Design of a paclitaxel prodrug conjugate for active targeting of an enzyme upregulated in breast cancer cells

Article abstract of DOI:10.1021/mp500128k

Breast cancer is the second most common cause of cancer-related deaths in women. Chemotherapy is an important treatment modality, and paclitaxel (PTX) is often the first-line therapy for its metastatic form. The two most notable limitations related to PTX

Full text of DOI:10.1021/mp500128k

Article
pubs.acs.org/molecularpharmaceutics
Design of a Paclitaxel Prodrug Conjugate for Active Targeting of an
Enzyme Upregulated in Breast Cancer Cells
Arpan Satsangi,*,†,‡ Sudipa S. Roy,§ Rajiv K. Satsangi,∥ Ratna K. Vadlamudi,§ and Joo L. Ong‡
†Joint Graduate Program in Biomedical Engineering, University of Texas at San Antonio and the University of Texas Health Science
Center at San Antonio, San Antonio, Texas 78249, United States
‡Department of Biomedical Engineering, University of Texas at San Antonio, San Antonio, Texas 78249, United States
§Department of Obstetrics and Gynecology, University of Texas Health Science Center at San Antonio San Antonio, Texas78229,
United States
∥RANN Research Corporation, San Antonio, Texas 78250, United States
ABSTRACT: Breast cancer is the second most common
cause of cancer-related deaths in women. Chemotherapy is an
important treatment modality, and paclitaxel (PTX) is often
the first-line therapy for its metastatic form. The two most
notable limitations related to PTX-based treatment are the
poor hydrophilicity of the drug and the systemic toxicity due
to the drug’s nonspecific and indiscriminate distribution
among the tissues. The present work describes an approach
to counter both challenges by designing a conjugate of PTX
with a hydrophilic macromolecule that is coupled through a
biocleavable linker, thereby allowing for active targeting to an
enzyme significantly upregulated in cancer cells. The resultant strategy would allow for the release of the active ingredient
preferentially at the site of action in related cancer cells and spare normal tissue. Thus, PTX was conjugated to the hydrophilic
poly(amdioamine) [PAMAM] dendrimer through the cathepsin B-cleavable tetrapeptide Gly-Phe-Leu-Gly. The PTX prodrug
conjugate (PGD) was compared to unbound PTX through in vitro evaluations against breast cancer cells and normal kidney cells
as well as through in vivo evaluations using xenograft mice models. As compared to PTX, PGD demonstrated a higher
cytotoxicity specific to cell lines with moderate-to-high cathepsin B activity; cells with comparatively lower cathepsin B activity
demonstrated an inverse of this relationship. Regression analysis between the magnitude of PGD-induced cytotoxic increase over
PTX and cathepsin B expression showed a strong, statistically significant correlation (r2 = 0.652, p < 0.05). The PGD conjugate
also demonstrated a markedly higher tumor reduction as compared to PTX treatment alone in MDA-MB-231 tumor xenograft
models, with PGD-treated tumor volumes being 48% and 34% smaller than PTX-treated volumes at weeks 2 and 3 after
treatment initiation.
KEYWORDS: paclitaxel−dendrimer conjugate, active targeting, biocleavable linker, cathepsin B, breast cancer, drug delivery
of PTX in patients with metastatic breast cancer has been
confirmed to be 56%.6,7 Therefore, PTX was chosen as a model
anticancer drug against the breast cancer cells in this study.
Anticancer treatment involving PTX faces several limitations.
The most notable challenge is its poor aqueous solubility and
the slow dissolution rate in bodily fluids. PTX is a diterpenoid
centered on a bulky and complex taxane ring with a number of
hydrophobic substituents. It lacks ionizable functional groups,
which otherwise would have allowed for possible solubilization
with moderation in pH and formation of salts and charged
complexes. These characteristics render the molecule a log P
value of about 3.96 and an aqueous solubility of less than 0.01
INTRODUCTION
■
Breast cancer accounts for the second largest number of cancer
deaths in the United States, with more than 230,000 new cases
diagnosed every year.1,2 While chemotherapy plays a central
role in cancer treatment, common anticancer agents demon-
strate well-known limitations, including systemic toxicity, poor
solubility in body fluids, and slow tumor uptake among other
pharmacological barriers. Recent nanotechnological advance-
ments are being applied to overcome these problems. This
research is one such effort toward targeted delivery and
improved efficacy of an anticancer agent commonly used for
the treatment of breast cancer.
The broadly cytotoxic class of drugs taxanes, including
paclitaxel (PTX) and docetaxel, is the most effective single-
agent drug used in breast cancer therapy and is considered the
first-line therapy in metastatic form of the disease.3,4 Among
taxanes, PTX is used more preferentially5 as the response rate
Received: February 12, 2014
Revised: April 30, 2014
Accepted: May 6, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/mp500128k | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics
Article
Scheme 1. Synthesis of the Paclitaxel−GFLG Conjugate
mg/mL.8−10 To overcome this issue, PTX is currently
administered as a 1.3% solution in Cremophor EL, a
polyethoxylated castor oil excipient known to generate severe
side effects such as bronchospasms, hypotension, hyper-
sensitivity, nephrotoxicity, and neurotoxicity.6,11−13 Though
pretreatment with corticosteroids and antihistaminics takes care
of majority of symptoms of hypersensitivity, 5−30% of patients
treated with the Cremophor EL formulation of PTX still
develop hypersensitivity and related problems.14
In recent years, innovative approaches have been investigated
using alternative drug carriers or pro-drug designs to enhance
drug bioavailability at the site of needed action;15 however,
many of these devices fail the “reality checks” of biological
systems, such as phagocytic clearance in the vascular system
and rejection by membrane transporters. In this context, an
attractive strategy is the synthesis of a stable derivative that may
possess better solubility in polar conditions and may release the
active ingredient at the site of action. Some solubility-
moderating polymeric conjugates of PTX have been designed,
and clinically useful outcomes are awaited, but the approach is
considered promising.16 Solubility of dendrimers, a class of
repetitively branched monodisperse polymers, can be tailored
by the manipulation of their terminal functional groups, and
therefore, they are very versatile and promising for the
modulation of solubility of a drug. Dendrimers, terminally
derivatized with hydrophilics (such as hydroxyl, amino, and
carboxyl groups), are water-soluble.17 Polyamidoamine
(PAMAM) dendrimer is terminated by primary amino group
and is commonly used for enhancing availability of carried drug
in aqueous conditions. They are smart macromolecules, and the
change in their conformations relative to their molecular weight
and size (1 nm per generation), terminal functional groups, and
external environment can be exploited for the purpose of an
efficient drug delivery.18 Their actions can be affected and
manipulated by the biochemical environment defined com-
monly by the polarity, pH, and ionic conditions of the
biological fluids around.19 Also, generation-4 (G4) PAMAM
dendrimers have been reported to demonstrate bioperme-
B
dx.doi.org/10.1021/mp500128k | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics
Article
Scheme 2. Synthesis of Paclitaxel−GFLG−Dendrimer (PGD or IX) Conjugate from Paclitaxel−GFLG Precursor and PAMAM
Dendrimer
ability.20 These factors allow PAMAM dendrimers to be an
effective vehicle to enhance the availability and delivery of
drugs, particularly PTX, in biological conditions.21
biomolecule that is overexpressed in the cancerous tissue in
question. Such a strategy of “active targeting” helps at least in
greatly reducing undesirable side effects to the rest of the
body.24−27
The other major problem encountered during PTX chemo-
therapy is its significant toxicity due to the nonspecific and
indiscriminate distribution of the drug among the tissues. The
primary mechanism of action of PTX is its induction of
apoptosis by binding to microtubules within a dividing cell
during mitosis, causing kinetic stabilization and inhibition of
microtubule depolymerization that leads to mitotic arrest and
the blockage of cell cycle progression.22−24 This mechanism is
nonspecific and results in serious unwanted consequences. One
way to possibly enhance availability of PTX in biofluids and to
introduce specificity at the same time is to couple PTX to
PAMAM dendrimer using a bioactive linker, which targets a
Cathepsin B, a lysosomal enzyme, is up-regulated in breast
cancer cells28,29 and in highly metastatic breast tumors30,31
compared to most other “normal” tissues. Barring exceptional
conditions, cathepsin B is not found extracellularly in normal
tissue; therefore, pro-drugs that target cathepsin B should
remain stable in circulation32 and should refrain from exhibiting
toxic affects before they are hydrolyzed by cathepsin B.
