lectin-mediated cytotoxicity and specificity of 5-fluorouracil conjugated with peanut agglutinin...
TRANSCRIPT
Lectin-mediated cytotoxicity and specificity of 5-fluorouracil conjugatedwith peanut agglutinin (5-Fu-PNA) in vitro
QIN CAI, & ZHI-RONG ZHANG
Key Laboratory of Drug Targeting and Novel Drug Delivery Systems, West China School of Pharmacy, Sichuan University,
No. 17, Block 3, Southern Renmin Road, Chengdu 610041, P.R. China
(Received 30 November 2004; revised 21 January 2005; accepted 30 March 2005)
AbstractIn order to take advantage of the biorecognition between lectin and carbohydrate for targeted drug delivery, the lectin ofpeanut (Arachis hypogaea) agglutinin (PNA) was coupled by fixing its amino groups to the carbodiimide-activated carboxylicgroups of 5-fluorouracil (5-Fu) derivative (N1-substituted 5-Fu acetate) to form 5-Fu-PNA conjugate. When the couplingreaction was carried out in the presence of D-galactose (D-gal, specific sugar for PNA), the affinity of PNA was maintainedafter its coupling to N1-substituted 5-Fu acetate, which was confirmed by the result of the haemagglutination test. Otherwise,PNA would lose its affinity after the cross-linking reaction. The cytotoxicity, specificity and selectivity of 5-Fu-PNA wereexamined on the human colorectal cancer cell line LoVo and the human normal liver cell line Chang using MTT assay.Compared with free drug, the active conjugate, which maintained the affinity of lectin, had similar cytotoxic effect on LoVocells with much lower cytotoxicity on Chang cells ðp , 0:05Þ: On the other hand, lower cytotoxic effects on LoVo cells wereobserved for the non-active conjugate even at higher drug concentrations. The cytotoxic effect of conjugate was specificbecause only the active conjugate could inhibit the growth of LoVo cells in a dose- and time-dependent manner as that of thefree drug. The achieved results indicate the significance to maintain the affinity of lectin for lectin-mediated cytotoxicity. Still,the potential of 5-Fu-PNA conjugate as a targeting agent for colorectal cancer needs to be further investigated in vivo.
Keywords: Lectin, 5-Fu-PNA conjugate, targeted drug delivery, cytotoxicity
Introduction
Cancer chemotherapy often causes unwanted side
effects due to the low selectivity of chemotherapeutic
agents for cancer cells. Drug targeting, which could
carry drugs directly to the target sites via specific ligands
against receptors expressed on malignant cells, is one
approach to overcome this problem. However, reali-
zation of targeted drug delivery was hampered by the
difficulties in finding appropriate carrier molecules. An
appropriate carrier molecule plays an important role for
a successful targeted delivery system.
Cancerous cells often express different glycans,
compared with their normal counterparts (Shanghal
and Hakamori 1990). Lectin recognizes glycans on cell
surface with a high degree of specificity. Therefore,
lectins are proposed tobe promising carrier molecules to
target drugs specifically to different cells and tissues
(Bruck et al. 2001, Yi et al. 2001, Wirth et al. 2002).
In literatures, lectin-mediated drug delivery systems
have showed their advantages (Russell-Jones et al. 1999,
Smart 2004, Jepson et al. 2004).
The lectin of peanut (Arachis hypogaea) agglutinin
(PNA) is a tetramer, carbohydrate-free protein
composed of four identical subunits exhibiting
a molecular weight of 110 kDa (Lotan et al.
1975). PNA has been extensively used as a probe to
detect malignant phenotype in several tissues as
it strongly recognizes the cancer specific T antigen
ISSN 1061-186X print/ISSN 1029-2330 online q 2005 Taylor & Francis Group Ltd
DOI: 10.1080/10611860500138505
Correspondence: Z. -R. Zhang, Key Laboratory of Drug Targeting and Novel Drug Delivery Systems, West China School of Pharmacy,Sichuan University, No. 17, Block 3, Southern Renmin Road, Chengdu 610041, P.R. China. Tel: 86 28 85501566. Fax: 86 28 85456898.E-mail: [email protected]
Journal of Drug Targeting, May 2005; 13(4): 251–257
Jour
nal o
f D
rug
Tar
getin
g D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y M
ichi
gan
Uni
vers
ity o
n 11
/03/
14Fo
r pe
rson
al u
se o
nly.
