A. Hatano et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
group is clearer than that of the trifluorocoumarin dye of probe 1.
Fig. 4b shows that probe 2 detected the difference between Caco-2 and
the other four cell lines. DNJ-Dans 2 selectively imaged the cell mem-
brane of only Caco-2; this probe did not stain any part of the HeLa cells,
Neuro-2a cells, N1E-115 cells, or A1 cells. Additionally, probe 2 pro-
vided the clearest fluorescent images among our synthetic probes. It
appears that DNJ-DEAC probe 3 does not stain any of the cell lines,
which is strange since the probe structures differ only in the fluor-
ophore located outside of the active site. Additional experiments using
probe 3 under alternate conditions are ongoing.
between three atoms of the substrate and three atoms at the surface of
the protein pocket. This simulation evaluates the distance and simila-
rities between the protein pocket and substrate. Second, a local
minimum search was performed that is reinitiated with different con-
formers. The receptor–substrate interactions account for van der Waals,
coulomb, hydrogen bond, and hydrophobic interactions. The di-
mensionless raw docking score that was used for receptor–substrate
docking is below. The “g” was a parameter set to 0.01 mol/kcal. The “E”
value shows the potential functions for each interaction, van der Waals
interaction (EvdW), coulomb interaction (Eelec), hydrogen bond (EH-bond
)
and hydrophobic interaction (EASA). Sraw = g(EvdW + Eelec + EH-
bond + EASA).
3. Conclusion
Cell-specific probes composed of linked protein recognition and
signal parts were utilized to recognize cell surface proteins. Three
probes were developed composed of an enzyme inhibitor and a con-
jugated fluorophore: DNJ-CF3 1, DNJ-Dans 2 and DNJ-DEAC 3. First,
the structure of the complex of these probes and 3TON derived from
human lumen α-glucosidase was simulated via MF myPresto. The PDB:
3TON complex structure between α-glucosidase and acarbose was ob-
tained via X-ray crystallography for comparison. Docking studies be-
tween α-glucosidase and each inhibitor were compared to the crystal
structure of acarbose and miglitol and found to be similar. Next, the
expression of α-glucosidase on the surface of Caco-2 cells was evaluated
via western blotting, which detects α-glucosidase as an antigen. The α-
glucosidase activity was analyzed with respect to its ability to hydrolyze
p-nitrophenyl α-D-glucopyranoside to α-D-glucopyranoside and p-ni-
trophenol. The results indicate that Caco-2 cells express α-glucosidase
on the membrane surface. The ability of probes 1–3 to detect the dif-
ference between HeLa cells, Neuro-2a cells, N1E-115 cells, A1 cells, and
Caco-2 cells was then examined. Cell-specific fluorescent imaging was
possible for probes 1 and 2. Caco-2 cells treated with probes 1 and 2
showed blue and green fluorescence, respectively, on the cell mem-
branes, but did not stain the cells inside. Probes 1 and 2 did not stain
any part of the HeLa cells, Neuro-2a cells, N1E-115 cells, or A1 cells, in
agreement with the well-known fact that there is α-glucosidase on the
surface membrane in Caco-2, but not on the surface membrane of the
other four cell types. While DNJ-DEAC probe 3 did not stain any of the
cells here, alternative condition studies are ongoing.
4.3. Western blotting31–33
After incubation of each sample, the samples were collected and
used in western blotting as described previously, with some modifica-
tions.32 The lysates were centrifuged, and protein content was de-
termined using a Bio-Rad protein assay (#500-0006JA, Bio-Rad Japan,
Tokyo, Japan) according to the manufacturer’s procedure. Protein ex-
tracts were separated on 10% SDS-polyacrylamide gels and transferred
to Immobilon transfer membranes (PVDF; Merck KGaA). The mem-
branes were washed and then incubated in blocking solution 2.0% skim
milk in TBS-T (0.14 M NaCl, 0.01 M Tris buffer pH 7.4, 0.1% Tween 20)
for 1 h at room temperature. The membranes were washed in TBS-T,
and then treated with anti-alpha-glucosidase I antibody (H-300, Santa
Cruz Biotechnology) at 1:200 dilution in TBS-T overnight at 4 °C. HRP-
conjugated anti-rabbit IgG antibody (Promega, W402B) was used as a
secondary antibody at 1:4000 dilution in TBS-T for 1 h at room tem-
perature. Western blotting experiments were performed at least twice.
