Riboflavin-targeted polymer conjugates for breast tumor delivery

Article abstract of DOI:10.1007/s11095-013-1024-5

Purpose: In breast cancer, a significant decrease in riboflavin (RF) serum levels and increase in RF carrier protein occurs, indicating a potential role of RF in disease progression. To evaluate RF's ability to serve as a targeting agent, mitomycin C (MMC)-conjugated N-(2-hydroxypropyl)methacrylamide (HPMA) copolymers were synthesized and targeted to the RF internalization pathway in human breast cancer cells. Methods: Competitive uptake studies were used to determine specificity of RF-targeted conjugates, and an MTT assay established the IC₅₀ for the conjugates. Endocytic mechanisms were investigated by confocal microscopy. Results: Studies revealed a high-affinity endocytic mechanism for RF-specific internalization of fluorescently-labeled conjugates in both MCF-7 and SKBR-3 cells, whereas folic acid-mediated endocytosis showed high specificity only in SKBR-3 cells. MMC internalization was significantly higher following nontargeted and RF-targeted MMC-conjugate administration compared to that of free MMC. Cytotoxic analysis illustrated potent IC ₅₀ values for RF-targeted MMC conjugates similar to free MMC. Maximum nuclear accumulation of MMC resulted from lysosomal release from RF-targeted and nontargeted MMC-conjugates following 6 h incubations, unlike that of free MMC seen within 10 min. Conclusion: Targeting polymer-MMC conjugates to the RF internalization pathway in breast cancer cells enabled an increase in MMC uptake and nuclear localization, resulting in potent cytotoxic activity.

Full text of DOI:10.1007/s11095-013-1024-5

Pharm Res
DOI 10.1007/s11095-013-1024-5
RESEARCH PAPER
Riboflavin-Targeted Polymer Conjugates for Breast Tumor
Delivery
Lisa M. Bareford & Brittany R. Avaritt & Hamidreza Ghandehari & Anjan Nan & Peter W. Swaan
Received: 10 January 2013 /Accepted: 4 March 2013
#
Springer Science+Business Media New York 2013
Conclusion Targeting polymer-MMC conjugates to the RF
internalization pathway in breast cancer cells enabled an in-
crease in MMC uptake and nuclear localization, resulting in
potent cytotoxic activity.
ABSTRACT
Purpose In breast cancer, a significant decrease in riboflavin
(RF) serum levels and increase in RF carrier protein occurs,
indicating a potential role of RF in disease progression. To
evaluate RF’s ability to serve as a targeting agent, mitomycin C
(MMC)-conjugated N-(2-hydroxypropyl)methacrylamide
(HPMA) copolymers were synthesized and targeted to the RF
internalization pathway in human breast cancer cells.
Methods Competitive uptake studies were used to determine
specificity of RF-targeted conjugates, and an MTT assay
established the IC50 for the conjugates. Endocytic mechanisms
were investigated by confocal microscopy.
Results Studies revealed a high-affinity endocytic mechanism
for RF-specific internalization of fluorescently-labeled conjugates
in both MCF-7 and SKBR-3 cells, whereas folic acid-mediated
endocytosis showed high specificity only in SKBR-3 cells. MMC
internalization was significantly higher following nontargeted and
RF-targeted MMC-conjugate administration compared to that
of free MMC. Cytotoxic analysis illustrated potent IC50 values
for RF-targeted MMC conjugates similar to free MMC.
Maximum nuclear accumulation of MMC resulted from lyso-
somal release from RF-targeted and nontargeted MMC-
conjugates following 6 h incubations, unlike that of free MMC
seen within 10 min.
.
.
KEY WORDS breast cancer nanomedicine
.
.
receptor-mediated endocytosis riboflavin targeted delivery
ABBREVIATIONS
DIPEA
DMEM Dulbecco’s modified eagle medium
DMF Dimethylformamide
DMSO Dimethyl sulfoxide
Diisopropyl ethylamine
DPBS
EDTA
FA
Dulbecco’s phosphate buffered saline
Ethylenediaminetetraacetic acid
Folic acid
FITC
HBSS
HPMA
Fluorescein isothiocyanate
Hank’s balanced salt solution
N-(2-hydroxypropyl)methacrylamide
LAMP-1 Lysosomal-associated membrane protein 1
MMC
MTT
Mitomycin C
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide
RCP
RF
Riboflavin carrier protein
Riboflavin
:
:
:
:
RME
TFA
THF
Receptor-mediated endocytosis
Trifluoroacetic acid
Tetrahydrofuran
L. M. Bareford B. R. Avaritt H. Ghandehari A. Nan
P. W. Swaan (*)
Department of Pharmaceutical Sciences
Center for Nanomedicine and Cellular Drug Delivery
University of Maryland Baltimore
20 Penn Street, Health Sciences Facility 2, Room 543
Baltimore, Maryland 21201, USA
e-mail: pswaan@rx.umaryland.edu
INTRODUCTION
Present Address:
H. Ghandehari
Chemotherapeutic agents lack drug specificity at the tumor site,
thereby significantly limiting their therapeutic utility. To con-
trol the indiscriminate diffusion of anticancer drug molecules,
macromolecular vehicles have been extensively investigated as
Department of Pharmaceutics & Pharmaceutical Chemistry
& of Bioengineering, Utah Center for Nanomedicine, Nano Institute
of Utah, University of Utah, Salt Lake City, Utah 84108, USA

Bareford et al.
an alternative delivery strategy. Conjugating chemotherapeu-
tics to biocompatible carriers offers many advantages including
enhanced drug solubility, increased tumor accumulation, and
reduced toxicity (1,2). Being primarily limited to endocytic
internalization, nanoscale drug conjugates are retained in
blood circulation longer, enabling passive accumulation in
tumor tissue by means of the enhanced permeability and
retention (EPR) effect (3) or by active targeting using
receptor-mediated endocytosis (RME). In addition, vesicular
internalization may potentially avoid cellular drug-resistance
mechanisms, such as efflux transporters, to permit higher cyto-
solic drug concentrations (4). Ligand-targeted therapeutics take
advantage of their intracellular localization to the acidic and
enzyme-rich endolysosomal compartments through biodegrad-
able spacers which are pH (5) and enzyme (6) sensitive.
Cell-surface vitamin-B receptors facilitate the efficient ab-
sorption of nutrients that are essential for normal cellular
growth and metabolism, but particularly important for the
exponential development of tumor cells (7). High-affinity
RME facilitates the efficient absorption of B-vitamins follow-
ing dietary intake (8) and certain vitamin receptors are signif-
icantly upregulated in cancer cells (7,9), allowing for possible
tumor-specific delivery of chemotherapeutics (10). However,
the extent of vitamin receptor overexpression and ligand-
selectivity varies by cancer type (11). For example, the folate
receptor (FRα) is overexpressed in over one-third of human
cancers (12) including 90% of ovarian cancers, but less than
25% of the various breast cancer forms (13). Overall, consid-
erable evidence supports the therapeutic utility of vitamin-
targeted drug conjugates.
Breast cancer remains the highest cause of malignancy-
related death for women in the United States (14). A specific
plasma protein marker for breast cancer, riboflavin carrier
protein (RCP), was identified following clinical studies that
revealed elevation in RCP serum levels (15) with simultaneous
decrease in RF levels (16) in breast cancer patients. This effect
is suggested to be the result of an upregulation in RF-specific
cell surface receptors and RF internalization similar to other
vitamin systems (7). While studies of avian-RCP suggest the
protein aids in the recruitment of RF to a RF-specific plasma
membrane associated receptor (17), the sequence of the hu-
man ortholog of RCP has yet to be elucidated, and no RF
receptor has been identified. Kinetic analysis of RF internal-
ization was conducted in a variety of cell lines where all
systems revealed a high-affinity, saturable mechanism that is
Na+-, potential-, and pH-independent, suggesting an RF-
specific carrier-mediated system (18). Further characterization
in human breast cancer cells demonstrated RF absorption to
occur predominantly via RME at physiological concentrations
(~ 12 nM) (19,20) prior to vesicular trafficking to subcellular
organelles (21). This data suggests a clinical application for RF
as a novel targeting agent for breast cancer-directed
chemotherapeutics.
To utilize RF targeting the water-soluble polymer N-(2-
hydroxypropyl)methacrylamide (HPMA) was employed as a
biocompatible drug delivery vehicle due to its versatile syn-
thetic chemistry (22–24). The highly effective breast cancer
compound mitomycin C (MMC) was selected because of its
significantly limited clinical utility (25). Both RF and folic
acid (FA) were used as targeting ligands to examine the
targeting potential of RF for the selective absorption of
polymeric-MMC conjugates in two breast cancer cell
models, MCF-7 and SKBR-3. The in-vitro efficacy of RF-
targeted MMC-conjugates was characterized by evaluating
cellular internalization, cytotoxicity, subcellular trafficking,
and drug stability.
MATERIALS AND METHODS
Chemicals
MMC was purchased from Bristol Laboratories (Evansville,
IN). Unless otherwise mentioned all other chemicals were
reagent grade and obtained from Sigma (St. Louis, MO).
Synthesis of Comonomers and Polymers
Synthesis of Monomers
HPMA comonomer (26) and peptide containing comonomers
N-methacryloyl-glycylphenylalanylleucylglycine p-nitro-
phenyl ester (MA-GFLG-ONp) (27), N-methacryloylgly-
cylglycine p-nitrophenyl ester (MA-GG-ONp) (28), and 5-[3-
(methacryloylaminopropyl)thioureidyl] fluorescein (MA-AP-
FITC) (29), were synthesized according to well established pro-
cedures. Methacryloyl-GFLG-MMC (MA-GFLG-MMC):
MA-GFLG-ONp was reacted with a reactive equivalent (eqv)
of MMC in N,N-dimethylformamide (DMF) at 4°C, followed
by the slow addition of an eqv of triethylamine. The mixture
was then left to react in the dark for 4 h at RT. After overnight
mixing at 4°C, the mixture was vacuum evaporated to remove
DMF and the resulting purple-colored solid product, MA-
GFLG-MMC, was extracted using diethyl ether precipitation.
Copolymerization
Experimental copolymers were synthesized by free radical
precipitation copolymerization of monomer subunits using
standardized reaction conditions (30). Fluorescently labeled
polymer conjugates were prepared by copolymerization of
HPMA (88 mol%), MA-GG-ONp (10 mol%), and MA-AP-
FITC (2 mol%) comonomers. Experimental drug polymer
conjugates were produced by copolymerization with HPMA
(85 mol%), MA-GG-ONp (10 mol%), and MA-GFLG-
MMC (5 mol%) comonomers. Control polymer conjugates

