(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



The conformational plasticity of structurally unrelated lipid transport proteins correlates with their mode of action [1]

['Sriraksha Srinivasan', 'Department Of Biology', 'University Of Fribourg', 'Fribourg', 'Andrea Di Luca', 'Daniel Álvarez', 'Departamento De Química Física Y Analítica', 'Universidad De Oviedo', 'Oviedo', 'Arun T. John Peter']

Date: 2024-08

Lipid transfer proteins (LTPs) are key players in cellular homeostasis and regulation, as they coordinate the exchange of lipids between different cellular organelles. Despite their importance, our mechanistic understanding of how LTPs function at the molecular level is still in its infancy, mostly due to the large number of existing LTPs and to the low degree of conservation at the sequence and structural level. In this work, we use molecular simulations to characterize a representative dataset of lipid transport domains (LTDs) of 12 LTPs that belong to 8 distinct families. We find that despite no sequence homology nor structural conservation, the conformational landscape of LTDs displays common features, characterized by the presence of at least 2 main conformations whose populations are modulated by the presence of the bound lipid. These conformational properties correlate with their mechanistic mode of action, allowing for the interpretation and design of experimental strategies to further dissect their mechanism. Our findings indicate the existence of a conserved, fold-independent mechanism of lipid transfer across LTPs of various families and offer a general framework for understanding their functional mechanism.

Funding: S.V. acknowledges support by the SNSF (PP00P3_194807) and by the European Research Council under the European Union’s Horizon 2020 research and innovation program (grant agreement no. 803952). G.D’A. acknowledges support by the Swiss Cancer League, KFS-4999-02-2020; by the EPFL institutional fund; and by SNSF (310030_184926). This work was supported by grants from the Swiss National Supercomputing Centre under projects ID s1030 and s1132. S.V. and A.J.P. acknowledge support from the Novartis Forschungsstiftung via a FreeNovation grant. D.A. acknowledges support from the Margarita Salas program 2021–2023 funded by Ministerio de Universidades (MU-21-UP2021-030-53773022). M.A.L. acknowledges support from the Foundation Suisse de Recherche sur le Maladies Musculaires (FSRMM). TH acknowledges support from Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung (310030_215134) and from European Joint Program on Rare Diseases (32ER30_187505). MAL acknowledges support from EMPIRIS foundation, Zürich. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data Availability: Input files for atomistic and coarse-grained MD simulations, structures of the LTD with the membrane binding interface, and representative apo-like and holo-like conformations of each LTD arising from clustering of the atomistic simulations, as well as the data presented in the Figures, protocols and scripts can be found at: https://doi.org/10.5281/zenodo.12728271 .

Copyright: © 2024 Srinivasan et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Here, we employ a computational-based alternative approach to characterize the conformational plasticity of multiple LTDs belonging to different families. We find that despite significant differences in 3D structure, small LTDs (<50 to 100 kDa) can adopt distinct conformations based on the presence or absence of a bound lipid. The ability to transition between these conformations has potential implications for their functional mechanism, including membrane binding, and suggests a conserved mechanism of lipid transfer across LTPs of diverse families.

While these studies were limited to individual LTPs, when viewed altogether, they suggest features that could be shared among many of them. Specifically, a potential role for LTP conformational plasticity, i.e., their ability to adopt multiple conformations, has been proposed for various LTPs belonging to distinct protein families, including Osh/ORP, Ups/PRELI, PITP, and START family members [ 11 , 13 , 16 – 29 ]. These works have proposed that conformational changes between apo-like and holo-like structures might regulate lipid uptake and release by individual LTPs, but whether this is a general mechanism that extends to most LTPs remains unexplored.

