Hemicellulosic Polysaccharides Mimics: Synthesis of Tailored

Subscriber access provided by University of Sussex Library

Article

Hemicellulosic Polysaccharides Mimics: Synthesis of Tailored Bottlebrush-Like Xyloglucan Oligosaccharide Glycopolymers as Binders of Nanocrystalline Cellulose Jing Chen, Christophe Travelet, Redouane Borsali, and Sami Halila Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.7b01056 • Publication Date (Web): 14 Sep 2017 Downloaded from http://pubs.acs.org on September 15, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biomacromolecules is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Hemicellulosic Polysaccharides Mimics: Synthesis of Tailored Bottlebrush-Like Xyloglucan Oligosaccharide Glycopolymers as Binders of Nanocrystalline Cellulose Jing Chen,†,‡,# Christophe Travelet,†,‡ Redouane Borsali†,‡ and Sami Halila*,†,‡ †

Univ. Grenoble Alpes, CERMAV, F-38000 Grenoble, France ‡

CNRS, CERMAV, F-38000 Grenoble, France

KEYWORDS : glycopolymers, xyloglucan, multivalent interactions, cellulose nanocrystals.

ABSTRACT : We report in this contribution that while low molecular weight hemicellulosic building blocks are known not to interact with cellulosic materials, their multivalent presentation on a polymeric scaffold significantly enhanced the binding interactions that are remarkably in the same range as those usually observed for lectin-carbohydrate interactions. We developed a poly(propargyl methacrylate) scaffold on which we conjugated, by “post-click” reaction, a variety of azide reducing-end functionalized xyloglucan oligosaccharides with controlled enzymatic-mediated rate of degalactosylation. Bottlebrush-like xyloglucan oligosaccharide glycopolymers (poly(XGOn)) were obtained and their self-assemblies in aqueous solution were

ACS Paragon Plus Environment

1

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 24

investigated using dynamic light scattering (DLS). We demonstrated that increasing the extent of degalactosylation promoted self-association of poly(XGOn), which we attribute to the appearance of hydrophobic domains. A sharp thermoresponsiveness, which corresponds to a decrease in aggregate size with increasing temperature, was observed when the extent of degalactosylation was 30% or greater. Importantly, isothermal titration calorimetry (ITC) and polarized/depolarized DLS revealed that poly(XGOn) exhibit a significant capacity to interact with nanocrystalline cellulose (NCC) surfaces particularly for the non-degalactosylated form, emphasizing the important role of galactosyl residues in the binding mechanism and in the 3dimensional structures of glycopolymers.

INTRODUCTION Polysaccharides are ubiquitous in Nature. They are present in various living species like plants, micro-organisms or mammals. Their roles are numerous and diverse acting as structural elements where they accommodate the mechanical properties of the cell wall,1 as bio-recognition sites where they are involved in cell interaction, growth and metabolism,2 or as texturing agents where they monitor the rheological properties of physiological fluids.3 And thanks to their abundant and renewable resources, polysaccharides are already extensively used in food, pharmaceutical, cosmetic and related biomedical industries. However processing of polysaccharides and finetuning of their properties are difficult to achieve owing to their variable sources, heterogeneous structures, high polydispersity and high molecular weight making them difficult to solubilize and to chemically modify with precision and reproducibility. In order to circumvent these drawbacks, some research groups have suggested to associate the power of polymer chemistry enabling precise synthesis of macromolecular architectures with carbohydrate building blocks to mimic the functional properties of natural polysaccharides.


2

Page 3 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

These kinds of glycopolymers are featured by presentation of multiple oligo- or mono-saccharide units as side chains covalently linked along the synthetic polymer scaffolds. While glycopolymers are widely exploited to promote the multivalency effect in protein-carbohydrate interaction4 in order to mimic biological functions of glycoproteins such as mucin5 and glycosaminoglycans such as heparin,6 only few reports have investigated this strategy for probing carbohydrate-carbohydrate interactions in spite of their significant roles in cell adhesion and communication7,8 or in biomechanical properties of plant primary cell walls.9 Inspired by the adaptive network of polysaccharides in plant primary cell wall of higher plant such as for xyloglucan (XG) hemicellulosic polysaccharides that are tightly adsorbed on cellulose microfibrils and act as a cross-linker through multiple non-covalent bonding (Figure 1a),10,11 we anticipate that glycopolymers made of xyloglucan oligosaccharides side chains derived of tamarind seeds XG, termed as poly(XGO), could interact with the high-surface area of bundles of rod-like NCC whiskers (Figure 1b) even if single subunits XGO are not able to strongly bind to cellulose.12 An encouraging result was reported by Q. Zhou et al.13 demonstrating that an amphiphilic triblock copolymer consisting in XGO-b-poly(ethylene glycol)-b-polystyrene was able to adsorb onto cotton NCC in a nonpolar organic medium such as toluene. On the other side, O. Ikkala et al.14 noted that dendronized glucose-based systems effectively interacted with NCC.