(Gal-1-3GalNAc-) (Hwang et al. 1999, Chacko and
Appukuttan 2001). In addition, PNA receptors were
highly expressed in gastric and intestinal (GI) cancer,
but was not expressed or weakly expressed on normal
tissues (Kuang and Lei 1993, Lu and Xu 1997). These
data suggest the excellent selectivity of PNA towards
GI cancer.
To exploit the lectin-sugar binding selectivity, PNA
was chosen as the carrier molecule for the targeting
delivery of cytostatic agents to the colorectal cancer
tissue. The 5-Fu was used as the model drug to obtain
a targeting conjugate. The conjugate was characteri-
zed with SDS-PAGE. The affinity of the conjugate
was tested using haemagglutination assay and the
drug/lectin molar ratio was determined with TNBS
method. In order to examine the effectiveness of the
targeting system in comparison to the free drug, the
cytotoxicity of the drugs were assayed on LoVo cells
using MTT assay. The cytotoxic selectivity of the free
drug and the active conjugate was detected on normal
Chang cells. The specificity of the active conjugate was
also examined using MTT assay.
Materials and methods
Chemicals and animals
N-(3-dimethylaminopropyl)-N0-ethyl carbodiimide
hydrochloride (EDC), N-hydroxy succinimide (NHS),
N-morpholino propanesulfonic acid (Mops), trinitro-
benzenesulfonic acid (TNBS) and 3-(4,5-dimethyl-2-
tetrazolyl)-2,5-diphenyl-2H tetrazolium bromide
(MTT), were purchased from Sigma (St. Louis, MO,
USA). The PNAwasobtained from VectorLaboratories
Inc. (Burlingame, CA, USA). Cell culture medium
RPMI-1640 and DMEM were from Gibco Co. (USA).
The 5-Fu was from Nantong pharmaceutical Co. Ltd.
(Jiangsu, P.R.China). D-galactose (D-gal) was supplied
by Amresco (USA). All other chemicals were of
analytical grade obtained commercially.
MiceC57BL/6were culturedbyExperimentalAnimal
Center of Huaxi, Sichuan University, P.R. China.
Conjugation of 5-Fu to PNA
N1-substituted 5-Fu acetate was synthesized in two
steps according to the literature (Dong and Yu 1996)
with modifications. First, 5-Fu (3.90 g), K2CO3
(2.76 g) and KI (0.5 g) were dissolved in 10 ml
of DMSO. After heated to 708C, the mixture
was treated with ethyl chloro acetate (3.68 g)
and stirred overnight. After evaporation of the
solvent, the residue was extracted with ethyl
acetate and purified by silica gel column chromato-
graph (ethyl acetate/cyclohexane ¼ 2:1) to obtain an
ethyl N1-substituted 5-Fu acetate (2.71 g, mp
158–1608C, 1H-NMR(CDCl3) d 8.77(1H,br,N3-
H), 7.20(1H,d,JH–F ¼ 5:25;C6-H), 4.27(2H,s,N1-
CH2), 4.26(2H,q,O-CH2)). Secondly, the ethyl
N1-substituted 5-Fu acetate were dissolved in 10 ml
of HCl and refluxed for 5 h. After the solvent was
evaporated off, the residue was recrystallized from
H2O.
Cross-linking of N1-substituted 5-Fu acetate to
PNA was carried out via EDC/NHS using a single-
step method as follows: N1-substituted 5-Fu acetate
was activated by reacting with EDC and NHS in a
buffered solution (0.05 M Mops, pH ¼ 5:5) for 3 h at
the room temperature to form an active ester
(Grabarek and Gergely 1990). The molar ratio of
the carboxylic acid groups, EDC and NHS was 1:5:2.
