Vapor pressure osmometry studies of osmolyte-protein interactions: Implications for the action of osmoprotectants in vivo and for the interpretation of "osmotic stress" experiments in vitro

TO interpret or to predict the responses of biopolymer processes in vivo an d in vitro to changes in solute concentration and to coupled changes in wat er activity (osmotic stress), a quantitative understanding of the thermodyn amic consequences of interactions of solutes and water with biopolymer surf aces is required. To this end, we report isoosmolal preferential interactio n coefficients (Gamma(mu 1)) determined by vapor pressure osmometry (VPO) o ver a wide range of concentrations for interactions between native bovine s erum albumin (BSA) and six small solutes. These include Escherichia coli cy toplasmic osmolytes [potassium glutamate (K(+)Glu(-)), trehalose], E. coli osmoprotectants (proline, glycine betaine), and also glycerol and trimethyl amine N-oxide (TMAO). For all six solutes, Gamma(mu 1) and the correspondin g dialysis preferential interaction coefficient Gamma(mu 1,mu 3) (both calc ulated from the VPO data) are negative; Gamma(mu 1,mu 3) is proportional to bulk solute molality (m(3)(bulk)) at least up to 1 oz (molal). Negative va lues of Gamma(mu 1,mu 3) indicate preferential exclusion of these solutes f rom a BSA solution at dialysis equilibrium and correspond to local concentr ations of these solutes in the vicinity of BSA which are lower than their b ulk concentrations. Of the solutes investigated, betaine is the most exclud ed (Gamma(mu 1,mu 3)/m(3)(bulk) = -49 +/- 1 m(-1)); glycerol is the least e xcluded (Gamma(mu 1,mu 3)/m(3)(bulk) = -10 +/- 1 m(-1)). Between these extr emes, the magnitude of Gamma(mu 1,mu 3)/m(3)(bulk) 3 3 decreases in the ord er glycine betaine >> proline >TMAO > trehalose approximate to K+Glu- > gly cerol. The order of exclusion of E, coli osmolytes from BSA surface correla tes with their effectiveness as osmoprotectants, which increase the growth rate of E. coli at high external osmolality. For the most excluded solute ( betaine), Gamma(mu 1,mu 3) provides a minimum estimate of the hydration of native BSA of approximately 2.8 x 10(3) H2O/ BSA, which corresponds to slig htly less than a monolayer (estimated to be similar to 3.2 x 10(3) H2O). Co nsequently, of the solutes investigated here, only betaine might be suitabl e for use in osmotic stress experiments in vitro as a direct probe to quant ify changes in hydration of protein surface in biopolymer processes. More g enerally, however, our results and analysis lead to the proposal that any o f these solutes can be used to quantify changes in water-accessible surface area (ASA) in biopolymer processes once preferential interactions of the s olute with biopolymer surface are properly taken into account.