Consequently, in this work cathepsin B was targeted as a
cellular enzyme that would cleave the drug attached to the
PAMAM dendrimer, given that the linker between them is its
substrate. One such substrate is a tetrapeptide, glycine-
C
dx.doi.org/10.1021/mp500128k | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics
Article
phenylananine-leucine-glycine (GFLG) that is cleaved by
cathepsin B and allows liberation of the attached active ligand
after endocytosis.33,34 Polymers with intense three-dimensional
(3D) molecular structure (such as dendrimers) make enzymatic
release sterically challenging compared to linear polymers such
as poly(ethylene glycol) (PEG) and N-(2-hydroxypropyl)
methacrylamide (HPMA). Therefore, use of linkers that
provide spacing from the dendrimer surface for easy access to
the reactive site on drug molecules and specificity for enzymatic
cleavage is essential. GFLG has been used as such a cleavable
linker for cathepsin B induced drug delivery,34,36 including the
cancer therapy.35,37
On the basis of the above information, it was logical to
explore if the conjugation of PTX to a PAMAM dendrimer
through a cathepsin B cleavable linker would result in the
synthesis of an active-targeting pro-drug version of PTX to
demonstrate an improvement in its in vitro efficacy on breast
cancer cells that have higher levels of enzyme and in vivo
efficacy on xenograft tumors. Therefore, this study reports the
synthesis of a pro-drug conjugate of PTX to a generation-4
PAMAM dendrimer via ester, amide, and peptide linkages,
incorporating a GFLG linker. The conjugate and the
intermediates prepared to reach to the product were
characterized using physicochemical techniques as needed.
The prodrug, paclitaxel−GFLG−dendrimer (PGD or IX) thus
synthesized was evaluated for its in vitro activities against breast
cancer cells and normal kidney cells, relative to those of PTX.
Confirmation of increased toxicity by PGD against breast
tumors in 3D, physiological conditions was performed using
xenograft mice models.
CH2), 3.8 (d, 1H, 3-CH), 4.20 and 4.33 (d + d, 2H, 20-CH2),
4.45 (dd, 1H, 7-CH), 4.97 (d, 1H, 5-CH), 5.5 (d, 1H, 2-CH),
5.7 (d, 1H, 2′-CH), 6.0 (dd, 1H, 3′-CH), 6.25 (t, 1H, 13-CH),
6.3 (s, 1H, 10-CH), 7.18 (d, 1H, N−H), 7.35 (m, 5H, 3′-
C6H5), 7.45−7.8 (m, 5H + 5H, 4′-C6H5 + 21-C6H5).
Paclitaxel Hemisuccinate, N-Hydroxysuccinimide
Ester (V, Scheme 1). Compound V was synthesized by the
method of Ryppa et al.40 Briefly, an equivalent mixture of III,
N-hydroxy-succinimide (IV, Scheme 1; Aldrich; 92 mg; 800
μM), and N,N-dicyclohexyl-carbodiimide (DCC; Aldrich) were
reacted in anhydrous tetrahydrofuran (Aldrich) overnight. The
mixture was precipitated with diethyl ether (Aldrich) and
filtered. The organic filtrate was rotary-evaporated to get a
residue, which was purified chromatographically on silica gel
using a mixture of hexane/ethyl acetate (1:2) as eluent. After
removal of solvent from the eluate, the product was obtained as
white residue (93% yield) with following properties: TLC Rf
(silica gel-G/hexane−ethyl acetate; 20:80) 0.80.
1H NMR (in CDCl3; referenced to δ 0.00 ppm for TMS,
used as IS): δ (ppm) 1.15 and 1.25 (s, 3H + 3H, 16-CH3 and
17-CH3), 1.7 (s, 3H, 19-CH3), 1.95 (s, 3H, 18-CH3), 2.15
(dd, 2H, 6-CH2), 2.25 (s, 3H, 29-CH3), 2.35 (d, 2H, 14-CH2),
2.45 (s, 3H, 31-CH3), 2.5−2.65 (complex m, 2H + 2H, X-CH2
and X′-CH2), 2.85−2.95 (complex m, 2H + 2H, N−CH2 and
N′-CH2), 3.82 (d, 1H, 3-CH), 4.20 and 4.35 (d + d, 2H, 20-
CH2), 4.45 (distorted t, 1H, 7-CH), 4.97 (dd, 1H, 5-CH), 5.5
(d, 1H, 2-CH), 5.74 (d, 1H, 2′-CH), 6.0 (dd, 1H, 3′-CH), 6.25
(t, 1H, 13-CH), 6.3 (s, 1H, 10-CH), 7.18 (d, 1H, N−H), 7.35
(m, 5H, 3′-C6H5), 7.45−7.75 (m, 5H + 5H, 4′-C6H5 + 21-
C6H5).
Paclitaxel−Hemisuccinamide Derivative of GFLG (VII,
Scheme 1). A mixture of III (Scheme 1, 764 mg; 800 μM), IV
(92 mg; 800 μM), and 1-ethyl, 3-(3-dimethylamino-propyl)-
carbodiimide hydrochloride (EDC-HCl; Aldrich; 153.6 mg;
800 μM) in an organic solvent was stirred overnight to form V
(Scheme 1). The reaction was proceeded in to the next step
without isolation and purification of V. GFLG tetrapeptide (VI,
Scheme 1; Biomatik Corp., Wilmington, DE; 305.6 mg; 800
μM) and N-methyl-morpholine (NMM; Aldrich; 60 μL) were
then introduced into the flask, and the reaction mixture was
further stirred at room temperature overnight. The product was
chromatographed on silica. After solvent removal and drying
the residue under vacuum, 915 mg (86% yield) of VII was
obtained with the following properties: TLC Rf (silica gel-G/
EtOAc/methanol; 90:10) 0.29; HPLC (chromatographic
conditions same as for III) RT = 2.5 min.
EXPERIMENTAL SECTION
■
Syntheses. Starting from PTX, the synthesis of PGD was
accomplished via five steps, the diagrammatic representation of
which is portrayed in Schemes 1 and 2. While Scheme 1
displays the synthesis of the paclitaxel−GFLG conjugate, its
further conjugation to PAMAM dendrimer is illustrated in
Scheme 2.
Paclitaxel Hemisuccinate (III, Scheme 1). Compound III
was prepared using the method of Majoras et al.38 Briefly, PTX
(I, Scheme 1; ChemieTek, Indianapolis, IN) was reacted with
an excess of succinic anhydride (II, Scheme 1; Aldrich,
Milwaukee, WI) in anhydrous methylene chloride in the
presence of a catalytic amount of anhydrous pyridine. After the
reaction was confirmed to be complete by thin layer
chromatographic (TLC) analysis, pyridine was neutralized
with dilute aqueous solution of hydrochloric acid and extracted.
The residue, recovered from the organic phase after solvent
evaporation, was recrystallized with acetone/water to get needle
like crystals of III in almost quantitative yield; mp 177−179
°C;39 TLC Rf (silica gel-G/hexane−ethyl acetate (all from
Aldrich, Milwaukee, WI); 20:80) 0.17; HPLC (5 μL of 1 mg/
mL sample was eluted isocratically with a mixture of water/
acetonitrile; 55:45 at 1.5 mL/min through a C-18 modified
silica particles of 5 μm size packed in a 3.9 × 300 mm Waters μ-
Bondapak HPLC column; monitored spectrophotometrically at
215 nm; chart speed 1 cm/min) RT = 6.0 min.
1H NMR (in DMSO-d6; referenced to δ 0.00 ppm for TMS
as IS): δ (ppm) 0.82 and 0.87 [d + d, 3H + 3H, Leu-
ω(CH3)2], 1.0 and 1.05 (s + s, 3H + 3H, 16-CH3 and 17-
CH3), 1.5 (s, 3H, 19-CH3), 1.55−1.7 (complex m, 5H, 6-CH2,
Leu-γCH, Leu-CH2), 1.8 (s, 3H, 18-CH3), 2.1 (s, 3H, 29-
CH3), 2.25 (s, 3H, 31-CH3), 2.42 (d, 2H, 14-CH2), 2.8 (dd,
2H, X-CH2), 3.0 (dd, 2H, X′-CH2), 3.60 (complex, 2H,
phenylalanine-CH2), 3.75 (complex, 1H, 3-CH), 4.0 [s, 4H, 2
X Gly-CH2)], 4.1 and 4.3 (d + d, 2H, 20-CH2), 4.5−4.6
(complex, 1H + 1H, 7-CH + Leu-αCH), 4.92 (distorted t, 1H
+ 1H, 5-CH + phenylalanine-α-CH), 5.3 (d, 1H, 2-CH), 5.4 (d,
1H, 2′-CH), 5.8 (complex m, 1H + 1H, 3′-CH + 13-CH), 6.3
(s, 1H, 10-CH), 7.16−8.16 (multiple m, 20H, aromatic-H + 5-
amidic N−H).
1H NMR [in CDCl3 (Aldrich); referenced to δ 0.00 ppm for
tetramethyl-silane (TMS; Aldrich) used as internal standard
(IS)]: δ (ppm) 1.15 and 1.25 (s, 3H + 3H, 16-CH3 and 17-
CH3), 1.7 (s, 3H, 19-CH3), 1.9 (s, 3H, 18-CH3), 2.20 (dd, 2H,
6-CH2), 2.25 (s, 3H, 29-CH3), 2.40 (dd, 2H, 14-CH2), 2.45
(s, 3H, 31-CH3), 2.6 and 2.75 (m, 2H + 2H, X-CH2 and X′-
N-Hydroxysuccinimide Ester of Paclitaxel−Hemisucci-
namide Derivative of GFLG (VIII, Scheme 2) and
Coupling of PTX to G4- PAMAM Dendrimer, Using
D
dx.doi.org/10.1021/mp500128k | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics
Article
GFLG as Linker (Collectively in One Unit Referred As
Paclitaxel−GFLG−Dendrimer; Abbreviated as PGD or IX,
Scheme 2). Compound VII (Scheme 2, 800 mg; 600 μM),
NHS (69 mg; 600 μM) and EDC-HCl (115 mg; 600 μM) were
reacted by stirring for 8 h at room temperature to form the N-
hydroxysuccinimide ester of paclitaxel−hemisuccinamide de-
rivative of GFLG (VIII, Scheme 2). The reaction was forwarded
to the next step without isolating and purifying VIII. Then, G4-
PAMAM dendrimer (Aldrich; 1.075 g; 75 μM) was introduced
to the above reaction mixture containing VIII. The resultant
mixture was again stirred at room temperature for overnight.
