Metabolically stabilized derivatives of phosphatidylinositol 4-phosphate: Synthesis and applications

Article abstract of DOI:10.1016/j.chembiol.2011.07.022

Phosphatidylinositol 4-phosphate (PtdIns(4)P) lipid is an essential component of eukaryotic membranes and a marker of the Golgi complex. Here, we developed metabolically stabilized (ms) analogs of PtdIns(4)P and the inositol 1,4-bisphosphate (IP₂) head group derivative and demonstrated that these compounds can substitute the natural lipid fully retaining its physiological activities. The methylenephosphonate (MP) and phosphorothioate (PT) analogs of PtdIns(4)P and the aminohexyl (AH)-IP₂ probe are recognized by the PtdIns(4)P-specific PH domain of four phosphate adaptor protein 1 (FAPP1). Binding of FAPP1 to the PtdIns(4)P derivatives stimulates insertion of the PH domain into the lipid layers and induces tubulation of membranes. Both ms analogs and IP₂ probes could be invaluable for identifying protein effectors and characterizing PtdIns(4)P-dependent signaling cascades within the trans-Golgi network (TGN).

Full text of DOI:10.1016/j.chembiol.2011.07.022

Chemistry & Biology
Article
Metabolically Stabilized Derivatives
of Phosphatidylinositol 4-Phosphate:
Synthesis and Applications
Ju He,1,7 Joanna Gajewiak,2,7 Jordan L. Scott,3,4,7 Denghuang Gong,5 Muzaffar Ali,1 Michael D. Best,5
Glenn D. Prestwich,2 Robert V. Stahelin,3,4,6 and Tatiana G. Kutateladze1,
*
1Department of Pharmacology, University of Colorado School of Medicine, Aurora, CO 80045, USA
2Department of Medicinal Chemistry, University of Utah, Salt Lake City, UT 84108, USA
3Department of Chemistry and Biochemistry
4The Walther Center for Cancer Research
University of Notre Dame, Notre Dame, IN 46556, USA
5Department of Chemistry, University of Tennessee, Knoxville, TN 37996, USA
6Department of Biochemistry and Molecular Biology, Indiana University School of Medicine-South Bend, South Bend, IN 46617, USA
7These authors contributed equally to this work
*Correspondence: tatiana.kutateladze@ucdenver.edu
DOI 10.1016/j.chembiol.2011.07.022
SUMMARY
position by a set of PI 4-kinases, including PI4KIIa, PI4KIIb,
and PI4KIIIb (D’Angelo et al., 2008). It can be further phosphory-
Phosphatidylinositol 4-phosphate (PtdIns(4)P) lipid lated to the bisphosphorylated isoform PtdIns(4,5)P2 through the
action of PI 5-kinases or hydrolyzed back to PtdIns by a Sac1
phosphatase. For a long time PtdIns(4)P was thought to be
is an essential component of eukaryotic membranes
and a marker of the Golgi complex. Here, we devel-
oped metabolically stabilized (ms) analogs of
PtdIns(4)P and the inositol 1,4-bisphosphate (IP2)
head group derivative and demonstrated that these
compounds can substitute the natural lipid fully re-
taining its physiological activities. The methylene-
phosphonate (MP) and phosphorothioate (PT)
merely a precursor of PtdIns(4,5)P2; however, in the last decade
numerous studies have shown that PtdIns(4)P itself functions as
a signaling molecule with its monophosphorylated head group
serving as a docking site for protein effectors.
PtdIns(4)P is recognized in the TGN by two families of effec-
tors. One family is comprised of soluble lipid-transfer proteins,
such as four phosphate adaptor protein 1 and 2 (FAPP1 and
FAPP2), oxysterol-binding protein (OSBP), and ceramide trans-
fer protein (CERT). Each of these effectors contains a PtdIns(4)
analogs of PtdIns(4)P and the aminohexyl (AH)-IP2
probe are recognized by the PtdIns(4)P-specific
PH domain of four phosphate adaptor protein 1 P-specific pleckstrin homology (PH) domain that binds to the
lipid head group, recruiting the host protein to the Golgi
(FAPP1). Binding of FAPP1 to the PtdIns(4)P deriva-
membranes. The recent reports reveal that selective interaction
with PtdIns(4)P is required for subcellular localization, activation,
and function of the effectors (D’Angelo et al., 2008). For example
tives stimulates insertion of the PH domain into the
lipid layers and induces tubulation of membranes.
Both ms analogs and IP2 probes could be invaluable
for identifying protein effectors and characterizing
PtdIns(4)P-dependent signaling cascades within
the trans-Golgi network (TGN).
secretory transport from the Golgi to the plasma membrane is
mediated by FAPPs. The FAPP proteins regulate the formation
and fission of post-Golgi vesicles, and these activities depend
on the association of the PH domain with PtdIns(4)P (Godi
et al., 2004). Likewise, recognition of PtdIns(4)P by the adaptor
and coat complexes, which comprise another family of the
INTRODUCTION
protein effectors, is essential for TGN-to-endosome trafficking
and the formation of clathrin-coated vesicles (Godi et al.,
Phosphatidylinositol 4-phosphate (PtdIns(4)P) is a monophos- 2004). The fundamental role of PtdIns(4)P in signaling, mem-
phorylated derivative of the phosphatidylinositol (PtdIns) lipid brane trafficking, and protein sorting necessitates the develop-
and an essential component of eukaryotic membranes. Among ment of synthetic analogs of this lipid that can be used in
the seven phosphorylated isoforms of PtdIns, known as phos- biochemical, structural, and functional assays to advance our
phoinositides (PIs), PtdIns(4)P is the most abundant and understanding of the biological processes occurring at the Golgi
comprises ꢀ0.05% of all phospholipids in mammalian cells membranes.