3

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 24

Figure 1. Graphical representation of interactions between (a) XG and (b) poly(XGO) with NCC. This work is first focused on the synthesis, structural characterization and self-assembly properties of bottlebrush-like poly(XGO) containing different galactosyl substitution pattern. Secondly we investigated their ability to interact with cotton NCC in aqueous medium using isothermal titration calorimetry (ITC) and polarized/depolarized dynamic light scattering (DLS).

EXPERIMENTAL SECTION Materials. Tamarind seeds xyloglucan was kindly provided by Megazyme (Megazyme International Ireland Ltd., Bray, Ireland). Cotton-derived cellulose nanocrystals (NCC) was prepared by sulfuric acid hydrolysis according to a reported procedure,15 to afford a 1 wt % stock suspension. Cellulase from Trichoderma reesei (C8546) was purchased from Sigma-Aldrich. βgalactosidase from Aspergillus Niger was purchased from Megazyme. 2-azidoethylamine was synthesized according to the literature.16 3-(Trimethylsilyl)propargyl alcohol, triethylamine, methacryloyl chloride, 4-cyanopentanoic acid dithiobenzoate (CAPDB), azobisisobutyronitrile (AIBN), tetrabutylammonium fluoride trihydrate (TBAF•3H2O), copper(II) sulfate pentahydrate (CuSO4•5H2O) and sodium ascorbate were purchased from Sigma-Aldrich. All solvents (analytical grade) were provided by CARLO ERBA (France) and used as received. Methods. NMR Spectroscopy. 1H NMR and DOSY spectra were recorded on a Bruker Avance DRX400 (400 MHz). FTIR Spectroscopy. Infrared spectra were recorded using a PerkinElmer spectrometer. The samples were analyzed by transmission from 400 to 4000 cm−1 (16 scans resolution 4). Analytical liquid chromatography. HPAEC-PAD analysis were carried out on a Dionex CarboPacTM PA100 column (4 mm x 250 mm), with a gold working electrode and using triple-


4

Page 5 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

pulse amperometry (pulse potentials and durations: E1 0.1 V, 400 ms; E2 0.7 V, 100 ms; E3 -0,1 V, 100 ms). XGOn were separated using a gradient elution (0–4 min : NaOH 100mM / NaOAc 60 mM; 4–20 min : linear gradient elution up to NaOH 100 mM / NaOAc 500 mM) at a flow rate of 1 mL min-1. The ratio of XGOn (XXXG/XXLG/XLXG/XLLG) residue was determined by comparison with the surface area of detector response. Gel Permeation Chromatography. GPC measurements were performed at 60 °C using 1260 Infinity GPC system (Agilent Technologies) (100 µL manual injection system, 1260 Agilent quaternary pump, 1260-MDS refractive index detector) equipped with two Agilent PolyPore PL1113-6500 columns (linear, 7.5 × 300 mm; particle size, 5 µm; exclusion limit, 200 – 2 000 000) in DMF containing lithium chloride (0.005 M) at the flow rate of 1.0 mL min–1. Atomic Force Microscopy. AFM was carried out using a PicoPlus microscope (Molecular Imaging, Inc., Tempe, AZ, USA). AFM samples were prepared using Si wafer deposited by glycopolymer solution in DMSO (0.01 g/L) followed by drying at 40 °C under reduced pressure overnight. Images were recorded in tapping mode. Isothermal Titration Calorimetry. 6 mg of glycopolymer ligands, poly(XGOn) were dissolved into 0.6 mL of deionized water and loaded in the injection syringe. ITC was performed with an ITC200 microcalorimeter (MicroCal Inc.) at 25 °C. Titration was performed with 29 of 2 µL injections of poly(XGOn) every 300 s in the NCC (1% wt) containing cell. Data were fitted with MicroCal Origin 7 software, according to standard procedures. Fitted data yielded the stoichiometry (n), the association constant (Ka), and the enthalpy of binding (∆H). Other thermodynamic parameters (i.e., changes in free energy, ∆G, and entropy, ∆S) were calculated from the equation ∆G = ∆H − T∆S = −RT ln Ka, in which T is the absolute temperature and R =


