Dong et al. (J. Biophotonics 13(2):e201900214, 2020) combined two-photon Laurdan generalized-polarization (GP) imaging of membrane phase with fluorescence correlation spectroscopy (FCS) of raft-associated probes in living breast cancer cells and proposed a self-assembly model in which β-estradiol-induced enlargement of liquid-ordered (Lo) raft domains is accompanied by an increased proportion of slow subdiffusion of the ganglioside GM1 probe cholera toxin B-subunit (CT-B), which vanishes upon raft disruption by methyl-β-cyclodextrin. The model, however, is inferential: raft domain size, number, and lifetime were never directly measured; the molecular pathway from estrogen-receptor activation to raft remodeling within minutes is unknown; the framework has not been tested in other cell types, with other receptors, or in other pathophysiological states; and the fidelity of CT-B and DiIC18 as raft reporters is unestablished. The open problem is to convert this inferred self-assembly model into a directly measured, mechanism-based account in the same living-cell system: quantify the nanoscale structure (domain size, number, lifetime) underlying the slow-subdiffusion state while disentangling raft-phase domains from actin fence-and-picket compartments and probe-crosslinking artifacts; and identify the molecular steps linking estrogen-receptor activation to Lo-domain enlargement.

Background

Lipid rafts are nanoscale (putatively 5–200 nm), transient, cholesterol- and sphingolipid-enriched domains of the plasma membrane that are hypothesized to recruit signaling proteins and laterally sort lipids (Lingwood and Simons, Science 2010). Because such domains are smaller than the diffraction limit of optical microscopy and dynamic on millisecond-to-second timescales, direct measurement of their size, number, and lifetime in living cells remains an unsolved methodological problem, and the raft concept itself is still debated (Sezgin et al., Nat. Rev. Mol. Cell Biol. 2017).

Two optical observables are standardly used to probe this organization. In Laurdan generalized-polarization (GP) imaging, the solvatochromic dye Laurdan shifts its emission spectrum with local lipid packing, and the ratiometric GP value maps the balance between liquid-ordered (Lo, tightly packed, raft-like) and liquid-disordered (Ld) membrane at each pixel; GP reports local packing order but not domain size, number, or lifetime (Issler et al., BBA Biomembranes 2026). In fluorescence correlation spectroscopy (FCS), fluorescence fluctuations in a diffraction-limited spot (~0.28 µm lateral waist in confocal mode) are autocorrelated to yield molecular diffusion times and the anomaly parameter α, where α = 1 indicates Brownian diffusion and α < 1 indicates subdiffusion, i.e., transient trapping or obstruction. Cholera toxin B-subunit (CT-B) binds pentavalently to the raft-associated ganglioside GM1 and is a widely used raft probe; DiIC18 is a dialkylcarbocyanine lipid analog with low raft affinity; methyl-β-cyclodextrin (MβCD) extracts cholesterol and is used to disrupt rafts. MCF-7 cells are estrogen-receptor-positive (ER+) breast cancer cells; MDA-MB-231 cells are estrogen-receptor-negative (ER−).

Dong et al. (2020) combined two-photon Laurdan GP imaging with confocal FCS in these living cells. They reported that (i) in MCF-7 cells, CT-B shows a slow-subdiffusion population (diffusion time t > 100 ms, α ≈ 0.62) that disappears after raft depletion by 1 mM MβCD for 15 min (α rising to ≈ 0.97), while DiIC18 diffusion remains essentially Brownian (α ≈ 1, diffusion times 2–90 ms) and is not markedly affected; (ii) 1 nM β-estradiol for 5 min shifts the GP order-peak center from 0.346 ± 0.006 to 0.378 ± 0.008 and increases the proportion of CT-B slow-subdiffusion events (α ≈ 0.61) in ER+ MCF-7, whereas the same treatment produces no such changes in ER− MDA-MB-231; and (iii) on this basis they proposed a self-assembly model in which hormone-induced enlargement of the Lo (raft) area increases the probability that CT-B is transiently trapped within the FCS detection volume, so that raft fusion and lipid sorting are read out as a coupled phase-diffusion signature.

The broader literature frames what this model leaves open. Stimulated-emission-depletion FCS (STED-FCS), which shrinks the detection spot tens of nanometers, showed that sphingolipids and GPI-anchored proteins are transiently trapped (~10–20 ms) in cholesterol-dependent complexes smaller than ~20 nm (Eggeling et al., Nature 2009), and scanning STED-FCS mapped the spatiotemporal heterogeneity of such interactions (Honigmann et al., Nat. Commun. 2014) — a nanoscale regime distinct from the t > 100 ms events recorded with a ~0.28 µm confocal waist. Ultrafast single-molecule tracking reveals hop diffusion of lipids in ~100 nm compartments formed by the actin-based membrane skeleton ('fence-and-picket' model), with ~10 ms residency times (Fujiwara et al., J. Cell Biol. 2023), providing a competing structural explanation for anomalous diffusion that requires no raft domains. Directly on the probe question, Day and Kenworthy (PLOS ONE 2012) showed that CT-B's confined diffusion is impeded by actin- and ATP-dependent processes and is unaffected by caveolae; Kabbani et al. (PNAS 2020) showed that structured clustering of GM1 is required for cholera-toxin-induced membrane curvature; and Arnold et al. (Nat. Commun. 2023) quantified how probe-scale binding reports nanometer crowding heterogeneities consistent with raft-like exclusion of bulky proteins. On the stimulus side, membrane-initiated estrogen signaling through palmitoylated membrane estrogen receptor-α (ERα, a lipid-anchored form of the classical estrogen receptor residing at the plasma membrane) and the G protein-coupled estrogen receptor (GPER) is an established framework (Mauvais-Jarvis et al., Endocr. Rev. 2022), but it has not been connected to Lo-domain enlargement or raft fusion in breast cancer cells. Adjacent methodological work shows that GP imaging and scanning/line-scan FCS can be combined in cancer cell models to quantify drug-induced lipid-order changes (Tzoneva et al., Biomolecules 2020), and current reviews of fluorescence and single-molecule methods conclude that comprehensive, minimally perturbative visualization of lipid domains in living cells is still an open challenge (Mori and Suzuki, Membranes 2025; Anselmo et al., Membranes 2025).

Problem Statement

The problem, as left open by Dong et al. (2020) and unresolved by subsequent literature, is to convert the inferred coupled phase-diffusion model of lipid raft self-assembly into a directly measured, mechanism-based account of how raft domains assemble and remodel in the living breast cancer cells studied by Dong et al., with two interlocking components:

  1. Structural quantification. Directly measure the size, number, and lifetime of the membrane domains underlying the slow-subdiffusion CT-B population whose proportion increases during β-estradiol-induced liquid-ordered (Lo) domain enlargement and raft fusion in living ER+ MCF-7 cells and which vanishes upon MβCD raft depletion, as reported by Dong et al. (2020), at spatial and temporal resolution matched to the putative nanoscale raft domains — and determine how much of this state is attributable to Lo-phase raft domains, as the self-assembly model assumes, versus actin fence-and-picket compartments versus CT-B/GM1 crosslinking artifacts. The current model infers 'raft fusion' from GP shifts and diffusion-event statistics without any direct structural observable; this is the load-bearing gap.

  2. Mechanism. Identify the molecular pathway by which estrogen-receptor activation produces Lo-domain enlargement and raft fusion within minutes in ER+ MCF-7 cells (and why the same stimulus is ineffective in ER− MDA-MB-231): which receptor species — for example ERα, its variants such as ERα36, or GPER — initiates the response, and which intermediate steps — membrane receptor localization or palmitoylation, cholesterol redistribution, coupling to the cortical cytoskeleton — connect receptor activation to the observed raft remodeling.

These components jointly support one scientific objective — an experimentally grounded mechanism of raft self-assembly in the living-cell membrane — and they are not separable in practice: the mechanistic pathway (2) cannot be tested without a structural readout that is independent of the probes' own reorganization of the membrane (1), and the structural attribution in (1) requires discriminating Lo-phase trapping from probe-crosslinking artifacts. Whether the coupled phase-diffusion signature generalizes to other cell types, other signaling receptors, and other pathophysiological conditions, and the stand-alone fidelity of CT-B/GM1 and DiIC18 as raft reporters, remain open context informing these two components rather than separate requirements of the answer. The problem is answered when the coupled phase-diffusion signature observed by Dong et al. in these living breast cancer cells is explained by directly measured domain structure and a demonstrated receptor-to-domain molecular pathway, with the conditions of the explanation in that system stated.