PNA was incubated with or without D-gal in Mops
(0.05 M, pH ¼ 7:5) solution for 0.5 h before its
addition to the active ester solution. Then the pH
was adjusted to 7.5 by stepwise addition of 1 M NaOH
solution. The cross-linking of active ester to PNA
lasted for 12 h at 48C. The reaction solution was
applied to a column of Sephadex G-25. The column
was eluted with a solution of PBS (0.01 M, pH 7.2) at
a rate of 5 ml/min. Gel filtration on Sephadex G-25
provided two peaks and the fraction of the first peak
was collected, tested and lyophilized. The property of
the conjugate was detected on SDS-PAGE, which was
performed at pH 8.9 in 15% acrylamide running gels
and 4% stacking gels in the presence of 0.1% sodium
dodecyl sulfate at 140 V for 4 h. The gels were stained
with Coomassie brilliant blue R250. The result was
analysed with the JEL-PRO software (Mediacyber-
netics, USA). The drug/lectin molar ratio was
determined by TNBS method (Morcol et al. 1997).
Haemagglutination test
The affinity of PNA after covalent coupling was
determined by haemagglutination test because the
lectin could agglutinate thymus cells from mice
C57BL/6 (Gu and Li 1983).
According to the requirements of the National Act
on the use of experimental animals (P.R. China), the
Sichuan University animal ethical experimentation
committee approved all procedure of the studies
in vitro.
Briefly, the thymus was excised after sacrificing and
rinsed in isotonic PBS (pH7.2). The thymus was
chipped into single cells and the cell concentration was
adjusted to 2% (w/v) with the same PBS ready to be
used. A measure of 25ml of the conjugate solutions for
each well was added to a “V-bottom” 96-well plate
and diluted successively. Thereafter, 25ml of the cell
suspension was added when shaking and the plate was
incubated for 1.5–2 h. Affinity was confirmed by
comparing the value of titre of dilution giving the last
visible agglutination against a positive control (PNA
solution) and negative control (25ml of the cell
suspension added with 25ml of PBS). Agglutination is
judged by the flocculation of cells, which appears as a
diffuse milk-white color. In the case of a negative
Q. Cai & Z.-R. Zhang252
Jour
nal o
f D
rug
Tar
getin
g D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y M
ichi
gan
Uni
vers
ity o
n 11
/03/
14Fo
r pe
rson
al u
se o
nly.
result, there is a bright milk-white spot in the bottom
of the well due to the sinking of non-agglutinated cells.
All tests were done in triplicate at 258C.
Cell culture
The human colon carcinoma cell line LoVo and
the human normal liver cell line Chang were obtained
from Shanghai Cell Institute, China Academy of
Sciences. LoVo cells and Chang cells were cultivated
in RPMI 1640 and DMEM, respectively with 10%
(v/v) fetal calf serum (FCS), penicillin (100 U ml21)
and streptomycin (100 U ml21). The cells were
cultured at 378C in a humidified 5% CO2/95% air
incubator. After growing to a confluent monolayer,
cells were passaged by removing the adherent cells
with trypsin (in buffered saline, pH 7.2). Cell viability
was assessed by means of the trypan blue exclusion
method.
In vitro cytotoxic effect of 5-Fu-PNA conjugate
MTT assay was employed to determine the number
of surviving cells. The cytotoxicity of 5-FU in vitro
was evaluated either as the free drug or the conjugated
form. MTT assay is based on the ability of the
mitochondrial enzymes in living cells to convert a
yellow MTT tetrazolium salt into a blue MTT
formazan thus the amount of formazan present is
proportional to the number of viable cells. MTT
assay detects living cells with a high degree of precision
(Gieni et al. 1995).
The experiment was processed as follows: Malig-
nant cells were continuously treated at various
concentrations. A single cell suspension was obtained
and the cell number (counted with a haemocytometer)
was adjusted to 4 £ 104 cells per ml with culture
medium containing 10% FCS. A measure of 100ml of
cell suspension were added into the wells of flat-
bottom 96-well plates. After the cells adhered, 10ml of
solution was added per well containing appropriate
amount of 5-FU either in free state or conjugated
form. The plates were incubated for periods of time
ranging from 24 to 120 h at 378C and the surviving
cells were assayed for reduction of MTT. Each sample
was assayed in triplicate at each time period. At
intervals of each experimental time, 20ml of MTT
solution (5 mg/ml) was added to each well and the
plate was placed in the incubator for 4 h. The culture
medium was removed by aspiration. Thereafter,
100ml of DMSO was added to each well to dissolve
the formazan crystals while slightly stirring the cells
using an automated shaker. The absorbance of the
suspension was measured at 570/630 nm on an ELISA
plate reader (Bio-Rad, Microplate Reader 550). All
the results were expressed as % cell survival of the
control (untreated) over time. Background absor-
bance levels, determined in wells containing no cells,
were subtracted from the experimental and control
values.
Specificity of 5-Fu-PNA conjugate
In order to confirm the specificity of PNA-mediated
cytotoxic effect of 5-Fu, PNA, the active conjugate or
the non-active conjugate was added to LoVo cells,
respectively. The active and non-active conjugate were
obtained with or without the D-gal protection when
prepared, and the drug/lectin molar ratio of the non-
active conjugate was higher than that of the active
conjugate (see Results section). In brief, the cells were
seeded into 96-well plate at a concentration of
4 £ 103 cells per well. A measure of 10ml of solution
containing PNA, the active conjugate or the non-
active conjugate at various concentrations was added,
respectively, into the adherent cells with equivalent
amounts of lectin for 48 h or at certain concentration
for different incubation times. Subsequently, 20ml of
MTT solution was added to each well and treated as
the above in vitro cytotoxicity assay.
Selectivity of 5-Fu-PNA conjugate
For assessment of the cytotoxic selectivity of the
conjugate, the normal Chang cells were used. The
single cell suspension of Chang cells was obtained and
the cell number was adjusted to 4 £ 104 cells/ml with
DMEM medium. A measure of 100ml of cell
suspension was seeded into each well in a flat-bottom
96-well plate. When the cells were adhered to the
plate, the free drug and the active conjugate with
equivalent amounts of 5-Fu were added at gradually
increased concentrations and maintained for 48 h at
378C. Then 20ml of MTT solution was added and
treated the same as the in vitro cytotoxicity assay.
Results
Preparation and characteristics of 5-Fu-PNA conjugate
To explore the potential of PNA as a carrier molecule for
targeted drug delivery of 5-Fu to colorectal cancer, the
cytostatic agent was attached covalently to the protein
through EDC/NHS by two-step zero-length cross-
linking. First, the drug was converted to the correspond-
ing carboxylic-acid derivative. The object was obtained
with a yield of 76%. As indicated by 1H-NMR (DMSO-
d6) d 13.21 (1H,br,COOH), 11.92(1H,s,N3-H),
8.08(1H,d,JH–F ¼ 6:8;C6-H), 4.36 (2H,s, N1-CH2),
mp 245–2478C, 5-Fu was converted to the correspond-
ing derivative of N1-substituted 5-Fu acetate.
To avoid cross-linking of PNA molecules by
EDC, the N1-substituted 5-Fu acetate was reacted
with EDC/NHS to form an active ester intermedi-
ate (Grabarek and Gergely 1990) prior to coupling.
Afterwards, the drug was conjugated to accessible
amino-residues of PNA forming an amide-bond to
Lectin-mediated cytotoxicity 253
Jour
nal o
f D
rug
Tar
getin
g D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y M
ichi
gan
Uni
vers
ity o
n 11
/03/
14Fo
r pe
rson
al u
se o
nly.
get 5-Fu-PNA conjugate. The result of SDS-PAGE
confirmed the covalent attachment of 5-Fu to PNA.
On the lane of conjugate, there was a new band
with a little higher molecular weight than that of
PNA (Figure 1), which migrated to a distance
corresponding to molecular weight of 33,
323 ^ 424 relative to markers of 116.66, 66.2, 45,
35, 25, 18.4 and 14.4 kDa. The number of moles
5-Fu/mol PNA resulted by the new band was
14.85, which is similar to the result obtained by
TNBS method.