5

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 24

8.314 J mol−1 K−1. Two independent titrations were performed for each poly(XGOn) tested. ITC figures were prepared using the Origin software provided with the apparatus. Dynamic Light Scattering. The polarized and depolarized DLS measurements were performed using an ALV/CGS-8FS/N069 apparatus (from ALV) equipped with an ALV/LSE-5004 multiple τ digital correlator with a 125 ns initial sampling time (from ALV) and a 35 mW red HeNe linearly polarized laser operating at a wavelength of 632.8 nm (from JDSU). The unfiltered suspensions NCC or {NCC + poly(XGOn)} in water were loaded in 10 mm diameter quartz cylindrical cells and maintained at a constant temperature of 25.0 ± 0.1 °C prior to measurement. Data were collected using the digital ALV correlator software at different scattering angles ranging from 50 to 130° (i.e. 1.12 × 10-2 ≤ q = (4πn/λ)sin(θ/2) ≤ 2.40 × 10-2 nm-1 where q represents the scattering vector modulus, n the refractive index of the pure solvent (water in this case), λ the wavelength of the incident light in vacuum and θ the scattering angle relative to the transmitted beam) by a step of 10°. At each angle, the counting time was typically 3 min in the polarized vertical position, and 90 min in the depolarized horizontal position of the GlanThompson prism polarizer. The relaxation time (τ) distributions were obtained using Contin analysis of the autocorrelation functions. The translational diffusion coefficients (Dtrans) of the cylindrically symmetric particles were extrapolated from the slopes of the relaxation frequency (1/τ) dependences on q2 in the polarized vertical (VV) position of the Glan-Thompson prism polarizer. And the rotational diffusion coefficients (Θrotat) of these particles were extrapolated from the intercepts of the 1/τ dependences on q2 in the depolarized horizontal (VH) position. Knowing these two parameters Dtrans and Θrotat allow then to determine the lengths (L) and crosssectional diameters (D) of the cylindrically symmetric particles using the Broersma equations.


6

Page 7 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Preparation of xyloglucan oligosaccharide (XGO) via enzymatic depolymerisation. The enzymatic depolymerization of tamarind seeds xyloglucan (XG) was carried out by commercial cellulase. In a typical experiment, 10 g of XG was dissolved in 500 mL of water. After adding 5 mL of NH4OAc buffer (pH=4.5, 1 M) and 1 g of cellulase (700 U), the mixture was incubated at 37 oC for 16 h. The cellulase enzyme was deactivated by boiling at 90 oC for 10 min, and then the solution was filtrated and then concentrated for dialysis (MWCO = 100~500 D, for 4 days). The final lyophilization afforded a while solid product with 90% of yield. Degalactosylation of XGO via β-galactosidase (XGOn). The kinetics of degalactosylation of XGO by β-galactosidase (A. Niger) was monitored by HPAEC-PAD. In a typical experiment, 200 mg of XGO was dissolved in 20 mL of water. After adding 10 U of β-galactosidase (A. Niger) and 0.2 mL of NH4OAc buffer (pH=4.5, 1 M), the mixture was incubated at 37 oC. After each time interval, several drops of mixture was performed HPAEC-PAD to observe the evolution of the composition of different mixture of XGO as a function of incubating time. Synthesis of azide-functionalized XGOn (XGOn-N3). Four XGO samples with different galactose removal ratios were azide end-functionalized (XGOn-N3) via reductive amination with 2-azidoethylamine. In a typical synthesis, 0.4 g (0.3 mmol) of XGOn (n=0, 30, 60, 90) and 564 mg (9 mmol) of NaBH3CN were dissolved in 9 mL of DMSO/AcOH (7v/3v). The mixture was heated to 65 oC and 258.3 mg (3 mmol) of 2-azidoethylamine was added. The solution was stirred at 65 oC for 2 h and the TLC was used to monitor the conversion until 100%. After precipitation in acetonitrile, the solid was recovered by centrifugation. The final lyophilization afforded a white solid product in quantitative yield. The products were characterized by 1H NMR and ESI-MS.


7

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 24

Synthesis of bottlebrush-like XGOn-glycopolymers, poly(XGOn), via CuAAC “click” reaction. In a typical synthesis, PPMA17 (1.83 mg, 0.015 mmol of PMA unit), XGOn-N3 (n=0, 30, 60, 90) (40 mg, 0.03 mmol), CuSO4•5H2O (3.8 mg, 0.015 mmol) and sodium ascorbate (12 mg, 0.06 mmol) were dissolved in 1 mL of DMSO. The mixture was degassed by Argon bubbles for 30 min, and then stirred at 50 °C for 4 days. The copper salt and excess of XGO-N3 were removed by dialysis (MWCO = 3500 D) for 3 days. The glycopolymers were freeze-dried to give white amorphous solids.

