Other Techniques Available

Density Measurement

The Centre is equipped with 2 Anton Paar DMA O2C precision densimeters for the determination of macromolecular partial specific volume (vbar) or solvent/solution density. For this procedure 1ml of sample is required. For vbar determinations mg/ml concentrations are necessary. In most circumstances it is also possible to calculate vbar from the amino acid composition of a peptide using the additivity rule as in the Biomols or Sednterp software.

Volume-Temperature Dispersion is the measurement of vbar as a function of temperature.
 

Refractometric Measurements

The refractive index increment can be measured by means of an Atago DD-5 precision differential refractometer. At least 0.5 ml of sample of accurately known solute concentration is required, together with several ml of exhaustively dialysed buffer solution. The maximum usable concentration is 20 mg/ml and the precision is ± 0.01 mg/ml in concentration units. This unit measures at a wavelength of 660nm.
 

Volume-Temperature Dispersion of Proteins

Introduction

As a general phenomenon, matter of any type tends to expand as the temperature is raised. Proteins are no exception, and their expansion on heating, in the range which might be loosely called ‘cold-room to ambient’, is well known and documented. To be more precise, the increase in partial specific volume (vbar) which occurs in that temperature range has been studied: the definition of vbar is strictly in terms of the partial derivative of the volume of the system with respect to change in concentration of the (protein) solute. The volume of the protein domain could be defined in a number of ways. The domain of the covalently-linked atomic volumes, as defined by electron density in crystallography, or (less readily) of the interactions in NMR, corresponds to what has been traditionally looked upon as the ‘anhydrous’ domain. In a simple model this domain of the covalently-linked atomic volumes would be accurately estimated by the quantity vbar.

However, solution properties of macromolecules, including hydrodynamic properties, are more readily related to domains such as the hydrodynamic volume (in specific terms Vs) or the excluded volume (u). The protein domain defined in these terms is always larger than that defined by vbar. Studies on the variation of the volume of the protein domain defined in these alternative ways are almost totally lacking, and could not readily be undertaken without a prior understanding of the way in which vbar varies with temperature.

Attention has been focussed for many years on the high temperature range, 40-100 degrees. Here the protein denatures, and it has long been appreciated that denaturation is accompanied by a significant fall in vbar. It has seemed to us to be of some interest to monitor changes in vbar in the physiological (warm-blooded) range. To achieve this by a method less tedious than repeated single densitometric determinations we have devised a new approach based upon differential use of the Kratky digital densimeter. This new approach has turned out to be more precise than we had expected, and the results which we have obtained in the range 30-40 degrees C show wholly unexpected features, as (to a lesser extent) do results in the 5-30 degree range. We term the study of the temperature dependence of macromolecular vbar ‘Volume Temperature Dispersion’ (VTD).

The precision measurement of Volume Temperature Dispersion (VTD)

The Kratky precision densimeter is capable of making measurements of extremely high precision (<1 in 10^7). However, the precision in vbar as measured conventionally is seriously limited by

(i) limitations in knowledge of the solute mass concentration

(ii) possible drift in temperature of the measurement cell between measurements being made on solution and on reference solvent

Thus, in performing individual determinations of vbar between 10 and 25 degrees C, one would have uncertainties arising from both (i) – due to possible surface adsorption of a multiply used sample, and (ii) in all 10 cases. The procedure would also be extremely tedious. Using a single-cell instrument, we have found it simple to avoid these problems when studying VTD, by recording first the oscillation times for pure solvent at a range of temperatures where each temperature is precisely known (from an accurate thermometer) but not precisely fixed. This approach enables much shorter equilibration times to be used. Values for the oscillation times at 1 degree intervals are then computed by sliding strip least squares fitting procedures. The same whole operation is then performed for the protein solution, enabling a precise estimate for vbar to be yielded for each temperature interval.

Obviously, the absolute values for vbar thus yielded remain uncertain by virtue of reason (i) above. However, it is the relative change in vbar which we are studying in VTD.