A direct consequence of this case-by-case modus operandi is that a plethora of concurring models have been put forward to explain the mechanism and specificity of lipid transport. Yet, these models are of limited transferability across different LTPs since they largely rely on specific observations either on protein structure (such as the presence of a lid [ 6 , 10 ], of electrostatic surface patches [ 11 , 12 ], or on the specificity of the lipid-binding cavity [ 13 ]) or on experimentally determined transport properties (such as counter-exchange between different lipid species and lipid-dependent transport rates) [ 14 , 15 ].

This picture is further complicated by the sheer number of existing LTPs. So far, hundreds of different LTPs, belonging to several distinct protein families [ 6 , 7 ], have been identified. This diversity possibly originates from the huge chemical variability of the lipid substrates and organellar membranes they bind to. While this has likely allowed fine-tuning the mechanism and specificity of lipid transport by LTPs, it has so far prevented the establishment of a common framework to understand and interpret the molecular steps underlying this process. To this extent, only a few studies have attempted to investigate in a high-throughput fashion the functional properties of LTPs, such as their membrane or lipid binding [ 8 , 9 ]. Rather, the investigation of individual LTPs using cellular biology or reconstitution approaches remains to date the most frequent strategy.

Despite growing interest in the nonvesicular lipid transport pathway, our mechanistic understanding of how LTPs perform their function is still largely incomplete. The only unifying feature of LTPs is a polar exterior and the presence of a lipid transfer domain (LTD) containing a hydrophobic cavity that encloses the lipid. This architecture, by shielding the hydrophobic lipid molecule from the aqueous environment of the cytoplasm, reduces the energetic cost of transferring a lipid between 2 membranes. Two main models of lipid transport by LTPs have been put forward: the shuttle model and the tunnel model. In the shuttle model, a small (<50 to 100 kDa) LTD travels between the donor and the acceptor organelle, cyclically taking up and releasing their substrate lipid [ 5 ]. In the tunnel model, a large (>100 kDa) LTD physically connects the 2 organelles, establishing a continuous hydrophobic pathway in which lipids can simply diffuse between the 2 membranes [ 5 ]. In both cases, however, a fine regulation of multiple mechanistic steps must be accurately tuned to achieve lipid transport with the correct directionality and rate, and how such complex coordination is achieved remains largely unclear [ 5 ].

Since lipid synthesis is not ubiquitous, but rather mostly localized to the endoplasmic reticulum (ER) [ 3 ], lipids must be rapidly transported between organelles to maintain lipid homeostasis and organellar identity. This is achieved via 2 main routes, the vesicular and nonvesicular pathways. In the vesicular pathway, cargo vesicles, originating from lipid remodeling processes mediated by coat proteins, travel from a donor organelle to an acceptor one, where the vesicle undergoes fusion [ 4 ]. This pathway is not only crucial for cellular exocytosis and endocytosis but also intracellularly along the secretory pathway [ 4 ]. Alternatively, in the nonvesicular pathway, trafficking of lipids between organelles is performed by lipid transfer proteins (LTPs), which solubilize lipids and facilitate their transport between 2 membranes. Nonvesicular lipid transport promotes a more rapid modulation of the lipid composition of organelles compared to vesicular trafficking and is crucial during stress conditions when vesicular trafficking is compromised [ 5 ].

Lipids are one of the key building blocks of eukaryotic cells, as they allow for the spatial and temporal organization of chemical reactions in different cellular compartments called organelles. Eukaryotic cells contain thousands of different lipid types, and each membrane-bound organelle possesses a characteristic lipid composition necessary for its proper functioning [ 1 ]. This compositional identity is crucial not only towards their functions but also to shape and regulate intracellular signaling and trafficking processes between them [ 2 ].

