Stephen Harding has been director of the National Centre for Macromolecular Hydrodynamics for the last 24 years, a period which has seen hydrodynamic methods (analytical ultracentrifugation, light scattering, viscometry etc.) come from almost oblivion to flourish as tools for biomolecular characterisation. He has published over 350 research papers and 9 books in the field of biomolecular hydrodynamics and applied biochemistry. In recognition of his discoveries he received a junior medal from the Royal Society of Chemistry in 1991 and a DSc from the University of Oxford in 2002. In 2010 he gave the prestigious annual Pfizer lecture at the University of Connecticut and in 2012 he was awarded by King Harald V den Kongelige Norske Fortjenstorden - the Royal Norwegian Order of Merit (Knight of the 1st class) - the highest award to a foreign non Head of State. His current focus is on biopolymer stability in response to bioprocessing, of particular interest to the Biopharma industry where he now draws much of his support.

Stability of Complex Carbohydrate Structures

Biofuels, Foods, Vaccines and Shipwrecks

By Stephen E. Harding

The Royal Society of Chemistry

Copyright © 2013 The Royal Society of Chemistry
All rights reserved.
ISBN: 978-1-84973-563-6

Contents

1 Carbohydrates: First Cousins of Water F. Franks, 1,
2 Enzymatic Stability of Starches C.-L. Lin and R. F. Tester, 13,
3 Enzymatic Degradation of Cell Wall Polysaccharides G.A. Tucker, 28,
4 Functional Components and Mechanisms of Action of 'Dietary Fibre' in the Upper Gastrointestinal Tract: Implications for Health T. Grassby, C.H. Edwards, M. Grundy, P.R. Ellis, 36,
5 Stability and Degradation Pathways of Polysaccharide and Glycoconjugate Vaccines C. Jones, 56,
6 Mucin Turnover A.P. Corfield, 68,
7 Viscometry, Analytical Ultracentrifugation and Light Scattering Probes for Carbohydrate Stability S.E. Harding, 80,
8 Stability of Pectin-Based Drug Delivery Systems G.A. Morris, G.G. Adams, S.E. Harding, J.D. Castile and A. Smith, 99,
9 Stability of Polysaccharide Complexes: The Effect of Media and Temperature on the Physical Characteristics and Stability of Chitosan-Triphosphate/Alginate Nanogels C.A. Schiitz, P. Käuper and C. Wandrey, 110,
10 Cellulose Crystallinity: Perspectives from Spectroscopy and Diffraction M. C. Jarvis, 125,
11 The Water Vapour Sorption Properties of Cellulose C.A.S. Hill and Y. Xie, 137,
12 Lignin Biosynthesis and Lignin Manipulation P. Daly, M. Maluk, M. Zwirek and C. Halpin, 153,
13 Bacterial degradation of Archaeological Wood in Anoxic Waterlogged Environments N.B. Pedersen, C.G. Bjordal, P. Jensen and C. Felby, 160,
Subject Index, 188,

CHAPTER 1

CARBOHYDRATES: FIRST COUSINS OF WATER

F. Franks

BioUpdate Foundation, 25 The Fountains, Ballards Lane, London N3 1NL asdi35@dsl.pipex.com

1 INTRODUCTION

Water has frequently been described as the most eccentric molecule in our ecosphere. Its many anomalous physical properties derive basically from the sp3 hybridisation of the oxygen orbitals, which gives rise to an almost tetrahedral, quadrupolar bond orientation, in which the oxygen atom is placed at the centre of the water tetrahedron and four charges, two -OH groups (positive) and two lone electron pairs (negative) are directed towards the vertices, as shown in Figure 1. That, in itself, is not a unique molecular feature; for instance SiO2 and GeO2 have similar tetrahedral configurations. The unique features of H2O are twofold: 1) its inability to form or participate in stable covalent bonds. Hydrogen bonding is thus the only method by which the molecule can interact with other molecules, including other water molecules, and 2) because of its quadrupolar nature with an equal number of proton donor and acceptor sites, H2O is thus amphipathic, and as such, it can participate in a wide range of reactions: proton donor/acceptor, oxidation/reduction, and hydrolysis/aggregation. To some extent, carbohydrates are able to participate in the same types of reactions, although the rates in fused (vitreous) carbohydrates will differ vastly from those in liquid aqueous solutions.