Riboflavin-Targeted HPMA Polymers
were prepared similarly, where MA-GFLG-ONp replaced
the drug-containing monomer MA-GFLG-MMC. For each
set of experimental conjugates, comonomers were solubilized
in acetone-dimethyl sulfoxide (DMSO < 10% total volume) at
12.5 wt % with 1.2 wt % radical initiator (2,2’-
azobisisobutyronitrile). Reaction mixtures were sealed under
nitrogen in an ampoule and left to react at 50°C for 24 h.
Precipitated polymers were dissolved in methanol and
reprecipitated in acetone-ether to obtain the pure product.
polymer-incorporated reactive Gly-Gly spacers (Fig. 1).
Fluorescently-labeled copolymers were prepared with either
RF (P1-RF) or FA (P1-FA) targeting moieties. Copolymers
both without (P2) and with (P3) drug were functionalized with
two different amounts of the RF targeting moiety (P2B/C and
P3B/C). Deprotected RFl or FAl were reacted overnight with
copolymer conjugates in DMSO containing an eqv of DIPEA.
Unreacted GG-ONp spacers were neutralized by 1-amino-2-
propanol hydrolysis. The resulting RF- and FA-copolymers
were dialyzed for 72 h in the dark to remove unconjugated
targeting moieties and then used for further characterization.
Neutral, nontargeted control copolymers were also prepared
for each of the copolymer formulations (P1, P2A, and P3A) by
ONp hydrolysis.
Synthesis and Polymer Functionalization of Targeting
Ligands
Synthesis of RF linker (RFl)
RF was amine-functionalized at its ribityl side chain to allow
for stable conjugation to polymer-incorporated Gly-Gly di-
peptide side chains. First, 10 mmol RF was reacted with
5 mmol succinic anhydride in DMSO including 5% (v/v)
pyridine overnight at 80°C to promote monosubstitution at
the primary hydroxyl position on the ribityl side chain of RF.
The monosubstituted product was purified with a silica gel
column using a methanol/chloroform gradient elution. Next,
the resulting RF succinamide was reacted with an eqv of O-
(N-succinimidyl)-N,N,N,N'-tetra-methyluronium tetrafluo-
roborate in DMF-dioxane solution containing 10% (v/v) wa-
ter and diisopropyl ethylamine (DIPEA) for 45 min. To this
mixture, 11 mmol of N-BOC-ethylenediamine in DMF was
added and left to react for 4 h. The product was concentrated
and purified using silica gel column chromatography using
5% MeOH in chloroform as the eluent. The resulting BOC-
protected functionalized RF was concentrated, washed with
acetone, and characterized (m/z 519 (MH+), decomposition
temperature 314–317°C).
Physicochemical Characterization of Experimental
Polymer Conjugates
UV–vis was used to measure the incorporation of MMC
(365 nm) in the drug-containing precursor and FITC
(495 nm) in the fluorescently labeled precursor using calibration
curves for MMC and FITC, respectively. The MA-GG-ONp
content of all precursors was assessed by absorbance measure-
ments (400 nm) of hydrolytically released ONp. Following
conjugation to the available ONp, absorbance measurements
of polymer-incorporated RF (455 nm) and FA (365 nm) were
determined using standard curves of the amine-functionalized
ligands. The average molecular weight (Mw) and molecular
weight distribution [polydispersity (n), calculated as av-
erage molecular weight (Mw)/number average molecular
weight (Mn)] were estimated by size exclusion chroma-
tography (SEC) on a Superose 12 column (10 mm×
30 cm) (Amersham Biosciences, Piscataway, NJ) using a
fast protein liquid chromatography (FPLC) system
(Amersham Biosciences) which was calibrated for poly
(HPMA) fractions of known molecular weights.
Synthesis of FA linker (FAl)
Following methods standardized by Luo et al. (31), 4.5 mmol
of folic acid was solubilized in tetrahydrofuran (THF), and
an excess of trifluoroacetic acid (TFA) was added drop wise.
The solution was stirred overnight at room temperature to
obtain N10-(trifluoroacetyl)pyrofolic acid. The pyrofolic ac-
id product was reacted with an eqv of N-BOC-
ethylenediamine in DMF at 40°C overnight. Folate-
ethylenediamine-BOC was then deprotected in neat TFA
to yield folate-ethylenediamine which was precipitated in
methanol to obtain the final FA-based targeting moiety
(m/z 355 (MH+), decomposition temperature 289–292°C).
Cell Culture
MCF-7 and SKBR-3 cells were obtained from American Type
Culture Collection (Manasas, VA), and maintained at 37°C,
under 5% CO2 in DMEM and McCoy’s 5a medium, respec-
tively, (Invitrogen Life Technologies, Carlsbad, CA)
supplemented with 10% fetal bovine serum, 1% nonessential
amino acids, 100 U/ml penicillin, and 100 μg/ml streptomycin.
HPMA-FITC Copolymer-Ligand Internalization
MCF-7 and SKBR-3 cells were seeded at 5×105 cells/ml
in 96-well plates and cultured until confluence prior to
experimentation. Cells were washed twice with 200 μl
pre-warmed HBSS, then incubated with 0–2,000 μg/ml
weight eqv of nontargeted (P1) and RF- (P1-RF) or FA-
Polymer Functionalization with Targeting Moieties
Experimental copolymers were functionalized by the addition
of targeting ligand following RFl or FAl conjugation to