RESULTS AND DISCUSSION “Grafting onto” method was preferred to synthesize poly(XGO) which consisted to “click” reducing-end azide-modified XGO onto pre-formed alkyne-grafted polymethacrylate backbone (Scheme 1). The XGO building blocks were prepared through a controlled cellulase-mediated depolymerization of XG to give a mixture of cellotetraose backbone regularly substituted by 3 xylosyl units themselves linked to 0, 1 or 2 terminal galactosyl units.18,19 The XGO mixture symbolized by using a single-letter code20: XXXG, XLXG, XXLG, and XLLG (Scheme 1) is readily obtained in a quantitative way and has the benefit to be structurally well-defined and of homogeneous molecular weight. Even if previous studies tend to demonstrate that the binding ability of XG with cellulose starts from at least twelve glucosyl units on the backbone, which corresponds roughly to the persistence length of XG, ca. 8 nm,21 we assume that multiple copies of XGO should offset this lack. Moreover since the substitution pattern of XG is another important parameter,22,23 we decided to design and study binding capacity of various poly(XGO) with different degalactosylation rate. Scheme 1. Schematic presentation of the preparation of bottlebrush-like XGO glycopolymers, poly(XGOn)


8

Page 9 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

The degalactosylation of XGO was performed using a commercially available Aspergillus Niger β-galactosidase (10 U) in highly pure form. The galactose removal was estimated according to high-performance anion exchange chromatography equipped with pulsed amperometric detection (HPAEC-PAD) and 1H NMR, both giving similar results (Table 1). HPAEC-PAD is a useful method enabling the quantification between the isomer XLXG (DP8) and XXLG (DP8) (Figure S2) while with integration of anomeric 1H, we were only able to calculate the relative proportion between XXXG (DP7), XXLG/XLXG (DP8) and XLLG (DP9) (Figure S4). Table 1. Composition of XGOn samples with different galactose removal ratios


9

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 24

XGO composition (%) by HPAECGalactose removal % Samp

Time

les

(h)

PAD by HPAECPAD

by 1H NMR

XXX

XLX

XXL

XLL

G

G

G

G

XGO 0

0

0

17

6

31

46

1

31

36

26

3

55

16

6

58

54

46

0

53

1

60

93

94

97

0

3

0

0

XGO 30

XGO 60

XGO 90

As observed in Figure 2, the enzyme preferentially hydrolyzed the Gal residue on the antepenultimate position thereby converting XLXG (DP8) to XXXG (DP7), and XLLG (DP9) to XXLG

(DP8).24

By

carefully

monitoring

the

A.

Niger

β-galactosidase-mediated

degalactosylation of XGO, it was possible to finely control, with the incubation time, the composition of the XGO mixture in order to obtain, for instance, almost exclusively XXXG (DP7) block after 60 h of incubation time. The enzymatic hydrolysis was stopped after 1, 6 and 60 h to produce a series of XGOn, termed as XGO30, XGO60 and XGO90 with roughly 30, 60 and 90% of removal of galactosyl residues, respectively.


10

Page 11 of 24

100

10 Unit of A. Niger

XXXG XLXG XXLG XLLG

80

60

%

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

40

20

0 0

10

20

30

40

50

60

70

80

90

100

110

Time (h)

Figure 2. The evolution of composition of XGO (DP7 (XXXG), DP8 (XLXG/XXLG) and DP9 (XLLG)) as a function of incubating time at 37 oC with 10 U β-galactosidase (A. Niger) The reducing-end modification of XGOn with an azide function (XGOn-N3) was accomplished in a straightforward manner using reductive amination method. The reactions were done with an excess of 2-azidoethylamine and reductant, NaBH3CN, in mixed solvents (DMSO/AcOH = 7/3, v/v, pH = 5) and led to 100% conversion in 2 h. XGOn-N3 was purified by simple precipitation in acetonitrile. All XGOn-N3 chemical structures were clearly confirmed by 1H NMR and ESI-MS (Figures S5 and S6). Poly(propargyl methacrylate) (PPMA) (Mn = 23600 g mol−1, Mw/Mn = 1.2), in which pendant “clickable” alkyne functions were along each repeating unit, was synthesized by RAFT polymerization according to a known procedure.17 The structural characteristics were consistent with those published elsewhere (Figures S7-S10).17,25 The subsequent postpolymerization modification of PPMA with series of XGOn-N3 was realized by copper(I)catalyzed azide–alkyne cycloaddition (CuAAC) in DMSO to afford corresponding poly(XGOn) ) in 69% to 78% isolated yields. The success of the click reaction was clearly evidenced in 1H NMR through the appearance of the characteristic triazolic proton at 8 ppm (Figure 3a).


11

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 24

Additionally, the 1H NMR spectra confirmed the presence of both blocks with methyl protons of PPMA at 1 ppm and H-sugars in the region 2.6-5.4 ppm (Figure 3a and Figure S11). Moreover, the diffusion-ordered NMR spectroscopy (DOSY) for poly(XGOn) confirmed the click coupling since a single translational diffusion coefficient (D = 6.3×10-5 cm2 s-1) was observed in Figure 3b, which was different from the XGO0 (D = 1.3×10-5 cm2 s-1, Figure S12). The lower diffusion coefficient for XGO0 is indicative of aggregation phenomena driven by intermolecular hydrogen bonding while for poly(XGO0) the aggregation is limited thanks to rather intramolecular hydrogen bonding between XGO0 side-chains in close proximity.

