A fluorescent inositol phosphate glycan stimulates lipogenesis in rat adipocytes by extracellular activation alone

Article abstract of DOI:10.1016/j.bmcl.2005.02.061

The chemical synthesis of 2,6-dideoxy-2-amino-6-mercaptoglucopyranosyl- (α1-6)-myo-inositol 1,2-cyclic phosphate and its conjugation with a lucifer yellow derivative are reported. The resulting fluorescent IPG analogue was able to stimulate lipogenesis in rat adipocytes despite the fact that it was not internalized into the cell. The results demonstrate that internalization of the IPG is not required for manifestation of its insulin-like effects.

Full text of DOI:10.1016/j.bmcl.2005.02.061

Bioorganic & Medicinal Chemistry Letters 15 (2005) 2023–2025
A fluorescent inositol phosphate glycan stimulates lipogenesis
in rat adipocytes by extracellular activation alone
David I. Turner, Nilanjana Chakraborty and Marc dÕAlarcao*
Michael Chemistry Laboratory, Department of Chemistry, Tufts University, Medford, MA 02155, USA
Received 28 January 2005; revised 17 February 2005; accepted 18 February 2005
Available online 16 March 2005
Abstract—The chemical synthesis of 2,6-dideoxy-2-amino-6-mercaptoglucopyranosyl-(a1–6)-myo-inositol 1,2-cyclic phosphate and
its conjugation with a lucifer yellow derivative are reported. The resulting fluorescent IPG analogue was able to stimulate lipogenesis
in rat adipocytes despite the fact that it was not internalized into the cell. The results demonstrate that internalization of the IPG is
not required for manifestation of its insulin-like effects.
Ó 2005 Elsevier Ltd. All rights reserved.
The inositol phosphate glycans (IPGs) are small, ionic
oligosaccharides that are released from the exterior
surface of insulin-sensitive cells in response to insulin
stimulation. IPGs isolated from the medium after
insulin-induced release or generated artificially by phos-
eral insulin-responsive enzymes, including c-AMP phos-
phodiesterase,6 pyruvate dehydrogenase phosphatase,7
protein phosphatase 2C,8 c-AMP-dependent protein ki-
nase,9 and adenylate cyclase6 in cell-free assays. On the
other hand, Muller and coworkers have demonstrated
that synthetic and yeast-derived IPGs are able to bind
to a 115 kDa protein on the exterior surface of rat adi-
pocytes and that this binding activates the pp59(Lyn) ki-
nase, leading to insulin-like effects in the cell.10
However, it has not been determined if surface binding,
or transport into the cell, or both are necessary and suf-
ficient for cell activation by IPGs.
phatidylinositol–specific phospholipase
C (PI–PLC)
cleavage of glycolipids on the cell surface are able to
stimulate insulin-sensitive cells upon exogenous addi-
tion, even in the absence of insulin. These facts have
led to the hypothesis that the IPGs are second messen-
gers of insulin action.1–4 Their extracellular release
may provide a mechanism for signal amplification, since
one molecule of insulin can result in the release of
numerous IPG molecules and could thereby activate
more than one cell.
We have been interested in studying the relationship be-
tween IPG transport and cell activation. To facilitate the
study of transport by optical techniques including fluo-
rescence microscopy and fluorescence assisted cell sort-
ing (FACS), it was desirable to have a fluorescent IPG
analogue. In this communication we describe the chem-
ical synthesis of a small IPG tethered to a lucifer yellow
fluorescent probe and report the finding that, while the
conjugate is competent to stimulate lipogenesis in native
rat adipocytes, it is not transported into the cell. We
conclude that, despite the fact that the natural IPGs
are transported into cells, this internalization is not re-
quired for activation of adipocytes.
The mechanism by which IPGs stimulate insulin-like
activities in sensitive cells, including even the site(s) of
IPG action in or on the cell, is still not fully understood.
Alvarez et al.5 convincingly demonstrated that a natural
IPG is actively transported into the cytoplasm of hepa-
tocytes in a process that is inhibited by cyanide and
therefore is probably energy dependent. This discovery
lead to the hypothesis that cell activation by IPGs oc-
curs once the IPG is in the cytosol where it modulates
the metabolic activity of the cell. This hypothesis is con-
sistent with the observations from several laboratories
that the IPGs are able to modulate the activities of sev-
We opted to prepare a pseudodisaccharide IPG ana-
logue since it is the structurally simplest (and syntheti-
cally least demanding) IPG analogue known to exhibit
insulin-like activity, albeit modestly.11 To permit easy
conjugation of the pseudodisaccharide with a fluoro-
phore or other moiety, the primary hydroxyl was replaced
Keywords: IPG; Inositol; Insulin.
*
Corresponding author. Tel.: +1 617 627 3686; fax: +1 617 627
3443; e-mail: marc.dalarcao@tufts.edu
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.02.061

2024
D. I. Turner et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2023–2025
with a thiol group. A thiol-modified IPG has been pre-
viously reported12 but we pursued a different synthetic
strategy with the pseudodisaccharide analogue. Lucifer
yellow was chosen as the fluorescent dye for the conju-
gate because it is dianionic and therefore is expected
to preclude nonspecific association of the conjugate with
cell surfaces and passive diffusion through the cell
membrane.
Conjugate 1 was evaluated for its ability to stimulate
lipogenesis in intact native rat adipocytes by a modifica-
tion18 of the published method.19 The stimulation of
lipogenesis as a function of concentration for conjugate
1, the parent pseudodisaccharide (2-deoxy-2-amino-
glucopyranosyl-(a1–6)-myo-inositol 1,2-cyclic phos-
phate, 7,11) and insulin are shown in Figure 1. The data
were fitted to a standard receptor-binding model (Eq.
1).
Our synthesis of fluorescent IPG 1 is shown in Scheme 1.
The known,13 selectively protected 2-azido-2-deoxyglu-
cose 2 was tosylated at the 6-position, then treated with
lithium benzylthiolate in THF to afford the protected
thiol 3. Removal of the silyl group at the anomeric posi-
tion with tetrabutylammonium fluoride and acetic acid,
followed by formation of the trichloroacetimidate by
SchmidtÕs method14 produced the glycosyl donor.
Trimethylsilyl triflate-induced glycosylation of known
Ã
activity ¼ ðactivitymax ½analyteŠÞ=ðKd þ ½analyteŠÞ: ð1Þ
Conjugate 1 was found to stimulate lipogenesis with a
maximal activity of 47% that of the maximal insulin re-
sponse (MIR, defined as the stimulation by 5 nM insu-
lin) and a Kd of 12.0 lM (R = 0.877). It is interesting
to note that the conjugate was better able to stimulate
lipogenesis than the parent 7 (maximal activity: 24%
MIR, Kd = 15.9 lM; R = 0.789) possibly because 1
bears a better resemblance to the IPG pharmacophore20
due to the presence of the sulfate groups distal to the
inositol.
inositol 415 (diethyl ether, 4 A molecular sieves,
˚
ꢀ42 °C) produced a-pseudodisaccharide 5, together
with the corresponding b-anomer (a:b, 2.4:1). The mix-
ture of anomers was not separated at this stage but was
treated with lithium hydroxide in THF–water to hydro-
lyze the cyclic carbonate producing the corresponding
anomeric diols. The a-anomer was isolated chromato-
graphically and treated with the reagent produced by
CH3OPOCl2 in pyridine,16 followed by sodium in liquid
ammonia to reduce the azide function and remove all of
the benzyl protective groups, in analogy to earlier IPG
syntheses.15 The deprotected IPG was obtained from
this reaction mixture as the disulfide 6, presumably
due to air oxidation during workup, in 76% yield (two
steps). Finally, coupling of 6 with the commercially
available lucifer yellow iodoacetamide at pH 7.1 in the
presence of tris(carboxyethyl)phosphine (TCEP) to re-
duce the disulfide in situ, produced fluorescent IPG 1.