Bareford et al.
Fig. 1 Synthesis of HPMA polymer constructs. HPMA copolymer (V) and dipeptide-containing monomer (W) functionalized with T1 or T2 were
copolymerized with FITC-conjugated monomer (X) or tetrapeptide-MMC monomer (Y) to produce the fluorescently-labeled and drug-containing polymer
conjugates, respectively. Neutral control polymers were produced to yield nontargeted (GG-OH) and drug-lacking (GFLG-OH, Z) test conjugates.
(P1-FA) targeted FITC-labeled conjugates at 37°C for 1 h.
Following incubations, cells were washed three times with
pre-chilled DPBS pH7.4 to remove uninternalized poly-
mers, followed by a 5 min wash with pre-chilled DPBS pH
3 to remove all membrane-associated fractions which asso-
ciate with membrane receptors at neutral pH. Cells were
then lysed for 2 h in 1 N NaOH. The intrinsic FITC fluo-
rescence (495/521 nm) within the conjugates was used to
determine the amount of cell-associated FITC from a stan-
dard curve and results were standardized to total protein
content. Intracellular FITC concentrations were then
converted to the amount of internalized conjugate using
the experimentally determined FITC content of the conju-
gate precursors. The affinity (km) and capacity (Jmax) of RF-
and FA-specific uptake was calculated using Michaelis-
Menten kinetics following correction for nonspecific uptake
of the nontargeted conjugate. Results were expressed as
mean standard deviation (SD).
Competitive Uptake Studies
Studies were conducted in Costar® 24-well plates (Fisher
Scientific, Pittsburg, PA) where MCF-7 cells were seeded at
a density of 5×104 cells/cm2 and used 3 days after seeding.
Competitive uptake studies were conducted to determine
the extent of RF-mediated uptake for targeted and
nontargeted conjugates. Cells were incubated for 30 min
with media solutions containing 10 nM [3H]-RF (Moravek
Biochemicals, Brea, CA) and 0–5,000 ng/ml wt eqv of RF-
targeted (P2B, P2C) or nontargeted (P2A) conjugates. Cells
were then washed three times with HBSS pH7.4 and once
with acidic HBSS pH3 for 10 min to remove all nonspecific
and membrane bound fractions. Following a 2 h cell lysis
with 1 N NaOH, cell-associated radioactivity was measured
with scintillation counting, and results were standardized to
total protein content. Statistical analysis was conducted
using a one-way analysis of variance (ANOVA) followed

Riboflavin-Targeted HPMA Polymers
by Newman-Keuls multiple comparison test. Results were
expressed as mean SD.
iteratively deconvolved for both channels and analyzed for
3D colocalized regions using Volocity (v. 3.6, Improvision,
Waltham, MA). The extent of colocalization between MMC
(405 nm), following its free and RF-targeted or nontargeted
conjugate administration, and protein markers (543 nm) was
determined by calculating the percent of total overlapping
volume between the two channels compared to the total
drug volume. The data represents the average of three re-
gions for each treatment, and statistical significance was
determined using Pearson’s correlation (PC).
Internalization of Free and Conjugated MMC
Non-radioactive uptake studies were conducted in a similar
manner as explained above, in order to measure the cellular
accumulation of MMC following passive and active inter-
nalization. MCF-7 cells were treated for 60 min with a
10 μg/mL MMC eqv of free and RF-targeted (P3B, P3C)
or nontargeted (P3A) conjugates and then washed to remove
un-internalized and membrane-associated MMC. In order
to determine the effect of compartmental neutralization on
the internalization of free and conjugated MMC, cells were
pretreated for 30 min with 25 μM monensin or 300 μM
primaquine (LKT Laboratories, St. Paul, MN) prior to
MMC uptake studies. Cells not treated with monensin or
primaquine were used as a control. Cell lysates were ana-
lyzed with UV spectroscopy to measure the absorbance of
intracellular MMC (365 nm) which resulted for each treat-
ment. The concentration of internalized MMC was extrap-
olated from a standard curve produced for this drug, and
the results are presented as a percentage of the MMC
internalization in the cells treated with the lysomotropic
agents versus the untreated control. Statistical analysis was
determined by ANOVA followed by Newman-Keuls multi-
ple comparison test. Results were expressed as mean SD.
In-Vitro Stability of HPMA Copolymer-RF-MMC
Conjugates
RF-containing HPMA copolymer-MMC conjugate (P3B) was
incubated in buffers designed to mimic the physiological and
lysosomal environments in order to study the hydrolytic and
enzymatic stability of spacers linking RF and MMC to the
copolymers (Table IV). To investigate hydrolytic stability,
1 mg/ml of conjugate was dissolved in phosphate buffers of
pH7.4 and 5.5, at 37°C. Enzymatic stability was determined
in phosphate buffer pH5.5 containing 2 mM EDTA, 10 mM
reduced glutathione, and 100 μg/ml human cathepsin B
(EMD Biosciences, La Jolla, CA). At regular time intervals
samples were taken and immediately stored at −80°C until
analysis. Prior to analysis, all samples were eluted on desalting
spin columns of 7 kD molecular weight cut off (Pierce
Biotechnology, Rockford, IL) to obtain the polymer fraction.
After elution, polymer and control samples were analyzed for
RF (445 nm) and MMC (365 nm) content using UV–vis. The
percent of bound RF and MMC remaining was determined
from the total content in the original copolymers. Time-
dependent release profiles were produced by nonlinear re-
gression to calculate the half-lives associated with RF and
MMC release in all incubation conditions.
Immunofluorescence and Confocal Microscopy
MCF-7 cells were seeded 3 days prior to use at 5×103
cells/cm2 in collagen-coated culture slides (BD Biosciences,
Bedford, MA). Cells were treated with 20 μg/mL MMC eqv
of free MMC and RF-targeted (P3B) or nontargeted (P3A)
MMC-conjugates at 37°C for 10 min, 30 min, and 6 h. In
order to prevent potential receptor saturation and modifi-
cation to RF intracellular kinetics, RF-targeted trafficking
was monitored with P3B conjugates which contain targeting
ligand. Cells were then fixed, permeabilized, and blocked
prior to immunolabeling for 2 h at RT with early endocytic
(Rab5 GTPase) or lysosomal (LAMP-1) protein markers.
Cells were then washed thoroughly and probed with
AlexaFluor546-labeled sheep anti-mouse IgG (Molecular
Probes, Eugene, OR) for 1 h. Nuclear labeling was
conducted for 1 h using propidium iodide. Fluorescent
samples were preserved using GelMount (Biomeda Corp.,
Foster City, CA) and visualized on a Nikon Eclipse TE2000
E inverted confocal laser scanning microscope equipped
with two fixed lasers, 405 and 543 nm. The 3D images were
acquired on Nikon EZ-C1 software (Gold v. 2.3, Image
Systems Inc., Columbia, MD) using a 60× oil objective for
parameters: 0.25 μm z-step, 6 μs scan dwell time, and the
average of two scans per channel. Each image was
Cytotoxicity Studies
The cytotoxicity of free and conjugated MMC was deter-
mined using the fluorescent (3-(4,5-Dimethylthiazol-2-yl)-
2,5-diphenyltetrazolium bromide (MTT) assay, which is a
standard colorimetric method for measuring the enzymatic
activity of live cells. MCF-7 cells were seeded at 5×104
cells/mL in 96 well plates (USA Scientific, Ocala, FL)
1 day prior to experimentation. Cells were treated in culture
medium for 72 h with 1–1,000 ng/mL MMC eqv of free
and RF-targeted (P3B, P3C) and nontargeted (P3A) conju-
gates of MMC, RF eqv for targeted conjugates without
MMC (P2B, P2C), or weight eqv for nontargeted conjugates
without MMC (P and P2A). Following treatment, 5 mg/mL
MTT was added to each well and incubated for 4 h. This
solution was removed from each well prior to resuspension
in DMSO for 1 h. Treatment solutions were analyzed for

