [θ]₂₂₂ value of −9.0 deg cm² dmol⁻¹ was calculated at 80 °C. Overall, once coordinated to Zn²⁺ and at pH = 7, the total destructuring of PGA 1–4 was not achieved up to 80 °C. As compared to PGA 1–4 in acidic conditions, PGA destructuring at pH = 7 and in presence of Zn²⁺, were almost similar in trend, except for PGA 1.

Fig. 3 Temperature as a destructuring trigger of PGA 1–4 once mixed with Zn²⁺ (pH 7, mixing ratio C_M/C_P = 1, molar ellipticities calculated from θ_obs at 222 nm).

To design smart metallopolypeptide polymers, a total destructuring of the macromolecule need to be achieved. Therefore, we further concentrated our “destructuring” effort in removing Zn²⁺ coordination by means of an external ligand competitor. With metallopolypeptides, the use of an external ligand would be an ideal chemical leverage to trigger physico-chemical properties, including solubility, functionality and or destructuring. We first used nitrilotriacetic acid, a tridentate ligand generally referred to as NTA,¹⁵ which is well known in biotechnology for its Ni²⁺ affinity (the dissociation constant for the NTA complex is 10⁻¹²). Its affinity for Zn²⁺ is slightly lower with a dissociation constant of 10⁻¹¹. We prepared solutions of PGA 1–4 at a neutral pH of 7 that were first coordinated to Zn²⁺ at a mixing ratio C_M/C_P = 1. After characterization by circular dichroism, sodium nitrilotriacetate was added in stoichiometric amount with respect to the metal salt. Resulting circular dichroism spectra are shown in Fig. 4 and S9.† For all the polymers, addition of NTA resulted in nearly full helix-to-coil transition: after NTA addition, CD spectra presented a similar positive maximum at λ = 218 nm as compared to CD spectra recorded at pH = 7 without metals (Fig. 4 top). This unprecedented result clearly evidenced the success of this destructuring strategy. Influence of the Na⁺ concentration was ruled out by desalting the sample after NTA addition or by adding NaCl at 20 mM before NTA addition (Fig. S10†). The induced helix-to-coil transition would therefore be attributed to the stronger affinity constant of the metal for the NTA ligand compared to the PGA polymer. We then compared the ligand-induced destructuring of PGA, once coordinated to Zn²⁺, with two other chemicals, iminodiacetic acid (IDA, the dissociation constant for the Zn²⁺ complex is 10⁻⁷) and glycine (GLY, the dissociation constant for the Zn²⁺ complex is 10⁻⁵). The purpose of this experiment was to evaluate the metal affinity required to trigger PGA helix-to-coil transition. As clearly evidenced on CD spectra, IDA was only able to induce destructuring with PGA 1, the lower molecular weight. For 2–4, the higher molecular weights or for 1–4 using GLY as a ligand, significant negative θ_obs values were found at 224 nm with or without ligands, indicating that metallopolypeptides remained structured in α-helix after ligand addition (Fig. S11†). It is important to note that Zn²⁺ might still be coordinated to PGA side chains after ligand addition. In the case of NTA, ¹³C NMR of the PGA side chains seems to indicate a signal broadening that could be attributed to residual coordination (Fig. S12†). Current studies are dedicated to better characterize this possible coordination bonding.

Fig. 4 Influence of external ligand competitors on the CD spectra of an aqueous solution of PGA 1 or 4 at pH 7 mixed with: (top) ZnSO₄ at a mixing ratio C_M/C_P = 1 and then NTA; (down) ZnSO₄ at a mixing ratio C_M/C_P = 1 and then IDA. [PGA] = 300 μM in monomer units.

Conclusions

As simplified analogues of natural proteins, synthetic polypeptides polymers constitute important tools to develop new biomaterial applications, particularly if smart systems can be designed. In this paper, we show that thanks to reversible coordination chemistry, synthetic polypeptides polymers can behave as metal responsive polymeric systems. The aim of this current work is the design of new smart metallopolypeptide that will find uses in (bio)diagnosis and in catalysis.

Acknowledgements

The authors wish to thank the Centre National pour la Recherche Scientifique (CNRS) for financial support. Authors also acknowledge the IDEX Uniti Emergence program for grant UT2014-593.