The product was dissolved in a mixture of acetonitrile/water
(2:1; 20 mL) and dialyzed against a mixture of acetonitrile/
water (2:1). After evaporation of acetonitrile, the purified
sample in residual water was lyophilized and then dried under
vacuum to get IX (Scheme 2) in 75% recovery, with the
following analytical characteristics.
M acetate, 0.01 M EDTA, and 0.05 M reduced glutathione)/
organic solvent mixture in the presence or absence (control) of
0.5 × 10−6 M papain. Samples were analyzed by HPLC after 24
h. Aliquots of both control and enzyme-digested samples were
analyzed at the Texas A&M University Laboratory for
Biological Mass Spectrometry (TAMU-LBMS; College Station,
TX) by the liquid chromatography−mass spectrometry (LC−
MS) analysis.
Cell Cultures. MDA-MB-231, MDA-MB-468, BT-20, and
T47-D human breast cancer cells along with mouse proximal
tubule kidney cells BUMPT were cultured in RPMI 1640
medium (Life Technologies, Carlsbad, CA). MDA-MB-435
human breast cancer cells and JEG-3 human choriocarcinoma
cells were maintained in DMEM (Life Technologies). Both
media were supplemented with 10% heat-inactivated FBS
(Hyclone Laboratories, Logan, UT), 1% Glutamax (Life
Technologies), and penicillin (100 IU/mL)/streptomycin
(100 μg/mL)/amphotericin B (0.25 μg/mL) (Life Technolo-
gies, 15240-062). All cell lines were maintained at 37 °C in a
humidified incubator in a 5% carbon dioxide atmosphere.
Western Blot Analysis. Cellular lysates for Western blot
analysis were prepared using previously described method-
ology.43 Briefly, cells were washed three times with phosphate-
buffered saline (PBS), treated with cold RIPA lysis buffer [50
mM Tris-HCl, 150 mM NaCl, 0.5% Nonidet P-40, 0.1 sodium
dodecyl sulfate (SDS), 0.1% sodium deoxycholate, 1× protease
inhibitor mixture (Roche Biochemical), and 1 mM sodium
vanadate] on ice for 20 min, and harvested by scraping. Lysates
were centrifuged at 14,000g for 15 min at 4 °C and supernatant
was collected. Protein concentrations were determined using
microbicinchoninic acid (BCA) assay (Thermo Fisher
Scientific, Rockford, IL) and proteins were stored at −80 °C.
Cell lysates containing equal amounts of protein (∼20 μg) were
electrophoresed on 8% SDS-polyacrylamide gels and trans-
ferred onto a 0.2 μm nitrocellulose membranes. Blots were
probed with appropriate antibodies. Actin and cathepsin B
proteins were probed using antiactin antibody produced in
rabbit (Sigma-Aldrich, St. Louis, MO) and anticathepsin B
monoclonal mouse antibody (Sigma-Aldrich), respectively, as
primary antibodies. After staining with appropriate secondary
antibodies, blots were developed using enhanced chemilumi-
nescence method.
Metabolic Activity Assay. The effect of PGD on
metabolic activity was determined by 4-[3-(4-iodophenyl)-2-
(4-nitrophenyl)-2H-5-tetrazolio]-1,3-benzene disulfonate
(WST-1) assay (Roche Diagnostics, Indianapolis, IN). Cells
(2 × 103 cells/well) were seeded in 96-well culture plates and
incubated for 24 h. Previous media was aspirated and
replenished with fresh RPMI 1640 culture medium supple-
mented with 5% FBS, 1% Glutamax, and penicillin (100 IU/
mL)/streptomycin (100 μg/mL)/amphotericin B (0.25 μg/
mL) (Life Technologies, 15240-062) and containing PTX or
PTX-equivalent PGD. To determine optimal treatment
concentrations to observe drug effects, MDA-MB-231 cells
were first treated with various concentrations of PTX or PTX-
equivalent PGD (0.01 nM to 1 μM). Subsequent analysis on a
panel of cancer cells and normal kidney cells were performed at
5 and 10 nM PTX or PTX-equivalent PGD, approximately two
to three times the IC50 concentration for PTX against MDA-
MB-231 cells. Media alone served as a blank, and plain,
unloaded PAMAM at equimolar concentration was the negative
control. After 72 h, 10 μL of WST-1 reagent was added to each
well and plates were incubated for an additional 4 h.
Analyses. (i) HPLC [chromatographed isocratically on a
3.9 mm × 300 mm, Waters μ-Bondapak column packed with 5
μm size, C18-derivatized silica particles, eluted with H2O/
acetonitrile (55:45; 1.5 mL/min) at a pressure of 3000 psi. The
eluted components were monitored spectrophotometrically at
215 nm: RT 8.7 min. (chart speed: 1 cm/min).
1
(ii) H NMR (in DMSO-d6: referenced to δ 0.00 ppm for
TMS as IS): δ (ppm) 0.80 and 0.85 [d + d, 3H + 3H, Leu-
ω(CH3)2], 1.0 (s + s, 6H, 16-CH3 and 17-CH3), 1.5 (s, 3H,
19-CH3), 1.55−1.7 (complex m, 5H, 6-CH2, Leu-γCH, Leu-
CH2), 1.8 (s, 3H, 18-CH3), 2.0 (s, xH, from dendrimer), 2.1
(s, 3H, 29-CH3), 2.24 (s, 3H, 31-CH3), 2.42 (d, 2H, 14-CH2),
2.7 (broad, xH, from dendrimer), 2.8 (dd, 2H, X-CH2), 2.9
(dd, 2H, X′-CH2), 3.1 (broad, xH, from dendrimer), 3.60−3.75
(complex m, 2H + 1H, phenylalanine-CH2 and 3-CH,
respectively), 3.9−4.05 [complex, 2H + 2H, 2 X Gly-CH2)],
4.1 and 4.25 (complexes, 2H, 20-CH2), 4.5 (m, 1H, Leu-αCH),
4.6 (t, 1H, 7-CH), 4.85−4.95 (complex m, 1H + 1H, 5-CH +
phenylalanine-α-CH), 5.4 (complex m, 1H + 1H, 2-CH and 2′-
CH), 5.87 (t, 1H, 13-CH), 6.15 (d, 1H, 3′-CH), 6.3 (s, 1H, 10-
CH), 7.06−8.4 (multiple m, amidic N-Hs + aromatic-Hs).
(iii) Estimation of a number of PTX units per molecule of
PGD: The quantitation of ligation of the paclitaxel−hemi-
succinimide derivative of GFLG (VII) units per dendrimer
molecule in PGD was done by NMR analysis using fumaric acid
as IS. Thus, a mixture of PGD (5.5 mg; 0.228 μmol) and
fumaric acid (Aldrich; 2.675 μg; 2.305 μmol) was taken in 600
1
μL of DMSO-d6 (Aldrich) and the H NMR spectrum was
scanned. The spectrum exhibited all of the signals recorded for
IX above, in addition to the one for methylene protons
[−(H)CC(H)−] of fumaric acid. The comparison of proton
integration of the chemical shift signal at δ6.3 (10-CH of PTX
moiety) relative to that of the chemical shift signal at δ6.6
[−(H)CC(H)− of fumaric acid] yielded the quantitation of
PTX moieties in the used amount of PGD, from which the
molecular units of VII per dendrimer molecule was established.
(iv) MALDI-TOF mass spectrometry analysis of PGD was
performed to support molecular weight determination using
NMR method at the University of Texas Health Science Center
at San Antonio (UTHSCSA) Department of Molecular
Medicine Mass Spectrometry core service.
Confirmation of Activity of a Cathepsin Homologue
on PGD. Enzymatic activity of papain, a homologue of
cathepsin B41,42 against the PGD was verified for hydrolytic
release of paclitaxel−GFLG from the conjugate. PGD conjugate
(1 mg/mL, n = 3) was reacted at 37 °C in reaction buffer (0.1
E
dx.doi.org/10.1021/mp500128k | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics
Article
Absorbance of the converted WST-1 product was read at 415
nm on a microplate reader (Biotek Synergy 2). The relative
IC50 data were calculated by GraphPad Prism 5 (GraphPad
Software, La Jolla, CA).
