Osh6 Revisited: Control of PS Transport by the Concerted Actions of PI4P and Sac1 Phosphatase

Osh6, a member of the oxysterol-binding protein–related protein (ORP) family, is a lipid transport protein that is involved in the transport of phosphatidylserine (PS) between the endoplasmic reticulum (ER) and the plasma membrane (PM). We used a biophysical approach to characterize its transport mechanism in detail. We examined the transport of all potential ligands of Osh6. PI4P and PS are the best described lipid cargo molecules; in addition, we showed that PIP2 can be transported by Osh6 as well. So far, it was the exchange between the two cargo molecules, PS and PI4P, in the lipid-binding pocket of Osh6 that was considered an essential driving force for the PS transport. However, we showed that Osh6 can efficiently transport PS along the gradient without the help of PI4P and that PI4P inhibits the PS transport along its gradient. This observation highlights that the exchange between PS and PI4P is indeed crucial, but PI4P bound to the protein rather than intensifying the PS transport suppresses it. We considered this to be important for the transport directionality as it prevents PS from returning back from the PM where its concentration is high to the ER where it is synthesized. Our results also highlighted the importance of the ER resident Sac1 phosphatase that enables the PS transport and ensures its directionality by PI4P consumption. Furthermore, we showed that the Sac1 activity is regulated by the negative charge of the membrane that can be provided by PS or PI anions in the case of the ER membrane.

. Response of the used techniques to the amount of the lipid of interest in the membrane of LUVs sensed by its biosensor. The calibration was acquired under the same conditions we used for the transport experements. a) FCCS read-out Gcc/GR as a function of the molar fraction of PI4P in the membrane, b) FCCS read-out Gcc/GR as a function of the molar fraction of PIP2 in the membrane, a) FRET read-out as a function of the molar fraction of PS in the membrane.
Evaluation and use of the FCCS data Fluorescence crosscorrelation spectroscopy (FCCS) is a microscopy technique that suits for monitoring interactions of fluorescently labeled species. It uses the confocal microscope to focus excitation light into a diffraction limited spot. Motion of fluorescently labeled particles through the focus translates into fluctuations in the detected signal. These fluctuations refer on the number of molecules in the focus and the speed of their diffusion through. If additionally, two differently labeled species are excited by two different overlapping laser beams and the resulting signal is spectrally split and detected on two detectors, the relatedness of the two fluctuation traces refers on the mutual interaction. The relatedness is quantified by so called cross-correlation function Gcc(t): where δI is a fluctuation of the green and the red signal, the brackets stand for temporal averaging. The correlation decays with time τ, which is the time lag between the green and the red signal. The interacting species are monitored at τ = 0, as at zero time lag, diffusion does not cause any decay of the correlation and its amplitude refers on the concentration of the interacting participants.
In the case of the experiments done here, the red labeled species are the LUVs containing the red fluorescent marker DiD. The green species are the CFP labeled biosensors for various lipids (PI4P, PIP2). During the transportation the lipid of interest is moved from the unlabeled LUVs to the red labeled LUVs and the biosensor goes along. For the crosscorrelation amplitude it holds: where cR and cG is concentration of the red labeled LUVs and the LUVs, from which the biosensor is moved by following the lipid transport, respectively. fGnR is the fraction of red LUVs bearing n molecules of the CFP-biosensor. fGm is the fraction of the unlabeled LUVs with m molecules of the CFP-biosensor. Veff GR is the effective volume of the two overlapping foci.
In order to keep the transport monitoring parameter independent of the red labeling (the concertation of the red dye can vary due to pipetting inaccuracies, for example), we always follow the ratio between the crosscorrelation and the red autocorrelation amplitude GR(0) ( % 0 = 1 ' ()) % * % ⁄ ). Eventually, the FCCS transport referring read-out is: It consists of the volume ratio of the red focus and the overlap of the two foci (geometry of the beams) and from the contribution of biosensor distribution between the labeled and unlabeled LUVs.
The distribution of the biosensor molecules among the given type of LUVs is assumed to be Poissonian. Fig. S2a illustrates how the biosensor molecules are distributed between the two different LUV types at 5, 25 and 60 % of the biosensor translocation. Independent of the ratio of the two types of LUVs and the total amount of the biosensor, the amplitude ratio Gcc/GR is directly proportional to the fraction of the translocated biosensor (Fig. S2b -red line). Additional FCCS parameter that refers on the transport is the amplitude of the green autocorrelation function. It follows the formula: Veff G is the volume of the green focal region. Fig. S2b (black curves) depicts three curves for different ratios of red and unlabelled LUVs. All three are parabolic showing that first, a decay is observed when the biosensor molecules start occupy more LUVs, and second, they start prevailing at the target LUVs, i.e. the amount of labelled particles drops which is accompanied by the increase in GG amplitude.
We have also used GG to visualize the transport of PI4P followed by its dephosphorylation in the target LUVs. If PI4P is only transported and the dephosphorylation does not occur, decay in GG can be observed (black solid curve in Fig. S2b and S2c). However, if the dephosphorylation of Sac1 occurs upon the transport, the biosensor is not transferred to LUVs but instead is released to the solution, this is manifested by the increase in the GG amplitude (red curve in Fig. S2c).