Bareford et al.
absorbance of a reduced purple formazan MTT product
(560 nm), which correlates to cell viability, and compared
with that of positive (untreated) and negative (100% MeOH
pretreated) controls. Dose–response curves were generated
using a sigmoidal dose–response nonlinear curve fitting
function (GraphPad Prism 5) to determine the concentra-
tion necessary to inhibit cell viability by 50% relative to
nontreated control cells (IC50 dose).
P3C) and control (P2B, P2C) conjugates exhibited an RF-
dependent increase in molecular weight, which can likely be
explained by differences in the hydrodynamic volumes of
the conjugates containing the water-soluble vitamin. The
polydispersity of all copolymer conjugates remained rela-
tively consistent (n=1.26–1.54).
Breast Cancer Cell Selectivity for Polymeric
Conjugates
RESULTS
The extent of RF-specific conjugate internalization was
measured in the human breast cancer cell lines MCF-7
and SKBR-3 and compared with that of FA-assisted conju-
gate uptake. The nonspecific endocytosis of nontargeted
FITC-labeled conjugates (P1) was subtracted from the
amount of RF- (P1-RF) and FA- (P1-FA) targeted FITC-
conjugates in order to determine the extent of RF- and FA-
specific internalization. RF-targeted conjugates showed ap-
preciable accumulation within MCF-7 (50.86 2.69 ng/mg
protein/60 min) and SKBR-3 (83.29 4.65) cells. FA-
targeting enabled a similar amount of conjugate incorpora-
tion into SKBR-3 cells (58.70 3.98) yet only half that in
MCF-7 cells (27.62 3.42). Conjugate internalization oc-
curred with high affinity for RF-specific uptake in both
MCF-7 (198.8 36.85 μg/ml) and SKBR-3 (306.1 30.03)
cells, unlike that of the FA-targeted conjugates which
showed similar affinity for internalization within SKBR-3
(150.4 19.28) but not MCF-7 (903.7 185.2) cells
(Table II). MCF-7 cells were chosen for subsequent
experiments due to the higher affinity for RF-targeted
conjugate internalization and the lack of affinity for FA-
targeted conjugates.
Characterization of Polymeric Conjugates
Novel copolymer conjugates containing RF and MMC were
synthesized to evaluate the targeting potential of RF for
macromolecular chemotherapeutics. UV spectroscopy was
used to analyze the FITC, MMC, and (GG)ONp content in
each polymerized conjugate, as well as to determine RFl
and FAl incorporation following conjugation to the
derivatized targeting conjugates. Polymer content of each
compound is listed in Table I.
An SEC/FPLC system was utilized for the separation
and UV quantification of the weight average molecular
weight (Mw) and polydispersity (n) of all final copolymer
conjugates. Fluorescently-labeled conjugates (P1, P1-RF,
P1-FA) showed similar molecular weight averages of 13–
14 kD. Experimental conjugates without (P2A, P2B, P2C)
and with (P3A, P3B, P3C) MMC displayed molecular
weight averages between 19–33 kD, with the exception of
HPMA homopolymer control (45 kD). The incremental
increases in RF content for both the drug-containing (P3B,
Table I Physicochemical Characterization of HPMA Copolymer Conjugates
Feed Comonomer Content
Samplea HPMA GG-
Polymer Characteristics
GFLG- GFLG- APMA- FITC (mmol/g RF (mmol/g FA (mmol/g MMC (mmol/g Estimated Mw/Mn
ONp OH
MMC
FITC
polymer)
polymer)
polymer)
polymer)
Mw (kD)
(n)
P1
88
88
10
10
10
0
0
0
0
0
5
5
5
0
0
0
0
0
0
0
0
0
0
5
5
5
2
2
2
0
0
0
0
0
0
0
0.12
0
0
0
13.0
14.6
14.3
44.7
24.0
19.4
20.9
24.7
30.2
33.6
1.42
P1-RF
P1-FA
P
0.12
0.56
0
0
0
1.30
1.37
1.51
1.49
1.54
1.50
1.42
1.48
1.26
88
0.12
0
0.51
0
0
100
85
0
0
P2A
P2B
P2C
P3A
P3B
P3C
10
10
10
10
10
10
0
0
0
0
85
0
0.48
0.84
0
0
0
85
0
0
0
85
0
0
0.61
0.61
0.61
85
0
0.46
0.87
0
85
0
0
a. The sample identification codes correspond to the following copolymer conjugates: P1 = HPMA-GGOH-APMAFITC; P1-RF = HPMA-GGRF-
APMAFITC; P1-FA = HPMA-GGFA-APMAFITC; P = HPMA homopolymer; P2A = HPMA-GGOH-GFLGOH; P2B = HPMA-GGRF↓-GFLGOH;
P2C = HPMA-GGRF↑-GFLGOH; P3A = HPMA-GGOH-GFLGMMC; P3B = HPMA-GGRF↓-GFLGMMC; P3C = HPMA-GGRF↑-GFLGMMC; RF↓ =
low (0.4 mmol/g polymer) RF content; RF↑ = high (0.8 mmol/g polymer) RF content

Riboflavin-Targeted HPMA Polymers
Table II Internalization Kinetics of HPMA Copolymer-FITC-Ligand Con-
jugates in Human Breast Cancer Cells
following the active internalization of the nontargeted
(P3A: 2.31 ng/mg protein) and RF-targeted (P3B: 5.13
and P3C: 7.65 ng/mg protein) conjugates, as compared to
that following the passive diffusion of free MMC
(1.32 ng/mg protein) (Fig. 2). The difference between
RF-targeted (specific + nonspecific) and nontargeted
(nonspecific) MMC internalization reflects the amount
of MMC internalized specifically by RF-mediated endo-
cytosis. The resulting differences (2.82 and 5.34 ng/mg
protein) indicate that RF-mediated conjugate internali-
zation is an efficient and selective method of drug entry
into breast cancer cells.
Treatment
Internalization Kinetics
km (μg/ml)
Jmax (ng/60 min)
MCF-7
P1-RF
P1-FA
SKBR-3
P1-RF
P1-FA
198.8 36.85
903.7 185.2
50.86 2.69
27.62 3.42
306.1 30.03
150.4 19.28
83.29 4.65
58.70 3.98
Following endocytosis, internalized material is contained in
membrane-bound vesicles which shuttle through an
endolysosomal pathway, characterized by increasing com-
partmental acidification. This cellular processing pathway re-
sults in the dissociation of receptor-ligand complexes and
subsequent receptor membrane recycling for continuous li-
gand internalization (33). Biochemical modification of vesicu-
lar pH was induced to determine the role of compartmental
acidification in the internalization of free and conjugated
MMC. The lysosomotropic agents monensin and primaquine
work through different mechanisms to neutralize the normally
acidic (pH5–6) subcellular compartments which govern the
intracellular trafficking of endocytosed material, namely the
endosomes, lysosomes, and golgi. The intracellular accumu-
lation of conjugated MMC, both RF-targeted and
nontargeted, was significantly decreased upon vesicular neu-
tralization (Fig. 3). The RF-mediated internalization of
MMC, for both RF-containing conjugates (P3B and P3C),
was abolished with monensin (1.98 and 1.60% untreated
control) and primaquine (1.18 and 1.14%) treatment,
which was similar for that of the nontargeted conjugate
(P3A) (7.76 and 6.11%). The uptake of free MMC was
RF-Competed Conjugate Internalization
In order to demonstrate active receptor-mediated targeting
of RF-containing conjugates, competitive internalization
studies with radiolabeled RF were conducted. Table III de-
picts the IC50 values associated with RF-competed internal-
ization of nontargeted (P2A) and RF-targeted (P2B, P2C)
conjugates. Incubation with the nontargeted P2A copoly-
mer exhibited negligible inhibition of radiolabeled RF up-
take within the concentration range tested. RF-targeted
polymers exhibited potent IC50 values, which were
positively-associated with increasing RF content (P2B=
198.4, P2C=127.2 ng/mL). Based on these data and earlier
studies (32), we concluded that functionalizing RF via its
ribityl side chain enabled the retention of RF binding affin-
ity for its cell-surface machinery.
Cellular Accumulation of MMC
To determine the extent of RF-mediated MMC internali-
zation, MCF-7 cells were incubated for 60 min with
10 μg/mL MMC eqv of free and RF-targeted (P3B, P3C)
or nontargeted (P3A) conjugates to allow both passive and
active internalization, respectively. Results revealed that
MMC cytosolic accumulation was significantly higher
P3C
P3B
P3A
Table III Competitive Internalization of RF against Free RF and RF-
targeted or Nontargeted Polymer Conjugates in MCF-7 Cells
MMC
0
3
6
9
Sample
Content
IC50 (ng/mL)
Intracellular MMC (ng/mg protein)
P2A
P2B
P2C
RF
HPMA-GGOH-GFLGOH
HPMA-GGRF(0.48)-GFLGOH
HPMA-GGRF(0.84)-GFLGOH
free RF
NAa
Fig. 2 Cellular accumulation of MMC following free and RF-targeted or
nontargeted conjugate administration. MCF-7 cells were incubated with
10 μg/mL MMC eqv of free MMC, nontargeted MMC conjugate (P3A), or
RF-targeted conjugates (P3B and P3C). The intracellular MMC concentra-
tion for each treatment was determined by UV analysis, calibrated to MMC
standards, and standardized to total protein content. * p<0.05, ** p<
0.001.
198.4 15.5
127.2 10.3
46.9 5.3
a Competitive uptake against RF was negligible within the concentration
range tested