For the conjugation of a drug molecule (i.e., PTX, here) to a
supporting carrier molecule (i.e., PAMAM dendrimer, here), an
important factor is the protection of reactive site(s) of drug
molecule and retention of natural steric factors around it. Since,
the direct attachment of drug molecule to carrier molecule may
introduce crowding (steric effects) on the reactive sites of drug,
invariably the drug molecules are attached to support molecule
via a spacer arm that keeps a sufficient distance between the
two to exclude the possibility of chemical damage to the
reactive sites and the introduction of steric factors in drug
molecule.44,45 To retain bioactivity, it is essential to protect
oxetane ring (covering C-4, C-5, and C-20), acetate group at C-
4, benzoyl or related group at C-2, and benzamido and phenyl
groups at C-3′ of PTX molecule (Figure 1).46 However, the
Apoptosis Assay. Annexin staining for measurement of
apoptosis was performed with the use of a kit (FITC Annexin V
Apoptosis Detection Kit II; BD Pharmingen, San Jose, CA).
Briefly, MDA-MB-231 cells were seeded at 106 cells/dish and
incubated for 24 h. The cells were then treated with 50 nM
PTX or PTX-equivalent PGD. Media alone served as blank.
After 48 h of treatment, cells were washed with cold PBS and
resuspended in 100 μL binding buffer. Cells were stained with 5
μL of FITC Annexin V and 5 μL of propidium iodide (PI) for
15 min in the dark before diluting with an additional 400 μL
binding buffer. Samples were analyzed by flow cytometry on a
FACS Calibur (BD Biosciences, San Diego, CA).
In Vivo Antitumor Efficacy. All animal experiments were
performed after obtaining approval from the UTHSCSA
Institutional Animal Care and Use Committee, and animals
were housed in accordance with the UTHSCSA institutional
protocol for animal experiments.
MDA-MB-231 cells were resuspended in 50% v/v serum-free
culture medium/Matrigel Matrix (BD Biosciences, San Jose,
CA). Approximately 2 × 106 MDA-MB-231 cells (100 μL)
were injected into the mammary fatpad of 6 to 7-week-old
female athymic nude mice. Once tumor masses in xenografts
reached an average volume of 90 mm3 (about 8 days after
innoculation), mice were randomly assigned to four groups (n
= 7): control (saline vehicle), PTX (10 mg/kg), PGD (40 mg/
kg or PTX-equivalent), and unloaded PAMAM dendrimer
(negative control; 24 mg/kg or PGD-equivalent). Mice were
given injections every other day intraperitoneally for a total of 3
injections. Two-dimensional tumor measurements were made
weekly after the first dose. The tumor volume was calculated
according to the following formula:
Figure 1. Chemical structure of paclitaxel.
volume = (short diameter2) × (long diameter) × 0.5
molecule contains three hydroxyl groups that may be
chemically modified for suitable structural manipulations. The
hydroxyl group at C-1 has no considerable chemical activity;
therefore, PTX prodrugs are generally designed by derivatizing
the hydroxyl groups at C-2′ and/or C-7,47 though the one at C-
2′ is significantly more reactive than the one at C-7.39 Hence,
PTX was conjugated to PAMAM via a GFLG linker by
derivatizing the hydroxyl group at C-2′ of the PTX. Thus, using
a known methodology, PTX (I) was converted to its
hemisuccinate (III) to acquire a terminal carboxy group on
the molecule. The newly acquired carboxy terminus at position
C-2′ of PTX was activated as N-hydroxysuccinimide (NHS)
ester (V) and was reacted to amino terminus of GFLG
tetrapeptide (VI). The carboxy-terminus of the resultant
paclitaxel−hemisuccinate derivative of GFLG (VII) was again
activated as NHS ester (VIII) to finally react to the amino
groups of PAMAM (G4) dendrimer to get the required PTX
conjugated dendrimer (IX or PGD) coupled by cathepsin B
cleavable GFLG.
The extent of tumor burden per animal was expressed in
cubic millimeters. Mice were sacrificed 24 days after treatment
initiation.
Statistical Analysis. Data are presented as the mean
standard deviation (SD). One-way analysis of variance tests
were used to investigate the differences between groups for
apoptosis in vitro experiments and two-way analysis of variance
tests for in vivo and metabolic activity in vitro experiments,
followed by post hoc tests with Bonferroni correction for
comparison between individual groups. Statistical significance
was established at p < 0.05. Data for metabolic activity assays
was investigated using Stata/IC 12.0 (StataCorp LP, College
Station, TX). All other experiments were analyzed using
GraphPad Prism 5 (GraphPad Software, La Jolla, CA).
RESULTS AND DISCUSSION
■
The goal of this work was to design a PTX-bound PAMAM
dendrimer, using a cathepsin B-cleavable spacer for active-
targeted delivery of PTX, and to examine the designed device
for its enzymatic release, cellular toxicity/apoptotic activity, and
capability to reduce tumor growth, compared to the activity of
PTX alone. The design of cathepsin B targeting GFLG linked
PTX−PAMAM conjugate was accomplished through a multi-
step chemical synthesis as presented in Schemes 1 and 2.
The identity of III was confirmed by its melting point being
in agreement with that of the reported one39 and by the
presence of the triplets at δ2.6 and δ2.75 ppm in addition to the
1
required peaks for PTX in its H NMR spectrum, confirming
the hemisuccinate derivatization of the OH group at C-2′ of
PTX. Also, the proton at C-2′ showed up at δ5.53 ppm instead
F
dx.doi.org/10.1021/mp500128k | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics
Article
Figure 2. LC−MS analysis of PGD conjugates treated without and with papain, a cathepsin B-like enzyme. LC chromatograms of samples (a)
without and (b) with papain are shown. A peak unique to papain-digested PGD sample was observed at 12.97 min, and (c) mass spectrometry
results show a molecular ion peak at 1350 indicative of paclitaxel−GFLG (VIII).
of the proton geminal to “unsuccinated” OH at the same
position in PTX at δ4.80 ppm. The formation of V was clearly
indicated by the chromatographic separation of V from III
when cospotted in a chromatogram over SiO2-G and developed
by hexane/ethyl acetate; 20:80 (Rfs: 0.17 and 0.80 for III and V,
respectively). The structural conformation of V was done by 1H
NMR as it recorded the presence of a complex multiplet at
δ2.85−2.95 for the two methylene groups (N−CH2 and N′-
CH2) of succinimide terminus, while keeping those for two
hemisuccinate methylenes as in III. The formation of VII was
similarly indicated by the chromatographic separation of VII
from V when cospotted in chromatogram over SiO2-G and
developed by ethyl acetate/methanol; 90:10 (Rfs: 0.92 and 0.29
for V and VII, respectively). In the 1H NMR spectral analysis of
VII, the absence of signals at δ2.85−2.95 for the two methylene
groups (N−CH2 and N′−CH2) of succinimide terminus found
in V, and instead, the presence of signals for leucine (at δ 0.80,
0.87, 1.55−1.7), for phenylalnine (at δ 3.60, 4.92), for glycine
(at δ 4.0), and those for amidic N−Hs confirms the
replacement of NHS group in V with the GFLG tetra-peptide
to form VII.
The formation of the final product (IX) was indicated by the
difference of HPLC retention time for IX compared to that for
VII when eluted on a C-18 RP column with H2O/acetonitrile
(55/45) at 1.5 mL/min; while the retention time for VII was
2.5 min, IX was retained for 8.7 min. The 1H NMR spectrum of
IX retained the signals for VII with the addition of the ones at
δ2.0, 2.7, and 3.1 (all specific to PAMAM dendrimer)
confirmed the conjugation of VII to PAMAM dendrimer to
form IX. Markedly improved solubility in polar solvents (a key
factor in determining bioavailability) by the PGD conjugate was
a strong physicochemical differentiator from the PTX molecule.
The number of PTX moieties attached per dendrimer entity
in the final product (IX or PGD) was determined by using
fumaric acid as an IS in 1H NMR scanning. NMR peaks at δ6.3
and δ6.6 were integrated and the resulting peak areas were
normalized per single proton. The ratio of the peak area of a
proton of fumaric acid (derived from the signal at δ6.6) relative
to the peak area of one proton from PTX (derived from the
signal at δ6.3) is equivalent to the ratio of moles of fumaric acid
over moles of PTX present. Given this, the total content of
PTX, and by extension paclitaxel−GFLG, in solution was
determined to be 1.662 μmol. The mass of paclitaxel−GFLG
was subtracted from the total mass PGD introduced, yielding
an estimated mass of PAMAM dendrimer. The ratio of the
moles of PTX to the moles of PAMAM dendrimer was 7.112,
suggesting approximately 7 PTX moieties were loaded per
dendrimer unit. The estimated molecular weight was
approximately 23,385 Da, as also confirmed by MALDI-TOF
mass spectrometry, which showed a broad peak centered
around 23−24 kDa. This ratio was used to create dilutions of
PGD to deliver specific molar amounts of PTX in further
biological experiments.