(Lemmon, 2008; Martin, 1998). It is found in highest concentra-
In this study we developed metabolically stabilized (ms) ana-
tions in membranes of the trans-Golgi network (TGN) and is logs of PtdIns(4)P and an inositol 1,4-bisphosphate (Ins(1,4)P2
commonly viewed as a marker of the Golgi complex. PtdIns(4)P or IP2) head group derivative suitable for outfitting with a reporter
is generated in the outer leaflet of the Golgi membranes via tag and demonstrated that these compounds can substitute the
phosphorylation of the inositol head group of PtdIns at the C4 natural lipid in peripheral protein recruitment and membrane
1312 Chemistry & Biology 18, 1312–1319, October 28, 2011 ª2011 Elsevier Ltd All rights reserved

Chemistry & Biology
Synthesis and Applications of PI4P Analogs
Figure 1. Synthesis of the ms PtdIns(4)P-MP, 8a
(R = C7H15) and 8b (R = C15H31), and PtdIns(4)P-
PT, 15a (R = C7H15) and 15b (R = C15H31), Analogs
of PtdIns(4)P
Top panel shows reagents and conditions: (A) CH3ONa,
CH3OH, 88%; (B) TfOCH2PO(OMe)2, BuLi, THF, 50%; (C)
TBAF$H2O, THF, 95%; (D) 6a, (R = C7H15) or 6b, (R =
C15H31), 1H-tetrazole, t-BuOOH, CH2Cl2, 88%–89%; (E)
Et3N, BSTFA, CH3CN; (F) TMSBr, CH2Cl2, 0ꢁC-rt; (G)
CH3OH, rt; (H) DOWEX [Na+] ion exchange resin, H2O,
92%–95%.
Bottom panel illustrates reagents and conditions: (A)
TBAF$H2O, THF, 83%; (B) TESCl, imidazole, CH2Cl2,
85%; (C) DIBHAL-H, CH2Cl2, ꢂ78ꢁC, 92%; (D) bis(2-cya-
noethyl)diisopropyl
phosphoramidite,
1H-tetrazole,
CH3CN, rt, then phenylacetyl disulfide, rt, 65%; (E) NH4F,
CH3OH, rt, 85%; (F) 6a (or 6b), 1H-tetrazole, t-BuOOH,
CH2Cl2, 67%–87%; (G) Et3N, BSTFA, CH3CN; (H) TMSBr,
CH2Cl2, 0ꢁC-rt; (I) CH3OH, rt; (J) DOWEX [Na+] ion
exchange resin, H2O, 80%–90%.
(MP) and phosphorothioate (PT) moieties, elim-
inating or reducing phosphate hydrolysis. To
prepare the PtdIns(4)P-MP and PtdIns(4)P-PT
analogs (Figure 1), we first employed the protec-
tion scheme developed by Kubiak and Bruzik
(2003). The 1-position of myo-inositol (1) was
silylated with the tert-butyldiphenylsilyl (TBDPS)
group to give intermediate 2, in which the
planned 4-MP position was protected as a
benzoate (Bz) derivative, and all remaining
hydroxyl groups were protected as methoxy-
methyl (MOM)-ethers. To prepare the PtdIns(4)
P-MP analogs (Figure 1, top panel), the
Bz group was then removed to provide the
1-O-(tert-butyldiphenylsilyl)-2,3,5,6-O-tetrakis-
(methoxymethylene)-myo-inositol (3). Installa-
tion of the MP group started with the synthesis
deformation. The advantage of applying the ms analogs in of dimethyl phosphonomethyltriflate (Hamilton and Roberts,
molecular and cell biology experiments is in their intrinsic ability 1999; Phillion and Andrew, 1986), which was coupled with
to resist dephosphorylation by PI 4-phosphatases. Thus, the compound 3 using n-BuLi (Minutolo et al., 2004). The TBDPS
signaling pathway involving activity of Sac1 can be separated group was removed by treating intermediate 4 with Bu4NF$H2O,
from the biological processes associated with binding of protein and compound 5 was produced in high yield (95%). Prior to the
effectors to PtdIns(4)P. Both ms analogs and IP2 probes could be coupling step, two 2-cyanoethyl phosphoramidites 6a and 6b
invaluable for identifying novel protein effectors and character- were prepared from 1,2-O-isopropylidene-sn-glycerol in five
izing and monitoring PtdIns(4)P-dependent signaling events at steps. Then, in the presence of 1H-tetrazole, alcohol 5 was
the TGN.
coupled with the phosphoramidite (Gajewiak et al., 2006; Huang
et al., 2007), followed by the mild oxidation with t-BuOOH to yield
fully protected intermediates 7a and 7b. Finally, the removal of
the cyanoethyl groups with triethylamine and bis(trimethylsilyl)
trifluoroacetamide (BSTFA) followed by removal of the MOM
and the methyl ester groups with TMSBr afforded the C8-
RESULTS AND DISCUSSION
Total SynthesisofthePtdIns(4)P-Methylenephosphonate
and PtdIns(4)P-Phosphorothioate Analogs
One of the major limitations in the use of natural or synthetic PIs PtdIns(4)-MP and C16-PtdIns(4)-MP (8a and 8b, respectively).