Bareford et al.
120
90
60
30
0
µ
Monensin [25 M]
Conjugate Trafficking Along the Endolysosomal
System
µ
Primaquine [300 M]
In order to illustrate the spatial and temporal trafficking
of free and polymer-conjugated MMC along vesicular
and nuclear compartments, the inherent fluorescence of
MMC was monitored for its 3D colocalization with fluo-
rescently labeled early endosome (Rab5), lysosome
(LAMP-1), and nuclear (propidium iodide) fluorescent
markers (Fig. 4), over a 6 h period. The extent of
colocalization was quantified at each time point using a
minimum of three regions using Pearson’s correlation
(PC) to calculate the percent of early endosome, lyso-
some, and nuclear fluorescent volumes which overlapped
with the total MMC fluorescent volume (Fig. 5). For
each marker, the percent overlap of nontargeted (P3A)
and RF-targeted (P3B) MMC-conjugate was compared to
that of free MMC, and significant differences were
expressed at each time point.
*
*
*
*
*
*
MMC
P3A
P3B
P3C
Fig. 3 Effect of endosomal neutralization on the internalization of free or
polymer-bound MMC. MCF-7 cells were treated with 25 μM monensin
(open bar) or 300 μM primaquine (solid bar) with 10 μg/mL MMC eqv of
free MMC, nontargeted MMC conjugate (P3A), and RF-targeted conju-
gates (P3B and P3C). Cell lysates were analyzed for intracellular MMC
content with UV spectroscopy. Results are presented as the percentage of
the untreated (monensin and primaquine-free) control. * p<0.001.
not significantly affected by either treatment. Results
confirm that the cellular uptake of polymer-conjugated
MMC is confined to endocytic processes.
Fig. 4 Subcellular co-localization of free and nontargeted- (P3A) or RF-targeted- (P3B) conjugate (blue) with early endocytic (Rab5 GTPase), lysosomal
(LAMP-1), and nuclear (propidium iodide) protein markers (red) in MCF-7 cells. The inset view defines the XY plane, the left panels represent the YZ, and
the top panels represent the XZ planes. The nontargeted MMC-conjugate (P3A) associated with an early endosomal compartment within 10 min (a),
followed by movement to lysosomal (b) and nuclear (c) compartments at 30 min and 6 h, respectively. The RF-targeted MMC-conjugate (P3B) was rapidly
endocytosed to allow for localization to early endosomes within 10 min (d) and lysosomes (e) within 30 min. Maximum MMC-nuclear overlap resulted
from longer incubation, up to 6 h, with RF-targeted MMC-conjugate (P3B) (f). Unconjugated MMC rapidly diffused, within 10 min, into cell nuclei (g) and
exhibited minimal fluorescent colocalization with endocytic markers (not shown).

Riboflavin-Targeted HPMA Polymers
Fig. 5 Quantification of the colocalization of free MMC and RF-targeted
ƒ
a
(P3B) or nontargeted (P3A) MMC-conjugates with early endosomal (a),
lysosomal (b), and nuclear (c) compartments in MCF-7 cells after 10 min
(open bar), 30 min (solid bar), or 6 h (striped bar). The extent of MMC
colocalization with each marker was determined using Pearson’s correla-
60
***
10 mins
30 mins
6 hrs
50
tion. Results are the mean
SD (n=3). Nontargeted (P3A) and RF-
40
30
20
10
0
targeted (P3B) MMC-conjugate colocalization with each marker was com-
pared to that of free MMC at each time point. * p<0.01, ** p<0.005,
*** p<0.001.
**
**
***
Following endocytic internalization, vesicles containing
MMC-conjugates displayed a time-dependent movement
along the endolysosomal system. Nontargeted MMC-
polymers rapidly associated with early endosomes within
10 min (Fig. 4a) where they remained consistently local-
ized (13.5 to 19.5%) over a 6 h period (Fig. 5a).
Similarly, RF-targeted drug conjugates immediately as-
sembled in apical endosomal compartments (Fig. 4d)
but to a significantly greater extent (~ 4-fold) than the
nontargeted complexes. Free MMC showed minimal re-
tention within early endosomes; however, results display
an increasing trend with longer exposure (Fig. 5a). Both
nontargeted and RF-targeted drug conjugates followed
similar intracellular sorting pathways ultimately leading
to lysosomal compartments where the highest extent of
colocalization occurred following 30 min (16.5 and
50.2%) of vesicular trafficking. Free MMC displayed
negligible accumulation (~ 3%) within lysosomes during
brief incubations, although longer exposure promoted the
pH-dependent ion-trapping (~ 10%) inside the acidic
organelles (Fig. 5b).
The prolonged retention of MMC-containing polymers
within digestive lysosomal compartments permitted the
enzymatic recognition of polymer-associated GFLG sub-
strates by cathepsin B and subsequent release of intact
MMC from these polymer-peptide linkers (34).
Immediately following release, MMC is directed along a
concentration gradient for diffusion into the cytosol from
which nuclear trafficking will occur. Once in the nucleus,
MMC is bioreductively activated to function as a potent
DNA-alkylating agent leading to a high degree of DNA
crosslinking and acute cytotoxic effects (35). A consider-
able amount of polymer-liberated MMC concentrated in
nuclei within 30 min; however, 6 h treatments supported
significantly higher nuclear accumulation of RF-targeted
(39.4%) and nontargeted (22.3%) conjugate-administered
MMC. Alternatively, incubation with free drug illustrated
maximum MMC-nuclear overlap during only 10 min
(20.2%) of passive drug diffusion at which time drug
equilibration between the intra- and extra-nuclear envi-
ronment suggestively prevented further nuclear localiza-
tion of drug molecules (Fig. 5c).
MMC
P3A
P3B
3
08
8
01
3
8
2
02
PC:
5
01
2
02
2
05
26
.0
00
03
+/- 0.
+/- 0.
027 +/- 0.
104 +/- 0.
195 +/- 0 193
135
.
050 +/- 0.
0.
493 +/- 0.208 +/- 0.068 +/- 0.
0.
0.
0.
0
0.
0.
0.
0.
b
***
60
50
40
30
20
10
0
***
*
MMC
P3A
P3B
8
3
9
7
6
4
3
1
PC:
23
0
0.
0.01
-
+/
- 0.01
- 0.01
-
- 0.01
- 0.02
- 0.09
- 0.09
+/
- 0.07
/
+/
+/
311
0.
0.032 +/
0.093 +/
0.045
0.071
0.034 +/
0.165 +
0.502
0.222 +/
c
60
50
40
30
20
10
0
***
**
***
*
MMC
P3A
4
P3B
6
8
5
3
01
9
PC:
8
5
05
1
- 0.02
- 0.00
- 0.01
- 0.
- 0.02
- 0.00
- 0.03
+/
- 0.
- 0.04
+/
+/
23
2
0.
167
.
.394 +/
0.224 +/ 0
0.202 +/
0.111 +/ 0.089 +/
0.116 +/ 0.164
0