The ability of PGD to release paclitaxel-GFLG in the
presence of papain, a homologue of cathepsin B, was evaluated
in a 24 h reaction. Isocratic HPLC analysis of the reaction
mixture showed the development of a new product peak in the
chromatogram at 2.5 min, relative to the control sample
(absence of papain). In subsequent analysis by LC−MS, the
above enzymatic reaction mixture and the related control
reaction mixture (without papain) were injected separately on a
reverse phase C18, 150 × 2.1 mm column, and the
chromatograms were developed for elution with the following
gradient system at 0.3 mL/min: from 0−5 min, the eluent was
5% acetonitrile in 1% aqueous acetic acid solution, then a
G
dx.doi.org/10.1021/mp500128k | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics
Article
gradient was run to bring the eluent composition to 100%
acetonitrile by 20 min. The chromatogram for the PGD
reaction with papain showed a peak at 12.97 min; this peak was
absent in the reaction without papain (control sample) (Figure
2). Electrospray ionization (ESI) mass spectrum analysis of the
content in the LC peak at 12.97 min revealed the molecular ion
peak to be at 1350. This confirms the formation of VIII in
papain degradation of PGD in a buffer solution that contains
sodium ion. Molecules with multiple oxygens, including PTX,
are known to form sodium adducts.48,49 Therefore, the
molecular ion peak for VIII shows the replacement of a proton
in VIII with a sodium ion, i.e., [M + Na]+ at m/e 1350. Thus,
PGD was shown to release paclitaxel−GFLG (VIII) in the
presence of cathepsin B-like enzyme.
To correlate the effect of PGD on metabolic activity with the
presence of cathepsin B, the levels of total cellular cathepsin B
were investigated in an array of cells by Western blot and
enzymatic assays. Six breast cancer cell lines (MDA-MB-231,
MDA-MB-468, MDA-MB-435, BT-20, BT-549, and T47D)
that demonstrate PTX sensitivity were chosen to represent the
diseased state. Breast cancer with choriocarcinomatous features
represents a distinct subset of the breast cancer spectrum, with
12−33% of female breast cancer patients expressing human
chorionic gonadotropin (hCG).50,51 As such, the human
choriocarcinoma cell line JEG-3 was chosen as a model to
represent such pathological features. Finally, renal clearance is
one of the primary methods of nanoparticle removal from the
body. Given its large blood flow and ability to concentrate toxic
compounds, the kidneys are also particularly sensitive to
xenobiotics.52 As such, BUMPT mouse kidney proximal tubule
cells were chosen as a model for nanoparticle behavior against
this critical normal tissue.
Cathepsin B was detected in all cells, an expected result given
its presence in the lysosome of normal cells (Figure 3).
However, marked differences were seen in its levels in the
normal BUMPT mouse kidney proximal tubule cell line relative
to that determined in the cancer cell lines. Western blot analysis
showed that cathepsin B had the highest presence in T47D
cells, followed by a medium level of presence in decreasing
order in JEG-3, MDA-MB-435, MDA-MB-468, MDA-MB-231,
and BT-20 cancer cells. Finally, the lowest presence was
observed in the BT549 cancer cell and BUMPT normal mouse
kidney proximal tubule cell line.
Figure 3. (A) Western blot analysis of cathepsin B in human breast
cancer, human choriocarcinoma, and normal immortalized mouse
kidney proximal tubule cells. Total cellular protein (20 μg) was
electrophoresed on 8% SDS-PAGE gel and blotted onto a nitro-
cellulose membrane. Cathepsin B detection was performed by using
anticathepsin B monoclonal mouse antibody, staining with horseradish
peroxidase-conjugated secondary antibody, and developing with
enhanced chemiluminescence method. (B) Densitometric analysis of
cathepsin B Western blot was performed using Adobe Photoshop
software (Adobe Systems, San Jose, CA), normalized to actin intensity,
and expressed relative to BUMPT values.
The effect of PGD on metabolic activity was studied.
Metabolic activity assays, such as the WST-1 assay, have been
previously established as indirect methods of assessing cytotoxic
activity of drugs.53,54 To ascertain an optimal treatment
concentration with which to study drug effects, MDA-MB-
231 cells were first treated with various concentrations of PTX
and PTX-equivalent PGD (Figure 4). The relative IC50
concentration is the halfway point between the top and bottom
plateaus of the concentration curve.55 When comparing the
IC50 of PTX relative to PGD, a marked leftward shift was
observed at 3.29 and 0.640 nM, respecitively. This suggested an
increased cytotoxic profile for PGD as compared to PTX alone.
A smaller concentration of PGD could deliver the same effect
as a five times larger concentration of PTX, thereby increasing
the efficiency of drug delivery. For further metabolic activity
assays, PTX concentrations near 2× and 3× IC50 of PTX (5 and
10 nM) were used. This treatment range has been shown
appropriate to evaluate drug therapies.56−58
Figure 4. Effect of drug concentrations on MDA-MB-231 metabolic
activity. Cells were exposed to PTX, PTX-equivalent PGD, and PGD-
equivalent PAMAM at different concentrations (0.01 nM to 1 μM) for
72 h. 4-[3-(4-Iodophenyl)-2-(4-nitrophenyl)-2H-5-tetrazolio]-1,3-ben-
zene disulfonate (WST-1) reagent was incubated for 4 h, and results
are reported here as control-normalized means standard deviation of
three independent experiments (n = 6). Nonlinear regression and
relative IC50 data was calculated by GraphPad Prism 5 (GraphPad
Software, La Jolla, CA).
A panel of 6 breast cancer cell lines, a choriocarcinoma cell
investigate the effect of PGD, and its unmodified predecessor
line, and a normal kidney cell line were used as models to
PAMAM, on metabolic activity (Figure 5). The unconjugated
H
dx.doi.org/10.1021/mp500128k | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics
Article
Figure 5. Effect of PGD on metabolic activity. 4-[3-(4-Iodophenyl)-2-(4-nitrophenyl)-2H-5-tetrazolio]-1,3-benzene disulfonate (WST-1) assay of
(a) MDA-MB-231, (b) MDA-MB-468, (c) MDA-MB-435, (d) BT-20, (e) T47D, (f) BT-549, (g) JEG-3, and (h) BUMPT cells were treated with 5
and 10 nM PTX, PTX-equivalent PGD, or PGD-equivalent PAMAM for 72 h. MDA-MB-231, MDA-MB-468, MDA-MB-435, BT-20, T47D, and
BT-549 are human breast cancer cell lines. JEG-3 is a human choriocarcinoma cell line. BUMPT is a mouse kidney proximal tubule cell line. Results
are reported as mean standard deviation of three independent experiments (n = 6). Statistically significant results are indicated by grouped lines (p
< 0.05).
I
dx.doi.org/10.1021/mp500128k | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics
Article
vehicle PAMAM served as a negative control. All cell lines
demonstrated that there was no statistical difference between
PAMAM treated (both 5 and 10 nM) samples and untreated
samples. Cancer cell lines that had a moderate-to-high
cathepsin B presence and enzymatic activity demonstrated
stronger response to PGD treatment than PTX alone. JEG-3,
MDA-MB-435, and BT-20 cell lines each had a statistically
significant (p < 0.05) difference between the PGD and PTX
response at both 5 and 10 nM concentrations. MDA-MB-231
and T47D cells demonstrated statistically significant differences
between PGD and PTX response at only 5 and 10 nM
concentrations, respectively, but not the other concentrations.
MDA-MB-468 cells demonstrated no statistical significant
metabolic activity response from PGD and PTX treatments at
either 5 or 10 nM concentrations. However, only PGD
treatment was statistically different from the untreated samples
at the 5 nM concentration.
Cells with the lowest cathepsin B presence, the cancer cell
line BT549 and the normal kidney cell line BUMPT,
demonstrated an inverse of this relationship between PTX
and PGD. The cytotoxic effect of PGD on metabolic activity
was reduced in BT549 cells and eliminated in BUMPT cells.
Treatment of cells with PTX still produced a statistically
significant decrease in metabolic activity. Thus, from a broad
perspective, cathepsin B levels correlate to PGD activity. Cells
with a moderate-to-high cathepsin B activity showed a
noticeable reduction in metabolic activity when treated with
PGD as compared to PTX, while cells with low cathepsin B
showed an inverse of this relationship. As cathepsin B is present
at a very basal level under tight regulation in most normal
tissues,59,60 with expectedly similar expression profiles to that
observed for BUMPT, these results suggest that PGD may
lower toxicity from PTX treatment to normal tissues while
maintaining or increasing therapeutic efficacy to cancer cells
overexpressing cathepsin B. In particular, the results indicate
potentially increased protection for the xenobiotic-sensitive
kidneys.
Figure 6. Regression analysis between magnitude of PGD-induced
cytotoxic improvement (metabolic activity differential) at 10 nM
treatment and normalized cathepsin B expression. At a 10 nM PTX-
equivalent treatment, PGD-induced mean metabolic activity was
subtracted from PTX-induced mean metabolic activity for each cell
line, resulting in a metabolic activity differential. This was plotted
against the respective cell line’s cathepsin B expression as determined
by densitometric analysis from Western blots.