The description of molecules composed of C, H and O as 'carbohydrates' provides a clue for some of their properties that are of particular significance to their industrial and medical applications. Another generic description is polyhydroxy compounds (PHC), which provides a further clue – water sensitivity/miscibility – to their usefulness. Added to this are their diverse functions in life processes, providing structures, energy storage, metabolic intermediates, recognition and defence (immune) systems and much else. The generic formula of many simple PHC compounds, when written as [FORMULA OMITTED], also suggests a close relationship between water and PHC molecules. Indeed, any material in our ecosphere that is composed mainly of C, H and O is either miscible with water, or at least sensitive to water, so that dry (anhydrous) organic materials do not exist in nature. Water has been described as the 'ubiquitous plasticizer' for all organic matter.

The actual relationship between water and PHCs at the molecular level lies in their almost identical oxygen orbital hybridization: -OH groups of the organic molecules structurally and energetically resemble those of water, i.e. in their interactions they are limited to hydrogen bonding and their bond orientations are also in the form of a tetrahedron. It indicates that PHC molecules in their crystalline state closely resemble ice and are held together (or apart) by weak interactions and, like ice, they can form chains, rings and infinite three dimensional networks. In their crystalline states they can also interact with water in the form of stoichiometric hydrates, e.g. glucose. H2O,α,α-trehalose.2H2O, β, β-trehalose.4H2O, raffmose.5H2O. Like inorganic hydrates, for example [Na.sub.2]C[O.sub.3].10[H.sub.2]O, they should be able to be dehydrated to lower hydration states and finally to anhydrous crystals, but unlike their inorganic counterparts, most of these hypothetical solid/solid transitions take place very slowly and proceed via long-lived non-crystalline (amorphous) intermediate states; characteristic transition temperatures cannot therefore be easily observed. Their phase transitions usually take place over very extended periods (months or years), and very few such systems have actually been studied directly in 'real time'. Therefore, the dehydrations appear to be irreversible in real time, so that the mixtures may consist of the higher and lower hydration states and the released water. The mixture will thus remain thermodynamically unstable and maintain an apparent stability due to exceedingly slow molecular motions, in the order of mm/century, until a spontaneous devitrification to the more stable, crystalline state, e.g. during annealing, occurs and is accompanied by more or less severe changes in their mechanical properties, such as tensile or torsional strength. Examples of such spontaneous devitrifications readily occur in metal alloys, in the crystallisation of cholesterol from its highly supersaturated solution in the gall bladder, or in the crystallisation of lactose in ice cream. The results may be of different orders of severity, ranging from loss of life to physical pain and minor inconvenience. Thus, several of the first ever jet-engined commercial airliners, the de Havilland Comet, disappeared during long-haul flights, when the fuselage disintegrated, resulting from metal stress and subsequent devitrification. The nucleation and biocrystallisation of cholesterol in the gall bladder results from a chemically minor error in the biosynthesis of bile salt molecules, whose natural function is the prevention of cholesterol precipitation from its supersaturated solution. The possibility of random devitrification of lactose in some cartons of ice cream, even during frozen storage, is still a subject of research.

Other possible sources of PHC dehydration problems include the decomposition of a previously unknown PHC hydrate, especially in pharmaceutical formulations, where such release of water during processing, e.g. mannitol. H2O -> mannitol + H2O, can result in major stability losses in the drug product. With our present sketchy knowledge of the close relationship between water and the hydrogen bond topology of PHCs, it is likely that more such presently unknown hydrates remain to be discovered.

It is also likely that future experimental data on crystalline PHC dehydration will remain to be of uncertain value, because of the extended periods required for these reactions to go to completion.

