Functional integration of a semi-synthetic azido-queuosine derivative into translation and a tRNA modification circuit.

Substitution of the queuine nucleobase precursor preQ1 by an azide-containing derivative (azido-propyl-preQ1) led to incorporation of this clickable chemical entity into tRNA via transglycosylation in vitro as well as in vivo in Escherichia coli, Schizosaccharomyces pombe and human cells. The resulting semi-synthetic RNA modification, here termed Q-L1, was present in tRNAs on actively translating ribosomes, indicating functional integration into aminoacylation and recruitment to the ribosome. The azide moiety of Q-L1 facilitates analytics via click conjugation of a fluorescent dye, or of biotin for affinity purification. Combining the latter with RNAseq showed that TGT maintained its native tRNA substrate specificity in S. pombe cells. The semi-synthetic tRNA modification Q-L1 was also functional in tRNA maturation, in effectively replacing the natural queuosine in its stimulation of further modification of tRNAAsp with 5-methylcytosine at position 38 by the tRNA methyltransferase Dnmt2 in S. pombe. This is the first demonstrated in vivo integration of a synthetic moiety into an RNA modification circuit, where one RNA modification stimulates another. In summary, the scarcity of queuosinylation sites in cellular RNA, makes our synthetic q/Q system a 'minimally invasive' system for placement of a non-natural, clickable nucleobase within the total cellular RNA.

Membrane destabilization and pore formation induced by the Synechocystis IM30 protein.

The inner membrane-associated protein of 30 kDa (IM30) is essential in chloroplasts and cyanobacteria. The spatio-temporal cellular localization of the protein appears to be highly dynamic and triggered by internal as well as external stimuli, mainly light intensity. The soluble fraction of the protein is localized in the cyanobacterial cytoplasm or the chloroplast stroma, respectively. Additionally, the protein attaches to the thylakoid membrane as well as to the chloroplast inner envelope or the cyanobacterial cytoplasmic membrane, respectively, especially under conditions of membrane stress. IM30 is involved in thylakoid membrane biogenesis and/or maintenance, where it either stabilizes membranes and/or triggers membrane-fusion processes. These apparently contradicting functions have to be tightly controlled and separated spatiotemporally in chloroplasts and cyanobacteria. IM30's fusogenic activity depends on Mg2+ binding to IM30; yet, it still is unclear how Mg2+-loaded IM30 interacts with membranes and promotes membrane fusion. Here, we show that the interaction of Mg2+ with IM30 results in increased binding of IM30 to native, as well as model, membranes. Via atomic force microscopy in liquid, IM30-induced bilayer defects were observed in solid-supported bilayers in the presence of Mg2+. These structures differ dramatically from the membrane-stabilizing carpet structures that were previously observed in the absence of Mg2+. Thus, Mg2+-induced alterations of the IM30 structure switch the IM30 activity from a membrane-stabilizing to a membrane-destabilizing function, a crucial step in membrane fusion.

Proton Leakage Is Sensed by IM30 and Activates IM30-Triggered Membrane Fusion.

The inner membrane-associated protein of 30 kDa (IM30) is crucial for the development and maintenance of the thylakoid membrane system in chloroplasts and cyanobacteria. While its exact physiological function still is under debate, it has recently been suggested that IM30 has (at least) a dual function, and the protein is involved in stabilization of the thylakoid membrane as well as in Mg2+-dependent membrane fusion. IM30 binds to negatively charged membrane lipids, preferentially at stressed membrane regions where protons potentially leak out from the thylakoid lumen into the chloroplast stroma or the cyanobacterial cytoplasm, respectively. Here we show in vitro that IM30 membrane binding, as well as membrane fusion, is strongly increased in acidic environments. This enhanced activity involves a rearrangement of the protein structure. We suggest that this acid-induced transition is part of a mechanism that allows IM30 to (i) sense sites of proton leakage at the thylakoid membrane, to (ii) preferentially bind there, and to (iii) seal leaky membrane regions via membrane fusion processes.

Cyanobacterial membrane dynamics in the light of eukaryotic principles.

Intracellular compartmentalization is a hallmark of eukaryotic cells. Dynamic membrane remodeling, involving membrane fission/fusion events, clearly is crucial for cell viability and function, as well as membrane stabilization and/or repair, e.g., during or after injury. In recent decades, several proteins involved in membrane stabilization and/or dynamic membrane remodeling have been identified and described in eukaryotes. Yet, while typically not having a cellular organization as complex as eukaryotes, also bacteria can contain extra internal membrane systems besides the cytoplasmic membranes (CMs). Thus, also in bacteria mechanisms must have evolved to stabilize membranes and/or trigger dynamic membrane remodeling processes. In fact, in recent years proteins, which were initially defined being eukaryotic inventions, have been recognized also in bacteria, and likely these proteins shape membranes also in these organisms. One example of a complex prokaryotic inner membrane system is the thylakoid membrane (TM) of cyanobacteria, which contains the complexes of the photosynthesis light reaction. Cyanobacteria are evolutionary closely related to chloroplasts, and extensive remodeling of the internal membrane systems has been observed in chloroplasts and cyanobacteria during membrane biogenesis and/or at changing light conditions. We here discuss common principles guiding eukaryotic and prokaryotic membrane dynamics and the proteins involved, with a special focus on the dynamics of the cyanobacterial TMs and CMs.

