The Svedberg Lecture 2017. From nano to micro: the huge dynamic range of the analytical ultracentrifuge for characterising the sizes, shapes and interactions of molecules and assemblies in Biochemistry and Polymer Science
The analytical ultracentrifuge (AUC) invented by T. Svedberg has now become an extremely versatile and diverse tool in Biochemistry and Polymer Science for the characterisation of the sizes, shapes and interactions of particles ranging in size from a few nanometres to tens of microns, or in molecular weight, M (molar mass) terms from a few hundred daltons to hundreds of megadaltons. We illustrate this diversity by reviewing recent work on (1) small lignin-like isoeugenols of M ~ 0.4–0.9 kDa for archaeological wood conservation, (2) protein-like association of a functional amino-cellulose M = 3.25 kDa, (3) a small glycopeptide antibiotic (M ~ 1.5 kDa) and its association with a protein involved in antibiotic resistance (M ~ 47 kDa), (4) tetanus toxoid protein TTP (M ~ 150 kDa) and (5) the incorporation of TTP into two huge glycoconjugates considered in glycovaccine development with molecular weight species in a broad distribution appearing to reach 100 MDa. In illustrating the diversity, we will highlight developments in hydrodynamic analysis which have made the AUC such an exciting and important instrument, and point to a potential future development for extending its capability to highly concentrated systems.
KeywordsLignin Amino-cellulose Vancomycin Tetanus toxoid Glycovaccines
It has been an honour for me to have been invited by the Organising Committee to give the 2017 Svedberg Lecture. The analytical ultracentrifuge (AUC) invented by Thé (Théodore) Svedberg has now become an extremely versatile and diverse tool in Biochemistry and Polymer Science for the characterisation of the sizes, shapes and interactions of particles ranging in size from a few nanometres to tens of microns, or in molecular weight (molar mass) terms from a few hundred daltons (g/mol) to hundreds of megadaltons (Scott et al. 2005).
The unique property of being a true solution or “matrix free” technique (without immobilisation onto a surface or ionization/vaporisation) and possessing an inherent fractionation ability without the need of a separation matrix (column or membrane), and for molecular weight/size determination, without the need for calibration standards renders it an essential tool in macromolecular Biochemistry and Polymer Science. It can be used for the analysis of interactions in solution (self-interactions or “self-association” and macromolecular-ligand interactions) in terms of stoichiometries and reversibility (especially when the molecular weights of the interacting species are known precisely from mass spectroscopy), and when used in combination with other techniques such as viscometry, nuclear magnetic resonance, light scattering and X-ray scattering, can provide information about the overall conformation (shapes) of macromolecules in free solution (García de la Torre and Harding 2013).
We illustrate this diversity by reviewing the seminal role played by the analytical ultracentrifuge in recently published studies we have been involved with: (1) lignin-like isoeugenols (of molecular weights M = 0.4–0.9 kDa) of potential importance in the conservation of archaeological wood, (2) protein-like self-association of a functional amino-cellulose (M = 3.25 kDa), (3) dimerization of a small (M = 1.449 kDa) glycopeptide antibiotic vancomycin of current interest in the field of understanding the mechanisms behind antimicrobial resistance, (4) oligomerisation of tetanus toxoid protein (M = 150 kDa) and its incorporation into (5) two huge glycoconjugate vaccines with a very broad distribution of material with molecular weights up to 100 MDa. In illustrating the diversity, we will highlight developments in hydrodynamic analysis which have made the AUC a relevant technology. We conclude by pointing to a potential future development for extending its capability to highly concentrated systems, relevant, for example, for the characterisation of the behaviour of monoclonal antibodies at the high concentrations they are used for administration.
Lignin-like isoeugenols (M ~ 0.4–0.9 kDa)
Native lignins were characterized by our laboratory (Alzahrani et al. 2016) and we showed by sedimentation equilibrium in the analytical ultracentrifuge (performed in dimethylsulfoxide), analysed using the SEDFIT-MSTAR algorithm developed with P. Schuck (Schuck et al. 2014) based on the M* function of Creeth and Harding (1982) that lignins from different wood sources (“alcell” and “kraft”) had weight average molecular weights of ~ 20 kDa (19 kDa alcell and 25 kDa kraft). Analysis of the distributions of molecular weight using the MULTISIG algorithm (Gillis et al. 2013) showed a relatively broad distribution for the alcell compared with the kraft, and by combining the molecular weights with intrinsic viscosity data it was possible to show using the ELLIPS1 algorithm (García de la Torre and Harding 2013) that both adopted a discoid structure of aspect ratio ~ 30:1 consistent with previous work.