for monitoring time-dependent phenomena in biochemical and The synthesis of the 4-PT analogs of PtdIns(4)P is shown in
biological applications is that they undergo rapid hydrolysis by the bottom panel of Figure 1. The 4-benzoyl-1-TBDPS-2,3,5,6-
phosphatases. To overcome this problem, we have designed tetrakis(MOM)-inositol (compound 2) was treated with TBAF to
metabolically stable analogs of PtdIns(4)P that are resistant to remove the 1-TBDPS ether and replace it with the triethylsilyl
dephosphorylation. We replaced the 4-phosphate group in the ether (TES). After debenzoylation of (10) with diisobutyl
inositol ring of the lipid with the alkoxymethylenephosphonate aluminum hydride (DIBAL-H), the PT group was introduced at
Chemistry & Biology 18, 1312–1319, October 28, 2011 ª2011 Elsevier Ltd All rights reserved 1313

Chemistry & Biology
Synthesis and Applications of PI4P Analogs
to bind PtdIns(4)P (Figure 2A) (Dowler et al., 2000; Godi et al.,
2004; Lenoir et al., 2010). The GST-fusion FAPP1 PH domain
was incubated with small unilamellar vesicles (SUVs) composed
of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC),
1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE),
1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoserine (POPS), and
either dipalmitoyl (C16)-PtdIns(4)P, C16-PtdIns(4)P-MP, or C16
-
PtdIns(4)P-PT. Following centrifugation the distribution of the
PH domain between the supernatant and pelleted fractions was
examined. As expected, in the absence of PIs, GST-FAPP1 PH
was found primarily in the soluble fraction; however, half of the
protein associated with SUVs containing unmodified C16
-
PtdIns(4)P, corroborating the findings that FAPP1 is specific for
PtdIns(4)P (Dowler et al., 2000). When C16-PtdIns(4)P-MP or
C16-PtdIns(4)P-PT lipids were incorporated in the vesicles,
almost all GST-FAPP1 PH was pelleted with SUVs, indicating
that the protein binds to the synthetic analogs stronger than it
binds to the unmodified lipid.
Binding affinities of the FAPP1 PH domain for the ms analogs
were measured by surface plasmon resonance (SPR) in 10 mM
HEPES (pH 7.4) containing 160 mM KCl (Figure 2). In these
experiments the active surface was coated with POPC/POPE/
C16-PtdIns(4)P-MP or POPC/POPE/C16-PtdIns(4)P-PT lipo-
somes, and the control surface was coated with POPC/POPE
liposomes. The FAPP1 PH domain was injected at varying
concentrations, and the equilibrium response (Req) was reached
in each reaction (Figure 2C). The Req values were subsequently
used to generate binding isotherms (Figure 2D). As summarized
in Figure 2B, the ms analogs were bound by the FAPP1 PH
domain more strongly (Kd values = 250 and 390 nM) than unmod-
ified C16-PtdIns(4)P (Kd = 460 nM). Thus, the synthetic deriva-
tives of PtdIns(4)P can substitute the natural lipid retaining its
ability to associate with protein effectors.
Figure 2. PtdIns(4)P-MP and PtdIns(4)P-PT Are Recognized by the
FAPP1 PH Domain
(A) The SDS-PAGE gel showing the partitioning of GST-fusion FAPP1 PH
between supernatant (S) and the pellet (P). SUVs contain either C16-PtdIns(4)P,
C16-PtdIns(4)P-MP, 8b or C16-PtdIns(4)P-PT, 15b.
The PtdIns(4)P-PT Analog Occupies the Binding Site
for the Natural Lipid
(B) Binding affinities of the wild-type and mutant FAPP1 PH domain were
measured by SPR (a) and NMR (b).
To assess if the modification of the lipid head group alters the
binding mechanism, we investigated interaction between the
FAPP1 PH domain and PtdIns(4)P-PT by NMR spectroscopy
(Figure 3). The 1H,15N heteronuclear single quantum coherence
(HSQC) spectra of uniformly 15N-labeled FAPP1 PH were
collected as a water-soluble dioctanoyl (C8) form of PtdIns(4)P-
PT was added stepwise (Figure 3A). Substantial chemical shift
changes induced in the PH domain by the analog confirmed
direct interaction. We identified the C8-PtdIns(4)P-PT-binding
(C and D) Representative SPR sensorgrams obtained in 10 mM HEPES (pH
7.4) containing 160 mM KCl (C) and binding isotherms (D) were used to
calculate Kd values for the interaction with PtdIns(4)P-MP, 8b and PtdIns(4)P-
PT, 15b. Because FAPP1 PH domain-PI interaction is dependent on the ion
strength of the buffer and pH (He et al., 2011), for proper comparison, binding
to the unmodified and modified lipid and IP compounds throughout the study
was investigated under similar conditions (described in detail in Experimental
Procedures).
the 4-position through the reaction with bis(2-cyanoethoxy)- site of FAPP1 PH through resonance perturbation analysis. The
(diisopropylamino)-phosphine in the presence of 1H-tetrazole, largest changes were observed for the K7, W8, Q16, F20, S28,
followed by treating with phenylacetyl disulfide. Removal of the Y29, G42-A47, C49, L63-I65, E68, Q69, and F71 residues of
TES with the weakly acidic reagent NH4F in methanol gave the the PH domain. These residues formed a contiguous patch on
advanced intermediate 13, which was then coupled with phos- the PH domain surface, outlining the pocket where C8-
phoramidites 6a and 6b as above to produce the protected PtdIns(4)P-PT is bound (Figure 3C). The recently determined
phospholipids 14a and 14b. Removal of the protective groups structure of this domain shows that it folds into a seven-stranded
yielded C8-PtdIns(4)-PT (15a) and C16-PtdIns(4)-PT (15b).
b-barrel capped by an a helix at one edge with the opposite
edge, formed by the b4 and b7 strands and three loops, remain-
ing open (He et al., 2011; Lenoir et al., 2010). Many residues
located in the b4 and b7 strands and some residues of the b1,
PtdIns(4)P-MP and PtdIns(4)P-PT Analogs Are Potent
Ligands of the FAPP1 PH Domain
To determine whether the ms analogs of PtdIns(4)P are recog- b2, b3, and b6 strands were the most perturbed due to binding
nized by a protein effector, we tested the lipids by liposome- to C8-PtdIns(4)P-PT, implying that the open edge of the b-barrel
binding assays using the PH domain of FAPP1 that was shown comprises the binding site for the lipid analog.