MDA-MB-231 breast cancer cells were chosen as model cells
for further studies due to their favorable response rates in the
metabolic activity assays, appropriateness as models of
malignant metastatic breast cancer phenotype, tumorigenicity
in vivo, and well-characterized nature with regards to PTX
response.62,63
To confirm if reduced metabolic activity by PGD conjugate
was a result of cell death via the apoptotic pathway, annexin V-
FITC staining was used to quantitatively determine the
percentage of cells within a population that were undergoing
apoptosis (Figure 7). MDA-MB-231 cells were treated for 48 h
with 50 nM PTX, PTX-equivalent PGD, and PAMAM (as
negative control). Our results indicate that PAMAM treatment
produced no significant difference (α = 0.05) in percent of cells
stained with annexin V compared to untreated controls (28%
versus 33%, respectively). By contrast, both PTX and PGD
This was verified by regression analysis between the
magnitude of PGD-induced cytotoxic improvement at 10 nM
treatment and cellular cathpesin B expression (Figure 6). PGD-
induced mean metabolic activity was subtracted from PTX-
induced mean metabolic activity for each cell line, resulting in a
metabolic activity differential that was plotted against the cell
line’s cathepsin B expression. The coefficient of determination,
r2, was 0.652 at p < 0.05. This suggested a strong, statistically
significant correlation between cathepsin B expression and
increased ability of PGD to induce cytotoxicity compared to
PTX (at 10 nM concentration).
Still, the correlation between cathepsin B expression and
PGD-induced cytotoxicity is not a perfect positive fit,
suggesting that factors other than cathepsin B expression
alone may contribute to targeting effectiveness. In particular,
recent work has suggested that PAMAM-based PTX conjugates
adversely affect cell behavior by PTX-dependent and PTX-
independent modes of action.61 Similarly, beyond a threshold-
level of total cellular cathepsin B activity, cell-dependent
alternative mechanisms may be responsible for the variability in
cell line response observed in our experiments, such as cell line-
influenced nanoparticle absorption capability and replication
rate. Alternatively, cell sensitivity to PTX may reflect on the
ability of PGD dendrimer to elicit a larger response. Such
mechanistic studies will be pursued in due course as they are
beyond the scope of the current work.
Figure 7. PGD exposure to MDA-MB-231 cells induces apoptosis.
Annexin V staining was conducted with the use of a kit (Annexin V-
FITC apoptosis detection kit; BD Pharmingen) and analyzed by flow
cytometry. MDA-MB-231 breast cancer cells were exposed to 50 nM
PTX, PTX-equivalent PGD, and PGD-equivalent PAMAM for 48 h.
Untreated blank samples served as control. Results are reported as
mean
standard deviation of three independent experiments.
Statistically significant results are indicated by grouped lines (p <
0.05).
J
dx.doi.org/10.1021/mp500128k | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics
Article
treatments elicited a strong response with nearly 80% of cells
analyzed staining positively for annexin V. Cells under PGD
treatment had slightly stronger response with 81% of cells
staining for annexin V as compared to 78% of cells under PTX
treatment; these results were statistically differentiable from
untreated cells (controls) but not from each other. The results
suggested that cell death from the apoptotic pathway was a key
component in the observed reduced proliferation rates in the
WST-1 assays. Further, given the similarity in annexin staining,
the underlying mechanisms responsible for apoptosis under
PGD treatment were not altered from that of PTX treatment.
While our studies confirmed the cytotoxic behavior of PGD
toward cancer cells during in vitro testing, two-dimensional cell
cultures cannot replicate the cell−cell and cell−matrix
interactions of the inherently three-dimensional system faced
by drug delivery platforms in vivo, let alone the drastically
different diffusion and transport properties.64 The behavior of
nanoparticles in vivo often differs significantly from the
predicted behavior derived from in vitro cell culture studies.65
As a result, to validate previous observations and discern new
insights, the effect of PGD as compared to PTX on tumor
growth was evaluated through in vivo xenograft studies using
MDA-MB-231 cells. Mice treated from all groups survived until
prescheduled termination of experiment 32 days after tumor
inoculation. No treatment showed external toxic or lethal side-
effects such as, furring, weakness, or fatigue.
Unloaded PAMAM dendrimer was tested as vehicle control
to verify that it did not contribute to antitumor activity
observed during PGD treatment. As can be seen (Figure 8),
unloaded PAMAM was statistically indistinguishable at every
measured time point from mice without treatment (saline
injections). By contrast, treatment with PGD began to show
reduced growth in tumor volume compared to control 1 week
after treatment initiation and had statistically significant
differences (p < 0.05) in tumor volume by weeks 2 and 3.
PGD-treated tumor volumes were up to 58% and 49% smaller
than controls for weeks 2 and 3, respectively. Mice treated with
PTX had consistently smaller tumor volumes than control
throughout the experiment as well. Yet, the tumor volumes
were not significantly different (p > 0.05) from control until 3
weeks after treatment initiation, when PTX-treated tumor
volumes were on average 23% smaller than controls. The
growth of tumors treated with PGD was slower compared to
those treated with PTX, resulting in statistically significant
differences in tumor volume between the pair in weeks 2 and 3
after treatment initiation. PGD-treated tumor volumes were
48% and 34% smaller than PTX-treated tumor volumes at
weeks 2 and 3 after treatment initiation, respectively. Thus,
there was a discernible impact from PGD treatment on slowing
the growth rate of MDA-MB-231 tumor volumes as compared
to PTX treatment, more so than anticipated from the in vitro
cell culture system. Gains in active targeting by the PGD seem
to be amplified due to enhanced permeation and retention
effects of nanoparticles observed under physiological con-
ditions. With an improved and sustained treatment schedule
along with higher drug delivery, it is anticipated that further
gains in tumor growth reduction are possible.
Figure 8. Effect of PGD dendrimer on human breast cancer (MDA-
MB-231) xenograft tumors in mice. Subcutaneous tumors were
generated by injecting 2 × 106 cells into the mammary fatpads of nude
mice. When tumor volumes reached 90 mm3, therapy was initiated and
administered every other day for three intraperitoneal injections total
using the following treatments: control (saline), PAMAM, PTX, and
PGD. Doses of 10 mg PTX/kg body weight were administered as PTX
alone or PTX-equivalent PGD. Tumor volumes were measured 1, 2,
and 3 weeks after therapy initiation and reported as mean standard
error of the mean (n = 7). Statistically significant results are indicated
by grouped lines (p < 0.05). Results are reported as two different
representations (a,b) of the same data set.
inhibitors may explore if increased cathepsin B expression alone
is directly involved in the release of paclitaxel and inducement
of cell death. As this work demonstrates, enzymes with similar
structural activity to cathepsin B (such as those belonging to
the papain family of enzymes) can effectively release PTX−
GFLG product. Such enzymes may have concurrent over-
expression with cathepsin B in malignant cancer cells; therefore,
cathepsin B expression alone may not be responsible for the
appreciable PGD induced cytotoxicity. Similarly, the current
work provided only limited investigation into the mode of
action of PGD activity on cells (whether PTX-dependent or
PTX-independent), absorption kinetics, and other rationales for
cell line variability in response to PGD above a certain
cathepsin B expression threshold. Future work should
mechanistically explore the nature of PGD activity. Finally,
the in vivo activity in this study was limited to investigation of
tumor reduction capabilities of PGD as compared to PTX.
Future studies would expand on this work to probe the
pharmacodynamics and distribution of PGD in vivo as it
compares to PTX alone.
This research, thus, provides a detailed account of the
synthesis and initial biological exploration of an active-targeting,
PTX-laden prodrug. Future studies, however, must overcome a
few limitations of the current work. While a correlation was
established in this work between in vitro increased cytotoxicity
by PGD and cathepsin B expression, future work using enzyme
CONCLUSIONS
■
A novel PTX conjugate of PAMAM dendrimer through a
cathepsin B cleavable GFLG peptide spacer (PGD) was
synthesized and characterized. PTX when delivered through
K
dx.doi.org/10.1021/mp500128k | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics
Article
polyalkylcyanoacrylate nanoparticles. J. Pharm. Sci. 1980, 69 (2), 199−
202.
(11) Bristol-Myers Squibb. Taxol (paclitaxel) Injection (patient
information included). http://packageinserts.bms.com/pi/pi_taxol.pdf
(accessed May 25, 2013).
this conjugate showed a better ability to reduce metabolic
activity (increased cytotoxicity) to an array of cancer cells than
when delivered as drug alone, in line with moderate-to-high
cathepsin B expression and activity in these cells. By contrast,
cells with low cathepsin B activity demonstrated the opposite
relationship. Cytotoxicity by the PGD conjugate was inducted
by the apoptotic pathway, similar to that of PTX treatment.
The PGD conjugate demonstrated markedly higher tumor
reduction in MDA-MB-231 xenograft mice models as
compared to PTX treatment alone, suggesting the successful
development of a cancer-specific active targeting delivery
vehicle.
(12) Merisko-Liversidge, E.; Sarpotdar, P.; Bruno, J.; Hajj, S.; Wei, L.;
Peltier, N.; Rake, J.; Shaw, J. M.; Pugh, S.; Polin, L.; Jones, J.; Corbett,
T.; Cooper, E.; Liversidge, G. G. Formulation and antitumor activity
evaluation of nanocrystalline suspensions of poorly soluble anticancer
drugs. Pharm. Res. 1996, 13 (2), 272−278.