The structure of the conjugate was confirmed by 1H
The ability of conjugate 1 to enter intact rat adipocytes
was checked by two methods: qualitatively by fluores-
cence microscopy, and quantitatively by flow cytometry.
Cells were isolated from epidydimal fat pads from male
Long–Evans rats by collagenase digestion,19 then di-
vided into two groups. One group of cells was incubated
with 40 lM conjugate 1 at 37 °C for 1 h, then viewed by
fluorescence microscopy or analyzed by flow cytometry.
No incorporation of fluorescence into adipocytes was
observed by either method. The second group of cells
was checked for viability by assaying for stimulation
of lipogenesis by either insulin or conjugate 1. These
cells showed results identical to those reported above,
establishing that the cells were alive and metabolically
active.
31
and P NMR and ESI MS (negative ion mode).17
SBn
O
BnO
BnO
OH
O
SBn
O
c, d, e
BnO
HO
a,b
N3
O
BnO
BnO
BnO
BnO
O
BnO
O
OTBS
NH2
OTBS
OBn
OBn
O
N3
N3
OBn
OBn
2
3
5
O
O
O
f, g, h
4
KO3S
SO3K
S
O
N
O
O
O
H3N+
HO
S
HO
i
HO O
P
O
HO
HO
O
OH
OH
HN
O
S
H3N+
HO O
O
HO
HO
OH
OH
O
O -
O
O
H3N+
HO O
P
O
OH
OH
O -
O
O
1
P
6
O -
O
Scheme 1. Synthesis of a fluorescent IPG. Reagents and conditions: (a) TsCl, pyridine (89%); (b) BnSLi, THF, 0 °C–25 °C (77%); (c) TBAF, AcOH,
THF (86%); (d) Cl3CCN, K2CO3; (e) 4, TMSOTf (10mol%); (f) LiOH, H 2O, THF (29% over three steps); (g) CH3OPOCl2, pyridine; (h) Na, NH3,
ꢀ78 °C; NH4Cl(s), ꢀ78 °C; CH3OH, ꢀ78 °C to 25 °C, (76% over two steps); (i) lucifer yellow iodoacetamide, 25 mM aq HEPES, pH 7.1, TCEP, 2 h.

D. I. Turner et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2023–2025
2025
Acknowledgements
We are grateful to Profs. Krishna Kumar, David Lee,
and Kyongbum Lee for insightful discussions. We also
thank Dr. Wendy Foulds Mathes, Ms. Monica Leibo-
vici, Dr. Allen Parmelee, and Mr. Viatcheslav Azev for
important assistance with the adipocyte isolations, flow
cytometry, and mass spectrometry, respectively. Fund-
ing for this work was provided by Tufts University.
The Tufts University NMR and mass spectrometry
facilities are supported in part by National Science
Foundation grants CHE 0320783 and CHE 9723772,
respectively.
References and notes
1. Jones, D. R.; Varela-Nieto, I. Int. J. Biochem. Cell Biol.
1998, 30, 313–326.
2. Stralfors, P. BioEssays 1997, 19, 327–335.
3. Field, M. C. Glycobiology 1997, 7, 161–168.
4. Varela-Nieto, I.; Leon, Y.; Caro, H. N. Comp. Biochem.
Physiol. B: Biochem. 1996, 115, 223–241.
5. Alvarez, J. F.; Sanchez-Arias, J. A.; Guadano, A.; Estevez,
F.; Varela, I.; Feliu, J. E.; Mato, J. M. Biochem. J. 1991,
274, 369–374.
6. Saltiel, A. R. Endocrinology 1987, 120, 967–972.
7. Lilley, K.; Zhang, C.; Villar-Palasi, C.; Larner, J.; Huang,
L. Arch. Biochem. Biophys. 1992, 296, 170–174.
8. Huang, L. C.; Heimark, D.; Linko, J.; Nolan, R.; Larner,
J. Biochem. Biophys. Res. Commun. 1999, 255, 150–156.
9. Villalba, M.; Kelly, K. L.; Mato, J. M. Biochim. Biophys.
Acta 1988, 968, 69–76.