Bareford et al.
Controlled Enzymatic Release of Intact MMC
The hydrolytic stability of RF-containing HPMA-MMC
conjugate (P3B) was tested in conditions representative of
extracellular, pH7.4, and lysosomal, pH5.5, environments.
The results demonstrate the hydrolytic resistance of both
Gly-Gly and Gly-Phe-Leu-Gly linkages for RF and MMC,
respectively, at neutral pH. Upon incubation in acidic buffer,
RF (t1/2=59.22 6.74 h) and MMC (t1/2=45.61 9.31 h)
were slowly released from their peptidyl spacers at a rate that
correlated with increasing peptide length. The enzymatic
stability of P3B conjugates were investigated in buffer of pH
5.5 containing cathepsin B in order to measure the efficiency
of drug release from the polymer-peptide enzyme substrate.
Data revealed the rapid release of MMC (t1/2=3.99 1.97 h)
from polymer-contained GFLG spacers, whereas RF
(t1/2=65.49 7.88 h) release occurred at a rate similar
to that in acidic medium. Results are reported in
Table IV.
Fig. 6 Cytotoxicity of free MMC and polymer conjugates after treatment with
free MMC (closed circle), nontargeted (P3A, closed square), or RF-targeted (P3B
and P3C, closed triangle and diamond) conjugates in MCF-7 cells. MMC-free
conjugates’ (P1, open circle; P2A open square; P2B, open triangle; P2C, open
diamond) cytotoxicity were negligible in the concentration range tested. Using
nonlinear sigmoidal dose response curve fitting (GraphPad Prism 5), the
mean
SD IC50 values obtained were 0.05 0.02, 0.61 0.03,
0.21 0.02, and 0.10 0.02 μg/mL for MMC, P3A, P3B, and P3C,
respectively.
MMC-Induced Cytotoxicity
the IC50 dose (0.21–0.10 μg/mL). P3A did exhibit apprecia-
ble MMC-induced cytotoxic effect (IC50=0.61 μg/mL), but
to a lesser extent than free and RF-targeted MMC. Although
the cytotoxic potency of RF-targeted MMC conjugates was
less than that of free MMC, apoptotic analysis following 72 h
treatments displayed equivalent cell death (> 95%).
Cytotoxicity of experimental and control polymers was deter-
mined using the MTT assay, where fluorescence detection of
a reduced formazan product is used to measure the enzymatic
activity in viable cells. In order to demonstrate that the cyto-
toxic effect of the MMC conjugates was not influenced by the
drug delivery vehicle, control polymers of HPMA (P) and RF-
targeted (P2B and P2C) or nontargeted (P2A) HPMA systems
without drug were included. All control polymers exhibited
negligible cytotoxicity within the concentration range tested,
suggesting the biocompatible nature of HPMA, RF, and
peptide linkers, in MCF-7 cells. MMC-induced cytotoxicity
was measured for free and RF-targeted (P3B and P3C) and
nontargeted MMC conjugates (P3A). High cytotoxic poten-
cies were exhibited by all MMC treatment groups (Fig. 6);
however, free MMC elicited the lowest effective dose (IC50=
0.05 μg/mL). This may be the result of the additional time
required for conjugate internalization, lysosomal drug release,
and nuclear drug localization. Increasing the RF content in
targeted polymers resulted in an RF-dependent decrease in
DISCUSSION
Pharmaceutical nanocarriers have been successfully employed
to enhance the locoregional delivery of conventional antican-
cer agents. Carrier conjugation prevents the promiscuous
diffusion of cytotoxic compounds into healthy and cancerous
tissues consequently prolonging the drugs’ residence time in
blood circulation and exposure to the EPR effect characteris-
tic of the neovasculature supporting solid tumors. Targeting
the macromolecular conjugates with the aid of ligands specific
to cell-surface receptor proteins allows for the highly-selective
delivery of therapeutics to target cancer cells. Unlike normal
tissues, the majority of tumor types overexpress certain vita-
min membrane receptors although the degree of
overexpression varies not only between each cancer form
but also among the various stages of tumor progression (7).
Research into vitamin-targeted chemotherapy has advocated
its promising clinical utility; however, the differential regula-
tion of vitamin receptors mandates further investigations to
determine the most efficacious targeting approach for each
type of cancer tissue. The present study was initiated to
investigate a novel targeting strategy using RF for chemother-
apeutic delivery to breast cancer tissue, where RF has been
recognized as an important regulatory factor. Recent analyses
determined that the high-affinity receptor-assisted
Table IV Hydrolytic and Enzymatic Release Rates of RF and MMC from
Polymer Peptide Linkages
Treatment Conditions
Release Half-life (t1/2, hr)
RF
MMC
pH7.4
NAa
NAa
pH5.5
59.22 6.74
65.49 7.88
45.61 9.31
3.99 1.97
pH5.5 + Cathepsin B
a Compound release was negligible (< 5%) during 72 h incubation

Riboflavin-Targeted HPMA Polymers
internalization of RF (17,19,36,37) is significantly upregulated
in breast cancer tissue (20), possibly explaining the dramatic
decrease in RF serum levels of breast cancer patients (16).
Preliminary studies were conducted to measure the
ligand-selectivity of RF-targeted polymers in an estrogen-
dependent (MCF-7) and independent (SKBR-3) breast can-
cer cell model and compared with that of the well charac-
terized targeting vitamin folic acid (5,38). Findings illustrate
the high-affinity internalization of RF-conjugates in both
MCF-7 and SKBR-3 cells, whereas FA-selectivity was
maintained only by SKBR-3 cells. The extent of polymer
internalization was comparable for the high-affinity
endocytic processes mediated by RF and FA; however, the
low-affinity incorporation of FA-polymers was considerably
less in MCF-7 cells (Table II). Similar variability in ligand-
specificity of breast cancer cells was previously demonstrated
for FA (9), cobalamin (39), and hyaluronan (40). Although
further investigations are necessary to be conclusive, results
from this comparative analysis suggest a highly reliable and
efficacious targeting role for RF in breast cancer-directed
drug delivery.
In order to measure the extent of RF-specific internaliza-
tion, P2 polymers were prepared for analysis in competitive
uptake studies with tritiated RF. RF-targeted polymers po-
tently inhibited radiolabeled-RF uptake in a manner positively
associated with polymer RF content (Table III). These results
ensured that RF retained high affinity for RF cell surface
machinery following polymer-peptide conjugation facilitating
the receptor-specific internalization of targeted conjugates.
Nontargeted conjugates showed negligible competition for
RF-specific internalization within the physiological RF con-
centration range suggesting internalization occurred predom-
inantly via a nonspecific or passive endocytic mechanism.
The biocompatible copolymer HPMA was employed as a
multifunctional vehicle which offers facile tailoring to suit
specific drug targeting needs (22). Clinically, it is important
that a macromolecular prodrug remain stable during blood
circulation, but the cytotoxic drug will need to be released
from carrier complexes following endocytic internalization.
Shuttling along cytoskeletal networks, vesicular cargo
trafficks to endolysosomal compartments where the large
complexes will be metabolized. An array of oligopeptides
have been designed for proteolytic degradation by lysosomal
proteases, particularly the tetrapeptide Gly-Phe-Leu-Gly
(GFLG) which serves as a substrate for cathepsin B, an
enzymatic protein located in the lysosome (41). In the pres-
ent study RF and MMC were stably attached to the dipep-
tide Gly-Gly and tetrapeptide GFLG, respectively, at
neutral pH. Although both RF and MMC were slowly freed
from their peptidyl spacers in acidic conditions, MMC re-
lease occurred slightly faster indicating the increased acces-
sibility of the longer tetrapeptide for hydrolysis. Previous
studies have indicated that a peptide spacer with a terminal
hydrophobic amino acid may enhance the hydrolytic stabil-
ity of the MMC linkage (34); therefore, additional studies
are necessary to select the peptidyl spacer which would
promote optimal resistance to extracellular hydrolysis yet
enable maximum lysosomal cleavage. The secondary amide
conjugation of MMC to the terminal glycine of polymer-
GFLG side chains permitted its efficient recognition and
rapid cleavage by the lysosomal enzyme cathepsin B
(Table IV) at a rate similar to that seen for previously
investigated poly-N-(2-hydroxyethyl)-l-glutamine (PHEG)-
polypeptide-MMC conjugates (42). Combined, data from
the in-vitro stability and fluorescent colocalization studies
suggest that the intracellular trafficking of RF-targeted
HPMA-MMC conjugates promotes the highly controlled
release of intact MMC and subsequent diffusion into nuclei
where it will exert its anticancer effect.
The in-vitro efficacy of RF-targeted MMC-conjugates (P3
polymers) was analyzed in MCF-7 cells, where RF-targeting
appears to be more sensitive. The anticancer agent MMC
was chosen as the proof-of-principle drug because it is an
extremely potent cytotoxin in breast cancer tissue (43) but is
rarely administered as a monotherapy due to the delayed
cumulative myelosuppression resulting from its unavoidable
accumulation in bone marrow (25). Low molecular weight
chemotherapeutics rapidly diffuse across cell membranes,
with the assistance of an intracellular concentration gradi-
ent, until equilibration between the extra- and intracellular
compartments at which point drug influx ceases.
Macromolecular delivery limits drug internalization to ac-
tive endocytic processes, which occur in a concentration-
independent manner afforded by cellular energy supplies.
Unlike the nonspecific pinocytic capture of nontargeted
complexes, ligand-targeted conjugates serve as high-affinity
substrates for specific membrane proteins which assist in
their preferential receptor-mediated endocytic entry.
Cellular uptake studies were conducted to measure the
cellular accumulation of MMC following 60 min treatments
with free and RF-targeted or nontargeted conjugates. The
active internalization of polymeric-MMC resulted in signif-
icantly higher intracellular MMC concentrations, most no-
tably for RF-targeted conjugates, compared to the passive
diffusion of free MMC (Fig. 2). Although endocytosis occurs
much slower than simple diffusion, the direct binding and
signaling events associated with receptor-mediated entry
promote the immediate internalization of ligand-
targeted drug conjugates. Similar results were previously
shown for macromolecular-MMC conjugates in a varie-
ty of cancer cell lines (34,44).
Several polymer-based delivery vehicles have been found
to disrupt the membrane integrity of epithelial cells (45). In
order to ensure the increase in cell-associated MMC resulted
from the active internalization of polymeric-MMC, cellular
uptake studies were also conducted in the presence of

