Aqueous Two-Phase System Formation Kinetics for Elastin-Like Polypeptides of Varying Chain Length Yanjie Zhang, Kimberly Trabbic-Carlson, Fernando Albertorio, Ashutosh Chilkoti, and Paul S. Cremer* Department of Chemistry, Texas A&M University, 3255 TAMU, College Station, Texas 77843, and Department of Biomedical Engineering, Duke University, Durham, North Carolina 27708 Received March 18, 2006; Revised Manuscript Received May 1, 2006 The kinetics of aqueous two-phase system (ATPS) formation for elastin-like polypeptides (ELP) with defined chemical composition and chain length was investigated by dark field microscopy in an on-chip format with a linear temperature gradient. Scattering intensities from peptide solutions in the presence and absence of sodium dodecyl sulfate (SDS) were recorded as a function of temperature and time, simultaneously. It was found that the formation of the ATPS for three ELPs of different molecular weights (36 075, 59 422, and 129 856 Da) in the absence of SDS followed a coalescence mechanism, and the rate constant and activation energy were independent of chain length. With the introduction of SDS into the ELP solutions, the rate constants were attenuated more strongly with increasing chain length. Moreover, the coalescence process in the presence of SDS showed non- Arrhenius kinetics as a function of temperature. For the two shorter ELPs, ATPS formation occurred via coalescence at all SDS concentrations and temperatures investigated. On the other hand, the coalescence process was greatly suppressed for the longest ELP at elevated temperatures and higher SDS concentrations. Under these circumstances, ATPS formation was forced to proceed via a mixed Ostwald ripening and coalescence mechanism. Introduction The formation of an aqueous two-phase system 1-3 com- mences when a macromolecular system spontaneously separates into two separate water-containing phases. Such phenomena often involve two polymers separating into distinct phases with different densities. ATPS formation may also occur in solutions consisting of a single polymer plus a high concentration of an appropriate salt 4-7 or even in the absence of salt for certain thermoresponsive polymers. 8-10 In this final case, ATPS forma- tion occurs above the polymer’s lower critical solution temper- ature (LCST). For these systems, a denser polymer-rich phase is formed on the bottom of the container, while the upper phase consists primarily of pure water. Since the phase boundary in an ATPS has very low interfacial tension, 11 these systems have been widely applied as tools for bioseparations without signifi- cant interfacial denaturing effects. 12-14 Despite the practical and fundamental significance of ATPS formation, there is almost no literature dealing with its formation kinetics because of the difficulties involved in data collection. To remedy this problem, we recently developed a temperature- gradient microfluidic technique 15-17 that allows ATPS formation to be followed in a high-throughput, low sample volume fashion. 18 This enabled us to explore the activation energy of the process for R-elastin, a thermoresponsive protein, which was prepared by the methods of Partridge et al. 19 It was found that there are two major mechanistic pathways through which ATPS formation could proceed: coalescence and Ostwald ripening. Coalescence is the direct fusion of one particle with another to make a larger one, while Ostwald ripening involves the growth of particles through the transfer of individual molecules between them such that bigger particles grow at the expense of smaller ones. 20 Each process has a different activation barrier associated with it. In the case of coalescence, the barrier comes from the removal of intervening solvent molecules between the particles which are being joined. On the other hand, Ostwald ripening involves the desorption of single macromolecules from shrinking particles and their deposition onto growing particles. The rate constants of ATPS formation for elastin are consistent with a coalescence mechanism, although the process can be forced to proceed through Ostwald ripening by the addition of sodium dodecyl sulfate to the solution. We suspected that the mechanism of ATPS formation might be dependent on the molecular weight of the polymer; however, the natural product, R-elastin, is a mixture of cross-linked materials that is highly polydisperse. Therefore, to obtain more control over the molecular weight, we employed elastin-like polypeptides. ELPs are based on a repetitive pentapeptide motif, Val-Pro-Gly-Xaa-Gly, where the guest residue, Xaa, is any amino acid except Pro. The sequence and chain length can be precisely controlled by genetic expression in bacteria using recombinant DNA techniques. 21,22 The work described herein was undertaken with three different molecular weight species (36 075, 59 422, and 129 856 Da) to investigate the effect of chain length on the kinetics of ATPS formation. It was found that coalescence in the absence of SDS is not dependent on the chain length of the ELPs; however, SDS increases the activation barrier for the coalescence process of longer-chain ELPs to a greater extent than for shorter ones. In fact, the ATPS formation process could be forced to go by Ostwald ripening at higher temperatures and SDS concentrations in the case of the 129 856 Da ELP, but not for the two smaller molecules. Experimental Section The ELPs employed herein come from a library of ELP[V5A2G3-n] where the repeat sequence of amino acids is the same. Members within the library differ only in their chain length, where n is the number of pentapeptides. For example, ELP[V5A2G3-90] consists of 90 pentapep- tides and has a repeat unit composed of 10 pentapeptides with the guest 2192 Biomacromolecules 2006, 7, 2192-2199 10.1021/bm060254y CCC: $33.50 © 2006 American Chemical Society Published on Web 06/20/2006