Cell homogenates were subjected to western blot analysis for α-gluco-
sidase expression using Image J software. β-Actin was detected to
monitor equal loading.
4.4. Enzymatic reactions34
The α-glucosidase activity of Caco-2 was analyzed using a modified
version of a published procedure.34 Caco-2 cells were seeded on each
well of 24-well plastic plates (1.5 × 105/well). The cells were cultured
in MEM medium (FBS(+), P/S(+)) for 1, 2, and 3 weeks at 37 °C. Prior
to enzyme assay, the culture medium was washed with PBS twice, and
then the cells were loaded with 1 mM of 4-nitrophenyl α-D-glucopyr-
anoside dissolved in 0.10 mol/L of PBS buffer (12 mmol/L of phos-
phate, 137 mmol/L of NaCl, 2.7 mmol/L of KCl, pH 7.4) and kept at
37 °C in a CO2 incubator. The absorption was measured at 400 nm using
a UV microplate reader (PerkinElmer EnSpire®) for each incubation
time (0, 10, 30, 60, and 90 min). The assay was performed in duplicate,
and the average value reported.
4. Experimental section
4.1. General
Chemicals were of the best grade available, supplied by Tokyo
Chemical Industries, Wako Pure Chemical, or Sigma-Aldrich Co. and
were used without further purification. Absorbance spectra for all
probes were collected at 37 °C with a JASCO V-630 Bio UV–visible
spectrometer. All compounds were synthesized in our laboratory, and
spectroscopic identification data for all compounds (1H and 13C NMR,
mass spectrometry, fluorescent spectrometry, etc.) were reported pre-
viously (ref 22). The cells were observed with a fluorescence micro-
scope (IX81, Olympus, Tokyo, Japan). Enzyme activity was measured
with a microplate reader (EnSpire®, PerkinElmer, Tokyo, Japan). HeLa
cells were derived from human cervical cancer tissue. Caco-2 cells were
derived from colon carcinoma. N1E-115 cells were derived from a
mouse neuroblastoma C1300 tumor. Neuro-2a cells were derived from
a mouse brain neuroblastoma C1300 tumor. A1 cells were derived from
the astrocyte of a newborn mouse brain.
4.5. Cell cultures
HeLa cells (1.0 × 105 cells) were cultured on glass plates in
Dulbecco’s modified Eagle’s medium (D-MEM) containing a fetal bovine
serum (5%) in a CO2 incubator at 37 °C. For microscopic observation, a
probe solution in DMSO was introduced into the cell culture medium,
and probes 1–3 were incubated overnight at 37 °C (11 µmol/L). The
culture medium was washed with PBS buffer and distilled water, and
then the cells were observed using a fluorescence microscope. Caco-2
cells from a human colon carcinoma were cultured similarly (4.3).
N1E-115, Neuro-2a, and A1 cell were cultured on glass plates, three
glass plates were placed in a 35 mm dish, and 1.0 × 105 cells were
seeded there. A laminin-coated dish and glass plates were used for the
N1E-115 cells. After 24 h, a medium exchange was carried out in order
to induce differentiation. Condition medium containing 1% DMSO was
used for the N1E-115 cells. E-MEM containing 2% FBS was used for the
Neuro-2a cells. D-MEM containing 2% FBS was used for the A1 cells.
4.2. Docking studies
Docking studies were performed using MF myPresto sievgene soft-
ware (FiatLux, Tokyo, Japan) with an AMBER-type molecular force
field. This simulation was performed using two basic criteria as follows.
First, a global minimum search was carried out that assumes contact
5
Hatano, Akihiko
