The Fusion Activity of IM30 Rings Involves Controlled Unmasking of the Fusogenic Core

Abstract

The inner membrane-associated protein of 30 kDa (IM30, also known as Vipp1) is required for thylakoid membrane biogenesis and maintenance in cyanobacteria and chloroplasts. The protein forms large rings of ∼2 MDa and triggers membrane fusion in presence of Mg2+. Based on the here presented observations, IM30 rings are built from dimers of dimers, and formation of these tetrameric building blocks is driven by interactions of the central coiled-coil, formed by helices 2 and 3, and stabilized via additional interactions mainly involving helix 1. Furthermore, helix 1 as well as C-terminal regions of IM30 together negatively regulate ring-ring contacts. We propose that IM30 rings represent the inactive form of IM30, and upon binding to negatively charged membrane surfaces, the here identified fusogenic core of IM30 rings eventually interacts with the lipid bilayer, resulting in membrane destabilization and membrane fusion. Unmasking of the IM30 fusogenic core likely is controlled by Mg2+, which triggers rearrangement of the IM30 ring structure.

IM30 IDPs form a membrane protective carpet upon super-complex disassembly

“…The core structure of IM30 proteins is the coiled-coil formed by the helices and 3 15,19,30,31 . This structure is stable even in the isolated, monomeric helix 2-3a fragment 30 ; thus, no additional interactions with other parts of the protein are required for secondary and tertiary structure formation of this fragment. In contrast, the isolated C-terminal region comprising helix 3b-7 is largely unstructured, albeit capable of forming stable dimers.…”

“…In fact, we recently observed that the isolated helix 2/3 coiled-coil does have an intrinsic propensity to dimerize 30 but does not form higher-ordered super-complexes. Thus, interactions involving both, the helix 2-3 coiled-coil as well as (at least) the region containing helices 4-6, are required for ring formation 30 .…”

“…Thus, IM30* dimerizes via residues located in the C-terminal domains, in line with our dimer models ( Supplementary Fig. 12), whereas the N-terminal region could mediate membrane interaction.In fact, stable membrane interaction of the isolated helix 2/3 fragment has been demonstrated recently 30 . Based on our dimer models ( Supplementary Fig.…”

“…IM30* (E83A, E84A, F168A, E169A, R170A, M171A), IM30_A227C (A272C) and IM30*_A227C (E83A, E84A, F168A, E169A, R170A, M171A, A227C) were generated via site-directed mutagenesis. The WTfragments IM30_H2-3a (amino acids 22-145) and IM30_H3b-7 (amino acids 147-267) were generated by PCR cloning 30 . IM30-CFP and IM30-Venus were generated by restriction digestion and T4 ligation of the CFP/Venus fragments into pRSET SynIM30 plasmids 43 .…”

Mg2+-binding shifts the IM30 activity from membrane protection to membrane destabilization

“…However, a fraction of the protein is attached to the chloroplasts' inner envelope or the cyanobacterial cytoplasmic membrane, as well as to thylakoid membranes (TMs) (4)(5)(6)(7). The activity of IM30 is broadly linked to the biogenesis/maintenance of TMs (recently reviewed in (8)), and two specific activities are currently studied and discussed in greater detail: (i) membrane protection (9)(10)(11)(12)(13)(14)(15) and (ii) membrane remodeling and repair (10,15,16). Protection and maintenance of the TM system are necessary to prevent energy dissipation and to maintain a proton gradient across the TM, a membrane system highly vulnerable to membrane damage (17).…”