Novel GFP-fused protein probes for detecting phosphatidylinositol-4-phosphate in the plasma membrane

ABSTRACT Phosphatidylinositol-4-phosphate (PI4P) plays a crucial role in cellular functions, including protein trafficking, and is mainly located in the cytoplasmic surface of intracellular membranes, which include the trans-Golgi network (TGN) and the plasma membrane. However, many PI4P-binding domains of membrane-associated proteins are localized only to the TGN because of the requirement of a second binding protein such as ADP-ribosylation factor 1 (ARF1) in order to be stably localized to the specific membrane. In this study, we developed new probes that were capable of detecting PI4P at the plasma membrane using the known TGN-targeting PI4P-binding domains. The PI4P-specific binding pleckstrin homology (PH) domain of various proteins including CERT, OSBP, OSH1, and FAPP1 was combined with the N-terminal moderately hydrophobic domain of the short-form of Aplysia phosphodiesterase 4 (S(N30)), which aids in plasma membrane association but cannot alone facilitate this association. As a result, we found that the addition of S(N30) to the N-terminus of the GFP-fused PH domain of OSBP (S(N30)-GFP-OSBP-PH), OSH1 (S(N30)-GFP-OSH1-PH), or FAPP1 (S(N30)-GFP-FAPP1-PH) could induce plasma membrane localization, as well as retain TGN localization. The plasma membrane localization of S(N30)-GFP-FAPP1-PH is mediated by PI4P binding only, whereas those of S(N30)-GFP-OSBP-PH and S(N30)-GFP-OSH1-PH are mediated by either PI4P or PI(4,5)P2 binding. Taken together, we developed new probes that detect PI4P at the plasma membrane using a combination of a moderately hydrophobic domain with the known TGN-targeting PI4P-specific binding PH domain.


Introduction
Phosphoinositides, which are derivatives of phosphatidylinositol (PI), are minor lipid components of eukaryotic membranes, but have important roles in many cellular functions, including serving as lipid markers of intracellular membrane organelles (Di Paolo and De Camilli 2006;Lemmon 2008). Phosphoinositides are not randomly distributed within intracellular membranes, but rather, each type has a unique cellular localization (Di Paolo and De Camilli 2006;Jang et al. 2009). For example, phosphatidylinositol 3-phosphate (PI3P) is localized to the cytoplasmic surface of early endosomes, while phosphatidylinositol 4phosphate (PI4P) is mainly enriched in the trans-Golgi network (TGN) of cells (D'Angelo et al. 2008).
PI4P is mostly known to serve as a precursor of phosphatidylinositol 4,5-bisphosphate (PI(4,5)P 2 ). However, PI4P plays many roles in the plasma membrane. For example, PI4P is involved in the plasma membrane targeting of many plasma membrane proteins through direct binding or electrostatic interactions (Hammond et al. 2012;Simon et al. 2016). Depletion of both PI4P and PI(4,5)P 2 simultaneously could change the localization of many plasma membrane-targeted proteins, such as the short forms of Aplysia phosphodiesterase 4 (PDE4) and K-Ras in animal cells (Jang et al. 2010;Hammond et al. 2012;Kim et al. 2014). In plant cells, PI4P is highly enriched in the plasma membrane and plays a key role in the formation of membrane surface charges that lead to the recruitment of a variety of membrane proteins (Simon et al. 2016). PI4P is indirectly involved in the enrichment of phosphatidylserine (PS) in the plasma membrane through oxysterol-binding protein (OSBP)-related protein 5 (ORP5) and ORP8 (Chung et al. 2015;Sohn and Korzeniowski 2018).
Until recently, although PI4P is located at the cytoplasmic surface of the TGN, plasma membrane, and endosome, most PI4P-binding probes were only localized to the TGN. For example, the PH domains of OSH1, four-phosphate-adaptor protein (FAPP), and oxysterol-binding protein (OSBP) are PI4P-binding proteins, but they require other binding proteins, including ADP ribosylation factor 1 (ARF1), to be localized to the TGN (Hanada et al. 2003;Godi et al. 2004;Roy and Levine 2004). Recently, we found that an N-terminal hydrophobic domain itself was not sufficient for specific membrane organelle targeting, but additional domains induced the stable plasma membrane or autopahgosome association of the ApPDE4 short-form (Kim et al. 2014;Lee et al. 2017). Therefore, we hypothesized that if the N-terminal hydrophobic domain of the ApPDE4 short-form, which itself was localized in the cytosol but helps the plasma membrane localization, was combined with the PI4P-binding PH domains, these novel PI4Pspecific probes could detect PI4P at the plasma membrane and endosome ( Figure 1A).
In this study, we found that if an N-terminal hydrophobic domain of ApPDE4 short-form is linked to the GFP-fused PH domain of OSBP, OSH1, and FAPP1, these proteins were localized to the plasma membrane as well as the TGN. The plasma membrane localization of modified OSBP and OSH1 is mediated by PI4P as well as PI(4,5)P 2 , whereas a modified PH domain FAPP1 was localized to the plasma membrane and endosome mainly via PI4P binding. Thus, we could develop new probes detecting PI4P at the plasma membrane using a combination of the moderately hydrophobic domain with the known PI4P-specific binding PH domain.