Figure 3. (a) 1H NMR spectrum in DMSO-d6 and (b) DOSY NMR spectrum in D2O of poly(XGO0) at 80 °C Further confirmation of the click coupling was provided by FTIR spectra that unambiguously showed the complete disappearance of characteristic azide and alkyne signals in glycopolymers (Figure 4).


12

Page 13 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Figure 4. FTIR spectra of PPMA, XGOn-N3 and poly(XGOn) glycopolymers Because the nitrogen atoms were only provided by XGOn side-chains the grafting efficiency as well as the molecular weight of poly(XGOn) were easily determined. Roughly, the efficiency of the click coupling was estimated about 70% with an average molecular weight of 80 000 g mol-1 (Table 2). Table 2. XGO grafting ratios on the PPMA backbone calculated from element analyses Grafting ratio Mn (g mol-1)

Samples

C/H/N

Poly(XGO0)

52.58/6.48/4.20

69.2%

78200

Poly(XGO30) 51.38/6.65/4.33

75.3%

80700

Poly(XGO60) 51.73/6.63/4.54

76.1%

78200

Poly(XGO90) 53.45/6.45/4.95

77.9%

75000

The morphology of the single brush-like poly(XGO0) was difficult to visualize by AFM due to its tendency to self-associate during drying process. However, in dilute conditions, the poly(XGO0) revealed extended worm-like shapes owing to steric hindrance between the densely grafted brush XGO side chains (Figure 5). The width of worm-like structure is around 50 nm which is much higher than single bottle-like glycopolymer (maximum 6 nm) suggesting the


13

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 24

bundle structure of aggregation. The extended conformation of poly(XGOn), which is radically different from the presumed random coil shape of native XG in solution, is of first importance since flatly bound XG segments are believed to preferentially tether to cellulose surface.26,27

Figure 5. AFM height image in tapping-mode of brush poly(XGO0) drop-casted from water solution (left) on mica with an enlarged picture and its corresponding cross section (right). The thermo-responsive properties of poly(XGOn) in water were investigated by DLS since some reports showed that degalactosylated XG (above 35%) forms thermoreversible gels as a result of hydrophobic interactions and formation of a very strong 3-dimensional network.28,29 First, Figure 6a showed that the increase of galactose removal promoted aggregation of poly(XGOn) (C = 1 g L-1), certainly due to emergence of hydrophobic microdomains. This result supports the observations reported much earlier by Hayashi et al. on XG polysaccharides30 where the sidechains containing galactose residues were identified to prevent XG from self-association.


14

Page 15 of 24

1000

(a)

at 25 oC at 50 oC

Rh (nm)

800

600

400

200

0 0

20

40

60

80

100

Galactose removal (%)

1200

(b)

poly(XGO60) poly(XGO90)

1000

800

Rh (nm)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

600

400

200 25

30

35

40

45

50

Temperature (oC)

Figure 6. Thermoresponsiveness of poly(XGOn) by DLS analyses At higher temperature (50 °C) the size of glycopolymer aggregates reduced for galactose removal ratios higher than 30% indicating either a better aggregates breakdown by water molecules or even more probably stemmed from a chain folding and compactness as a result of dehydration process (Figure S13). The opposite behavior of thermo-responsiveness compared to degalactosylated XG polysaccharides could be attributed to a higher contribution of intramolecular interactions between densely grafted degalactosylated XGO side chains rather than intermolecular ones. A carefully study of the Rh variation for poly(XGO60) and poly(XGO90) as a function of temperature in water has been carried out (Figure 6b). Interestingly, the data showed that Rh decreased, not linearly with rising temperature, but sharply from a given critical temperature (Tc), by a factor 2. Moreover, the Tc values increased when


15

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 24

increasing the galactose removal ratio: Tc(poly(XGO90)) = 43 °C versus Tc(poly(XGO60)) = 37 °C. This means that the Tc value can be finely tuned with the control of degalactosylation rate. Next, we assessed the interaction abilities at 25°C of poly(XGOn) solutions added to a cottonderived NCC suspension using ITC (Figure S14) to get information about the thermodynamic binding parameters including affinity constant (Ka), binding stoichiometry (n) and heat of binding (enthalpy; ∆H), and indirectly the entropy (∆S) as listed in Table 3. Table 3. Thermodynamic binding parameters for XGOn-based glycopolymers in presence of NCC in water and 25 °C Samples

n

Ka (µM-1)

∆H (kcal mol-1)

-T∆S (kcal mol-1)

poly(XGO0)

9.5

2.98

-435.8

426.1

poly(XGO30)

24.1

0.64

-256.7

248.5

poly(XGO60)

23.4

0.49

-202.7

194.9

poly(XGO90)