Figure 1. Stimulation of lipogenesis in rat adipocytes. Isolated native
rat adipocytes were incubated with 6-[3H]-glucose (0.55 mM) and
various concentrations of the analyte for 1 h. Incorporation of tritium
into lipids was measured and is expressed as a percent of the maximal
insulin response (%MIR). Black circles: 2-deoxy-2-amino-glucopyr-
anosyl-(a1–6)-myo-inositol 1,2-cyclic phosphate, 711 (0–60 lM); blue
squares: compound 1 (0–60 lM); red diamonds: insulin (0–500 nM).
Each data point is the average of at least five replicates. Error bars
represent 1 SD.
10. Muller, G.; Jung, C.; Frick, W.; Bandlow, W.; Kramer, W.
Arch. Biochem. Biophys. 2002, 408, 7–16.
11. Plourde, R.; dÕAlarcao, M.; Saltiel, A. R. J. Org. Chem.
1992, 57, 2606–2610; Dietrich, H.; Espinosa, J. F.; Chiara,
J. L.; Jimenez-Barbero, J.; Leon, Y.; Varela-Nieto, I.;
Mato, J. M.; Cano, F. H.; Foces-Foces, C.; Martin-
Lomas, M. Chem. Eur. J. 1999, 5, 320–336.
12. Lindberg, J.; Stralfors, P.; Konradsson, P. Tetrahedron
2002, 58, 4245–4254.
13. Kinzy, W. Ph.D. Thesis; University of Konstanz: Kon-
stanz, 2000.
14. Grundler, G.; Schmidt, R. R. Liebigs Ann. Chem. 1984,
1826–1847.
15. Jaworek, C. H.; Iacobucci, S.; Calias, P.; dÕAlarcao, M.
Carbohydr. Res. 2001, 331, 375–391.
16. Smrt, J.; Catlin, J. Tetrahedron Lett. 1970, 11, 5081–5082.
17. Both compounds 1 and 6 were purified by size exclusion
chromatography on a G-10column, eluting with water,
then lyophylized. The yield in the coupling of 6 with the
lucifer yellow iodoacetamide, was variable and often low
(ꢁ2%), almost certainly because the tris(carboxy-
ethyl)phosphine was competing for the iodoacetamide. A
recent study suggests that this problem may be solved by
using the maleimide derivative rather than the iodoacet-
amide derivative of the dye as the coupling partner:
Shafer, D. E.; Inman, J. K.; Lees, A. Anal. Biochem. 2000,
282, 161–164.
To rule out the possibility that stimulation of lipogenesis
by conjugate 1 was due to a degradation product pro-
duced during the incubation with adipocytes, the buffer
used during the incubation was collected at the end of
the incubation period and analyzed by HPLC on a
G-10sizing column, eluting with water. We found no
evidence of any degradation product in the incubation
buffer by this analysis. These chromatography condi-
tions successfully separated conjugate 1 from both of
the smaller compounds, that is, parent disaccharide 7
and unconjugated lucifer yellow iodoacetamide. We
cannot rule out the possibility that a degradation prod-
uct is formed at concentrations below the detection limit
of this technique, but the possibility that a degradation
product present at such a low concentration is responsi-
ble for the observed stimulation of lipogenesis is remote,
particularly since the parent disaccharide 7 was less
active than conjugate 1, even when present at the same
concentration.
We conclude that despite the known ability of an IPG to
be actively transported into an insulin sensitive cell,5 at
least for conjugate 1, such internalization is not required
for stimulation of lipogenesis in rat adipocytes. We are
currently exploring if larger synthetic IPGs are trans-
ported into rat adipocytes.15
18. Chakraborty, N.; dÕAlarcao, M., submitted.
19. Rodbell, M. J. Biol. Chem. 1964, 239, 375–384.
20. Frick, W.; Bauer, A.; Bauer, J.; Wied, S.; Muller, G.
Biochemistry 1998, 37, 13421–13436.