Self-Organized Amphiphiles Are Poor Hydroxyl Radical Scavengers in Fast Photochemical Oxidation of Proteins Experiments

Abstract

Analysis of membrane protein topography using fast photochemical oxidation of proteins (FPOP) has been reported in recent years but is still underrepresented in literature. Based on the hydroxyl radical reactivity of lipids and other amphiphiles, it is believed that the membrane environment acts as a hydroxyl radical scavenger decreasing effective hydroxyl radical doses and resulting in less observed oxidation of proteins. We found no significant change in bulk solvent radical scavenging activity upon the addition of disrupted cellular membranes up to 25600 cells/μL using an inline radical dosimeter. We confirmed the nonscavenging nature of the membrane in bulk solution with the FPOP results of a soluble model protein in the presence of cell membranes, which showed no significant difference in oxidation with or without membranes. The use of detergents revealed that, while soluble detergent below the critical micelle concentration is a potent hydroxyl radical scavenger, additional detergent has little to no hydroxyl radical scavenging effect once the critical micelle concentration is reached. Examination of both an extracellular peptide of the integral membrane protein bacteriorhodopsin as well as a novel hydroxyl radical dosimeter tethered to a Triton X-series amphiphile indicate that proximity to the membrane surface greatly decreases reaction with hydroxyl radicals, even though the oxidation target is equally solvent accessible. These results suggest that the observed reduced oxidation of solvent-accessible surfaces of integral membrane proteins is due to the high local concentration of radical scavengers in the membrane or membrane mimetics competing for the local concentration of hydroxyl radicals.

Keywords: FPOP; covalent labeling; hydroxyl radical protein footprinting; membrane proteins; structural biology.

Figure 1.

Cell membrane radical scavenging activity as measured by adenine dosimetry using an inline dosimeter. Different concentration of isolated cellular membrane showed no significant changes in hydroxyl radical scavenging.

Figure 2.

FPOP of myoglobin in the absence (blue) and the presence of 6400 cells/μL membrane (orange) from MS analysis. No significant difference was shown in peptide oxidation between these two conditions (p ≤ 0.05)

Figure 3.

Detergent micelles are poor hydroxyl radical scavengers. The black line represents the fluorescence intensity of ANS dye, indicating the CMC of the detergent. The red line represents hydroxyl radical scavenging activity by measuring ΔAbs_265nm of adenine with inline dosimeter. (A) Triton X-405 scavenges when detergent is free in solution, but once the CMC is reached additional detergent does not add to the scavenging capacity of the solution. (B) Triton X-100 shows a similar pattern, with the concentration region where additional detergent ceases to compete with adenine for the hydroxyl radical shifting to lower concentrations due to the lower CMC of Triton X-100.

Figure 4.

MS result for oxidation level of extracellular BR peptide 72–80 from (blue, inset) integral membrane BR protein and (orange) soluble digested BR peptide FPOP at the same concentration. The proximity of the solvent accessible peptide to the membrane greatly reduces the oxidation of the peptide compared to the peptide diffusing free in solution.

Figure 5.

Micelle-tethered hydroxyl radical dosimeter. (A) Structure of TX405A conjugate. Triton X-405 amphiphile is amide conjugated to the adenine hydroxyl radical dosimeter through a ribose linker. (B) Static model cross-section of a micelle of a 9:1 mixture of Triton X-100 and TX405A. The hydroxyl radical dosimeter is shown in red.

Figure 6.

ΔAbs₂₆₅ results from inline dosimeter with X-405 conjugate in comparison with free X-405 during FPOP experiment. (A) ANS fluorescence assay of a 9:1 mixed micelle of Triton X-100 and Triton X-405. The CMC is identified around 0.2 mM total detergent concentration. (B) Measurement of adenosine hydroxyl radical dosimetry for (left) 2 mM soluble adenosine in the presence of 20 mM 9:1 Triton X-100:Triton X-405 and (right) a mixed micelle of 2 mM TX405A and 18 mM Triton X-100. The result showed a significant difference between the two groups (p ≤ 0.05).

Figure 7.

Conceptual model of the observed reduction in oxidation of membrane-associated analytes. Any hydroxyl radical formed within diffusion distance of the membrane-associated analyte will encounter a very high local concentration of self-associated amphiphile. Even though the rate constant of oxidation of the self-organized amphiphile is much lower than for the free soluble amphiphile, the local concentration is so high that a large number of the local hydroxyl radical is consumed by the membrane or membrane-mimetic.