The affinity of the conjugate was tested by
haemagglutination test due to the agglutination of
the thymus cells from mice C57BL/6 by PNA (Gu and
Li 1983). The active conjugate was obtained with
D-gal protection. Titre value for the active conjugate
was in the same order as that obtained for PNA test
solution thus confirming lectin’s activity. The
obtained conjugate without D-gal protection was
non-active, which exhibited negative results as early
as the first dilution. The drug/lectin molar ratio
determined by TNBS of the active conjugate and the
non-active conjugate were 38.42 and 62.13%,
respectively.
In vitro cytotoxic effect of conjugate on LoVo cells
In order to provide comparability of the obtained data,
concentrations of 5-Fu of the active conjugate applied
to MTT test were equivalent to that of the free drug.
All results are expressed as % cell survival of the
control (untreated) cells over time. Results presented
in Table I and Figure 2 show the survival cells treated
by the free drug and the active conjugate at different
concentrations for 48 and 120 h, respectively. It could
be observed that there was a specific time- and
concentration-dependent antiproliferative effect of the
free drug and the active conjugate. Little toxic effect
on the cells was observed exposing the cells to the
active conjugate for 24 h. However, the cell cytotox-
icity was observed with time going on (Figure 2).
Specificity of 5-Fu-PNA conjugate
The cytotoxicity of PNA, the active conjugate and the
non-active conjugate were all tested on LoVo cells to
identify the specificity of the active conjugate. As the
concentration went up to 60mg/ml of PNA, no
significant inhibition effects were observed in the cells
either treated with PNA alone or the non-active
conjugates. On the other hand, the active conjugate
exhibited high cytotoxicity to the cells (see Figures 3
and 4). Therefore, the maintenance of affinity is of
great importance for the conjugate to exhibit its
cytotoxic effect. Only when lectin retained its affinity
did the conjugate inhibit the growth of cancer cells.
Cytotoxic selectivity of 5-Fu-PNA conjugate
In the present study, PNA was chosen to be a targeting
carrier to reduce the side effect of 5-Fu to normal
Figure 1. Vertical electrophoresis of PNA and conjugates on the
polyacrylamide gels. The direction of migration was from the top.
Electrophoresis was performed in 15% acrylamide running gels and
4% stacking gels in the presence of 0.1% sodium dodecyl sulfate at
pH 8.9 at 140 V for 4 h. The gels were stained for protein with
Coomassie brilliant blue R250. The result was analysed with the
JEL-PRO software (Mediacybernetics, USA). Lane 1: PNA; Lane2:
active conjugate; Lane 3: protein Markers of 116.66 kDa, 66.2 kDa,
45 kDa, 35 kDa, 25 kDa, 18.4 kDa and 14.4 kDa.
Table I. Cytotoxic effect of the free drug and the active conjugate on LoVo cells for 120 h treatment at various concentrations.
Free drug
(concentration mm)
Conjugate
(concentration mM)
Time (h) 1 2 4 8 1 2 4 8
24 95.75 ^ 5.33 92.66 ^ 8.32 73.05 ^ 4.45 57.52 ^ 3.36 99.71 ^ 8.17 89.07 ^ 6.80 75.37 ^ 5.82 68.06 ^ 9.01
48 85.45 ^ 9.64 69.86 ^ 7.00 40.91 ^ 9.58 23.13 ^ 6.56 95.73 ^ 6.49 76.81 ^ 9.35 62.65 ^ 7.71 47.63 ^ 12.81
72 67.53 ^ 12.90 55.20 ^ 8.78 16.68 ^ 6.92 10.76 ^ 9.02 84.26 ^ 11.6 59.06 ^ 4.73 30.89 ^ 8.81 21.54 ^ 15.44
96 60.48 ^ 11.58 43.35 ^ 5.86 2.24 ^ 1.55 28.67 ^ 6.27 61.28 ^ 8.94 47.64 ^ 9.97 14.09 ^ 7.11 5.32 ^ 3.28
120 53.10 ^ 8.27 35.49 ^ 5.24 0.80 ^ 0.37 A 50.23 ^ 7.59 39.96 ^ 6.96 4.51 ^ 3.64 22.78 ^ 1.41
All the values are means of surviving LoVo cells assessed by MTT assay. Data are expressed as a % of the control values (untreated
cells)(mean ^ SD, n ¼ 3). A: results not determined.