Human Claudin-7 cis-Interactions Are Not Crucial for Membrane-Membrane (Trans-) Interactions.

Human Claudin-7 (Cldn7) is a member of the Claudin (Cldn) superfamily. In vivo, these proteins form tight junctions, which establish constricted connections between cells. Cldns oligomerize within the membrane plane (= cis-interaction), and also interact with Cldns from adjacent cells (= trans-interaction). Interactions of Cldns are typically studied in vivo and structural analyses of isolated Cldns are limited. Here, we describe heterologous expression in E. coli and purification of human Cldn7, enabling in vitro analyses of the isolated protein using detergent and model membrane systems. Cldn7 exists as a monomer, hexamer, and various higher oligomers in micelles. While only limited unfolding of the protein was observed in the presence of the anionic detergent sodium dodecyl sulfate, decreased ionic strength did affect Cldn7 cis-interactions. Furthermore, we identified two amino acids which mediate electrostatic cis-interactions and analyzed the impact of disturbed cis-interaction on trans-contacts via atomic force microscopy and monitoring Förster resonance energy transfer between fluorescently labeled Cldn7-containing proteoliposomes. Our results indicate that Cldn7 cis-oligomerization might not be a prerequisite for establishing trans-contacts.

Binding and/or hydrolysis of purine-based nucleotides is not required for IM30 ring formation.

IM30, the inner membrane-associated protein of 30 kDa, is conserved in cyanobacteria and chloroplasts. Although its exact physiological function is still mysterious, IM30 is clearly essential for thylakoid membrane biogenesis and/or dynamics. Recently, a cryptic IM30 GTPase activity has been reported, albeit thus far no physiological function has been attributed to this. Yet, it is still possible that GTP binding/hydrolysis affects formation of the prototypical large homo-oligomeric IM30 ring and rod structures. Here, we show that the Synechocystis sp. PCC 6803 IM30 protein in fact is an NTPase that hydrolyzes GTP and ATP, but not CTP or UTP, with about identical rates. While IM30 forms large oligomeric ring complexes, nucleotide binding and/or hydrolysis are clearly not required for ring formation.

GTP hydrolysis by Synechocystis IM30 does not decisively affect its membrane remodeling activity.

The function of IM30 (also known as Vipp1) is linked to protection and/or remodeling of the thylakoid membrane system in chloroplasts and cyanobacteria. Recently, it has been revealed that the Arabidopsis IM30 protein exhibits GTP hydrolyzing activity in vitro, which was unexpected, as IM30 does not show any classical GTPase features. In the present study, we addressed the question, whether an apparent GTPase activity is conserved in IM30 proteins and can also be observed for IM30 of the cyanobacterium Synechocystis sp. PCC 6803. We show that Synechocystis IM30 is indeed able to bind and hydrolyze GTP followed by the release of Pi. Yet, the apparent GTPase activity of Synechocystis IM30 does not depend on Mg2+, which, together with the lack of classical GTPase features, renders IM30 an atypical GTPase. To elucidate the impact of this cryptic GTPase activity on the membrane remodeling activity of IM30, we tested whether GTP hydrolysis influences IM30 membrane binding and/or IM30-mediated membrane fusion. We show that membrane remodeling by Synechocystis IM30 is slightly affected by nucleotides. Yet, despite IM30 clearly catalyzing GTP hydrolysis, this does not seem to be vital for its membrane remodeling function.

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

Members of the phage shock protein A (PspA) family, including the inner membrane-associated protein of 30 kDa (IM30), are suggested to stabilize stressed cellular membranes. Furthermore, IM30 is essential in thylakoid membrane-containing chloroplasts and cyanobacteria, where it is involved in membrane biogenesis and/or remodeling. While it is well known that PspA and IM30 bind to membranes, the mechanism of membrane stabilization is still enigmatic. Here we report that ring-shaped IM30 super-complexes disassemble on membranes, resulting in formation of a membrane-protecting protein carpet. Upon ring dissociation, the C-terminal domain of IM30 unfolds, and the protomers self-assemble on membranes. IM30 assemblies at membranes have been observed before in vivo and were associated with stress response in cyanobacteria and chloroplasts. These assemblies likely correspond to the here identified carpet structures. Our study defines the thus far enigmatic structural basis for the physiological function of IM30 and related proteins, including PspA, and highlights a hitherto unrecognized concept of membrane stabilization by intrinsically disordered proteins.

Functional Implications of Multiple IM30 Oligomeric States.

The inner membrane-associated protein of 30 kDa (IM30), also known as the vesicle-inducing protein in plastids 1 (Vipp1), is essential for photo-autotrophic growth of cyanobacteria, algae and higher plants. While its exact function still remains largely elusive, it is commonly accepted that IM30 is crucially involved in thylakoid membrane biogenesis, stabilization and/or maintenance. A characteristic feature of IM30 is its intrinsic propensity to form large homo-oligomeric protein complexes. 15 years ago, it has been reported that these supercomplexes have a ring-shaped structure. However, the in vivo significance of these ring structures is not finally resolved yet and the formation of more complex assemblies has been reported. We here present and discuss research on IM30 conducted within the past 25 years with a special emphasis on the question of why we potentially need IM30 supercomplexes in vivo.