Preparation and use of a photoactivatable glucose-6-phosphate analogue

Article abstract of DOI:10.1016/S0960-894X(00)00043-3

A benzophenone-containing derivative of glucose-6-phosphate, 6-[(3-([2,3-³H₂]-p-benzoyl dihydrocinnamidoylpropyl-1-oxy)phosphoryl]-D-glucopyranose ([³H]BZDC-Glc-6-P) was synthesized and employed to photoaffinity label proteins on intact rat liver microsomes. The use of a non-photoactivatable, UV-transparent desoxy analogue of BZDC, named p-benzyldihydrocinnamoyl (BnDC), is introduced as a general method to achieve competition when hydrophilic ligands are modified with hydrophobic photophores. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00043-3

Bioorganic & Medicinal Chemistry Letters 10 (2000) 535±539
Preparation and Use of a Photoactivatable Glucose-6-Phosphate
Analogue
Jirong Peng, a Pei Yan Chen, b Richard B. Marchase b and Glenn D. Prestwich a,*
aDepartment of Medicinal Chemistry, The University of Utah, 30 South 2000 East, Room 201, Salt Lake City, Utah 84112-5820, USA
bDepartment of Cell Biology, University of Alabama at Birmingham, Birmingham, Alabama 35294, USA
Received 6 December 1999; accepted 12 January 2000
AbstractÐA benzophenone-containing derivative of glucose-6-phosphate, 6-[(3-([2,3-3H2]-p-benzoyl dihydrocinnamidoylpropyl-1-
oxy)phosphoryl]-d-glucopyranose ([3H]BZDC-Glc-6-P) was synthesized and employed to photoanity label proteins on intact rat
liver microsomes. The use of a non-photoactivatable, UV-transparent desoxy analogue of BZDC, named p-benzyldihydro-
cinnamoyl (BnDC), is introduced as a general method to achieve competition when hydrophilic ligands are modi®ed with hydro-
phobic photophores. # 2000 Elsevier Science Ltd. All rights reserved.
The homeostatic regulation of blood glucose levels is in
part dependent upon the release of glucose from the
liver.1 Surprisingly, hepatic glucose production, derived
from either gluconeogenesis or glycogenolysis, requires
that glucose-6-phosphate (Glc-6-P) be imported into the
lumen of the endoplasmic reticulum (ER) of hepato-
cytes and cleaved there by a glucose-6-phosphatase.2
The glucose units are then re-released to the cytoplasm
and ®nally exported across the hepatocyte plasma
membrane to the bloodstream. This transformation of
Glc-6-P to glucose is thus dependent upon a glucose-6-
phosphatase with an intralumenal active site and a
Glc-6-P transporter, which is responsible for the trans-
location of Glc-6-P from the cytoplasm to the lumen of
the ER where cleavage takes place.
with GSD 1b, mutations in this protein were detected.
Furthermore, linkage analysis mapped the GSD 1b
locus5 and the putative transporter6 to human chromo-
some 11q23. Subsequent work has substantiated the
localization of the glucose-6-phosphatase active site to
the lumen of the ER7 and established the functional
signi®cance of the Glc-6-P transporter.8
However, the tissue distribution of the Glc-6-P trans-
porter is broader than the distribution of glucose-6-
phosphatase.6 Interestingly, subjects with GSD 1b show
de®ciencies in neutrophil behavior9, perhaps related to
defects in Ca2+ signaling.10,11 In addition, the import of
Glc-6-P into microsomes from both liver and other
tissue has been demonstrated and shown to be accom-
panied by an increase in levels of sequestered Ca2+ 12
.
Until recently, this multi-component substrate-trans-
port model was controversial. However, recent data and
analyses of a family of human genetic disorders have
substantiated its validity. Mutations in the glucose-6-
phosphatase gene3 were found in glycogen storage dis-
ease (GSD)-type 1a patients, but not in a closely related
form of the disease (GSD 1b), in which transport
appears to be defective but normal phosphatase activity
is seen when the ER membrane is permeabilized. Com-
plementing this ®nding, a human cDNA that encodes a
46 kDa transmembrane protein homologous to bacter-
ial Glc-6-P transporters was identi®ed.4 In two patients
Therefore, it appears that the Glc-6-P transporter could
have signi®cance beyond its role in glucose homeostasis.
