Single-fluorescent molecules of the ganglioside probes, but not the DOPE probe, were recruited to induced large rafts in cell PMsThe ganglioside probes partitioned into Lo phase in GPMVs, but it is still unknown if these probes partition into raft domains in living cell PMs. To solve this issue, we performed single-fluorescent molecule imaging of ATTO594-GM3 and ATTO594-GM1 (termed as 594-GM3 and 594-GM1, respectively) in and outside of CD59 patches (used as a marker of raft regions) induced by crosslinking with primary and secondary antibodies. Each ganglioside probe was maintained within the boundaries of CD59 patch and the bulk region, rather than passing through (Figure 2). In contrast, non-raft unsaturated phospholipid 594-DOPE was primarily located outside of the CD59 patch (Figure 2). We found that 594-GM3 and 594-GM1 were enriched in the CD59 patches by 4- to 10-fold, while 594-DOPE was almost completely excluded3).Stabilization of GPI-anchored protein clusters and homodimers through ganglioside recruitmentWe next investigated raft formation under more physiological conditions. Upon ligation, a GPI-anchored protein CD59 formed stable oligomers that recruit and activate intracellular signaling molecules. These CD59 oligomers have been termed “CD59-cluster signaling rafts”. To examine if these CD59 oligomers really form rafts, we performed simultaneous dual-color single-molecule observations of fluorescent ganglioside probes (594-GM3 and 594-GM1) and the stable CD59 oligomers (Figure 3A). As a result, both 594-GM3 and 594-GM3 were recruited to the stable CD59 clusters for approximately 100 ms before separating, meanwhile 594-DOPE co-localized with the CD59 clusters for much shorter periods of time. After cholesterol depletion, the durations of co-localization were reduced to about a half. These results indicated that the co-localization of gangliosides and CD59 was not due to an interaction between the head group regions of gangliosides and CD59, but was rather due to raft-lipid interactions.In the steady-state cells, CD59 formed transient homodimers with a lifetime of 160 ms. Next, to investigate whether CD59 homodimers form small rafts, we simultaneously observed single molecules of CD59 and the ganglioside probes in the steady-state cell PMs. We found that the ganglioside probes were transiently recruited to CD59 homodimers for about 80 ms, and to CD59 monomers for about 50 ms. The non-raft phospholipid, DOPE co-localized with CD59 monomers and homodimers for only 40 ms, shorter than the co-localization lifetimes of the ganglioside probes. Transient interactions between the ganglioside probes and both CD59 monomers and homodimers may be due to interactions between glycans and/or lipid moieties. These results support our previous results indicating that CD59 clusters and CD59 homodimers are stabilized by raft-lipid interactions3)(Figure 3B).Unraveling of regulation mechanisms of receptor activity by gangliosidesKenichi G. N. Suzukisugar residues. When the gangliosides were conjugated with hydrophobic dyes such as TMR and ATTO647N, the probes partitioned into liquid disordered (Ld) phase, and their behaviors were very different form those of the parental gangliosides.Figure 2. Typical simultaneously-obtained, merged total internal reflection microscopic images of CD59 patches and single molecules of 594-GM3 and 594-DOPE in T24 cell PMsAnalysis of temporal confinement of ganglioside probes by high-speed observations of single-molecule probesPrevious reports have shown that ganglioside probes were temporarily confined in small domains (< 20 nm in diameter) in the steady-state cell PMs for 10~20 ms. However, ganglioside probes used in these earlier studies, partitioned into non-raft fractions (non-DRM and Ld phase). Therefore, to examine whether gangliosides are transiently trapped in small domains, we performed single-fluorescent molecule imaging of our ganglioside probes in the PM of 4 cell types at high-temporal resolution (2000 frames/sec). We found that fractions of time in which diffusion was constantly confined to small areas (the immobile fraction) were rarely found for all three tested molecules in each of the cell types. The proportions of time spent in the 100 nm diameter domains were estimated to be 1.7~3.1% for 594-S9-GM3 and GM1 and 4.6~5.2% 48
元のページ ../index.html#48