References

  1. Dong S, Tang X, Wang J, Zhang Z, Chen J, Lin Y, Xie S, Wang Z, Yang H. 'Self-assembly of lipid rafts revealed by fluorescence correlation spectroscopy in living breast cancer cells.' Journal of Biophotonics 13(2):e201900214 (2020; published online 5 December 2019). DOI: 10.1002/jbio.201900214. https://doi.org/10.1002/jbio.201900214
  2. Lingwood D, Simons K. 'Lipid Rafts As a Membrane-Organizing Principle.' Science 327(5961):46-50 (2010). DOI: 10.1126/science.1174621. https://doi.org/10.1126/science.1174621
  3. Sezgin E, Levental I, Mayor S, Eggeling C. 'The mystery of membrane organization: composition, regulation and roles of lipid rafts.' Nature Reviews Molecular Cell Biology 18(6):361-374 (2017). DOI: 10.1038/nrm.2017.16. https://doi.org/10.1038/nrm.2017.16
  4. Eggeling C, Ringemann C, Medda R, Schwarzmann G, Sandhoff K, Polyakova S, Belov VN, Hein B, von Middendorff C, Schönle A, Hell SW. 'Direct observation of the nanoscale dynamics of membrane lipids in a living cell.' Nature 457(7233):1159-1162 (2009). DOI: 10.1038/nature07596. https://doi.org/10.1038/nature07596
  5. Day CA, Kenworthy AK. 'Mechanisms Underlying the Confined Diffusion of Cholera Toxin B-Subunit in Intact Cell Membranes.' PLoS ONE 7(4):e34923 (2012). DOI: 10.1371/journal.pone.0034923. https://doi.org/10.1371/journal.pone.0034923
  6. Honigmann A, Mueller V, Ta H, Schoenle A, Sezgin E, Hell SW, Eggeling C. 'Scanning STED-FCS reveals spatiotemporal heterogeneity of lipid interaction in the plasma membrane of living cells.' Nature Communications 5:5412 (2014). DOI: 10.1038/ncomms6412. https://doi.org/10.1038/ncomms6412
  7. Komura N, Suzuki KGN, Ando H, Konishi M, Koikeda M, Imamura A, Chadda R, Fujiwara TK, Tsuboi H, Sheng R, Cho W, Furukawa K, Furukawa K, Yamauchi Y, Ishida H, Kusumi A, Kiso M. 'Raft-based interactions of gangliosides with a GPI-anchored receptor.' Nature Chemical Biology 12(6):402-410 (2016). DOI: 10.1038/nchembio.2059. https://doi.org/10.1038/nchembio.2059
  8. Fujiwara TK, Takeuchi S, Kalay Z, Nagai Y, Tsunoyama TA, Kalkbrenner T, Iwasawa K, Ritchie KP, Suzuki KGN, Kusumi A. 'Development of ultrafast camera-based single fluorescent-molecule imaging for cell biology.' Journal of Cell Biology 222(8):e202110160 (2023). DOI: 10.1083/jcb.202110160. https://doi.org/10.1083/jcb.202110160
  9. Kabbani AM, Raghunathan K, Lencer WI, Kenworthy AK, Kelly CV. 'Structured clustering of the glycosphingolipid GM1 is required for membrane curvature induced by cholera toxin.' Proceedings of the National Academy of Sciences USA 117(26):14978-14986 (2020). DOI: 10.1073/pnas.2001119117. https://doi.org/10.1073/pnas.2001119117
  10. Arnold DP, Xu Y, Takatori SC. 'Antibody binding reports spatial heterogeneities in cell membrane organization.' Nature Communications 14:2884 (2023). DOI: 10.1038/s41467-023-38525-2. https://doi.org/10.1038/s41467-023-38525-2
  11. Mauvais-Jarvis F, Lange CA, Levin ER. 'Membrane-Initiated Estrogen, Androgen, and Progesterone Receptor Signaling in Health and Disease.' Endocrine Reviews 43(4):720-742 (2022). DOI: 10.1210/endrev/bnab041. https://doi.org/10.1210/endrev/bnab041
  12. Tzoneva R, Stoyanova T, Petrich A, Popova D, Uzunova V, Momchilova A, Chiantia S. 'Effect of Erufosine on Membrane Lipid Order in Breast Cancer Cell Models.' Biomolecules 10(5):802 (2020). DOI: 10.3390/biom10050802. https://doi.org/10.3390/biom10050802
  13. Mori T, Suzuki KGN. 'Single-Molecule Imaging and Super-Resolution Microscopy of Lipid Domains in Cell Membranes Using Lipid-Binding Proteins and Fluorophore-Conjugated Lipid Analogs.' Membranes 15(10):317 (2025). DOI: 10.3390/membranes15100317. https://doi.org/10.3390/membranes15100317
  14. Anselmo S, Bonaccorso E, Gangemi C, Sancataldo G, Conti Nibali V, D'Angelo G. 'Lipid Rafts in Signalling, Diseases, and Infections: What Can Be Learned from Fluorescence Techniques?' Membranes 15(1):6 (2025). DOI: 10.3390/membranes15010006. https://doi.org/10.3390/membranes15010006
  15. Issler T, Kerek E, Prenner EJ. 'The evolution of steady-state laurdan fluorescence: From model membranes to cellular applications.' Biochimica et Biophysica Acta (BBA) - Biomembranes 1868:184562 (2026). DOI: 10.1016/j.bbamem.2026.184562. https://doi.org/10.1016/j.bbamem.2026.184562
  16. Shao W, Dong S, Tang X, Chen J, Yang H. 'Highly sensitive detection of DNA fragments by fluorescence correlation spectroscopy.' Proc. SPIE 11548, Optics in Health Care and Biomedical Optics X (2020). DOI: 10.1117/12.2575144. https://doi.org/10.1117/12.2575144