Turbidity as a probe of tubulin polymerization kinetics: A theoretical and experimental re-examination Damien Hall a, * , Allen P. Minton b a Department of Chemistry, Cambridge University, Cambridge CB2 1EW, UK b Section on Physical Biochemistry, National Institute of Diabetes, Digestive, and Kidney Disease, National Institutes of Health, Bethesda, MD 20892, USA Received 6 May 2005 Available online 28 July 2005 Abstract We report here an examination of the validity of the experimental practice of using solution turbidity to study the polymerization kinetics of microtubule formation. The investigative approach proceeds via numerical solution of model rate equations to yield the time dependence of each microtubule species, followed by the calculation of the time- and wavelength-dependent turbidity generated by the calculated distribution of rod lengths. The wavelength dependence of the turbidity along the time course is analyzed to search for generalized kinetic regimes that satisfy a constant proportionality relationship between the observed turbidity and the weight concentration of polymerized tubulin. An empirical analysis, which permits valid interpretation of turbidity data for distributions of microtubules that are not long relative to the wavelength of incident light, is proposed. The basic correctness of the simulation work is shown by the analysis of the experimental time dependence of the turbidity wavelength exponent for microtubule formation in taxol-supplemented 0.1 M Pipes buffer (1 mM GTP, 1 mM EGTA, 1 mM MgSO 4 , pH 6.4). We believe that the general findings and principles outlined here are applicable to studies of other fibril-forming systems that use turbidity as a marker of polymerization progress. Crown Copyright Ó 2005 Published by Elsevier Inc. All rights reserved. Keywords: Tubulin; Microtubules; Kinetics; Polymerization; Characterization Introduction The polymerization of tubulin to form microtubules has been studied using a large number of experimental techniques. These techniques have included various forms of light scattering, sedimentation analysis, dye and drug binding, GTP use, fluorescence, video and elec- tron microscopy, and viscosity measurement [1]. The utility of the various experimental techniques relies on their ability to equate the observed signal with a set ex- tent of polymerized tubulin. In a landmark article, Gas- kin et al. [2] showed that the turbidity recorded from a steady-state mixture of microtubules was approximately linearly proportional to the total mass of protein recov- ered from the pelletable fraction of a microtubule/tubu- lin mixture. They noted that the wavelength exponent of the turbidity generated by steady-state solutions of microtubules was closer to the negative third power than to the negative fourth power. This experimental observa- tion was lent theoretical support in an appendix by Berne [3]. BerneÕs approximation suggested that for a randomly oriented distribution of long rods having a width much smaller than the wavelength of light (k/10), the observed wavelength exponent should be equal to 3 when the average length of the rod-like structures is greater than the wavelength of light (length > k). Largely because of the success of the Gaskin et al. study [2], the turbidity measurement has become the principal method for the study of not only microtubule formation but also many 0003-2697/$ - see front matter. Crown Copyright Ó 2005 Published by Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2005.07.011 * Corresponding author. Fax: +44 1223 763 418. E-mail address: drh32@cam.ac.uk (D. Hall). www.elsevier.com/locate/yabio Analytical Biochemistry 345 (2005) 198–213 ANALYTICAL BIOCHEMISTRY