One approach to the further study of the Glc-6-P
transporter is to utilize radioactive photoactivatable
probes to selectively label it. Photoactivatable deriva-
tives of chlorogenic acid, a highly-speci®c competitive
inhibitor of Glc-6-phosphatase activity in intact, but not
disrupted, rat and human microsomes, were used to
label a high-anity binding site in intact rat liver
microsomes.13 We describe herein the synthesis of a
tritium-labeled p-benzoyldihydrocinnamoyl (BZDC)
derivative14 of Glc-6-P. The benzophenone photophore
probes o€er signi®cant advantages for photoanity
labeling in biological systems,15 particularly in terms of
synthesis, handling, and eciency of covalent modi®cation
*Corresponding author. Tel.: +1-801-585-9051; fax: +1-801-585-
9053; e-mail: gprestwich@deans.pharm.utah.edu
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00043-3

536
J. Peng et al. / Bioorg. Med. Chem. Lett. 10 (2000) 535±539
of the target proteins.16 In addition, we introduce the
utilization of the novel desoxy p-benzyldihydrocinna-
moyl (BnDC) group as a non-activatable, UV-trans-
parent mimic of the BZDC photophore. The use of
these probes with rat liver microsomes, however, pro-
duced an unexpected result. Rather than preferentially
labeling an integral membrane protein with the char-
acteristics of the Glc-6-P transporter, two proteins of
molecular sizes 34 and 40 kDa were labeled with the
[3H]BZDC Glc-6-P probe. These proteins were dis-
lodged from ER microsomes by treatment with salt and
localized to the aqueous portion of a phase separation
with Triton X-114. Thus, [3H]BZDC Glc-6-P appears to
selectively modify two peripheral proteins of unknown
function associated with the ER membranes, but fails to
covalently modify the Glc-6-P transporter.
photoinert Glc-6-P probes. Coupling 5 with [3H]BZDC-
NHS (6) gave [3H]BZDC-Glc-6-P (1) and reaction of
5 with BnDC-NHS reagent 8 gave the non-photo-
activatable BnDC-Glc-6-P (2).
Photoanity Labeling of Intact Microsomal Proteins
with [3H]BZDC-Glc-6-P
Microsomes were prepared from rat liver, and then
fractionated by density gradient centrifugation to enrich
for microsomes from the smooth endoplasmic reticulum
(SER), rough endoplasmic reticulum (RER), and the
Golgi apparatus.18 A fraction enriched in mitochondria
was also prepared. Each fraction was incubated with
[3H]BZDC Glc-6-P, photoanity labeled, separated by
SDS-PAGE, and processed for ¯uorography. Cooma-
sie-blue-stained gels demonstrated that the pattern of
major proteins found in each of the three organellar-
speci®c microsomal preparations were distinct and dif-
ferent from that of the mitochondrial fraction (Fig. 2).
The ¯uorograms (Fig. 2B) showed labeling primarily of
proteins at 34 and 40 kDa. The labeling of each of the
two bands was enhanced in preparations enriched for
SER (lane 2) and Golgi (lane 4). The bands were not
seen in fractions enriched for the RER or mitochondrial
fractions (lanes 3 and 5).
Synthesis of BZDC-Glc-6-P (1) and its
Non-Photoactivatable Analogue BnDC-Glc-6-P (2)
(Fig. 1)
The unsaturated succinimido-p-benzoylcinnamate (7),
previously prepared by DCC coupling of p-benzoyl-
cinnamic acid with N-hydroxysuccinimide,14 was more
conveniently synthesized using diphenyl succinimidyl
phosphate.17 Catalytic hydrogenation (10% Pd/C in
EtOAc) of 7 reduced the double bond and also con-
verted the benzophenone carbonyl group into a meth-
ylene group, giving succinimido-p-benzyl-dihydrocin-
namate (BnDC) (8) in 69% yield. Benzyl 2,3,4-tri-O-
benzyl-a-d-arabino-hexopyranoside (3) was prepared by
sequential treatment of d-glucose with benzyl alcohol
and benzaldehyde followed by benzylation and cleavage
of the benzylidene acetal with lithium aluminum
hydride-aluminum chloride. Intermediate 3 could be
obtained as white crystals from d-glucose with an over-
all yield of ca. 5%. Reaction of 3 with benzyloxy-[N-
Cbz-3-amino-1-propyloxy](N,N-diisopropylamino)phos-
phine and 1H-tetrazole in CH2Cl2, followed by oxidation
with mCPBA, a€orded fully protected intermediate 4.