HEK293 T cell culture and confocal microscopy
HEK293 T cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% (v/v) fetal bovine serum and penicillin/streptomycin in a humidified atmosphere of 5% (v/v) CO 2 at 37°C. Cells were seeded in a sticky-Slide 8-well system (Catalog #: 80828, Ibidi, Martinsried, Germany), in order to reach 40%-60% confluency on the day of imaging. In the period 24-26 h before imaging, the cells were transfected with DNA constructs using calcium phosphate or Lipofectamine 2000 (Life Technologies, Carlsbad, CA, USA). The relative amount of each construct was empirically determined based on the relative expression of each construct combination. Cells were then analyzed using an inverted Zeiss LSM-700 confocal scanning laser microscope operated by ZEN software (Carl Zeiss). Most images were derived from live cells.

Results and discussion
Generation of modified PI4P binding probes localized to the plasma membrane We first examined the cellular localization of the GFPfused PH domain of OSBP (GFP-OSBP-PH), OSH1 (GFP-   Figure 1C). Interestingly, although S(N30)-GFP alone was localized to the cytosol, S(N30)-GFP-OSBP-PH, S (N30)-GFP-OSH1-PH, and S(N30)-GFP-FAPP1-PH were clearly localized to the plasma membrane as well as the TGN ( Figure 1B, lower and C). In the case of S(N30)-GFP-P4M-SidM, it was localized to the plasma membrane more than GFP-P4M-SidM. Meanwhile, S(N30)-GFP-CERT-PH was localized to only the TGN, as was GFP-CERT-PH. Taken together, our results showed that the addition of the N-terminal hydrophobic domain of the ApPDE4 short-form into the PI4P-binding domains could induce plasma membrane targeting as well as retain TGN targeting.

Effect of an acute depletion of phosphoinositides on cellular localization of modified PI4P-binding domains
Next, to determine whether phosphoinositides are involved in the plasma membrane localization of these  probes, antimycin A, an ATP synthesis inhibitor, was used to deplete cellular lipid derivatives generated by various lipid kinases, including various phosphoinositides such as PI4P, PI(4,5)P 2 , and PI(3,4,5)P 3 within cells (Kim et al. 2014). As shown in Figure 2A, in the presence of antimycin A, the plasma membrane localization of S(N30)-GFP-OSBP-PH, S(N30)-GFP-OSH1-PH, S(N30)-GFP-FAPP1-PH, and S(N30)-GFP-P4M-SidM was disrupted (Figure 2A). In the case of S(N30)-GFP-FAPP1-PH, there was weak localization to the nuclear membrane in the presence of antimycin A (Figure 2A). Taken together, these results indicate the possibility that the plasma membrane localization of GFP-fusion proteins is probably mediated by phosphoinositide binding.
Overall, this study demonstrates that by adding the Nterminal hydrophobic domain of the short form of ApPDE4 to the known TGN-targeted PI4P binding domain, a probe capable of PI4P detection in the plasma membrane was generated. These constructs will be useful for future PI4P research.

Disclosure statement
No potential conflict of interest was reported by the authors.