(13) Constantinides, P. P.; Lambert, K. J.; Tustian, A. K.; Schneider,
B.; Lalji, S.; Ma, W.; Wentzel, B.; Kessler, D.; Worah, D.; Quay, S. C.
Formulation development and antitumor activity of a filter sterilizable
emulsion of paclitaxel. Pharm. Res. 2000, 17 (2), 175−182.
(14) Weiss, R. B.; Donehower, R. C.; Wiernik, P. H.; Ohnuma, T.;
Gralla, R. L.; Trump, D.; Baker, J. R.; VanEcho, D. A.; VonHoff, D. D.;
Leyland-Jones, B. Hypersensitivity reactions from taxol. J. Clin. Oncol.
1990, 8, 1263−1268.
AUTHOR INFORMATION
■
Corresponding Author
*(A.S.) E-mail: arpsatsangi@gmail.com.
(15) Surapaneni, M. S.; Das, S. K.; Das, N. G. Designing paclitaxel
drug delivery systems aimed at improved patient outcomes: Current
status and challenges. ISRN Pharmacol. 2012, No. 623139.
(16) Sohn, J. S.; Jin, J. I.; Hess, M.; Jo, B. W. Polymer prodrug
approaches applied to paclitaxel. Polym. Chem. 2010, 1, 778−792.
(17) Sigma-Aldrich. Solubility Properties of PAMAM Dendrimers.
http://www.sigmaaldrich.com/materials-science/nanomaterials/
dendrimers/solubility.html (accessed May 28, 2013).
(18) Rolland, O.; Turrin, C. O.; Caminade, A. M.; Majoral, J. P.
Dendrimers and nanomedicine: multivalency in action. New J. Chem.
2009, 33, 1809−1824.
(19) Nanjwade, B. K.; Bechra, H. M.; Derkar, G. K.; Manvi, F. V.;
Nanjwade, V. K. Dendrimers: Emerging polymers for drug-delivery
systems. Eur. J. Pharm. Sci. 2009, 38, 185−196.
(20) Tomalia, D. A.; Christensen, J. B.; Boas, U. (2012) The
Dendridic State. In Dendrimers, Dendrons, and Dendridic Polymers:
Discovery, Applications and Future; Cambridge University Press,
Cambridge, UK, 2012; pp 92−95.
(21) Teow, H. M.; Zhou, Z.; Najlah, M.; Yusof, S. R.; Abbott, N. J.;
D’Emanuelea, A. Delivery of paclitaxel across cellular barriers using a
dendrimer-based nanocarrier. Int. J. Pharm. 2013, 441, 701−711.
(22) Wang, T. H.; Wang, H. S.; Soong, Y. K. Paclitaxel-induced cell
death: Where the cell cycle and apoptosis come together. Cancer 2000,
88 (11), 2619−2628.
(23) Luo, Y.; Ziebell, M. R.; Prestwich, G. D. A hyaluronic acid-taxol
antitumor bioconjugate targeted to cancer cells. Biomacromolecules
2000, 1 (2), 208−218.
(24) Gelderblom, H.; Verweij, J.; Nooter, K.; Sparreboom, V.
Cremophor EL the drawbacks and advantages of vehicle selection for
drug formulation. Eur. J. Cancer. 2001, 37, 1590−1598.
(25) Gao, Z.; Lukyanov, A. N.; Singhal, A.; Torchilin, V. P.
Diacyllipid-polymer micelles as nanocarriers for poorly soluble
anticancer drugs. Nano Lett. 2002, 2 (9), 979−982.
(26) Liu, C.; Strobl, J. S.; Bane, S.; Schilling, J. K.; McCracken, M.;
Chatterjee, S. K.; Rahim-Bata, R.; Kingston, D. G. I. Design, synthesis,
and bioactivities of steroid-linked taxol analogues as potential targeted
drugs for prostate and breast cancer. J. Nat. Prod. 2004, 67 (2), 152−
159.
(27) Hayashi, Y.; Skwarczynski, M.; Hamada, Y.; Sohma, Y.; Kimura,
T.; Kiso, Y. A. Novel approach of water-soluble paclitaxel prodrug with
no auxiliary and no byproduct: Design and synthesis of isotaxel. J. Med.
Chem. 2003, 46 (18), 3782−3784.
(28) Elliott, E.; Sloane, B. F. The cysteine protease cathepsin B in
cancer. Perspect. Drug Discovery Des. 1996, 6, 12−32.
(29) Mohamed, M. M.; Sloane, B. F. Cysteine cathepsins:
multifunctional enzymes in cancer. Nat. Rev. Cancer. 2006, 6, 764−
775.
(30) Nouh, M. A.; Mohamed, M. M.; El-Shinawi, M.; Shaalan, M. A.;
Cavallo-Medved, D.; Khaled, H. M.; Sloane, B. F. Cathepsin B a
potential prognostic marker for inflammatory breast cancer. J. Transl.
Med. 2011, 9, 1.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Support for this research was provided, in part, by the
Translational Science Training (TST) Across Disciplines
program at the University of Texas Health Science Center at
San Antonio, funded by the University of Texas System’s
Graduate Programs Initiative. The laboratory space and
facilities were provided by the Department of Biomedical
Engineering at the University of Texas at San Antonio, by the
Department of Obstetrics and Gynecology at the University of
Texas Health Science Center at San Antonio and by RANN
Research Corporation, San Antonio, Texas.
REFERENCES
■
(1) American Cancer Society. Cancer Facts and Figures 2013. http://
www.cancer.org/research/cancerfactsstatistics/cancerfactsfigures2013/
index (accessed May 20, 2013).
(2) National Cancer Institute. Common Cancers Type. http://www.
cancer.gov/cancertopics/types/commoncancers (accessed May 20,
2013).
(3) Piccart-Gebhart, M. J.; Burzykowski, T.; Buyse, M.; Sledge, G.;
Carmichael, J.; Luck, H. J.; Mackey, J. R.; Nabholtz, J. M.; Paridaens,
R.; Biganzoli, L.; Jassem, J.; Bontenbal, M.; Bonneterre, J.; Chan, S.;
Basaran, G. A.; Therasse, P. Taxanes alone or in combination with
anthracyclines as first-line therapy of patients with metastatic breast
cancer. J. Clin. Oncol. 2008, 26 (12), 1980−1986.
̈
(4) Fauzee, N. J. S.; Dong, Z.; Wang, Y. L. Taxanes: Promising Anti-
Cancer Drugs. Asian Pacific J. Cancer Prevent. 2011, 12, 837−851.
(5) Sparano, J. A.; Wang, M.; Martino, S.; Jones, V.; Perez, E. A.;
Saphner, T.; Wolff, A. C.; Sledge, G. W.; Wood, W. C.; Davidson, N.
E. Weekly paclitaxel in the adjuvant treatment of breast cancer. New
Engl. J. Med. 2008, 358, 1663−1671.
(6) Singla, A. K.; Garg, A.; Aggarwal, D. Paclitaxel and its
formulations. Int. J. Pharm. 2002, 235, 179−192.
(7) Holmes, F. A.; Walters, R. S.; Theriault, R. L.; Forman, A. D.;
Newton, L. K.; Raber, M. N.; Buzdar, A. U.; Frye, D. K.; Hortobagyi,
G. N. Phase II trial of taxol, an active drug in the treatment of
metastatic breast cancer. J. Natl. Cancer Inst. 1991, 83, 1797−1805.
(8) Straubinger, R. M. Biopharmaceutics of paclitaxel (Taxol):
formulation, activity and pharmacokinetics. In Taxol: Science and
Applications; Suffness, M., Ed.; CRC Press: Boca Raton, FL, 1995; pp
237−258.
(9) Couvreur, P.; Tulkens, P.; Roland, M.; Trouet, A.; Speiser, P.
Nanocapsules: a new type of lysosomotropic carrier. FEBS Lett. 1977,
84 (2), 323−326.
(10) Couvreur, P.; Kante, B.; Lenaerts, V.; Scailteur, V.; Roland, M.;
Speiser, P. Tissue distribution of antitumor drugs associated with
L
dx.doi.org/10.1021/mp500128k | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics
Article
̂
(31) Bremer, C.; Tung, C. H.; Bogdanov, A.; Weissleder, R. Imaging
of differential protease expression in breast cancers for detection of
aggressive tumor phenotypes. Radiology 2002, 222 (3), 814−818.
(32) Dubowchik, G. M.; Firestone, R. A.; Padilla, L.; Willner, D.;
Hofstead, S. J.; Mosure, K.; Knipe, J. O.; Lasch, S. J.; Trail, P. A.
Cathepsin B-labile dipeptide linkers for lysosomal release of
doxorubicin from internalizing immunoconjugates: Model studies of
enzymatic drug release and antigen-specific in vitro anticancer activity.
Bioconjugate Chem. 2002, 13, 855−869.
(33) Malugin, A.; Kopeckova, P.; Kopecek, J. Liberation of
doxorubicin from hpma copolymer conjugate is essential for the
induction of cell cycle arrest and nuclear fragmentation in ovarian
carcinoma cells. J. Controlled Release 2007, 124, 6−10.