Q. Cai & Z.-R. Zhang254
Jour
nal o
f D
rug
Tar
getin
g D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y M
ichi
gan
Uni
vers
ity o
n 11
/03/
14Fo
r pe
rson
al u
se o
nly.
cells. Human normal Chang cells were used to
investigate the selectivity of the active conjugate. The
cells were treated either by the free drug or the active
conjugate for 48 h at 378C. The results were indicated
in Figure 5.
Discussion
PNA, as the carrier molecule, was conjugated with
5-Fu via an amide linkage. The affinity of PNA was
preserved after its coupling to 5-Fu when D-gal was
used for protection. Without D-gal protection, PNA
will lose its affinity after cross-link reaction, which is
due to the formation of amide bonds between
carboxylic acids of N1-substituted 5-Fu acetate and
amines of PNA. However, the free amino group at C-2
of PNA is required for its affinity to bind sugar (Lotan
et al. 1975).
SDS, which is an amphipathic molecule and
protein’s strong denaturing agent, binds to most
of proteins with the same ratio of 1.4:1 and causes
proteins to assume a rod-like shape. The large negative
charge that the SDS imparts masks the protein’s
intrinsic charge so that SDS-treated proteins tend to
have identical charge-to-mass ratios and similar
shapes. Therefore, SDS-PAGE separates proteins in
order of their molecular masses because of gel
filtration effects. The relative mobilities of proteins
on such gels vary linearly with the logarithm of their
molecular masses. In addition, SDS treatment
disrupts the noncovalent interactions between
protein’s subunits so SDS-PAGE measures the
molecular masses of the subunits rather than that of
the intact protein. Since PNA has four identical
subunits, there was only one band in the lane of PNA
in SDS-PAGE figure. After 5-Fu molecules were
conjugated to part of the subunits of PNA,
the molecular weight of which was increased.
Consequently, there was a new band in the lane of
conjugate with a slightly higher molecular weight than
that of PNA (as seen from Figure 1)
For treatment both with free drug and the active
conjugate, growth inhibition was observed in the
colorectal cancer cells under tested concentrations.
The cancer cell line showed a specific concentration-
and time-dependent sensitivity to drugs. The active
conjugate had similar cell killing ability to the free
drug. The survival rate was 48.48% for cells treated
with the free drug at the concentration of 4mM for
48 h, while the active conjugate inhibited the growth of
LoVo cells with a survival rate of 62.78% at the same
condition. It seems that the conjugated drug was less
active than free drug. Following are the underlying
reasons. Biomembrane is selectively permeable, i.e. it
allows certain substrates to cross by passive diffusion
or by active transport process, while restricting the
transfer of others. The passive diffusion is mainly
responsible for small molecules, while the active
process, which involves energy and sometimes
receptors as well, is to transport some special materials
such as large molecules. The 5-Fu-PNA conjugates
are water soluble and are of high molecular weight so
that they are unable to diffuse across biomembranes.
Their uptake into cells should be a receptor-mediated
Figure 2. Cytotoxic effect of the free drug and the active conjugate
on LoVo cells (A) Concentration-dependent curves for 48 h
treatment at gradually changed concentrations. (B) Time-
dependent curves for 120 h consecutive treatment at the
concentration of 4mM of 5-Fu. The viability of LoVo cells was
assessed by MTT assay after continuous exposure to the free drug
and the active conjugate. Data are expressed as % cell survival of the
control cells (untreated cells) (mean ^ SD, n ¼ 3).
Figure 3. The inhibition effect of PNA, the active conjugate and
the non-active conjugate on LoVo cells for 48 h treatment at various
concentrations. The viability of LoVo cells was assessed by MTT
assay after continuous exposure to the free PNA and the conjugates.
Data are expressed as % cell survival of the control cells (untreated
cells) (mean ^ SD, n ¼ 3).
Lectin-mediated cytotoxicity 255
Jour
nal o
f D
rug
Tar
getin
g D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y M
ichi
gan
Uni
vers
ity o
n 11
/03/
14Fo
r pe
rson
al u
se o
nly.
active pathway, which could be speculated from
the result of specificity experiment. Thereby, the
relative rate of uptake into cells became the primary
factor influencing the cytotoxicity of the conjugate.