1314 Chemistry & Biology 18, 1312–1319, October 28, 2011 ª2011 Elsevier Ltd All rights reserved

Chemistry & Biology
Synthesis and Applications of PI4P Analogs
Figure 3. The PtdIns(4)P-PT-Binding Site of the
FAPP1 PH Domain
(A, D, and F) Superimposed 1H,15N HSQC spectra of
15N-labeled FAPP1 PH collected as C8-PtdIns(4)P-PT
(C8-PI4P-PT), 15a (A) C8-PtdIns(4)P (D), or C4-PtdIns(4)P
(F) was titrated in. The spectra are color coded according
to the concentration of the lipid.
(B and E) The histograms show normalized chemical shift
changes induced in the backbone amides of the PH
domain by C8-PtdIns(4)P-PT, 15a and C8-PtdIns(4)P.
(C) Residues that display significant chemical shift change
in (B) are labeled on the FAPP1 PH domain surface and
colored red, orange, and yellow for large, medium, and
small changes, respectively.
(G) Differences in NMR resonance perturbations in the
FAPP1 PH domain upon binding to C8-PtdIns(4)P-PT, 15a
and C8-PtdIns(4)P.
(H) Overlay of the crystal structure of the ligand-free
FAPP1 PH domain (3RCP) with that of the PtdIns(3,4,5)P3-
bound GRP1 PH domain (1FGY). For clarity, in the GRP1
complex, only PtdIns(3,4,5)P3 is shown as
a stick
model and colored yellow. The 4-phosphate group of
PtdIns(3,4,5)P3 is in purple.
weaker, indicating that the hydrophobic con-
tacts involving the acyl chains of PtdIns(4)P
are essential (Figure 2B).
Because the PT analog contains a PT group
in place of a phosphate, we sought to identify
the binding pocket for the 4-phosphate group
in the complex. Despite numerous attempts
to cocrystallize the FAPP1 PH domain bound
to either unmodified PtdIns(4)P or the lipid
analogs, we were unable to obtain crystals of
the complex. We note that currently no single
structure of any protein bound to PtdIns(4)P is
available, and therefore, how this lipid is coordi-
nated by an effector remains unknown. A careful
comparison of patterns of the chemical shift
changes induced by the unmodified lipid and
the analog provided some details on the 4-
phosphate group position. Residues located
on the side of the b-barrel that is further away
from the long b1-b2 loop were perturbed to
a greater extent by the analog (Figures 3G and
3H). This suggests that the 4-phosphate group
is positioned in the interior of the b-barrel and
oriented away from the b1-b2 loop, in a manner
similar to that of how a trisphosphorylated PI,
We compared chemical shift changes induced in the PH PtdIns(3,4,5)P3, is bound by the PH domain of GRP1 (Ferguson
domain upon addition of the PT analog and unmodified C8- et al., 2000; Lietzke et al., 2000).
PtdIns(4)P. Generally, similar in directions and the magnitude
resonance perturbations (Figures 3A and 3D) and comparable Total Synthesis of the IP2-Aminohexyl Probe
binding affinities (Figure 2B) suggested that the PT group does Another limitation of the use of natural and/or unmodified lipids is
not compromise this interaction, and the binding mode is the difficulty in visualizing and tracing them in biological studies.
conserved. In contrast, association of the FAPP1 PH domain We designed and tested a hybrid probe of the PtdIns(4)P
with a short-chain, dibutanoyl (C4)-PtdIns(4)P lipid was 9-fold head group that carries an aminoalkyl chain in place of the
Chemistry & Biology 18, 1312–1319, October 28, 2011 ª2011 Elsevier Ltd All rights reserved 1315

Chemistry & Biology
Synthesis and Applications of PI4P Analogs
Figure 4. Synthesis of the IP2-AH, 23 Probe
The reagents and conditions are: (K) BOMCl, DIEA, DCE, reflux, 24 hr, 57%; (L) DIBAL-H, CH2Cl2, ꢂ78ꢁC, 1.5 hr, 91%; (M) (BnO)2PN(Ipr)2, 1H-tetrazole, CH2Cl2,
CH3CN, room temperature, 18 hr, then m-CPBA, ꢂ60ꢁC, 1 hr, 91%; (N)TBAF, DMF, room temperature, 12 hr, 86%; (O) 21, 1H-tetrazole, CH2Cl2, CH3CN, room
temperature, 18 hr, then m-CPBA, ꢂ60ꢁC, 1 hr, 73%; (P) H2, Pd(OH) 2, room temperature, 3 days, 100%.
diacylglycerol moiety of the lipid and is water soluble. This probe PtdIns(4)P-MP, PtdIns(4)P-PT, and IP2-AH Retain
can be further modified through the coupling to afluorescent, bio- Activities of the Natural Lipid
tinylated, or photoaffinity tag, which is necessary for the detec- The b1-b2 loop of the FAPP1 PH domain was shown to insert into
tion and characterization of the PtdIns(4)P-related processes. PtdIns(4)P-containing membranes and membrane-mimetics
The soluble aminohexyl (AH) IP2 probe (23) was synthesized as causing deformation of the lipid layer (He et al., 2011; Lenoir
shown in Figure 4 (Gong et al., 2009). To protect the 4-position, et al., 2010). To determine whether the biological activities are
intermediate 16 was first produced in four steps from myo- preserved for the PtdIns(4)P derivatives, we tested C16
-
inositol using the camphor resolution protocol (Kubiak and Bru- PtdIns(4)P-MP, C16-PtdIns(4)P-PT, and IP2-AH in monolayer
zik, 2003). The other four hydroxyl groups were subsequently penetration and membrane tubulation assays. As shown in Fig-
protected with benzyloxymethyl (BOM) moieties to produce 17. ure 6, the FAPP1 PH domain was unable to significantly pene-
The benzoyl group at the 4-position was then removed to yield trate a POPC/POPE (80:20) monolayer in the absence of
18, and phosphotriester 19 was generated via phosphoramidite PtdIns(4)P (surface pressure [pc] was ꢀ24 mN/m). However,
chemistry. Deprotection of the silyl group at the 1-position to 20 when 5% C16-PtdIns(4)P-MP or C16-PtdIns(4)P-PT was incorpo-
followed by coupling with phosphoramidite reagent 21 yielded rated in the lipid monolayer (or IP2-AH was injected together with
fully protected intermediate 22, which was quantitatively depro- FAPP1), the pc values increased to ꢀ33, 34, and 32 mN/m,
tected via hydrogenolysis to produce IP2-AH (23).