Hydrogenolytic deprotection of 4 gave precursor 5,
which was employed to make the photoactivatable and
One approach to determine whether this labeling was
dependent upon the Glc-6-P portion of the probe,
labeling with [3H]BZDC Glc-6-P was compared to that
seen with another BZDC coupled probe, [3H]BZDC-
inositol (1,4,5) trisphosphate (IP3).19±22 The [3H]BZDC-
IP3 probe selectively labeled a 60-kDa band, while the
Glc-6-P analogue again labeled predominantly bands at
34 and 40 kDa (Fig. 2C). Each of these experiments
used 800 mg of protein and 0.5 mCi of radioligand with a
60 min irradiation time. The speci®city of photolabeling
by the [3H]BZDC Glc-6-P probe was determined by
competitive displacement experiments using 100 mM
Glc-6-P, unlabeled BZDC Glc-6-P, or the photoinert
BnDC Glc-6-P (2). The presence of 1 mM of either
unlabeled 1 or 2 completely displaced the photocovalent
Figure 1. Synthesis of [3H]BZDC-Glc-6-P (1) and BnDC-Glc-6-P (2).

J. Peng et al. / Bioorg. Med. Chem. Lett. 10 (2000) 535±539
537
Figure 2. [3H]BZDC Glc-6-P labeling of rat liver subcellular fractions. Panel A: Coomassie-blue staining of 10% SDS-PAGE of 15 mg protein per
lane: lane 1, total microsomes; lane 2, SER; lane 3, RER; lane 4, Golgi; lane 5, mitochondria. Panel B: Fluorography of photoanity labeling
reaction (800 mg protein per 0.5 mCi 1). Panel C: Comparison of ¯uorograms for [3H]BZDC Glc-6-P labeling (lane 1) and [3H]BZDC-IP3 (lane 2)
labeling of SER microsomes. Panel D: Fluorogram showing speci®city of [3H]BZDC Glc-6-P labeling of SER: no competitor (lane 1), 100 mM Glc-
6-P (lane 2), 1 mM BZDC Glc-6-P (lane 3), 1 mM BnDC-Glc-6-P (lane 4).
modi®cation of the 32 and 40 kDa proteins (Fig. 2D). In
contrast, 100 mM Glc-6-P was only partially e€ective in
suppressing the photolabeling. This suggests that the
photophore is modifying a hydrophobic region of the
target proteins20,21 proximal to the Glc-6-P recognition
site.
protein.7 The larger of the photolabeled proteins was
however, not detected by an antibody speci®c for Glc-6-
phosphatase. This antibody did nonetheless detect a
protein found exclusively in the hydrophobic portion of
the Triton X-114 phase separation.
Recent studies have demonstrated that the Glc-6-P
transporter is an integral ER membrane protein with a
Mr of approximately 46 kDa.4,6 In order to determine if
the two proteins labeled preferentially by the photo-
anity probe were themselves integral membrane pro-
teins, we conducted phase separation experiments with
Triton X-114, a method that permits discrimination
between hydrophobic and hydrophilic proteins.23 Inte-
gral membrane proteins are found exclusively in the
detergent phase, while hydrophilic proteins are recov-
ered in the aqueous phase. Integral membrane proteins
can be separated from peripheral proteins associated
with the ER using sodium carbonate.24 After Na2CO3
treatment, integral membrane proteins are retained in
the pellet and peripheral proteins move to the super-
natant. When the ER vesicles were treated with Triton
X-114, the two photolabeled proteins were found
exclusively in the aqueous phase. These two proteins
were also recovered in the supernatant following
Na2CO3 pretreatment. These data (not shown) suggest
that the two proteins are peripherally associated with
the cytoplasmic surfaces of the ER.