Results

Lipid transport domains in solution display a conformational equilibrium that is modulated by the presence of bound lipids Since our data indicate a relationship between LTDs’ local and collective motions and membrane binding (Figs 2 and 3), we next opted to further characterize the conformational landscape of LTDs. To do so, we computed the population distribution of the projections of the first principal component (PC1) for all LTDs in our dataset, and we clustered the resulting conformations using a density-based automatic procedure [43] (Fig 4A). Simulations of 6 replicas for 500 ns each were performed to converge the distribution of PC1. Our results indicate that in the apo form, the proteins sample a diverse conformational landscape as shown by the multimodal distribution of the populations of PC1 (Fig 4C, orange histograms). Notably, and despite significant differences in the conformational landscape of the various LTDs, density-based automated clustering is able to distinguish at least 2 distinct clusters for each protein (Fig 4C, bar plots). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. Bound lipids modulate the conformational landscape of LTDs. (a) Protocol to characterize protein dynamics from atomistic simulations of the protein in solution. The first principal component PC1 was determined, and the resulting conformations were clustered using a density-based automatic procedure. (b) Simulations of 6 replicas for 500 ns each were performed to converge the distribution of PC1. (c) The population distributions of PC1 from apo (orange) and holo (purple) simulations of the protein are shown. The clusters of the distributions are indicated by the line above the histogram, with the black dot (labelled C1, C2, and C3) representing the cluster centers. Bar plots in the center indicate the relative apo and holo populations of each cluster. Box plots on the right indicate cavity volumes for apo (orange) and holo (purple) forms of the protein. Holo-forms of the protein could not be simulated for GM2A and LCN1 due to the lack of lipid-bound crystal structures. The data underlying the graphs shown in the figures can be found in https://doi.org/10.5281/zenodo.12728271. https://doi.org/10.1371/journal.pbio.3002737.g004 The alternation between different conformations is a hallmark of enzyme dynamics, where it generally correlates with protein activity [44]. To push further this parallelism, we next investigated the effect of the bound lipid on an LTD’s conformational landscape, since in classical enzymology the presence of a bound substrate is generally known to restrict enzyme dynamics and stabilize the protein in a specific conformation [45]. To determine if the presence of the bound lipid would alter the conformational preference of the LTDs in our dataset, for all cases in which a bound lipid was co-crystallized together with the protein (10 out of 12 proteins in our dataset), atomistic simulations in the holo-form were performed and analyzed following the same protocol used for the apo-forms. Notably, the comparison of the projections along the PC1 from the holo-form with that of the apo-form shows that, for all proteins, the presence of the bound lipid shifts the population distributions along the PC (Fig 4C, purple histograms). The residue-wise contribution to PC1 for simulations in the apo form, holo form, and when taken together is shown in Fig D in S1 Text. In detail, the presence of a bound lipid appears to stabilize the protein in one specific conformation (Fig 4C, bar plots). In most cases, this conformation is also sampled by the protein in its apo form, but always with a lower frequency than the corresponding holo simulations, indicating that the presence of a bound lipid indeed “locks” the LTD in a specific 3D structure, akin to substrate binding in enzyme dynamics. To further characterize this conformational landscape, we next evaluated the structural properties of representative conformers belonging to the 2 main clusters emerging from the PC analysis. To do so, we first computed the root mean square deviation (RMSD) of structures corresponding to the 2 extremes of the clusters along the PC1 spectrum (Fig E in S1 Text). For the LTDs in our dataset, this value varies between 2.3 and 7.1 Å, indicating that the range of structural difference between clusters is highly variable. Next, we computed the volume of the hydrophobic cavity of the structures sampled in our MD simulations for both apo and holo trajectories. For almost all LTDs investigated here, we could quantify significant differences in cavity volume for the 2 conditions (apo versus holo) (Fig 4C and Fig F in S1 Text). In almost all cases, when no lipid is found inside the protein, its hydrophobic cavity shrinks, thus reducing its size (Fig 4C and Fig F in S1 Text). Consistent with the behavior of the entire protein (Fig 4C), the cavity in the holo state generally adopts a more compact distribution of conformations (Fig F in S1 Text, purple), while, in the absence of bound lipids, the cavity exhibits larger variance in its volume (Fig F in S1 Text, orange).

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002737

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.