2 THE CONCEPT OF PHC HYDRATION AND ITS MEASUREMENT

To study binary mixtures of molecules which differ substantially in size and shape but closely resemble each other in their interactions (hydrogen bonds), certain simplifications are necessary. In the past, the concept of 'bound water' was developed to provide explanations to the above problems. In its simplest form it consisted of covering the exposed area of the PHC molecule with notionally spherical water molecules and counting the minimum number that would completely cover the PHC. Eventually experimental techniques, such as neutron scattering became available to verify (or discount) such simple approaches. They proved useful to establish the geometry of ion hydration and the assignment of credible hydration numbers, especially for monatomic ions.

However, even with molecules such as mono-, di-, or oligosaccharides in their aqueous mixtures, and leaving aside polymeric PHCs, such approaches can hardly be considered as satisfactory. In addition, the PHC in solution often exists as mixtures of two or more anomeric states. Finally, the sciences of physics and physical chemistry do not recognise the existence of different species of H2O, bound or free. In order to provide a measure of credibility for such a distinction, certain criteria must be established as to how 'bound' can be defined for the water-PHC system. At its simplest, these criteria are based on structure, energetics and dynamics:

(i) Structure: measured in terms of distances, orientations and Euler angles, by X-ray or neutron diffraction and circular dichroism

(ii) Energetics: the existence of special, probably favourable interaction energies of certain hydrogen bonds in specific locations, measured (with difficulty) by thermodynamic P-V-T properties, calorimetry and volumetry; also in principle by infrared or Raman spectroscopy

(iii) Dynamics: exchange rates of water molecules considered to be bound, their rotational and/or translational diffusion rates, measured by relaxation techniques, such as dielectric relaxation or nuclear magnetic resonance (NMR). A critical evaluation of the above, and other techniques, as applied to studies of PHC-water systems, is given in ref. 8.

3 HYDROGEN-BOND NETWORKS IN PHC-WATER SYSTEMS

In PHC-water mixtures, the following hydrogen bonds may be considered, where OW and OC refer to oxygen atoms of water and a PHC molecule respectively, and the colon ":" refers to a proton acceptor site:

WO-H -> :OC CO-H -> :OW WO-h -> :OW CO-H -> :OC etc.

The study of hydrogen bonding in its various aspects must centre on the behaviour of the exchangeable proton and details of the bond geometry. Unfortunately, spectroscopy has for long been obsessed with the desirability of narrow signals, and it was for many years a habit to exchange -O1H protons by -O2H, thereby simplifying the spectra, but also removing the spectral features that provide the information about hydrogen bonds. Bearing in mind that bond lengths and angles in water are almost identical with those in PHCs, direct studies of hydration must in most cases be subject to major experimental problems.

In attempts to study interactions between sugar -OH groups and water, Symons and co-workers first observed that over narrow pH ranges and at low temperatures, high resolution 1H NMR signals due to sugar -OH groups could be resolved at 1 – 3 ppm downfield from the H2O signal, measured at 100 MHz. For the identification of a structural influence on hydrogen bonding in binary sugar-water mixtures, compare hypothetical hydration systems in hexoses and pentoses in their α- with β-anomeric forms, respectively. Anomers differ only in the configuration of the -OH groups on C(1) (axial versus equatorial). At ordinary temperatures the proton transfer WO-H -> :OC(1) is fast, and any hydration difference between the α and β-anomers could for a long time not be directly measured. However, as shown in Figure 2, by inserting an α- or β-hexopyranose molecule into the notional tetrahedral 'structure' of liquid water, but without causing strain in the hydrogen bonded network, it is apparent that a significantly better hydration fit is provided for the β-C(1)O-H interactions with water, than can be produced with the α-anomer.

In order to establish the nature and possible effects of any such differences, an extensive series of experimental studies was initiated by a group of scientists at the Unilever Colworth Research Laboratories during the 1970s. The methodologies and results are summarised in ref. 8. The techniques included calorimetry, PVT properties, dielectric relaxation (frequency and time domain) and NMR relaxation (13C, 1H, 2H, 17O) of several mono- and oligosaccharides in aqueous solution. It was a happy coincidence that at the same time nucleation and crystallisation processes of water at subzero temperatures were under intensive study, and we developed practical methods for achieving the undercooling of liquid water down to ca. -40°C, without the necessity of adding cryoprotective agents. This made possible the slowing down and the resolution of individual steps in coupled multistep processes.