Conclusion
We sought to selectively label the Glc-6-P transporter
with the [3H]BZDC Glc-6-P analogue using SER-enri-
ched microsomes. Indeed, the probe selectively labeled
two proteins that are preferentially associated with the
SER and Golgi, but which had molecular masses dis-
tinct from that of the Glc-6-P transporter. The identity
of these peripheral, not integral, membrane proteins has
not yet been determined. The selectivity of the labeling
was established using the non-photoactivatable and UV
transparent BnDC group, since the free Glc-6-P ligand
lacking the hydrophobic moiety failed to competitively
displace the photolabeling of the 34 and 40 kDa pro-
teins. Even though soluble Glc-6-P was ine€ective as a
competitive inhibitor, the absence of labeling of these
two proteins by [3H]BZDC IP3 suggests that the Glc-6-P
head group must indeed be in part responsible for the
recognition of [3H]BZDC Glc-6-P by the labeled pro-
teins. In other words, the basis for the anity between
these proteins and [3H]BZDC Glc-6-P involves both the
hydrophobic photophore and the soluble Glc-6-P head
group. This property has previously been exploited with
analogues of Ins(1,4,5)P3 and Ins(1,3,4,5)P4, in which
anity for the lipid-containing phosphoinositide ligands
was subsequently demonstrated.16,19,20 Although Glc-6-
P is not a constituent of any known hydrophobic or
lipid-containing biochemical structure, the linker mod-
i®ed Glc-6-P is congruent (epimeric at one OH group)
with Man-6-P-phosphoethanolamine, the site of pro-
tein attachment to glycosylphosphatidylinositol (GPI)
anchored proteins. While the identities of the 34 and
40 kDa proteins remain to be determined, it is an
intriguing possibility that they could be proteins that
interact with the attachment site for GPI-anchored,
non-integral membrane proteins.
It appeared that the primary proteins labeled by the
[3H]BZDC Glc-6-P probe did not include the Glc-6-P
transporter. Often, failure to photolabel a target with an
externally-appended label suggests that the photophore
or attachment chemistry interferes with the interaction
between the normal ligand and the target protein.16 We
thus considered four other known Glc-6-P binding pro-
teins: phosphoglucomutase (62 kDa), glucose-6-phos-
phatase (39 kDa), Glc-6-P-dehydrogenase (72 kDa),
and phosphoglucosisomerase (132 kDa). Among these,
Glc-6-phosphatase has a molecular mass similar to one
of the proteins labeled with [3H]BZDC Glc-6-P,
although it had been shown to be an integral membrane

538
J. Peng et al. / Bioorg. Med. Chem. Lett. 10 (2000) 535±539
concentrated; the brown residue was dissolved in H2O, ®ltered
Acknowledgements
through a 0.45 mm ®lter, passed through a Chelex (Na+ form)
column, and concentrated to give the product 5 as a white
solid (0.543 g, 99%) as a mixture of a,b isomers (a:b 1:1.3). 1H
NMR (D2O) d 5.06 (d, J=4 Hz, H1 of a isomer), 4.49 (d, J=8
Hz, H1 of b isomer), 3.70-4.15 (m, 4H), 3.05±3.60 (m, 4H),
2.99 (dd, J=7, 7 Hz, 2H), 1.85 (dddd, J=6, 6, 6, 6 Hz, 2H).
Succinimido-p-benzoylcinnamate (7). To a suspension of p-
benzoylcinnamic acid (0.100 g, 0.396 mmol) and diphenyl
succinimidyl phosphate (0.206 g, 0.595 mmol) in CH3CN (15
mL) was added Et3N (0.060 g, 0.595 mmol) and stirred (rt, 16
h). The reaction mixture was then concentrated and the crude
product was puri®ed by radial chromatography (EtOAc:hex-
ane, 1:2, v:v) to yield unsaturated product 7 (0.137 mg, 99%)
We thank the NIH for Grant No. NS 29632 to G.D.P.
and The Juvenile Diabetes Foundation (995002) to
R.B.M. for ®nancial support. Earlier synthetic work on
this project was performed by Mr. O. Thum.
References and Notes
1. Nordlie, R. C.; Foster, J. D.; Lange, A. J. Annu. Rev.
Nutrition 1999, 19, 379.
2. Arion, W. J.; Ballas, L. M.; Lange, A. J.; Wallin, B. K. J.
Biol. Chem. 1976, 251, 4891.
1
as white crystals. H NMR (CDCl3) d 7.45±7.90 (m, 9H), 7.98
3. Lei, K.-J.; Shelly, L. L.; Lin, B.; Sidbury, J. B.; Chen, Y.-T.;
Nordlie, R. C.; Chou, J. Y. J. Clin. Invest. 1995, 95, 234.
4. Gerin, I.; Veiga-da-Cunha, M.; Achouri, Y.; Collet, J.-F.;
Van Schaftingen, E. FEBS Lett. 1997, 419, 235.