Bareford et al.
lysosomotropic agents which are known to significantly ham-
per the endocytic internalization and intracellular sorting of
receptor-bound ligands. Following internalization, vesicular
cargo is guided through an endosomal maturation process,
characterized by increasing intravesicular acidity (pH7–5) (8).
Endosomal acidification promotes the dissociation of ligand-
receptor complexes to allow for the tightly-regulated intracel-
lular processing of individual ligands and the efficient receptor
recycling to apical membranes (20). Modification to this in-
tracellular event results in the cytosolic build-up of
endocytosed complexes which limits membrane-receptor
availability and prevents further ligand internalization
(46–48). Biochemical modification with both monensin and
primaquine completely abolished (> 95%) the cellular accu-
mulation of polymeric-MMC while the passive uptake of free
MMC remained unaffected (Fig. 3), ruling out the possibility
of intracellular conjugate leaking.
limited to breast cancer tissue in an in-vivo system, and the
IC50 values of the conjugates are still within an acceptable
range for therapy.
In summary, RF-targeted therapeutics show considerable
potential for the treatment of breast cancer. RF-targeted
macromolecules exhibited high affinity for the RF-specific
cell-surface machinery in MCF-7 cells, enabling the en-
hanced internalization of MMC. RF-targeted conjugates
shuttled MMC along the intracellular endolysosomal net-
work for the bioresponsive release and nuclear trafficking of
MMC. Although RF-mediated internalization occurs slower
than the passive diffusion of free drug, this process is far
more selective (52). Data from these studies advocates fur-
ther in-vitro and in-vivo characterizations in order to optimize
the targeting efficiency of the highly-selective breast cancer
targeting moiety RF.
The intracellular trafficking of the HPMA-MMC conju-
gates was monitored using confocal microscopy (Fig. 4) and
quantified for the extent of fluorescent 3D colocalization of
drug with endocytic and nuclear compartments (Fig. 5). RF-
targeted and nontargeted MMC-conjugates displayed similar
trafficking profiles along endolysosomal machinery; however,
the extent of MMC colocalization with all compartments was
significantly greater for RF-directed polymers. Following ve-
sicular internalization, MMC-conjugates quickly shuttled
through early endosomal compartments prior to their locali-
zation in lysosomal compartments within 30 min. The diffu-
sion of free MMC resulted in its rapid nuclear localization
within 10 min, during which its association with early
endosomes and lysosomes was negligible. Interestingly, the
nuclear accumulation of free MMC decreased during longer
incubations and displayed an increasing trend in endosome
and lysosome localization. MMC diffusion into acidic organ-
elles may have resulted from its prolonged cystolic residence
(49), where the ionization of weak bases, such as MMC, will be
trapped (50). Although conjugates required longer incu-
bation, up to 6 h, for the lysosomal cleavage of MMC
from the polymer-peptide backbone, the diffusion of
polymer-released MMC led to a significantly higher
extent of nuclear localization.
CONCLUSION
Ligand-targeted therapeutics offer promising potential in the
future of chemotherapy, as indicated by the increasing amount
of these systems in Phase 2 and 3 clinical trials (2,22). Each
system has been uniquely developed, via targeting ligand selec-
tion, for the highly selective delivery of anticancer agents to
certain cancer types. The present study for the first time dem-
onstrates the potential utility of RF as a targeting agent for
breast cancer-directed therapeutics. Comparative analysis of
the breast cancer cell selectivity for previously investigated
targeting ligands indicates a more specific and effective
targeting pathway for the RF system. Future investigations will
aim to optimize the physicochemical parameters of RF-
targeted systems, including drug and RF content.
Furthermore, we would evaluate the RF-specific biorecognition
and internalization of drug conjugates within other tissue sys-
tems, including the liver and small intestines, where RF cellular
processing is also enhanced.
REFERENCES
Cytotoxic analysis of all control polymers ensured the
biocompatibility of our RF-containing HPMA-peptide de-
livery system. Upon comparison with previously investigated
MMC-conjugates (44,51), nontargeted HPMA copolymer-
MMC conjugates exhibited equivalent cytotoxic potency
(IC50 ~ 0.6 μg/mL) in breast cancer cells. The RF-assisted
internalization of MMC-conjugates significantly enhanced
MMC sensitivity in MCF-7 cells in a manner which was
positively associated with increasing polymer-RF content
(IC50=0.21–0.1 μg/mL). Although unconjugated MMC
was the most potent cytotoxic treatment in these cells
(Fig. 6), it is important to note that this effect would not be
1. Andersson L, Davies J, Duncan R, Ferruti P, Ford J, Kneller S, et al.
Poly(ethylene glycol)-poly(ester-carbonate) block copolymers carrying
PEG-peptidyl-doxorubicin pendant side chains: synthesis and evalu-
ation as anticancer conjugates. Biomacromolecules. 2005;6:914–26.
2. Duncan R. The dawning era of polymer therapeutics. Nat Rev
Drug Discov. 2003;2:347–60.
3. Tanaka T, Shiramoto S, Miyashita M, Fujishima Y, Kaneo Y.
Tumor targeting based on the effect of enhanced permeability and
retention (EPR) and the mechanism of receptor-mediated endocy-
tosis (RME). Int J Pharm. 2004;277:39–61.
4. Mamot C, Drummond DC, Hong K, Kirpotin DB, Park JW.
Liposome-based approaches to overcome anticancer drug resis-
tance. Drug Resist Updat. 2003;6:271–9.

