Publication Date

April 2016

Advisor(s)

Donald Oliver

Major

Molecular Biology&Biochemistry

Language

English (United States)

Abstract

It has previously been demonstrated that the DEAD motor protein SecA ATPase interacts dynamically with the heterotrimeric SecYEG channel complex to promote protein translocation in E. coli. In particular both peripherally and integrally-bound membrane states of SecA have been observed previously, and the latter state has been suggested to be critical for preprotein insertion into the SecYEG channel and/or channel opening and activation. However, the precise mechanistic linkage between integral membrane SecA and the different SecYEG channel states remains largely unknown. In order to address this question and elucidate potential structural changes in integral membrane SecA that are requisite for protein transport and channel dynamics, we performed a proteolytic probing study on isolated membranes from cells grown under three different sets of physiological conditions: (1) untreated growing cells containing wild-type translocation activity; (2) cells whose SecYEG channel complex was jammed with an OmpA-GFP translocation intermediate that results in a complete protein transport blockage; and (3) cells whose SecYEG channel complex was freed of any protein cargo by treatment with the protein translation-initiation inhibitor Kasugamycin, thereby capturing the resting state of the translocon. Peripheral SecA was separated from integral membrane SecA via heparin treatment, and both fractions were subsequently proteolyzed and analyzed by Western blotting. In addition, post proteolysis, the integral membrane fraction was also treated with urea followed by membrane sedimentation, in order to distinguish strongly-bound SecA proteolytic fragments within the channel or lipid bilayer from weakly-bound SecA proteolytic fragments within the channel interior or associated with lipid head groups. We observed increased levels of 45- and 30-kD fragments upon trypsinolysis of peripheral and integral SecA under jamming conditions, compared to the unjammed state. Both of these fragments were extracted from membranes by 6M urea, suggesting that they are not lipid anchored, although association with the hydrophilic channel interior remains a possibility. Channel clearing appears to enrich peripheral SecA in a conformation which favors high levels of the 45-kD fragment and much less of the 30-kD fragment, perhaps due to the lack of preprotein substrate for SecA interaction in this case. In addition, we wished to further characterize the unresolved question of the depth of integral SecA membrane insertion. To this end, we performed preliminary troubleshooting experiments for a technique by which periplasmically exposed regions of integral SecA can be labeled by a membrane-impermeable NHS-biotin compound. Future studies will be directed at optimizing this biotinylation technique and/or identifying whether one of the aforementioned SecA proteolytic fragments is inserted into the channel as well as the identification of the corresponding region of SecA utilizing region-specific SecA antibodies. In total, our study should contribute towards the creation of a SecA membrane topology map during active or resting translocation states and the relation of this information to the credibility of the currently proposed SecA mechanistic models.

Share

COinS
 

© Copyright is owned by author of this document