Notes and references

1. (a) J. Huang and A. Heise, Chem. Soc. Rev., 2013, 42, 7373; (b) M. A. Quadir, M. Martin and P. T. Hammond, Chem. Mater., 2014, 26, 461; (c) T. J. Deming, Chem. Rev., 2016, 116, 786.
2. (a) M. A. C. Stuart, W. T. S. Huck, J. Genzer, M. Muller, C. Ober, M. Stamm, G. B. Sukhorukov, I. Szleifer, V. V. Tsukruk, M. Urban, F. Winnik, S. Zauscher, I. Luzinov and S. Minko, Nat. Mater., 2010, 9, 101; (b) X. Fu, Y. Ma, J. Sun and Z. Li, RSC Adv., 2016, 6, 70243; (c) M. G. Mohamed, J.-H. Tu, S.-H. Huang, Y.-W. Chiang and S.-W. Kuo, RSC Adv., 2016, 6, 51456.
3. G. D. Fasman, Poly-alpha-amino acids, Decker, New York, 1967.
4. A. Carlsen and S. Lecommandoux, Curr. Opin. Colloid Interface Sci., 2009, 14, 329.
5. C.-T. Yang, Y. Wang, C. W. Frank and Y.-C. Chang, RSC Adv., 2015, 5, 86113.
6. (a) M. Rinaudo and A. Domard, J. Am. Chem. Soc., 1976, 98, 6360; (b) M. Nagasawa and A. Holtzer, J. Am. Chem. Soc., 1964, 86, 538.
7. R. C. Parker, K. Applegate and L. J. Slutsky, J. Phys. Chem., 1966, 70, 3018.
8. (a) A. L. Jacobson, Biopolymers, 1964, 2, 207; (b) M. Palumbo, A. Cosani, M. Terbojevich and E. Peggion, Macromolecules, 1978, 11, 1271.
9. (a) H. Maeda, T. Hiramatsu and S. Ikeda, Bull. Chem. Soc. Jpn., 1986, 59, 587; (b) H. Maeda, J. Nakajima, K. Oka, K. Ooi and S. Ikeda, Int. J. Biol. Macromol., 1982, 4, 352; (c) N. Kono and A. Ikegami, Biopolymers, 1966, 4, 823; (d) K. Oka, H. Maeda and S. Ikeda, Int. J. Biol. Macromol., 1983, 5, 342.
10. (a) H. Takesada, H. Yamazaki and A. Wada, Biopolymers, 1966, 4, 713; (b) N. Imai and J. A. Marinsky, Macromolecules, 1980, 13, 275; (c) T. Masujima and K. Yamaoka, Biopolymers, 1980, 19, 477; (d) T. Masujima, Bull. Chem. Soc. Jpn., 1983, 56, 838.
11. T. Hiraoki, M. Kaneko and K. Hikichi, Polym. J., 1979, 11, 397.
12. (a) K. Kuroiwa, T. Arie, S. Sakurai, S. Hayami and T. J. Deming, J. Mater. Chem. C, 2015, 3, 7779; (b) K. Kuroiwa, Y. Masaki, Y. Koga and T. J. Deming, Int. J. Mol. Sci., 2013, 14, 2022; (c) C. Cai, L. Wang and J. Lin, Chem. Commun., 2011, 47, 11189.
13. Y. Kato, N. Umemoto, Y. Kayama, H. Fukushima, Y. Takeda, T. Hara and Y. Tsukada, J. Med. Chem., 1984, 27, 1602.
14. C. Krejtschi and K. Hauser, Eur. Biophys. J., 2011, 40, 673.
15. R. Puchta, N. van Eikema Hommes, R. Meiera and R. van Eldikis, Dalton Trans., 2006, 3392.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra19753a

‡ It is to note that PGA 2–4 did not evidence helix structuring in presence of CoSO₄ or NiSO₄. Outside a possible decrease of the aqueous pH solutions after metal coordination, other parameters explaining the conflict include the use of PGA having higher molecular weight or the use of different inorganic salts to prepare the complexes, for instance CoCl₂ or NiCl₂.

---