The technique proved to be particularly useful for the separation and analysis of individual steps in complex biochemical processes, e.g. bioluminescence. We applied the techniques of deep undercooling to the study of proton residence times and exchange rates for several sugar-water systems and obtained time resolved 1H spectra from which proton residence times and activation energies of the proton transfer C(1)O-H -> :OW could be calculated for glucose and a number of related sugars, in particular the corresponding pentose anomers of xylose. It had already been observed that the chemical shifts for xylose are almost identical with those of glucose (although at different pH values), indicating that the CH2OH group on C(5) of xylose affects neither the acidity nor the solvation of the anomeric -OH. This conclusion is further confirmed by the very similar anomeric equilibrium compositions of the two sugars in aqueous solution. On the other hand, the positions of the ring -OH groups markedly affect the -OH shifts. This is seen by comparing the pentoses xylose and ribose (in ribose the -O(3)H is axial) with the hexoses glucose and mannose (axial-O(2)H), confirming previous suggestions that the hydration of sugars is subject to a high degree of stereospecificity.

Dissimilarities in the thermodynamic solution properties, the free energy ΔG and its T and P derivatives, of ribose and xylose also become apparent from a comparison of their osmotic coefficients (the second virial coefficient in the osmotic pressure equation). Whereas aqueous solutions of ribose are close to ideal, xylose exhibits marked positive deviations from ideal behaviour, indicating a net repulsion between xylose molecules.

The above discussion of pairs of single, but related sugar molecules in a hypothetical aqueous environment may seem laborious and convoluted, but it points to distinct differences in the interactions between water and chemically very similar carbohydrate molecules. These features have been classified as examples of 'specific hydration', i.e. they are exclusively observed in aqueous solutions and are related to differences in the stereochemical details of hydrogen bonds between water and specific -OH sites of PHC molecules.

4 PRACTICAL ASPECTS OF THE 'FIRST COUSIN' RELATIONSHIP

The isolated molecule and hypothetical water structure approaches are of limited practical value. Rather, the bulk properties of highly concentrated, supersaturated mixtures, i.e. the material science aspects, need to be considered. Tests must be applied whether the special relationship between PHC molecules and tetrahedrally 'structured' water molecules yields useful information about bulk properties such as undercooling, supersaturation, vitrification, crystallisation, viscous flow, elastic moduli, structural and chemical stability, etc.

Fortunately, here again there exists an excellent series of publications that probe such relationships. Basically, the experiments describe the rotational and translational diffusion of water relative to those of different parts of the PHC molecules, as the concentration increases to beyond the saturation value. The techniques encompassed 13C NMR for the PHCs and -OD, -CH2OD and D2O, a so-called "hydration phase", by 2H NMR over a temperature range (220-350 K) and concentrations (10-70% w/w) for sucrose, α,α-trehalose, maltose and α-D-methyl glucoside. The following three questions were addressed, namely: 1. How do PHCs affect the dynamics of water? 2. How do the dynamics of PHC molecules depend on the concentration? 3. What conditions determine the transition (glass transition Tg) into the vitreous state?

In summary, some of the results of the Girlich thesis can be discussed in terms of three distinct concentration domains (Figure 3).

4.1 The case of the concentration, c ≤ 30%

PHC molecules perform uncorrelated motions, and the solution viscosity exhibits a minimal concentration dependence. The dynamics of water molecules are governed by cooperative fluctuations within the hydrogen-bonded water network. In undercooled solutions the network increasingly inhibits the orientations and the rotational correlation time (τ) increases steeply with decreasing temperature. Hydrostatic pressure distorts the network and, in the metastable phase, it inhibits the generation of ordered domains with linear H-bonds; molecular motions thus remain rapid. Dissolved PHC molecules, because of their incompatible -OH orientations, produce a similar effect. This facilitates the undercooling of PHC solutions far into the metastable region, resulting in eventual vitrification, rather than crystallisation.

(Continues...)Excerpted from Stability of Complex Carbohydrate Structures by Stephen E. Harding. Copyright © 2013 The Royal Society of Chemistry. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.