5. Annabi, B.; Hiraiwa, H.; Mans®eld, B. C.; Lei, K. J.; Uba-
gai, T.; Polymeropoulos, M. H.; Moses, S. W.; Parvari, R.;
Hershkovitz, E.; Mandel, H.; Fryman, M.; Chou, J. Y. Am. J.
Hum. Genet. 1998, 62, 400.
(d, J=16 Hz, 1H), 6.70 (d, J=16 Hz, 1H), 2.90 (s, 4H).
BnDC-NHS (8). Ester 7 (0.100 g, 0.285 mmol) was dissolved in
EtOAc (10 mL) and hydrogenated over 10% Pd/C at 50 psi
for 24 h. The mixture was ®ltered through Celite, con-
centrated, and crude product puri®ed by radial chromato-
graphy (EtOAc:hexane, 1:2, v:v) and a€orded product 8 (0.67
1
g, 69%) as a white solid. H NMR (CDCl3) d 7.10±7.35 (m,
9H), 3.95 (s, 2H), 2.95±3.05 (m, 2H), 2.80±2.95 (m, 2H), 2.82
(s, 4H). 13C NMR (CDCl3) d 169.1, 167.8, 139.4, 136.7, 129.0,
128.7, 128.3, 128.2, 125.9, 41.5, 32.4, 29.8, 25.4. BnDC-Glc-6-P
(2). Aminopropyl Glc-6-P (5) (0.029 g, 0.077 mmol) was dis-
solved in TEAB bu€er (5 mL, 0.2 M, pH 8.0), and a solution
of BnDC-NHS (8) (0.026 g, 0.077 mmol) in DMF (5 mL) was
added. After stirring at rt for 24 h, the reaction was con-
centrated in vacuo, the residue was dissolved in water, loaded
onto a DEAE column, and eluted with linear gradient (0±100
mM) of TEAB bu€er (pH 8.0). Fractions containing organic
phosphate were combined, concentrated, redissolved in water,
and passed through a Chelex (Na+ form) column. Con-
centration of the ®ltrate under vacuum yielded product 2
(0.025 g, 65%) as a glass-like solid. NMR showed it was a
mixture of a,b isomers (a:b 1:1.4). 1H NMR (D2O) d 6.80±7.15
(m, 9H), 5.05 (d, J=4 Hz, H1 of a isomer), 4.48 (d, J=8 Hz,
H1 of b isomer), 3.50±4.00 (m, 5H), 3.56 (s, 2H), 3.20±3.40 (m,
3H), 3.05±3.20 (m, 1H), 2.97 (dd, J=7, 7 Hz, 2H), 2.63 (dd,
J=7, 7 Hz, 2H), 2.48 (s, 1H), 2.25 (dd, J=7, 7 Hz, 2H), 1.43
(m, 2H). 31P NMR (D2O) d 4.1. [3H]BZDC-Glc-6-P (1). A
solution of [3H]BZDC-NHS (6) (1.0 mCi, 0.025 mmol) in
EtOAc (0.30 mL) was transferred to a 0.5 mL plastic cen-
trifuge tube and dried with N2 ¯ow. To the centrifuge tube was
then added a solution of amine 5 (0.042 mg, 0.125 mmol) in a
mixture (0.020 mL) of 1:1 DMF:TEAB bu€er (0.25 M, pH
8.0). After stirring at rt for 24 h, the reaction was concentrated
under vacuum to dryness. The resulting residue was dissolved
in water (0.5 mL), loaded onto a small DEAE column, and
eluted with linear gradient (0-0.10 M) of TEAB bu€er (pH
8.0). Radioactivity was eluted at ca. 0.05 M TEAB to give
radioligand 1 (0.14 mCi, 14% radiochemical yield).
