CORDIS - EU research results

Fluorescent biosensors of organelle morphology to study the nuclear envelope dynamics during cell division

Final Report Summary - FLUOMORPH (Fluorescent biosensors of organelle morphology to study the nuclear envelope dynamics during cell division)

Eukaryotic cells are compartmentalized into membrane-bound organelles, ensuring highly specialized and essential functions such as protection of the genome or protein synthesis and packaging. Membranes surrounding organelles have unique characteristics that define organelle signature. These include lipid and protein composition and membrane shape. These membrane signatures serve as recruitment signals for organelle components and thus participate in their functions. In particular, it is now clear that membrane curvature can be sensed with high specificity by specialized protein domains. This can be used to coordinate events in space and time by triggering the recruitment of specific factors at a precise location and step of a membrane remodelling process. For instance, Nuclear Pore Complexes (NPCs) are giant protein complexes inserted in the nuclear envelope (NE) at fusion sites between the inner (INM) and outer nuclear membranes (ONM). The formation of new NPCs in the NE implies the remodelling of the INM and ONM, their local fusion, and the recruitment of more than 500 pore components at NPC assembly sites. We have previously shown that a membrane curvature sensor motif called ALPS (Amphipathic Lipid Packing Sensor) located in the NPC component Nup133 is required for the recruitment and assembly of NPC components. This suggests that curvature generated at INM and ONM fusion sites helps coordinate the sequence of protein recruitment involved in NPC assembly.
Importantly, most organelles are not delineated by a homogenous membrane, but rather harbour a complex morphology. In particular, some organelles are made of distinct subcompartments exhibiting specific shapes. Distinct functions are associated to these structural domains. For instance, the endoplasmic reticulum (ER) is an intricate network of cisternae and tubules and the Golgi apparatus is composed of stacked cisternae surrounded by trafficking vesicles.
Moreover, organelle architecture is highly dynamic and can undergo dramatic rearrangements in response to cellular changes like cell division or differentiation. To add another layer of complexity, membranes continuously exchange between organelles, raising the question of how their identity is established and maintained. In higher eukaryotes, mitotic NE breakdown is a spectacular example of membrane rearrangement and exchange: indeed, during mitosis, the NE is disassembled. During this process, the NE proteins reside in the ER, suggesting that the two compartments merge during mitosis. This major influx of membranes in the ER, together with the global change in cell shape, most likely impacts the balance between ER cisternae and tubules. This point is currently controversial as contradictory studies show that the ER is mostly tubular or mostly cisternal during mitosis. Being a starting point for NE post-mitotic assembly, ER mitotic morphology has implications in this process mechanisms. Accordingly, two models have been proposed: recruitment of tubules to the decondensing chromatin surface, followed by flattening, or direct recruitment of ER cisternae at the chromatin surface.
The methods available so far to render organelle morphology require highly specialized techniques and are most often incompatible with high-rate image acquisition. The aim of this proposal was to develop a method to distinguish organelle substructures in live cells using a simple bi-dimensional confocal imaging setup. Fluorescent probes of organelle morphology will be designed based on the striking membrane sensing properties of a family of amphipathic helices called ALPS motifs. To establish the usefulness of these tools, we proposed to study the dynamics of the nuclear envelope (NE) and endoplasmic reticulum (ER) during mitosis. Our objectives were three-fold:
1) Understand the molecular determinants, both in membrane and ALPS sequence, governing the specificity of ALPS motifs towards their target membranes. Most ALPS-containing proteins are targeted to curved regions of the early secretory pathway, but within distinct membrane compartments and regions of distinctive topologies. It is not clear how ALPS-containing proteins discriminate between the numerous curved regions present in cells. This is a pre-requisite to engineer fluorescent probes with exquisite specificity for organelle domains of interest. Beyond this technical goal is the biological question of the role of membrane curvature in defining membrane signatures and its role in accurate targeting of membrane components.
2) Design fluorescent probes with specificity towards ER cisternae, ER tubules and the INM. These probes directed against distinct organelle domains will be useful tools to gather topographical information about membrane compartments, in a 2D imaging set-up. This will lead to faster imaging rates and less photo-toxicity, two important caveats in 3D imaging.
3) Study the dynamics of the ER and the NE during mitosis and post-mitotic NE assembly.

To understand the molecular determinants of membrane specificity of ALPS-containing proteins, we have designed fluorescent proteins made of a fluorescent moiety and an ALPS motif. We took two ALPS motifs out of their natural protein environment and grafted them to non-related fluorescent proteins. While the constructs are not related sequence-wise to the original proteins, they mimic their overall architectures. In this context, the chimeras recapitulated the membrane curvature sensitivity and selectivity of the original ALPS-containing proteins, showing that the sequence of the surrounding protein is not involved in membrane selectivity. We then focused on Nup133 ALPS motif and made point mutations within the ALPS sequence to change its hydrophobic or charge content. These changes had an effect on membrane affinity, but not on specificity towards curvature or membrane compartment. However, switching backbones, we showed that the way ALPS motifs are presented to cellular membranes plays a key role in their localization. In particular, the presence of an artificial dimeric coiled-coil domain increases ALPS motifs sensitivity to membrane curvature. It can even potentiate a non-ALPS amphipathic helix into a membrane curvature sensor. During this part of the study, we also developed new analysis methods and molecular tools interesting for the scientific community interested in the ER morphology and function. These results were published in PLoS ONE.
Then, based on this study, we optimized our initial construct and designed a green fluorescent probe specific for ER tubules. We also made ER- and NE- specific constructs, available as red or green fluorescent proteins. We established stable cell lines expressing different combinations of these fluorescent probes.
Finally, we studied the behaviour of these cells during mitosis. The probes we designed enabled us to visualize differential behaviour for ER tubules and ER cisternae at the decondensing chromatin surface during pot-mitotic NE assembly. This suggests that the NE reforms from combined ER tubule reshaping and flat cisternae recruitment. We are now studying the behaviour of NE proteins relative to ER tubules or ER cisternae probes. We are also developing a quantitative analysis method. This will allow us to gather multiple sets of data and understand how membrane and protein recruitment is coordinated in space and time during NE formation.