respectively, indicating that binding to the PtdIns(4)P derivatives
induces robust insertion of the PH domain into the monolayer.
The strong dependence of insertion on the association with the
The Aminoalkyl Group in IP2-AH Does Not Alter Binding
We examined whether the binding properties of Ins(1,4)P2 are analogs was substantiated by the fact that the R18A mutant of
affected by the presence of the AH group in the IP2-AH probe. FAPP1 PH that lost its ability to bind C16-PtdIns(4)P-MP and
We recorded 1H,15N HSQC spectra of the 15N-labeled FAPP1 C16-PtdIns(4)P-PT in SPR experiments (Figure 2B) was also
PH domain while titrating IP2-AH or unmodified Ins(1,4)P2 into incapable of penetration (Figure 6). Because the incorporation
the NMR sample (Figure 5). A similar pattern of chemical shift of 5% C16-PtdIns(4)P increases the pc value to 33 mN/m (He
changes upon addition of either compound demonstrated that et al., 2011), we concluded that the ms analogs and the IP2 probe
IP2-AH and Ins(1,4)P2 are bound in the same way. The Kd values, do not compromise the penetrating ability of the FAPP1 effector.
measured by NMR, revealed that the FAPP1 PH domain equally
associates with IP2-AH and Ins(1,4)P2, exhibiting a 20 mM affinity membrane tubulation (Figure 7). The POPC/POPE (80:20),
(Figure 5G). Thus, the aminoalkyl moiety of IP2-AH does not have POPC/POPE/C16-PtdIns(4)P (75:20:5), POPC/POPE/C16
We next explored if the derivatives of PtdIns(4)P stimulate
-
a significant effect on binding, and this probe can effectively PtdIns(4)P-MP (75:20:5), and POPC/POPE/C16-PtdIns(4)P-PT
substitute for Ins(1,4)P2. The low affinity of the PH domain to (75:20:5) membrane sheets were generated and treated with
the unmodified or modified inositol head group in comparison the lipophilic dye FM 2-10 to facilitate imaging using a confocal
with the affinity to the intact PtdIns(4)P lipid indicated that a size- microscope. In agreement with previous observations, injecting
able hydrophobic component is necessary for strong interaction. FAPP1 PH to the POPC/POPE sheets did not cause detectable
We note that although the observed resonance perturbations changes in the bilayer morphology (data not shown). In contrast,
were generally less pronounced than those seen due to binding addition of the protein to the C16-PtdIns(4)P-, C16-PtdIns(4)P-
to PtdIns(4)P, the binding site for an isolated head group of the MP-, or C16-PtdIns(4)P-PT-containing sheets rapidly induced
lipid remains unchanged and is positioned at the open edge of tubulation of the membrane. Tubulation was also induced
the b-barrel (Figures 5C and 5F).
by injecting the PH domain preincubated with IP2-AH. The
1316 Chemistry & Biology 18, 1312–1319, October 28, 2011 ª2011 Elsevier Ltd All rights reserved

Chemistry & Biology
Synthesis and Applications of PI4P Analogs
Figure 5. The IP2-AH-Binding Site of the FAPP1 PH Domain
(A and D) Superimposed 1H,15N HSQC spectra of 15N-labeled FAPP1 PH collected during titration with IP2-AH, 23 (A) or IP2 (D). The spectra are color coded
according to the concentration of the ligand (inset).
(B and E) The histograms show normalized chemical shift changes induced in the backbone amides of the PH domain by IP2-AH, 23 (B) or IP2 (E).
(C and F) Residues that display significant chemical shift change in (B) and (E) are labeled on the FAPP1 PH domain surface and colored red, orange, and yellow
for large, medium, and small changes, respectively.
(G) Binding affinities of FAPP1 PH measured by NMR.
deformation of all tested membrane sheets was abolished when PtdIns(4)P-mimicking compounds could be instrumental
the R18A mutant of FAPP1 PH impaired in PtdIns(4)P binding for examining the cell processes occurring within the TGN
was used, underscoring the critical role of the PH-PtdIns(4)P and identifying protein effectors specific for PtdIns(4)P.
interaction for the biological activities of FAPP1.