18. Experimental details for biochemistry. Preparation of sub-
cellular fractions. Total liver microsomes were isolated from
livers of female Sprague±Dawley rats (250±300 g) by homo-
genization at 4 ꢀC with a Polytron, speed set at 5, using a 20-s
burst per 1 g tissue in 8 volumes of bu€er A containing 1 mM
HEPES, pH 7.5, 0.3 M sucrose, 1 mM dithiothreitol, and
1 mM phenylmethylsulfonyl ¯uoride. The homogenate was
then centrifuged at 10000Âg for 30 min. The resulting super-
natant was centrifuged at 100000Âg for 60 min, the pellet
resuspended in bu€er A, and recentrifuged at 100,000Âg for
60 min. The ®nal pellet was dissolved in bu€er B (25 mM Tris/
HCl pH 7.4, 1 mM EDTA, 1 mM K2HPO4). To isolate SER
and RER, 10 mL of supernatant resulting from the 10,000Âg
centrifugation was ¯oated over layers of 1.4 mL of 0.6 M
sucrose±15 mM CsCl and 6 mL of 1.3 M sucrose±15 mM CsCl
and centrifuged at 80,000Âg for 1.5 h at 4 ꢀC. SER was
6. Lin, B. C.; Annabi, B.; Hiraiwa, H.; Pan, C. J.; Chou, J. Y.
J. Biol. Chem. 1998, 273, 31656.
7. Pan, C. J.; Lei, K. J.; Annabi, B.; Hemrika, W.; Chou, J. Y.
J. Biol. Chem. 1998, 273, 6144.
8. Hiraiwa, H.; Pan, C. J.; Lin, B.; Moses, S. W.; Chou, J. Y.
J. Biol. Chem. 1999, 274, 5532.
9. Gitzelmann, R.; Bosshard, N. U. Eur. J. Pediatr. 1993, 152,
S33±S38.
10. Kilpatrick, L.; Garty, B. Z.; Lundquist, K. F.; Hunter, K.;
Stanley, C. A.; Baker, L.; Douglas, S. D.; Korchak, H. M. J.
Clin. Invest. 1990, 86, 196.
11. Korchak, H. M.; Garty, B. Z.; Stanley, C. A.; Baker, L.;
Douglas, S. D.; Kilpatrick, L. Eur. J. Pediatr. 1993, 152, S39.
12. Chen, P. Y.; Csutora, P.; Veyna-Burke, N. A.; Marchase,
R. B. Diabetes 1998, 47, 874.
13. Kramer, W.; Burger, H. J.; Arion, W. J.; Corsiero, D.;
Girbig, F.; Weyland, C.; Hemmerle, H.; Petry, S.; Haber-
mann, P.; Herling, A. Biochem. J. 1999, 339, 629.
14. Olszewski, J. D.; Dorman, G.; Elliott, J. T.; Hong, Y.;
Ahern, D. G.; Prestwich, G. D. Bioconjugate Chem. 1995, 6,
395.
15. Dorman, G.; Prestwich, G. D. Biochemistry 1994, 33, 5661.
16. Prestwich, G. D.; Dorman, G.; Elliott, J. T.; Marecak, D.
M.; Chaudhary, A. Photochem. Photobiol. 1997, 65, 222.
17. Experimental details for synthesis: Protected Glc-6-P ana-
logue (4). A solution of benzyl 2,3,4-tri-O-benzyl-a-d-arabino-
hexopyranoside (3) (1.00 g, 1.85 mmol), benzyloxy-[(N-Cbz-3-
amino-1-propyl)oxy](N, N-diisopropylamino)phosphine (0.826
g, 1.85 mmol) and 1H-tetrazole (0.259 g, 3.70 mmol) in
CH2Cl2 (50 mL) was stirred at rt under Ar for 2 h. The reac-
tion mixture was cooled to 40 ꢀC and a solution of mCPBA
(57%, 1.12 g, 3.7 mmol) in CH2Cl2 (20 mL) was added via
cannula. After stirring at 40 ꢀC for 30 min, the mixture was
warmed to rt, washed (satd aq Na2SO3, satd aq NaHCO3,
dried (MgSO4), concentrated, and puri®ed on SiO2 (EtOAc:
1
hexane, 3:2, v:v) to give 4 (1.47g, 88%) as a colorless oil. H
NMR (CDCl3) d 7.20±7.40 (m, 30H), 5.15±5.25 (m, 1H), 4.75±
5.10 (m, 7H), 3.95±4.30 (m, 4H), 3.75±3.85(m, 1H), 3.40±3.55
(m, 2H), 3.21 (ddd, J=6, 6, 6 Hz, 2H), 1.79 (dddd, J=6, 6, 6,
6, Hz, 2H). 13C NMR (CDCl3) d 156.3, 139.5, 137.9, 137.8,
136.9, 136.4, 135.7, 135.5, 128.5, 128.4, 128.3, 128.1, 127.9,
127.8, 127.7, 127.6, 127.5, 95.4, 95.3, 81.6, 79.7, 79.6, 76.9,
75.6, 75.0, 72.8, 69.5, 69.3, 69.2, 69.1, 66.4, 66.1 (m), 65.0, 64.9,
36.9, 29.8. Aminopropyl Glc-6-P (5). Precursor 4 (1.37 g, 1.52
mmol) in MeOH (40 mL) and H2O (10 mL) was hydrogenated
at 50 psi over 10% Pd/C for 24 h. The mixture was ®ltered and

J. Peng et al. / Bioorg. Med. Chem. Lett. 10 (2000) 535±539
539
collected at the lower interface and RER from the pellet.