After uptake into cells, the drug needs to be released
from the conjugate and to transport across the cellular
compartment membrane such as lysosome and
endosome into the cytoplasm. Thus, the rate and
efficiency of the release and the rate of diffusion may
lead to the decreased cytotoxicity of the conjugate.
According to Figure 2B, it took 96 h for the free drug
at the concentration of 8mM to kill almost all the cells
while for the active conjugate, it took 120 h.
Compared with the free drug, there was a time-
delayed effect of the active conjugate since it took
some time for the drugs to be released from the
conjugate and to transport across the biomembrane of
cellular compartment before they began to work.
Instead of the non-active conjugate, D-gal was used
to assess the specific binding of conjugate to LoVo
cells in our preliminary experiments. However, D-gal
could inhibit the growth of LoVo cells (data not
shown). Consequently, the non-active conjugate was
chosen in the specificity binding experiment. As a
result, the carrier protein alone had little effect on the
growth of LoVo cells, which is in accordance with the
literature (Kiss et al. 1997). The active conjugate with
affinity showed parallel cell killing effect to the free
drug. Meanwhile, little inhibition was detected after
the non-active conjugate was added to the cancer cells.
Therefore, it should be lectin that mediated the
conjugate delivery into cells. It can also be inferred
that the affinity and specificity of targeting ligands play
an important role in successful cellular delivery of the
conjugate.
A good targeting delivery system should identify the
target tissue or cells from the non-target ones. In order
to examine the selectivity of the conjugate, Chang cells
were employed. In this study, 5-Fu-PNA conjugate
exhibited much higher selectivity than the free drug
since it killed colorectal cancer cells LoVo selectively
and exhibited weak effect on the growth inhibition of
the normal liver cells Chang. However, the free drug
killed both.
Although 5-Fu-PNA conjugate was less cytotoxic
than the free 5-Fu to LoVo cells in vitro, the antitumor
effect of the conjugate in vivo may be higher than that
of free drug because it has much higher selectivity,
which might be of benefits to the accumulation of
conjugates at the tumor site. Such cases were reported
in literatures on studies of monoclonal antibody-drug
conjugate (Deguchi et al. 1986, Shawler et al. 1988).
However, the antitumor effect of 5-Fu-PNA in vivo
still needs further investigation.
Conclusion
In this study, we synthesized lectin-mediated targeting
system of 5-Fu-PNA, which fulfilled the principal
requirements for targeted systems such as binding
specificity, low toxic effect on normal cells, etc. Based
upon these optimistic results, 5-Fu-PNA conjugate
may be a promising target system, which should be
further investigated in vivo for site-specific delivery to
colon carcinoma.
Acknowledgements
This work was sponsored by the National Natural Fund
for Distinguished Young Scholars (No. 39925039) of
People’s Republic of China.
References
Bruck A, Abu-Dahab R, Borchard G, Schafer UF, Lehr CM. 2001.
Lectin-functionlilzled liposomes for pulmonary drug delivery:
Interaction with human alveolar epithelial cells. J Drug Target
9:241–251.
Figure 4. The inhibition effect of the active and the non-active
conjugate on LoVo cells for 120 h treatment at the concentration of
40mg PNA. The viability of LoVo cells was assessed by MTT assay
after continuous exposure to the activeand the non-active conjugates.
Data are expressed as % cell survival of the control cells (untreated
cells) (mean ^ SD, n ¼ 3).
Figure 5. The inhibition effect of the free drug and the active
conjugate on Chang cells for 48 h treatment at various
concentrations. The cells survival of Chang cells was assessed by
MTTassay after continuous exposure to the free drug and the active
conjugates. Data are expressed as % cell survival of the control cells
(untreated cells) (mean ^ SD, n ¼ 3).
Q. Cai & Z.-R. Zhang256
Jour
nal o
f D
rug
Tar
getin
g D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y M
ichi
gan
Uni
vers
ity o
n 11
/03/
14Fo
r pe
rson
al u
se o
nly.
Chacko BK, Appukuttan PS. 2001. Peanut (Arachis hypogaea) lectin
recognizes a-linked galactose, but not N-acetyl lactosamine
in N-linked oligosaccharide terminals. Int J Biol Macromol
28:365–371.