(34) Kurtoglu, Y. E.; Mishra, M. K.; Kannan, S.; Kannan, R. M. Drug
release characteristics of PAMAM dendrimer−drug conjugates with
different linkers. Int. J. Pharm. 2010, 384, 189−194.
(35) Borgman, M. P.; Ray, A.; Kolhatkar, R. B.; Sausville, E. A.;
Burger, A. M.; Ghandehari, H. Targetable HPMA copolymer−
aminohexylgeldanamycin conjugates for prostate cancer therapy.
Pharm. Res. 2009, 26 (6), 1407−1418.
(36) Duncan, R. The dawning era of polymer therapeutics. Nat. Rev.
Drug Discovery 2003, 2, 347−360.
(37) Kopecek, J.; Kopeckova, P.; Minko, T.; Lu, Z. R. HPMA
copolymer−anticancer drug conjugates: design, activity, and mecha-
nism of action. Eur. J. Pharm. Biopharm. 2000, 50, 61−81.
(38) Majoros, I. J.; Myc, A.; Thomas, T.; Mehta, C. B.; Baker, J. R.
PAMAM dendrimer-based multifunctional conjugate for cancer
therapy: synthesis, characterization, and functionality. Biomacromole-
cules 2006, 7 (2), 572−9.
(39) Deutsch, H. M.; Glinski, J. A.; Hernandez, M.; Haugwitz, R. D.;
Narayanan, V. L.; Suffness, M.; Zalkow, L. H. Synthesis of congeners
and prodrugs. 3. Water-soluble prodrugs of taxol with potent
antitumor activity. J. Med. Chem. 1989, 32 (4), 788−792.
(40) Ryppa, C.; Mann-Steinberg, H.; Biniossek, M. L.; Satchi-
Fainaro, R.; Kratz, F. In vitro and in vivo evaluation of a paclitaxel
conjugate with the divalent peptide E-[c(RGDfK)2] that targets
integrin ανβ3. Int. J. Pharm. 2009, 368, 89−97.
(41) Yang, N.; Ye, Z.; Li, F.; Mahato, R. I. HPMA polymer-based
site-specific delivery of oligonucleotides to hepatic stellate cells.
Bioconjugate Chem. 2009, 20 (2), 213−221.
(42) Zhou, Y.; Yang, J.; Kopecek, J. Selective inhibitory effect of
̌
HPMA copolymer-cyclopamine conjugate on prostate cancer stem
cells. Biomaterials 2012, 33 (6), 1863−72.
(51) Filho, O. G.; Miiji, L. N. O.; Vainchenker, M.; Gordan, A. N.
Breast cancer with choriocarcinomatous and neuroendocrine features.
Sao Paulo Med. J. 2001, 119 (4), 154−155.
(52) L’Azou, B.; Jorly, J.; On, D.; Sellier, E.; Moisan, F.; Fleury-Feith,
J.; Cambar, J.; Brochard, P.; Ohayon-Courtes
nanoparticles on renal cells. Part. Fibre Toxicol. 2008, 5, 22.
(53) Kaczirek, K.; Schindl, M.; Weinhausel, A.; Scheuba, C.; Passler,
̀
, C. In vitro effects of
̈
C.; Prager, G.; Raderer, M.; Hamilton, G.; Mittlbock, M.; Siegl, V.;
̈
Pfragner, R.; Niederle, B. Cytotoxic activity of camptothecin and
paclitaxel in newly established continuous human medullary thyroid
carcinoma cell lines. J. Clin. Endocrinol. Metab. 2004, 89 (5), 2397−
2401.
(54) Awasthi, N.; Zhang, C.; Schwarz, A. M.; Hinz, S.; Wang, C.;
Williams, N. S.; Schwarz, M. A.; Schwarz, R. E. Comparative benefits
of Nab-paclitaxel over gemcitabine or polysorbate-based docetaxel in
experimental pancreatic cancer. Carcinogenesis 2013, 34 (10), 2361−
2369.
(55) GraphPad. http://www.graphpad.com/support/faqid/1566/
(accessed June 2013).
(56) Keenan, J.; Liang, Y.; Clynes, M. Two-deoxyglucose as an anti-
metabolite in human carcinoma cell line RPMI-2650 and drug-
resistant variants. Anticancer Res. 2004, 24, 433−440.
(57) Ashikaga, T.; Yoshida, Y.; Hirota, M.; Yoneyama, K.; Itagaki, H.;
Sakaguchi, H.; Miyazawa, M.; Ito, Y.; Suzuki, H.; Toyoda, H.
Development of an in vitro skin sensitization test using human cell
lines: The human cell line activation Test (h-CLAT) I. Optimization
of the h-CLAT protocol. Toxicol. In Vitro 2006, 20, 767−773.
(58) Zhang, X.; Stecklein, S.; Behbod, F.; Cohen, M. S. A New
Mechanism Of Action For Withaferin A In Breast Cancer: Induction Of
Proteasomal-degradation of Notch1 and Notch3 Proteins; 7th Annual
Academic Surgical Congress, 2012; Control # ASC20120846.
(59) Gondi, C. S.; Rao, J. S. Cathepsin B as a cancer target. Expert
Opin. Ther. Targets 2013, 17 (3), 281−91.
(60) Kominami, E.; Tsukahara, T.; Bando, Y.; Katunuma, N.
Distribution of cathepsins B and H in rat tissues and peripheral
blood cells. J. Biochem. 1985, 98 (1), 87−93.
(61) Cline, E. N.; Li, M. H.; Choi, S. K.; Herbstman, J. F.; Kaul, N.;
Meyhofer, E.; Skiniotis, G.; Baker, J. R.; Larson, R. G.; Walter, N. G.
̈
Paclitaxel-conjugated PAMAM dendrimers adversely affect micro-
tubule structure through two independent modes of action.
Biomacromolecules 2013, 14 (3), 654−64.
(62) Lacroix, M.; Leclercq, G. Relevance of breast cancer cell lines as
models for breast tumours: an update. Breast Cancer Res. Treat. 2004,
83 (3), 249−89.
(63) Nakayama, S.; Torikoshi, Y.; Takahashi, T.; Yoshida, T.; Sudo,
T.; Matsushima, T.; Kawasaki, Y.; Katayama, A.; Gohda, K.;
Hortobagyi, G. N.; Noguchi, S.; Sakai, T.; Ishihara, H.; Ueno, N. T.
Prediction of paclitaxel sensitivity by CDK1 and CDK2 activity in
human breast cancer cells. Breast Cancer Res. 2009, 11, R12.
(64) Lee, J.; Lilly, G. D.; Doty, R. C.; Podsiadlo, P.; Kotov, N. A. In
vitro toxicity testing of nanoparticles in 3D cell culture. Small 2009, 5
(10), 1213−21.
(43) Vadlamudi, R. K.; Wang, R. A.; Mazumdar, A.; Kim, Y. S.; Shin,
J.; Sahin, A.; Kumar, R. Molecular cloning and characterization of
PELP1, a novel human coregulator of estrogen receptor a. J. Biol.
Chem. 2001, 276, 38272−38279.
(44) Khandare, J.; Minko, T. Polymer−drug conjugates: Progress in
polymeric prodrugs. Prog. Polym. Sci. 2006, 31, 359−397.
(45) Vaidya, A. A.; Lele, B. S.; Kulkarni, M. G.; Mashelkar, R. A.
Enhancing ligand-protein binding in affinity thermoprecipitation:
elucidation of spacer effects. Biotechnol. Bioeng. 1999, 64, 418−25.
(46) Kingston, D. G. I. Taxol: the chemistry and structure-activity
relationships of a novel anticancer agent. Trends Biotechnol. 1994, 12,
222−227.
(65) Sayes, C. M.; Reed, K. L.; Warheit, D. B. Assessing toxicity of
fine and nanoparticles: comparing in vitro measurements to in vivo
pulmonary toxicity profiles. Toxicol. Sci. 2007, 97 (1), 163−80.
(47) Skwarczynski, M.; Hayashi, Y.; Kiso, Y. Paclitaxel prodrugs:
Toward smarter delivery of anticancer agents. J. Med. Chem. 2006, 49
(25), 7253−7269.
(48) Mortier, K. A.; Zhang, G.-F.; Van Peteghem, C. H.; Lambert, W.
E. Adduct formation in quantitative bioanalysis: Effect of ionization
conditions on paclitaxel. J. Am. Soc. Mass Spectrom. 2004, 15, 585−592.
(49) Hou, L.; Yao, J.; Zhou, J. Simultaneous LC−MS analysis of
paclitaxel and retinoic acid in plasma and tissues from tumor-bearing
mice. Chromatographia 2011, 73, 471−480.
̈
(50) Erhan, Y.; Ozdemir, N.; Zekioglu, O.; Nart, D.; Ciris, M. Breast
carcinomas with choriocarcinomatous features: Case reports and
review of the literature. Breast J. 2002, 8 (4), 244−248.
M
dx.doi.org/10.1021/mp500128k | Mol. Pharmaceutics XXXX, XXX, XXX−XXX