Furthermore, the ms analogs are resistant to hydrolysis
and, thus, can be used as a tool to distinguish between the
signaling pathways involving activity of PI 4-phosphateses
and the cellular processes associated with binding of
SIGNIFICANCE
In this study we report the synthesis and biochemical and protein effectors to PtdIns(4)P.
functional characterization of derivatives of the PtdIns(4)P
lipid. We demonstrate that metabolically stabilized analogs
EXPERIMENTAL PROCEDURES
of PtdIns(4)P and the inositol 1,4-bisphosphate head group
retain the natural lipid binding and membrane tubulation
Synthesis of MP and PT Analogs of PtdIns(4)P
properties and, therefore, can replace PtdIns(4)P in bio-
logical and biochemical applications. These synthetic
The complete experimental details for the synthesis and characterization of
PtdIns(4)P-MP 8a (R = C7H15) and 8b (R = C15H31) and PtdIns(4)P-PT 15a
Figure 6. PtdIns(4)P-MP, 8b, PtdIns(4)P-PT, 15b, and IP2-AH, 23, Induce Membrane Penetration by the FAPP1 PH Domain
(A and B) Insertion of the FAPP1 PH domain into a POPC/POPE (80:20) monolayer (open circles), a POPC/POPE/PtdIns(4)P-MP (75:20:5) monolayer (open
squares), and a POPC/POPE/PtdIns(4)P-PT (75:20:5) monolayer (open triangles).
(C) Insertion of FAPP1 PH, preincubated with IP2-AH, into a POPC/POPE monolayer (open squares). The data on R18A FAPP1 PH are shown as filled squares and
triangles in (A)–(C). All measurements were performed using the subphase 10 mM HEPES (pH 7.4) containing 160 mM KCl.
Chemistry & Biology 18, 1312–1319, October 28, 2011 ª2011 Elsevier Ltd All rights reserved 1317

Chemistry & Biology
Synthesis and Applications of PI4P Analogs
Liposome-Binding Assay
The liposome-binding assays were performed as described (He et al., 2009).
Briefly, solutions of POPC, POPE, and POPS, containing either C16-PtdIns(4)P,
C16-PtdIns(4)P-MP, C16-PtdIns(4)P-PT (50:20:15:15), or no PI, were dissolved
in CHCl3:MeOH:H2O (65:25:4) and dried under vacuum. The lipids were resus-
pended in 800 ml of 20 mM MOPS, 100 mM KCl (pH 7.0), and passed 19 times
through an extruder with a 1 mm membrane. Liposomes were collected
by centrifugation at 25,000 3 g for 20 min and finally resuspended in 95 ml
of buffer by vortexing. Liposomes were incubated with 10 ml of 1.7 mg/ml
GST-FAPP1 PH domain for 1 hr at room temperature and then collected again
by centrifugation. The liposome pellets were separated from supernatant and
resuspended in 100 ml of buffer. The pelleted and supernatant fractions were
analyzed by SDS-PAGE.
SPR Measurements
The SPR experiments were carried out at 25ꢁC in 10 mM HEPES (pH 7.4)
containing 160 mM KCl (He et al., 2008; Hom et al., 2007). POPC/POPE/
C16-PtdIns(4)P (75:20:5), POPC/POPE/C16-PtdIns(4)P-MP (75:20:5), POPC/
POPE/C16-PtdIns(4)P-PT (75:20:5), and POPC/POPE (80:20) vesicles were
spread at 5 ml/min over the active and control surfaces until 6000 resonance
unit (RU) response was achieved. Equilibrium SPR measurements were per-
formed by injecting wild-type or mutated FAPP1 PH at a flow rate of 5 ml/min
to provide sufficient time for R values of the association phase to reach
equilibrium (Req). Sensorgrams were obtained using five or more different
concentrations of the protein (within a 10-fold range of Kd values) and cor-
rected for the refractive index change by subtracting the control surface
response. The Req values were plotted versus the total protein concentration
(P0), and Kd values were determined by a nonlinear least-squares analysis of
the binding isotherms using the equation: Req = Rmax/(1 + Kd/P0). The experi-
ments were performed in triplicate for C16-PtdIns(4)P and C16-PtdIns(4)P-MP,
and in duplicate for C16-PtdIns(4)P-PT.
NMR Spectroscopy
The 1H,15N HSQC spectra of 0.1–0.2 mM 15N-labeled FAPP1 PH domain were
recorded at 25ꢁC on a Varian INOVA 600 MHz spectrometer. Lipid and
Ins(1,4)P2 binding was characterized by monitoring chemical shift changes
in the 1H,15N HSQC spectra of the FAPP1 PH domain in 25 mM Bis-Tris,
150 mM NaCl (pH 6.5) as C4-PtdIns(4)P, C8-PtdIns(4)P, C8-PtdIns(4)P-PT,
Ins(1,4)P2, and IP2-AH were added stepwise. The normalized chemical shift
changes were calculated using the equation [(DdH)2+(DdN/5)2]0.5, where DdH
and DdN are 1H and 15N chemical shift changes in parts per million (ppm).
Significant changes in the resonances were judged to be greater than the
average plus 0.5, 0.8, 0.8, and 1.0 standard deviation for the titration of
C8-PtdIns(4)P-PT, C8-PtdIns(4)P, Ins(1,4)P2, and Ins(1,4)P2-AH, respectively.
Spectra were processed with NMRPipe (Delaglio et al., 1995) and analyzed
using CCPN (Vranken et al., 2005) and nmrDraw. The Kd values were calcu-
lated by a nonlinear least-squares analysis using the equation:
Figure 7. Binding of the FAPP1 PH Domain to PtdIns(4)P-MP, 8b,
PtdIns(4)P-PT, 15b, and IP2-AH, 23, Stimulates Membrane Tubu-
lation
Membrane sheets, POPC/POPE/PtdIns(4)P (75:20:5) (A), POPC/POPE/
PtdIns(4)P-MP (75:20:5) (B), POPC/POPE/PtdIns(4)P-PT (75:20:5) (C), and
POPC/POPE (80:20) in the presence of IP2-AH (D), labeled with FM 2-10 dye
are shown before or after addition of 1 mg/ml FAPP1 PH or the R18A mutant in
10 mM HEPES (pH 7.4) containing 160 mM KCl. Images shown were taken
after 5 min incubation with the protein on a Zeiss LSM 710 confocal micro-
scope using a 633 oil objective. Scale bars, 50 mM.