Riboflavin-Targeted HPMA Polymers
5. Reddy JA, Westrick E, Vlahov I, Howard SJ, Santhapuram
HK, Leamon CP. Folate receptor specific anti-tumor activity of
folate-mitomycin conjugates. Cancer Chemother Pharmacol.
2006;58:229–36.
27. Kopeček J, Rejmanová P, Strohalm J, Ulbrich K, Říhová B,
Chytrý V, Lloyd JB, Duncan R. Synthetic polymeric drugs. US
Patent. 5, 037, 883. Aug 6, 1991. p. 883.
28. Rejmanova P, Labsky J, Kopecek J. Aminolyses of monomeric and
polymeric p-nitrophenyl esters of methacryloylated amino acids.
Makromol Chem. 1977;178:2159–68.
29. Omelyanenko V, Kopeckova P, Gentry C, Kopecek J.
Targetable HPMA copolymer-adriamycin conjugates.
Recognition, internalization, and subcellular fate. J Control
Release. 1998;53:25–37.
30. Mitra A, Nan A, Ghandehari H, McNeill E, Mulholland J, Line
BR. Technetium-99m-Labeled N-(2-hydroxypropyl)
methacrylamide copolymers: synthesis, characterization, and in vivo
biodistribution. Pharm Res. 2004;21:1153–9.
31. Luo J, Smith MD, Lantrip DA, Wang S, Fuchs PL. Efficient
syntheses of pyrofolic acid and pteroyl azide, reagents for the
production of carboxyl-differentiated derivatives of folic acid. J
Am Chem Soc. 1997;119:10004–13.
32. Phelps MA, Foraker AB, Gao W, Dalton JT, Swaan PW. A novel
rhodamine-riboflavin conjugate probe exhibits distinct fluores-
cence resonance energy transfer that enables riboflavin trafficking
and subcellular localization studies. Mol Pharm. 2004;1:257–66.
33. Apodaca G, Katz LA, Mostov KE. Receptor-mediated transcytosis
of IgA in MDCK cells is via apical recycling endosomes. J Cell Biol.
1994;125:67–86.
34. Demarre A, Soyez H, Schacht E, Shoaibi MA, Seymour LW,
Rihova B. Synthesis and evaluation of macromolecular prodrugs
of Mitomycin-C. J Control Release. 1995;36:87–97.
35. Pan SS, Iracki T, Bachur NR. DNA alkylation by enzyme-
activated mitomycin C. Mol Pharmacol. 1986;29:622–8.
36. D’Souza VM, Bareford LM, Ray A, Swaan PW. Cytoskeletal
scaffolds regulate riboflavin endocytosis and recycling in placental
trophoblasts. J Nutr Biochem. 2006;17:821–9.
37. Foraker AB, Ray A, Da Silva TC, Bareford LM, Hillgren
KM, Schmittgen TD, et al. Dynamin 2 regulates riboflavin
endocytosis in human placental trophoblasts. Mol Pharmacol.
2007;72:553–62.
38. Guoand W, Lee RL. Receptor-targeted gene delivery via folate-
conjugated polyethylenimine. AAPS Pharm Sci. 1999;1:E19.
39. Rachmilewitz B, Sulkes A, Rachmilewitz M, Fuks Z. Serum
transcobalamin II levels in breast carcinoma patients. Isr J Med
Sci. 1981;17:874–8.
40. Auvinen P, Tammi R, Parkkinen J, Tammi M, Agren U,
Johansson R, et al. Hyaluronan in peritumoral stroma and malig-
nant cells associates with breast cancer spreading and predicts
survival. Am J Pathol. 2000;156:529–36.
41. Nichifor M, Schacht EH, Seymour LW. Macromolecular
prodrugs of 5-fluorouracil .2. Enzymatic degradation. J Control
Release. 1996;39:79–92.
6. Duncan R. Designing polymer conjugates as lysosomotropic
nanomedicines. Biochem Soc Trans. 2007;35:56–60.
7. Russell-Jones G, McTavish K, McEwan J, Rice J, Nowotnik D.
Vitamin-mediated targeting as a potential mechanism to increase
drug uptake by tumours. J Inorg Biochem. 2004;98:1625–33.
8. Bareford LM, Swaan PW. Endocytic mechanisms for targeted
drug delivery. Adv Drug Deliv Rev. 2007;59:748–58.
9. Parker N, Turk MJ, Westrick E, Lewis JD, Low PS, Leamon CP.
Folate receptor expression in carcinomas and normal tissues de-
termined by a quantitative radioligand binding assay. Anal
Biochem. 2005;338:284–93.
10. Garnett MC. Targeted drug conjugates: principles and progress.
Adv Drug Deliv Rev. 2001;53:171–216.
11. Qian ZM, Li H, Sun H, Ho K. Targeted drug delivery via the
transferrin receptor-mediated endocytosis pathway. Pharmacol
Rev. 2002;54:561–87.
12. Paulos CM, Reddy JA, Leamon CP, Turk MJ, Low PS.
Ligand binding and kinetics of folate receptor recycling in
vivo: impact on receptor-mediated drug delivery. Mol
Pharmacol. 2004;66:1406–14.
13. Salazarand MD, Ratnam M. The folate receptor: what does it
promise in tissue-targeted therapeutics? Cancer Metastasis Rev.
2007;26:141–52.
14. Castrellonand AB, Gluck S. Chemoprevention of breast cancer.
Expert Rev Anticancer Ther. 2008;8:443–52.
15. Karande AA, Sridhar L, Gopinath KS, Adiga PR. Riboflavin
carrier protein: a serum and tissue marker for breast carcinoma.
Int J Cancer. 2001;95:277–81.
16. Vaidya SM, Kamlakar PL, Kamble SM. Molybdenum, xanthine
oxidase and riboflavin levels in tamoxifen treated postmenopausal
women with breast cancer. Indian J Med Sci. 1998;52:244–7.
17. Mason CW, D’Souza VM, Bareford LM, Phelps MA, Ray A,
Swaan PW. Recognition, co-internalization, and recycling of an
avian riboflavin carrier protein in human placental trophoblasts. J
Pharmacol Exp Ther. 2006;317:465–72.
18. Yonezawa A, Masuda S, Katsura T, Inui KI. Identification and
functional characterization of a novel human and rat riboflavin
transporter, RFT1. Am J Physiol Cell Physiol. 2008;295:C632–41.
19. Huangand SN, Swaan PW. Involvement of a receptor-mediated
component in cellular translocation of riboflavin. J Pharmacol Exp
Ther. 2000;294:117–25.
20. Bareford LM, Phelps MA, Foraker AB, Swaan PW. Intracellular
processing of riboflavin in human breast cancer cells. Mol Pharm.
2008;5:839–48.
21. D’Souza VM, Foraker AB, Free RB, Ray A, Shapiro PS, Swaan
PW. cAMP-Coupled riboflavin trafficking in placental tropho-
blasts: a dynamic and ordered process. Biochemistry.
2006;45:6095–104.
22. Kopecek J, Kopeckova P, Minko T, Lu Z. HPMA copolymer-
anticancer drug conjugates: design, activity, and mechanism of
action. Eur J Pharm Biopharm. 2000;50:61–81.
23. Noriand A, Kopecek J. Intracellular targeting of polymer-bound
drugs for cancer chemotherapy. Adv Drug Deliv Rev.
2005;57:609–36.
24. Rihovaand B, Kubackova K. Clinical implications of N-(2-
hydroxypropyl)methacrylamide copolymers. Curr Pharm
Biotechnol. 2003;4:311–22.
25. Bradner WT. Mitomycin C: a clinical update. Cancer Treat Rev.
2001;27:35–50.
42. Hoste K, De Winne K, Schacht E. Polymeric prodrugs. Int J
Pharm. 2004;277:119–31.
43. Winski SL, Swann E, Hargreaves RH, Dehn DL, Butler J, Moody
CJ, et al. Relationship between NAD(P)H:quinone oxidoreductase
1 (NQO1) levels in a series of stably transfected cell lines and
susceptibility to antitumor quinones. Biochem Pharmacol.
2001;61:1509–16.
44. Cheung RY, Rauth AM, Ronaldson PT, Bendayan R, Wu XY. In
vitro toxicity to breast cancer cells of microsphere-delivered mito-
mycin C and its combination with doxorubicin. Eur J Pharm
Biopharm. 2006;62:321–31.
45. Chen R, Yue Z, Eccleston ME, Williams S, Slater NK.
Modulation of cell membrane disruption by pH-responsive
pseudo-peptides through grafting with hydrophilic side chains. J
Control Release. 2005;108:63–72.
46. Mukherjee S, Ghosh RN, Maxfield FR. Endocytosis. Physiol Rev.
1997;77:759–803.
26. Strohalmand J, Kopecek J. Poly N-(2-hydroxypropyl)
methacrylamide. 4. Heterogeneous polymerization. Makromol
Chem. 1978;70:109–18.

Bareford et al.
47. Maalmi M, Strieder W, Varma A. Ligand diffusion and receptor
mediated internalization: Michaelis-Menten kinetics. Chem Eng
Sci. 2001;56:5609–16.
48. King AC, Hernaez-Davis L, Cuatrecasas P. Lysomotropic amines
cause intracellular accumulation of receptors for epidermal growth
factor. Proc Natl Acad Sci U S A. 1980;77:3283–7.
distribution of chemotherapeutic agents in vitro. Biochem
Pharmacol. 2003;66:1207–18.
51. Nomura T, Saikawa A, Morita S, Sakaeda Kakutani T, Yamashita
F, Honda K, et al. Pharmacokinetic characteristics and therapeutic
effects of mitomycin C-dextran conjugates after intratumoural
injection. J Control Release. 1998;52:239–52.
49. Watts PL, Plumb JA, Courtney JM, Scott R. Sensitivity of cell lines
to mitomycin C. Br J Urol. 1996;77:363–6.
50. Mahoney BP, Raghunand N, Baggett B, Gillies RJ. Tumor acidity,
ion trapping and chemotherapeutics. I. Acid pH affects the
52. David A, Kopeckova P, Minko T, Rubinstein A, Kopecek J.
Design of a multivalent galactoside ligand for selective targeting
of HPMA copolymer-doxorubicin conjugates to human colon
cancer cells. Eur J Cancer. 2004;40:148–57.