Deguchi T, Chu TM, Leong SS, Horosowicz JS, Lee C-L. 1986.
Effect of methotrexate-monoclonal anti-prostatic acid phospha-
tase antibody conjugate on human prostate tumour. Cancer Res
46:3751–3755.
Dong JC, Yu LM. 1996. Synthesis of ethyl 5-fluorouracil acetate
derivatives. Chin J Med Chem 6:288–289, 302.
Gieni RS, Li Y, HayGlass KT. 1995. Comparison of [3H] thymidine
incorporation with MTT- and MTS-based bioassays for human
and murine IL-2 and IL-4 analysis Tetrazolium assays provide
markedly enhanced sensitivity. J Immun Methods 187:85–93.
Grabarek Z, Gergely J. 1990. Zero-length cross-linking procedure
with the use of active esters. Anal Biochem 185:131–135.
Gu BQ, Li H. 1983. The purification and specificity of peanut
agglutinin and its effect on the classification of thymus.
J Shandong Univ Med Sci 1:16–18.
Hwang IR, Nahm DH, Cho SN, Longenecker BM, Rao RK,
Park IS. 1999. Anti-Tantibodies and peanut-agglutinin-binding
glycoproteins in sera of patients with gastric cancer. J Cancer Res
Clin Oncol 125:582–587.
Jepson MA, Clark MA, Hirst BH. 2004. M cell targeting by lectins:
A strategy for mucosal vaccination and drug delivery. Adv Drug
Deliv Rev 56:511–525.
Kiss R, Camby I, Duckworth C, De Deckers R, Salmon I, Pasteels J-L,
Danguy A, Yeaton P. 1997. In vitro influence of Phaseolus vulgaris,
Griffonia simplicifolia, concanavalin A, wheat germ, and peanut
agglutinins on HCT-15, LoVo, SW837 human colorectal cancer
cell growth. Gut 40:253–261.
Kuang XW, Lei FS. 1993. Histochemical study of large bowel
carcinoma and chronic ulcerative colitis by peanut agglutinin.
Tumor 13:209–212.
Lotan R, Skuteslsky E, Danon D, Sharon N. 1975. The purification,
composition, and specificity of the anti-Tlectin from peanut
(Arachis hypogaea). J Biol Chem 250:8518–8523.
Lu LH, Xu P. 1997. Variety of glycoprotein in the development
of gastric cancer and its clinical significance. Prac J Cancer
12:31–33.
Morcol T, Subramanian A, Velander WH. 1997. Dot-blot analysis
of the degree of covalent modification of proteins and antibodies
at amino groups. J Immuno Methods 203:45–53.
Russell-Jones GJ, Veitch H, Arthur L. 1999. Lectin-mediated
transport of nanoparticles across Caco-2 and OK cells. Int
J Pharm 190:165–174.
Shanghal A, Hakamori S. 1990. Molecular changes in carbohydrate
antigens associated with cancer. Bioessays 12:223–230.
Shawler DL, Johnson DE, Sweet MD, Myers LJ, Tudor SD, Beidler
DE, Koziol JA, Dillman RO. 1988. Preclinical trials using an
immunoconjugate of T101 and methotrexate in an athymic
mouse/human T-cell tumor model. J Biol Response Modifiers
7:608–618.
Smart JD. 2004. Lectin-mediated drug delivery in the oral cavity.
Adv Drug Deliv Rev 56:481–489.
Wirth M, Gerhardt K, Wurm C, Gabor F. 2002. Lectin-mediated
drug delivery: Influence of mucin on cytoadhesion of plant
lectins in vitro. J Control Rel 79:183–191.
Yi SM, Harson RE, Zabner J, Welsh MJ. 2001. Lectin binding and
endocytosis at the apical membrane of human airway epithelia.
Gene Ther 8:1826–1832.
Lectin-mediated cytotoxicity 257
Jour
nal o
f D
rug
Tar
getin
g D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y M
ichi
gan
Uni
vers
ity o
n 11
/03/
14Fo
r pe
rson
al u
se o
nly.