McHale et al. (2016, 2017) have been seeking to develop lignin replacements for the decayed wood using isoeugenols (monomer M ~ 160 Da, comparable to that of an amino-acid or carbohydrate residue) which are structurally very similar to lignins, built up from very similar monomer units (Fig. 2c), but of lower molecular weight (low enough to be absorbed into the wood), but have to be made to polymerize in situ using a peroxidase in the presence of hydrogen peroxide H2O2. In this early study, in situ polymerized materials in dimethyl sulfoxide (DMSO) were extracted and analysed using SEDFIT-MSTAR (Schuck et al. 2014) showing a moderate degree of association to give molecular weights (Fig. 2d) corresponding to ~ 3 and 6 mers (M ~ 400–900 Da); so clearly, this is a step in the right direction. More recently, the synthesis of polymers of M ~ 1600 Da has been achieved and further work is ongoing in the bid to extend the polymerization to 3–4000 Da.
Amino-celluloses (M ~ 3.25–13 kDa)
The related polycationic polysaccharide chitosan—a partially deacetylated form of chitin (poly-N-acetyl glucosamine) has also been considered. Here the starting materials are usually too large and the chitosans have to be depolymerised to molecular weights M < 4 kDa, and a recent study by Wakefield et al. (2018) successfully used a combination of UV radiation and hydrogen peroxide to bring the weight average molecular weight—as assessed by sedimentation equilibrium analysis—to within this range. Again, measurements were done in aqueous solution. The task is to find an amino-cellulose or chitosan of suitable molecular weight that is soluble in non-aqueous medium and with the desired properties to penetrate the wood and be capable of interacting with and reinforcing the fibrils within the wood, and repolymerisation within the wooden structure.
Dimerization of the glycopeptide antibiotic vancomycin (M ~ 1.5 kDa) and its interaction with the membrane protein VanS (M ~ 47 kDa)
Dimerisation of the tetanus toxoid protein (M ~ 150 kDa)
Capsular polysaccharides from Haemophilus influenzae type b (weight average M w ~ 1200 kDa) and glycoconjugates (M w ~ 7300 kDa) with TTP
The polysaccharides from both bacteria were shown to adopt flexible coil conformations with low persistence lengths Lp < 10 nm on the basis of a global minimization method known as HYDFIT developed by Ortega and García de la Torre (2007). HYDFIT involves the minimization of a target function ∆ for Lp and the mass per unit length ML based on the Yamakawa–Fuji relations linking the sedimentation coefficient with molecular weight and corresponding Bohdanecky relations linking the intrinsic viscosity with molecular weight (Fig. 8b). Similar flexible coil conformations with low persistence lengths were also found for the polysaccharides activated with a cross linker for conjugation—and also their glycoconjugates with TTP—showing that they adopted the conformation properties of the polysaccharides and not the TTP. This was confirmed by a method known as conformation zoning (Fig. 8c), which involves the combination of the sedimentation coefficient and Gralen coefficient κs (from the concentration dependence of s) developed by Pavlov et al. (1997). The combined s and κs values are only commensurate with a random coil conformation, confirming the HYDFIT results (Fig. 8b).
Future trends—concentrated systems
I hope I have been able to provide a snapshot of the great diversity of sizes and types of substance that can be successfully characterised by the analytical ultracentrifuge. The examples I have focused on have all referred to dilute solution conditions, and these are well within the capabilities of the current optical systems available on commercially available instrumentation, most notably the Beckman-Coulter Optima XL-I analytical ultracentrifuge with its dual system of UV absorbance and Rayleigh interference. In my presentation I have not really touched on concentrated systems, and this is becoming increasingly important particularly in the field of monoclonal antibody research, a key area for drug development. We have been involved in antibody research for over three decades. This started with collaborations with researchers at USB Celltech and Prof. Dennis Burton’s group at the University of Sheffield and then the Scripps Institute at La Jolla. Using bead modelling developed by J. García de la Torre and co-workers, the first hydrodynamic model for IgE was produced correctly predicting the cusp shape conformation (Davis et al. 1990) and by 2007 papers were appearing (see e.g. Nobbmann et al. 2007; Lu et al. 2008) reporting the use of the analytical ultracentrifuge for assessing the stability of antibody preparations to processing (freeze–thaw and storage at elevated temperature), and in conjunction with other techniques such as dynamic light scattering (Nobbmann et al. 2007). Now the focus is very much on concentrated systems because of the high concentrations (80 mg/ml and higher) that are considered for administration. Unfortunately, these concentrations are out of reach of the current commercially available optical systems, although, oddly, the Schlieren (refractive index gradient) optical system—available in the original Svedberg analytical ultracentrifuge (Lloyd 1974) may be useful in this regard. Figure 1 shows such an image obtained with this system and the famous Beckman Model E had this—and concentrations > 80 mg/ml were possible. There may be a case for reintroduction of this type of system. The ability to measure at such high concentrations comes at a price, and issues of thermodynamic (or hydrodynamic) non-ideality caused by co-exclusion and polyelectrolyte effects can become serious. In the past, non-ideality phenomena were sometimes seen as advantageous as it was possible to use the second thermodynamic virial coefficient B to estimate the triaxial shape of a macromolecule using relations worked out by Rallison and Harding (1985). Indeed if B could be accurately calculated from knowledge of the overall structure of a macromolecule, then it could be eliminated as a variable in the thermodynamic equations—and this was made possible via the COVOL programme (Harding et al. 1999; Harding 2013). Finding corresponding relations for the equivalent hydrodynamic non-ideality coefficients at high concentration is still a major challenge.
I would like to thank all my colleagues—too numerous to list—with whom it has been a great pleasure working with over the last 40 years, and in particular Drs. Colin Blake and Professor D.C. Phillips of the Laboratory of Molecular Biophysics at the University of Oxford, Professor Arthur Rowe, Universities of Leicester and Nottingham, Dr. J. Michael Creeth, University of Bristol and Dr. Paley Johnson, Department of Biochemistry at the University of Cambridge. I would like also to thank Dr. M. Phillips-Jones for providing Fig. 4a, b and for her input into the significance of the VanS system. This work was supported in part by the UK Biotechnology and Biological Sciences Research Council [grant number BB/L025477/1].
- Abdelhameed AS, Adams GG, Morris GA, Almutairi FM, Duvivier P, Conrath SE, Harding SE (2016a) A glycoconjugate of Haemophilus influenzae type b capsular polysaccharide with tetanus toxoid protein: hydrodynamic properties mainly influenced by the carbohydrate. Sci Rep 6:22208CrossRefPubMedPubMedCentralGoogle Scholar
- Harding SE (2017) Science and the Vikings. The Hakon Hakonsson Lecture 2016. Largs and District Historical Society, Largs, pp 20–26Google Scholar
- Lloyd PH (1974) Optical methods in ultracentrifugation, electrophoresis, and diffusion: with a guide to the interpretation of records. Clarendon Press, OxfordGoogle Scholar
- Lu Y, Harding SE, Rowe AJ, Davis KG, Fish B, Varley P, Gee C, Mulot S (2008) The effect of a point mutation on the stability of IgG4 as monitored by analytical ultracentrifugation. J Pharm Sci 97:948–957Google Scholar
- Moscoso CG, Sun Y, Poon S, Xing L, Kan E, Martin L, Green D, Lin F, Vahlne AG, Barnett S, Srivastava I, Cheng RH (2011) Quaternary structures of HIV Env immunogen exhibit conformational vicissitudes and interface diminution elicited by ligand binding. Proc Natl Acad Sci USA 108:6091–6096CrossRefPubMedGoogle Scholar
- Phillips-Jones MK, Channell G, Kelsall CJ, Hughes CS, Ashcroft AE, Patching SG, Dinu V, Gillis RB, Adams GA, Harding SE (2017a) Hydrodynamics of the VanA-type VanS histidine kinase: an extended solution conformation and first evidence for interactions with vancomycin. Sci Reports 7:46180CrossRefGoogle Scholar
- Scott DJ, Harding SE, Rowe AJ (eds) (2005) Analytical ultracentrifugation: techniques and methods. Royal Society of Chemistry, Cambridge, UK Google Scholar
- Stöckel J, Meinl E, Hahnel C, Malotka J, Seitz R, Drexler K, Wekerle H, Dornmair K (1994) Refolding of human class II major histocompatibility complex molecules isolated from Escherichia coli: assembly of peptide-free heterodimers and increased refolding-yield in the presence of antigenic peptide. J Biol Chem 269:29571–29578PubMedGoogle Scholar
- Wakefield JMK, Gillis RB, Adams GG, McQueen CMA, Harding SE (2018) Controlled depolymerisation assessed by analytical ultracentrifugation of low molecular weight chitosan for use in archaeological conservation. Eur Biophys J. https://doi.org/10.1007/s00249-018-1290-6
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.