(R = C7H15) and 15b (R = C15H31) are provided in the Supplemental Experi-
mental Procedures available online.
ꢀ
ꢁ
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
Ddmax ð½Lꢃ + ½Pꢃ + KdÞ ꢂ ð½Lꢃ + ½Pꢃ + KdÞ ꢂ4½Pꢃ½Lꢃ
2½Pꢃ
Protein Expression and Purification
Dd =
A DNA fragment encoding human FAPP1 PH (aa 1–99) was cloned into a pET-
28a vector (He et al., 2011). The unlabeled and 15N-labeled wild-type and
mutated proteins wereexpressedin E. coliRosetta (DE3) pLysSinLB or minimal
media supplemented with 15NH4Cl. Bacteria were harvested by centrifugation
after induction with isopropyl-1-thio-b-D-galactopyranoside (IPTG) (0.1 mM)
at room temperature for 6 hr. The cells were lysed by sonication in lysis buffer
(50mMHEPES[pH7.6], 300mMNaCl,5 mMb-mercaptoethanol, 10%glycerol,
and a protease inhibitor cocktail). The His-tagged proteins were purified on
a Talon resin column, and the His tag was cleaved with Thrombin. The proteins
were further purified by size exclusion chromatography on a Superdexꢀ 75
column in either Bis-Tris or HEPES and concentrated in Millipore concentrators.
The GST-FAPP1 PH domain was expressed in E. coli BL21(DE3) cells and
purified on a glutathione Sepharose column as described (He et al., 2011).
where [L] is concentration of PtdIns(4)P or IP2, [P] is concentration of the PH
domain, Dd is the observed chemical shift change, and Ddmax is the normalized
chemical shift change at saturation.
Monolayer Measurements
The insertion of FAPP1 PH into a phospholipid monolayer was investigated by
measuring the change in surface pressure (p) of the invariable surface area
upon addition of the protein (He et al., 2008; Lee et al., 2006). A lipid monolayer
containing various combinations of phospholipids was spread onto the
subphase composed of 10 mM HEPES, 160 mM KCl (pH 7.4) until the desired
initial surface pressure (p0) was reached. After stabilization of the signal
(ꢀ5 min), 10 mg of FAPP1 PH was injected into the subphase. The surface pres-
sure change Dp was examined for 15 min. To test IP2-AH, Dp of a POPC:POPE
monolayer was monitored as a mixture of FAPP1-PH preincubated for 15 min
with an equimolar amount of IP2-AH was gradually added.
PCR Mutagenesis
Site-directed mutagenesis of the FAPP1 PH domain was performed using
a QuikChange Site-Directed Mutagenesis Kit (Stratagene). The sequence of
the R18A mutant was confirmed by DNA sequencing.
1318 Chemistry & Biology 18, 1312–1319, October 28, 2011 ª2011 Elsevier Ltd All rights reserved

Chemistry & Biology
Synthesis and Applications of PI4P Analogs
Membrane Tubulation Assays
Gong, D., Bostic, H.E., Smith, M.D., and Best, M.D. (2009). Synthesis of
modular headgroup conjugates corresponding to all seven phosphatidylinosi-
tol polyphosphate isomers for convenient probe generation. European J. Org.
Chem. 4170–4179.
Glass coverslips (22 3 40 mm) were cleaned by sonication in 1% 73 (MP
Biomedicals). After vigorous rinses and sonication in distilled water to remove
any trace of detergent, coverslips were washed with 100% ethanol and dried
under N2. To generate membrane sheets, 1 ml lipid solution in chloroform
(10 mg/ml) was spotted on each coverslip and dried under N2 for 30 min to re-
Hamilton, C.J., and Roberts, S.M. (1999). Synthesis of fluorinated phospho-
noacetate derivatives of carbocyclic nucleoside monophosphonates and
activity as inhibitors of HIV reverse transcriptase. J. Chem. Soc., Perkin
Trans. 1 1051–1056.
move traces of chloroform. Lipids (POPC:POPE [80:20], POPC/POPE/C16
-
PtdIns(4)P [75:20:5], POPC/POPE/C16-PtdIns(4)P-MP [75:20:5], or POPC/
POPE/C16-PtdIns(4)P-PT [75:20:5]) were prehydrated for 20–30 min in a small
chamber and then fully rehydrated by adding 20 ml of buffer (10 mM HEPES [pH
7.4], 160 mM KCl, 10 mM FM 2-10). With the chamber mounted on a Zeiss
LSM710 microscope stage, 10 ml of protein solution (1 mg/ml) was injected
into the chamber. The deformation of membrane sheets into tubules was de-
tected using laser excitation at 488 nm and monitoring emission above
510 nm. To test IP2-AH, it was first preincubated for 15 min with FAPP1-PH
at an equimolar concentration, and then the mixture was injected to a POPC:
POPE membrane sheet.
He, J., Haney, R.M., Vora, M., Verkhusha, V.V., Stahelin, R.V., and
Kutateladze, T.G. (2008). Molecular mechanism of membrane targeting by
the GRP1 PH domain. J. Lipid Res. 49, 1807–1815.
He, J., Scott, J.L., Heroux, A., Roy, S., Lenoir, M., Overduin, M., Stahelin, R.V.,
and Kutateladze, T.G. (2011). Molecular basis of phosphatidylinositol 4-phos-
phate and ARF1 recognition by the FAPP1 PH domain. J. Biol. Chem. 286,
18650–18657.
He, J., Vora, M., Haney, R.M., Filonov, G.S., Musselman, C.A., Burd, C.G.,
Kutateladze, A.G., Verkhusha, V.V., Stahelin, R.V., and Kutateladze, T.G.