These fractions were resuspended and centrifuged at
100000Âg for 60 min. To isolate mitochondria, liver homo-
genate was centrifuged at 600Âg for 10 min. The resulting
supernatant was centrifuged at 10300Âg for 10 min. The pellet
was suspended in bu€er C containing 5 mM HEPES, pH 7.4,
250 mM mannitol, 0.5 mM EGTA, 0.1% BSA and then
loaded onto a cushion of 30% Percoll in bu€er D, containing
25 mM HEPES, pH 7.4, 225 mM mannitol, 1 mM EGTA,
0.1% BSA. This was centrifuged at 95000Âg for 30 min. The
lower part of the dense, brownish yellow mitochondrial band
was collected and washed with bu€er C. The ®nal centrifuga-
tion at 6300Âg for 10 min. Photoanity labeling and electro-
phoresis. Irradiations were conducted in a 96-well plate under
dark conditions. To each well was added 43 mL (800 mg of
protein) of a subcellular fraction, 50 mL of 2Âbu€er B, and 7
mL of [3H]BZDC-Glc-6-P. They were incubated on ice for 10
min, and then photoactivated for 60 min using a 100-W long
wavelength (365 nm) UV lamp. Next, 6ÂSDS sample bu€er
was added to the samples and they were separated on 10%
SDS-PAGE. Following ®xation, the [3H]BZDC-Glc-6-P
labeled gels were prepared for ¯uorography using Entensify
(NEN Life Science Products, Boston, MA), dried, and placed
on X-ray ®lm at 80 ꢀC for 7±10 days. Treatment of mem-
branes with Triton X-114 or Na2CO3. Samples (1 mg/mL) in
bu€er containing 10 mM Tris/HCl, pH 7.4, 150 mM NaCl,
2% Triton X-114 were kept on ice for 5 min and subsequently
incubated at 30 ꢀC for 3 min. Mixtures were then centrifuged
at 3000Âg for 10 min. The resulting sample was resolved into
aqueous and detergent-rich phases, which were then separated
by SDS-PAGE for further analysis. Aliquots of smooth ER
(1 mg/mL) were treated with Na2CO3. The resulting pellet and
supernatant were collected and analyzed by SDS-PAGE.
Immunoblot analysis. Equal aliquots of Triton X-114 and
Na2CO3 treated samples were fractionated by SDS-PAGE and
electrotransferred to PVDF membrane for immunodetection
by an antiserum speci®c for glucose-6-phosphatase. Electro-
transferred proteins on PVDF membrane were incubated with
sheep antiserum against rat Glc-6-Pase, and visualized with
goat anti-sheep antibodies conjugated with horseradish per-
oxidase in conjunction with detection by luminescence using a
commercial ECL kit.
19. Prestwich, G. D. Acc. Chem. Res. 1996, 29, 503.
20. Chaudhary, A.; Chen, J.; Gu, Q.-M.; Witke, W.; Kwiat-
kowski, D. J.; Prestwich, G. D. Chemistry & Biology 1998, 5,
273.
21. Mourey, R. J.; Estevez, V. A.; Marecek, J. F.; Barrow, R. K.;
Prestwich, G. D.; Snyder, S. H. Biochemistry 1993, 32, 1719.
22. Mehrotra, B.; Elliott, J. T.; Chen, J.; Olszewski, J. D.;
Pro®t, A. A.; Chaudhary, A.; Fukuda, M.; Mikoshiba, K.;
Prestwich, G. D. J. Biol. Chem. 1997, 272, 4237.
23. Bordier, C. J. Biol. Chem. 1981, 256, 1604.
24. Borgese, N.; Aggujaro, D.; Carrera, P.; Pietrini, G.; Bas-
setti, M. J. Cell Biol. 1996, 135, 1501.