20.3

1.21

-346.3

336.7

In the first set of experiments the heat of dilution of poly(XGOn) solutions were determined without showing significant exothermic signals. In order to get valuable thermodynamic data of interactions, we estimated the molar accessible surface concentration of aqueous cotton NCC bundles suspension (1% w/w) from their average dimensions (6 × 6 × 180 nm3)31 that gave a concentration of 2.08×10-4 mM. All poly(XGOn) showed an enthalpically favorable interaction with NCC (∆H varying between −435.8 to −202.7 kcal mol−1) whereas the entropic contribution to the binding mechanism was negative, with the −T∆S values being in the range from 426.1 to 194.9 kcal mol−1. Enthalpy-entropy compensation appears to be a general phenomenon usually observed for lectin-carbohydrate interactions that can be explained in terms of reorganization of water structure around the binding site of the NCC and the poly(XGOn). It should be pointed out the micromolar affinities Ka, indicating that poly(XGOn) strongly bind to NCC, are remarkably


16

Page 17 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

in the same range than typical lectin-carbohydrate interactions.32 An interesting trend was observed : with increasing the galactose removal up to 60%: enthalpy cost increased while entropy penalties decreased resulting in an overall decrease of affinity. This enthalpy-entropy compensation reflects the “classical” hydrophobic effect. In fact, the release of hydrophobic domains, stemmed from the polar galactose units removal, decreased the enthalpy and improved the entropy. This compensation tends to reflect the lower desolvation cost of the enhanced hydrophobic domains of degalactosylated glycopolymers. The only deviation from this trend was poly(XGO90) that displays stronger affinity and heat capacity for NCC. One possible explanation could be an unfolding of globular poly(XGO90) upon interaction with NCC. The binding stoichiometry for poly(XGO0) (n = 9.5) compared to poly(XGO30,

60, 90

) (average n = 22.6)

emphasized that further binding sites were reached with degalactosylated glycopolymers certainly through additional interactions with hydrophobic surfaces of NCC.33 Interestingly, Buleon et al.34 reported that a mixture of XGO monomer and dimer weakly interacts with NCC according to exothermic (at low concentration, H-bonding interactions dominate) and endothermic (at high concentration, hydrophobic interactions dominate) processes. Our study demonstrates the importance of multipresentation of XGO side chains for a much better affinity with surface of NCC and of 3-D glycopolymer architectures (closely related to the degalactosylation rate) on mode of interactions. To complete this study, the dynamical behavior in salt-free suspensions of rodlike cotton NCC bundles, in absence or presence of poly(XGOn) having various extents of degalactosylation, was probed by polarized/depolarized DLS (Figure 7) according to a procedure described by de Souza Lima et al.35,36 The shape parameters in aqueous solution were extracted, i.e., length (L) and cross-sectional diameter (D) and thus aspect ratio (L/D), and determined from the translational


17

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 24

(Dtrans) and rotational (Θrotat) diffusion coefficients.37 We found that cotton NCC had bundle dimensions of L = 246 nm, D = 23 nm and L/D = 10.7, that were very close to previously reported data.36 In mixture with poly(XGOn), two series of behavior were observed. For the first one, associated to poly(XGO0) or poly(XGO30), the light scattering study evidenced that the length remained unaffected while the diameter increased significantly with the extent of degalactosylation up to D = 58 nm. These values indicated that those glycopolymers adsorbed onto the surface of individual cotton NCC leading to a thicker layer for poly(XGO30) due to its self-associative ability. In the second series, beyond 30% of degalactosylation rate, the apparent length as well as the diameter of cotton NCC interacting with poly(XGO60) and poly(XGO90) increased dramatically and clearly demonstrated an aggregation phenomenon owing to enhanced inter-chain interactions of these glycopolymers. Moreover, the self-assembled systems became less anisotropic (L/D ~ 4), suggesting that well organized NCC bundles were not present.


18

Page 19 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Figure 7. (a) and (b) DLS relaxation frequency (1/τ) dependences on squared scattering vector modulus (q2) in the polarized (VV)/depolarized (VH) positions of the Glan-Thompson prism polarizer. (c) Length (L), cross-sectional diameter (D) and aspect ratio (L/D) of NCC alone and in mixture with poly(XGOn) calculated from polarized/depolarized DLS data with schematic illustration describing the shape parameters Then, the slopes and the intercepts of the relaxation frequency (1/τ) dependences versus squared scattering vector modulus (q2) in the polarized/depolarized positions, respectively, of the Glan-Thompson prism polarizer led to the Dtrans and Θrotat values, respectively (Figure 7a and 7b). A dramatic decrease by a factor ca. 7 of Θrotat values with increasing the extent of


19

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 24

degalactosylation of poly(XGOn) (compared to the less moderate factor ca. 3 for Dtrans values) was observed and revealed that rotational motion and to a less extent, the translational motion, of anisotropic NCC was strongly influenced by the adsorption of glycopolymers as the extent of degalactosylation increased. A similar effect was found in case of (bio)polymers adsorbed onto NCC : the rotational motion is more strongly affected by the adsorption than the translational motion.38 The amplitude restriction of the rotational motion is consistent with the aggregate formation since rod-like NCC cannot rotate through a significant angle because of the presence of neighboring NCC.

