Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Please do not adjust margins
ChemComm
Page 3 of 4
DOI: 10.1039/C7CC07666E
Journal Name
COMMUNICATION
Therefore, the fluorescence quenching mechanism of PDI-Gal6 receptor (MR, mannose-selective), respectively, were used for
could be ascribed to both dynamic and static processes. The targeted imaging. Human cervical cancer (HeLa) and lung
direct fluorescence of the PDI-based glycoclusters could not cancer (A549) cell lines without the receptor expression were
directly be exploited since it was too weak (results not shown). used as control. The expression level of ASGPr and MR of all
The fluorescence quantum yield of PDI-Gal6 was determined to cells were confirmed by real-time quantitative polymerase
be 0.005 (reference: Rhodamine 6G), which is much lower chain reaction (Fig. S5). We determined that whereas DCM
than that of DCM alone (0.20) in PBS.
alone was unselectively internalized by the cancer cells, PDI-
Gal6/DCM (Fig. 4a and 4c) and PDI-Man6/DCM (Fig. 4b and 4d)
showed a predominant fluorescence production in Hep-G2 and
MDA-MB-231, respectively. Although the PDI-Man6/DCM led
to a weaker fluorescence increase in cells, (1) the accuracy of
this measurement (***P<0.001) and (2) the selectivity for the
MDA-MB-231 cell line makes it a reliable result.
Fig. 3. Fluorescence response of PDI-Gal6/DCM (5/30 μM) in the presence of (a)
different serum proteins including human serum albumin (HSA) (7.5 μM), fetal bovine
serum (FBS) (14 μL, from cesarean section fetal bovine) and bovine serum albumin (BSA)
(7.5 μM), and (b) different lectins including soybean agglutinin (SBA, 7.5 μM),
concanavalin A (Con A, 7.5 μM) and peanut agglutinin (PNA, 7.5 μM) with time. Cartoon
of the interaction mode of PDI-Gal6/DCM with (c) serum proteins and (d) lectins. All
fluorescence measurements were tested in Tris-HCl buffer (20 mM, pH 7.4) with an
excitation wavelength of 460 nm.
Fig. 4. Fluorescence imaging (a) and quantification (c) of different cancer cells with
DCM (10 μM) or PDI-Gal6/DCM (30/10 μM). Fluorescence imaging (b) and
quantification (d) of different cancer cells with DCM or PDI-Man6/DCM (***P<0.001).
Hep-G2: human liver cancer cell line; HeLa: human cervical cancer cell line; A549:
human lung cancer cell line; MDA-MB-231: human breast cancer cell line. Excitation
channel: 460–490 nm, emission: 580–650 nm; the cell nucleus was stained by Hoechst
Since DCM has been reported to have strong affinity to
serum proteins (SPs),37 we subsequently added different SPs to
the solution of PDI-Gal6/DCM. The result showed that
different SPs including human serum albumin (SBA), bovine
serum albumin (BSA) and fetal bovine serum (FBS) sharply 33342; scale bar: 100 μm.
recovered the fluorescence of the glyco-dot (Fig. 3a and S4),
To corroborate the carbohydrate receptor-targeting ability
suggesting the dissociation of the material because of the
competitive binding of DCM to the hydrophobic cavity of SPs
(Fig. 3c).38 Interestingly enough, the presence of lectins
including soybean agglutinin (SBA, N-acetyl galactosamine
selective), concanavalin A (Con A, mannose/glucose selective)
and, in particular, peanut agglutinin (PNA, a galactose selective
lectin) did not produce the fluorescence of PDI-Gal6/DCM (Fig.
3b and S4). This suggests the stability of the supramolecular
glyco-dot even in the presence of a selective lectin that
recognizes the external monosaccharide cluster of the
supramolecular architecture (Fig. 3d). This sets the basis for
fluorescence cell imaging given that (1) SPs are abundant
intracellularly to trigger the glyco-dot fluorescence,39,40 and
that (2) the selective interaction of cell-surface receptors with
glyco-dots would not disrupt the dots, preserving the integrity
of the nano-glycomaterial upon endocytosis.
of the glyco-dots, a knockdown assay was carried out. The
ASGPr and MR level was largely reduced in Hep-G2 and MDA-
MB-231 cells through RNA interference, producing sh-ASGPr
and sh-MR cells, respectively. We observed that the
fluorescence of PDI-Gal6/DCM (Fig. 5a and 5c) and PDI-
Man6/DCM (Fig. 5b and 5d) in sh-ASGPr and sh-MR was
significantly lower than that in Hep-G2 and MDA-MB-231,
respectively. This is in agreement with the significantly
reduced ASGPr (Fig. 5d) and MR (Fig. 5f) level in sh-ASGPr and
sh-MR, respectively, suggesting the receptor-targeting
property of the glyco-dots. Meanwhile, the pre-incubation of
increasing free galactose and mannose with Hep-G2 and MDA-
MB-231 suppressed the fluorescence of PDI-Gal6/DCM and
PDI-Man6/DCM (Fig. S6), respectively. These data suggest that
the supramolecular glyco-dots can be successfully used for
targeted cell imaging based on selective CPIs.
Subsequently, we tested the targeted fluorescence imaging
of the glyco-dots for cancer cells that express carbohydrate
receptors. Human liver cancer (Hep-G2) and breast cancer
To analyze the structure-activity relationship as regards the
valency of glycoclusters, we compared the imaging ability of
PDI-Gal6/DCM and PDI-Gal4/DCM for Hep-G2. The result
showed that the imaging ability of PDI-Gal6/DCM was stronger
than that of PDI-Gal4/DCM (Fig. S7), probably because of the
(MDA-MB-231)
cell
line
that
overly
expresses
asialoglycoprotein (ASGPr, galactose-selective) and mannose
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins