Proteins mixed-mode interactions

In reality, many proteins demonstrate mixed mode interactions (e.g., additional hydrophobic or silanol interactions) with a column, or multiple structural conformations that differentially interact with the sorbent. These nonideal interactions may distribute a component over multiple gradient steps, or over a wide elution range with a linear gradient. These behaviors may be mitigated by the addition of mobile phase modifiers (e.g., organic solvent, surfactants, and denaturants), and optimization (temperature, salt, pH, sample load) of separation conditions. 

Fig. 2. Schematic representation of the retention dependencies for peptides or proteins chromatographed on mixed-mode support media. The figure illustrates four case histories for the dependency of the logarithmic capacity factor (log it ) on the mole fraction, f, of the displacing species. As the contact area associated with the solute-Ugand interaction increases, the slopes of the log k versus f plots increase resulting in a narrowing of the elution window over which the solute will desorb. Cases (a) and (b) are typically observed for the RP-HPLC of polar peptides and small, polar globular proteins whilst cases (c) and (d) are more representative of the RP-HPLC behaviour of highly hydrophobic polypeptides and non-polar globular proteins, respectively.

Schafer et al. used several spectroscopic techniques to characterize the surface species on phosphate-modified zirconia particles. Their results show that phosphate merely adsorbs on the surface of zirconia under the mildest phosphate concentration, i.e., neutral pH, room temperature, and short contact times. However, at acidic pH and higher temperarnres, esterification of the phosphate with surface hydroxyls takes place as the kinetic barriers are overcome. The solid NMR studies clearly show the presence of covalently bound phosphate. This phosphate modification effectively blocks the sites responsible for the strong interaction of certain Lewis bases with the zirconia surface, resulting in a more biocompatible stationary phase. Unlike fluoride-modified zirconia, phosphate-modified zirconia behaves as a classic cation exchanger and not as a mixed-mode medium analogous to hydroxyapatite, despite spectroscopic evidence of zirconium phosphate formation on the surface. This limits the applicability of the supports, as most proteins and enzymes are anionic at neutral pH. Nevertheless, its ability to separate proteins with high p/ values still deserves much attention. The preparative-scale separation of murine IgGs from a fermentation broth demonstrates the utiUty of the supports for solutes that are retained. 

The second mode of toxicity is postulated to involve the direct interaction of the epidithiodiketopiperazine motif with target proteins, forming mixed disulfides with cysteine residues in various proteins. Gliotoxin, for example, has been demonstrated to form a 1 1 covalent complex with alcohol dehydrogenase [13b, 17]. Epidithiodi-ketopiperazines can also catalyze the formation of disulfide bonds between proxi-mally located cysteine residues in proteins such as in creatine kinase [18]. Recently, epidithiodiketopiperazines have also been implicated in a zinc ejection mechanism, whereby the epidisulfide can shuffle disulfide bonds in the CHI domain of proteins, coordinate to the zinc atoms that are essential to the tertiary structure of that domain, and remove the metal cation [12d, 19],... 

Inactivation of phage T2 in p-lactamase solution by QAC-treated beads is shown in Figure 4. The initial T2 titer was 3.0 x 10 PFU/ml. A quantity of 0.8 g of beads were mixed with 10 ml of p-lactamase solution. Fifteen percent of the viruses survived this treatment. The amount of total protein in the solution was 80% of the initial value after the adsorption process, while the recovery of p-lactamase activity was at least 70%. It was the purpose of this experiment to demonstrate that QAC-treated beads can effectively remove viruses from a protein solution without significantly losing the activity of the protein. Optimal adsorption condition and mode of operation ought to be determined by studying the interactive effects of pH, ionic strength, and temperature of the solution, with the specific t)qjes of virus and protein of interest. 

There are two principal problems with penetration experiments the adsorption characteristics of the protein have to be understood, and the amount of protein that adsorbs to the interface when lipid is present has to be determined. Previously, most researchers used the change in film pressure (Atr) as a measure of the amount of protein that interacted with the lipid monolayer. However, this approach implicitly assumes that the adsorption of protein can be described by Gibbs adsorption equation, but as pointed out by Colacicco (6), this is invalid for proteins which adsorb irreversibly. Because the surface concentration of protein is unknown, radiolabeled proteins have been used (8, 9, 10). This work has been concerned exclusively with highly water-soluble proteins whose prime mode of interaction with monolayers (and bilayers) is electrostatic. In these cases a simple description of the packing in the mixed lipid-protein films was impossible (6). 

---