CONCLUSION Inspired by carbohydrate/protein interactions that are of extremely low affinity but compensated in nature by a multivalent presentation of the carbohydrate ligands at the cell surface,

we

demonstrate

that

similarly the

polyvalent

presentation

of

xyloglucan

oligosaccharides grafted onto a polymer backbone bind strongly (Ka in sub-micromolar concentration) with rod-like NCC, thus mimicking the native hemicellulosic xyloglucan polysaccharide. These well-controlled structures of glycopolymers were obtained from a simple and versatile synthetic method using a click post-polymerization modification with a series of XGOs of various and finely tuned enzymatic-mediated degalactosylation rate. We put in evidence a sharp thermoresponsiveness of glycopolymers with a galactose removal higher than 30% where a chain extended-to-collapsed globule phase transition was observed at well-defined critical temperature. This behavior is opposite as compared to native tamarind seeds degalactosylated xyloglucan, that forms hydrogels, likely due to limitation of chain mobility in glycopolymers. Next, ITC and polarized/depolarized DLS studies, probing the interactions of


20

Page 21 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

glycopolymers with NCC, revealed that galactose removal unfavorably disturbed interaction strengths and induced aggregation phenomena. Our biomimetic approach brings a significant contribution for cellulose fiber surface modification through supramolecular functionalization with building block oligosaccharides bearing glycopolymers. This validated concept opens the way to the development of advanced biosourced glycomaterials facilitating their biotechnological and biomedical applications.

ASSOCIATED CONTENT Supporting Information. The Supporting Information is available free of charge on the ACS Publications website at http://pubs.acs.org. Characterization data of the synthesized compounds and additional Figures S1−S14

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected] ORCID Sami Halila: 0000-0002-9673-1099 Present Addresses #

Cixi Institute of Biomedical Engineering, Ningbo Institute of Materials Technology &

Engineering, Chinese Academy of Sciences, Ningbo 315201, China. Author Contributions


21

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 24

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes The authors declare no competing financial interest.

ACKNOWLEDGMENTS We acknowledge financial support from CNRS, Université Grenoble Alpes, PolyNat Carnot Institute (N° CARN 007-01) and the European Union’s Seventh Framework Programme (FP7/2007-2013) under Grant Agreement No. 603519. The authors thank Emilie Gillon for her help in the ITC measurements and Dr. Anne Imberty for helpful interpretations and discussions. We acknowledge Laurine Buon and Dr. Claire Boisset for their assistance in chromatography techniques and Dr. Yingjie Liao for performing AFM images. We are very grateful to Dr. Laurent Heux for providing us the cotton NCC suspensions. The authors acknowledge the Mass Spectrometry and NMR platform of ICMG FR2607.

REFERENCES (1) Cosgrove, D. J. Nat. Rev. Mol. Cell Biol. 2005, 6, 850–861. (2) Seeberger, P.H.; Werz, D.B. Nature 2007, 446, 1046–1051. (3) Shelke, N. B.; James, R.; Laurencin, C.T.; Kumbar, S. G. Polym. Adv. Technol. 2014, 25 448–460. (4) Yilmaz, G.; Becer, C. R. European Polym. J. 2013, 49, 3046–3051. (5) Godula, K.; Bertozzi, C. R. J. Am. Chem. Soc. 2012, 134, 15732−15742. (6) Oh, Y. I.; Sheng, G. J.; Chang, S.-K.; Hsieh-Wilson, L. C. Angew. Chem., Int. Ed. 2013, 52, 11796−11799. (7) Handa, K.; Hakomori, S.-I. Glycoconj. J. 2012, 29, 627–637. (8) de la Fuente, J. M.; Eaton, P.; Barrientos, A. G.; Menéndez, M.; Penadés, S. J. Am. Chem. Soc. 2005, 127 (17), 6192–6197. (9) Pauly, M.; Albersheim, P.; Darvill, A.; York, W. S. Plant J. 1999, 20, 629–639. (10) Carpita, N. C.; Gibeaut, D. M. Plant J. 1993, 3, 1–30. (11) Somerville, C.; Bauer, S.; Brininstool, G.; Facette, M.; Hamann, T.; Milne, J.; Osborne, E.; Paredez, A.; Persson, S.; Raab, T.; Vorwerk, S.; Youngs, H. Science 2004, 306, 2206–2211