(2009). Membrane insertion of the FYVE domain is modulated by pH.
Proteins 76, 852–860.
SUPPLEMENTAL INFORMATION
Hom, R.A., Vora, M., Regner, M., Subach, O.M., Cho, W., Verkhusha, V.V.,
Stahelin, R.V., and Kutateladze, T.G. (2007). pH-dependent binding of the
Epsin ENTH domain and the AP180 ANTH domain to PI(4,5)P2-containing
bilayers. J. Mol. Biol. 373, 412–423.
Supplemental Information includes Supplemental Experimental Procedures
and can be found with thisarticleonline at doi:10.1016/j.chembiol.2011.07.022.
ACKNOWLEDGMENTS
Huang, W., Zhang, H., Davrazou, F., Kutateladze, T.G., Shi, X., Gozani, O., and
Prestwich, G.D. (2007). Stabilized phosphatidylinositol-5-phosphate
analogues as ligands for the nuclear protein ING2: chemistry, biology, and
molecular modeling. J. Am. Chem. Soc. 129, 6498–6506.
This research was supported by grants from the National Institutes of Health
(to G.D.P. and T.G.K.), the American Heart Association (to R.V.S. and
T.G.K.), and the National Science Foundation (to M.D.B.). J.L.S. is an NIH
training fellow, and T.G.K. is an Independent NARSAD Investigator.
Kubiak, R.J., and Bruzik, K.S. (2003). Comprehensive and uniform synthesis of
all naturally occurring phosphorylated phosphatidylinositols. J. Org. Chem. 68,
960–968.
Received: May 17, 2011
Revised: July 7, 2011
Lee, S.A., Kovacs, J., Stahelin, R.V., Cheever, M.L., Overduin, M., Setty, T.G.,
Burd, C.G., Cho, W., and Kutateladze, T.G. (2006). Molecular mechanism of
membrane docking by the Vam7p PX domain. J. Biol. Chem. 281, 37091–
37101.
Accepted: July 8, 2011
Published: October 27, 2011
Lemmon, M.A. (2008). Membrane recognition by phospholipid-binding
REFERENCES
domains. Nat. Rev. Mol. Cell Biol. 9, 99–111.
D’Angelo, G., Vicinanza, M., Di Campli, A., and De Matteis, M.A. (2008). The
multiple roles of PtdIns(4)P—not just the precursor of PtdIns(4,5)P2. J. Cell
Sci. 121, 1955–1963.
Lenoir, M., Coskun, U., Grzybek, M., Cao, X., Buschhorn, S.B., James, J.,
Simons, K., and Overduin, M. (2010). Structural basis of wedging the Golgi
membrane by FAPP pleckstrin homology domains. EMBO Rep. 11, 279–284.
Delaglio, F., Grzesiek, S., Vuister, G.W., Zhu, G., Pfeifer, J., and Bax, A. (1995).
NMRPipe: a multidimensional spectral processing system based on UNIX
pipes. J. Biomol. NMR 6, 277–293.
Lietzke, S.E., Bose, S., Cronin, T., Klarlund, J., Chawla, A., Czech, M.P., and
Lambright, D.G. (2000). Structural basis of 3-phosphoinositide recognition
by pleckstrin homology domains. Mol. Cell 6, 385–394.
Dowler, S., Currie, R.A., Campbell, D.G., Deak, M., Kular, G., Downes, C.P.,
and Alessi, D.R. (2000). Identification of pleckstrin-homology-domain-contain-
ing proteins with novel phosphoinositide-binding specificities. Biochem. J.
351, 19–31.
Martin, T.F.J. (1998). Phosphoinositide lipids as signaling molecules: common
themes for signal transduction, cytoskeletal regulation, and membrane traf-
ficking. Annu. Rev. Cell Dev. Biol. 14, 231–264.
Minutolo, F., Antonello, M., Barontini, S., Bertini, S., Betti, L., Danesi, R.,
Gervasi, G., Giannaccini, G., Papi, C., Placanica, G., et al. (2004).
Phosphonomethylphosphorylmethyl(oxy)-analogues of geranylgeranyl di-
phosphate as stable and selective geranylgeranyl protein transferase inhibi-
tors. Farmaco 59, 887–892.
Ferguson, K.M., Kavran, J.M., Sankaran, V.G., Fournier, E., Isakoff, S.J.,
Skolnik, E.Y., and Lemmon, M.A. (2000). Structural basis for discrimination
of 3-phosphoinositides by pleckstrin homology domains. Mol. Cell 6, 373–384.
Gajewiak, J., Xu, Y., Lee, S.A., Kutateladze, T.G., and Prestwich, G.D. (2006).
Synthesis and molecular recognition of phosphatidylinositol-3-methylene-
phosphate. Org. Lett. 8, 2811–2813.
Phillion, D.P., and Andrew, S.S. (1986). Synthesis and reactivity of diethyl
phosphonomethyltriflate. Tetrahedron Lett. 27, 1477–1480.
Godi, A., Di Campli, A., Konstantakopoulos, A., Di Tullio, G., Alessi, D.R., Kular,
G.S., Daniele, T., Marra, P., Lucocq, J.M., and De Matteis, M.A. (2004). FAPPs
control Golgi-to-cell-surface membrane traffic by binding to ARF and
PtdIns(4)P. Nat. Cell Biol. 6, 393–404.
Vranken, W.F., Boucher, W., Stevens, T.J., Fogh, R.H., Pajon, A., Llinas, M.,
Ulrich, E.L., Markley, J.L., Ionides, J., and Laue, E.D. (2005). The CCPN data
model for NMR spectroscopy: development of a software pipeline. Proteins
59, 687–696.
Chemistry & Biology 18, 1312–1319, October 28, 2011 ª2011 Elsevier Ltd All rights reserved 1319