22

Page 23 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

(12) Vincken, J. P.; Dekezier, A.; Beldman, G.; Voragen, A. G. J. Plant Physiol. 1995, 108, 1579–1585. (13) Zhou, Q.; Brumer, H.; Teeri. T. T. Macromolecules 2009, 42, 5430–5432. (14) Majoinen, J.; Haataja, J. S.; Appelhans, D.; Lederer, A.; Olszewska, A.; Seitsonen, J.; Aseyev, V.; Kontturi, E.; Rosilo, H.; Osterberg, M.; Houbenov, N.; Ikkala O. J. Am. Chem. Soc. 2014, 136 (3), 866–869. (15) Dong, X. M.; Kimura, T.; Revol, J. F.; Gray, D. G. Langmuir 1996, 12, 2076–2082. (16) Rittera, S. C.; König B. Chem. Commun. 2006, 45, 4694–4696. (17) Quemener, D.; Le Hellaye, M.; Bissett, C.; Davis, T. P.; Barner-Kowollik, C.; Stenzel, M. H. J. Polym. Sci., Part A: Polym. Chem. 2008, 46, 155–173. (18) Mishra, A.; Malhotra, A. V. J. Mater. Chem. 2009, 19, 8528–8536. (19) Halila, S.; Manguian, M.; Fort, S.; Cottaz, S.; Hamaide, T.; Fleury, E.; Driguez, H. Macromol. Chem. Phys. 2008, 209, 1282–1290. (20) Fry, S. C.; York, W. S.; Albersheim, P.; Darvill, A.; Hayashi, T.; Joseleau, J.-P.;Kato, Y.; Lorences, E. P.; Maclachlan, G. A.; McNeil, M.; Mort, A. J.; Grant Reid, J. S.; Seitz, H. U.; Selvendran, R. R.; Voragen, A. G. J.; White, A. R. Physiol. Plant. 1993, 89, 1–3. (21) Muller, F.; Manet, S.; Jean, B.; Chambat, G.; Boue, F.; Heux, L.; Cousin, F. Biomacromolecules 2011, 12(9), 3330–3336. (22) Sampedro, J.; Gianzo, C.; Iglesias, N.; Guitian, E.; Revilla, G.; Zarra, I. Plant Phys. 2012, 158, 1146–1157. (23) Benselfelt, T.; Cranston, E. D.; Ondaral, S.; Johansson, E.; Brumer, H.; Rutland, M. W.; Wagberg, L. Biomacromolecules 2016, 17(9), 2801–2811. (24) York, W. S.; Harvey, L. K.; Guillen, R.; Albersheim, P.; Darvill, A. G. Carbohydr. Res. 1993, 248, 285–301. (25) Bertrand, A.; Stenzel, M.; Fleury, E.; Bernard, J. Polym. Chem. 2012, 3, 377–383. (26) Hanus, J.; Mazeau, K. Biopolymers 2006, 82(1), 59–73. (27) Park, Y. B.; Cosgrove D. J. Plant Cell Physiol. 2015, 56(2), 180–194. (28) Shirakawa, M.; Yamatoya, K.; Nishinari, K. Food Hydrocolloids 1998, 12, 25–28. (29) Nisbet, D. R.; Crompton, K. E.; Hamilton, S. D.; Shirakawa, S.; Prankerd, R. J.; Finkelstein, D. I.; Horne, M. K.; Forsythe, J. S. Biophysical Chemistry 2006, 121, 14–20. (30) Hayashi, T.; Takeda, T.; Ogawa, K.; Mitsuishi, Y. Plant Cell Physiol. 1994, 35, 893–899. (31) Elazzouzi-Hafraoui, S.; Nishiyama, Y.; Putaux, J.-L.; Heux, L.; Dubreuil, F; Rochas, C. Biomacromolecules, 2008, 9, 57–65. (32) Dam TK1, Brewer CF. Chem. Rev. 2002, 102 (2), 387–430. (33) .Mazeau, K.; Wyszomirski, M. Cellulose, 2012, 19, 1495–1506. (34) Lopez, M.; Bizot, H.; Chambat, G.; Marais, M. -F.; Zykwinska, A.; Ralet, M. -C.; Driguez, H.; Buleon, A. Biomacromolecules, 2010, 11, 1417–1428. (35) de Souza Lima, M. M.; Borsali, R. Langmuir 2002, 18, 992–996. (36) de Souza Lima, M. M.; Wong, J. T.; Paillet, M.; Borsali, R.; Pecora, R. Langmuir 2003, 19, 24–29. (37) Pecora, R. J. Chem. Phys. 1968, 48, 4126–4128. (38) Sim, J. H.; Dong, S.; Röemhild, K.; Kaya, A.; Sohn, D.; Tanaka, K.; Roman, M.; Heinze, T.; Esker, A. R. J. Colloid Interface Sci. 2015, 440, 119–125. For Table of Contents Only


23

Biomacromolecules

Page 24 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment

24

Hemicellulosic Polysaccharides Mimics: Synthesis of Tailored

Recommend Documents