Development of novel bisphosphonate prodrugs of doxorubicin for targeting bone metastases that are cleaved pH dependently or by cathepsin B: Synthesis, cleavage properties, and binding properties to hydroxyapatite as well as bone matrix

Article abstract of DOI:10.1021/jm300493m

Bone metastases are a frequent cause of morbidity in cancer patients. The present palliative therapeutic options are chemotherapy, hormone therapy, and the administration of bisphosphonates. The affinity between bisphosphonates and the apatite structure of bone metastases is strong. Thus, we designed two low-molecular-weight and water-soluble prodrugs which incorporate a bisphosphonate group as a bone targeting ligand, doxorubicin as the anticancer agent, and either an acid-sensitive bond (1) or a cathepsin B cleavable bond (3) for ensuring an effective release of doxorubicin at the site of action. Cleavage studies of both prodrugs showed a fast release of doxorubicin but sufficient stability over several hours in human plasma. Effective binding of prodrug 1 and 3 was demonstrated with hydroxyapatite and with native bone. In orientating toxicity studies in nude mice, the MTD of 1 was 3-fold higher compared to conventional doxorubicin, whereas 3 showed essentially the same MTD as doxorubicin.

Full text of DOI:10.1021/jm300493m

Article
pubs.acs.org/jmc
Development of Novel Bisphosphonate Prodrugs of Doxorubicin for
Targeting Bone Metastases That Are Cleaved pH Dependently or by
Cathepsin B: Synthesis, Cleavage Properties, and Binding Properties
to Hydroxyapatite As Well As Bone Matrix
Katrin Hochdorffer,† Khalid Abu Ajaj,†,§ Cynthia Schafer-Obodozie,‡ and Felix Kratz*,†
̈
̈
†Division of Macromolecular Prodrugs, Tumor Biology Center, Breisacher Straße 117, 79106 Freiburg, Germany
‡ProQinase GmbH, Breisacher Straße 117, 79106 Freiburg, Germany
S
* Supporting Information
ABSTRACT: Bone metastases are a frequent cause of morbidity in cancer patients.
The present palliative therapeutic options are chemotherapy, hormone therapy, and the
administration of bisphosphonates. The affinity between bisphosphonates and the
apatite structure of bone metastases is strong. Thus, we designed two low-molecular-
weight and water-soluble prodrugs which incorporate a bisphosphonate group as a
bone targeting ligand, doxorubicin as the anticancer agent, and either an acid-sensitive
bond (1) or a cathepsin B cleavable bond (3) for ensuring an effective release of
doxorubicin at the site of action. Cleavage studies of both prodrugs showed a fast
release of doxorubicin but sufficient stability over several hours in human plasma.
Effective binding of prodrug 1 and 3 was demonstrated with hydroxyapatite and with
native bone. In orientating toxicity studies in nude mice, the MTD of 1 was 3-fold
higher compared to conventional doxorubicin, whereas 3 showed essentially the same
MTD as doxorubicin.
INTRODUCTION
■
Bone cancer is manifested as: (a) primary bone cancer which
directly develops from bone and bone tissue (e.g., cartilage,
osteoblasts, or osteoclasts) and (b) secondary bone tumors
which originate in other sites and form metastases in the
skeleton. The latter occur far more often than primary bone
cancer. The annual incidence of cancer in the United Stated in
2011 was estimated to be 1.6 million cases, with approximately
3000 new cases of primary bone cancer. In striking contrast,
secondary bone cancer will occur in up to 50% of patients.1
Bone metastases are a frequent cause of morbidity in cancer
patients and are usually incurable.2 Bone metastases are
predominant in breast (65−75%), prostate (65−75%), thyroid
(60%), lung (30−40%), and kidney cancer (20−25%).3 In the
majority of cases, bone metastases cause pathologic fractures,
severe pain, and life-threatening hypercalcaemia. Patients who
are diagnosed with bone metastases generally cannot be treated
curatively, e.g., only 20% of patients with breast cancer are still
alive five years after the diagnosis of bone metastases.4
The present options for treating bone metastases are surgery,
radiotherapy, chemotherapy, hormone therapy, and the
administration of bisphosphonates. These therapeutic options
can improve the quality of life in some cases but without
prolonging survival. Bisphosphonates possess a high accumu-
lation in bone and bone metastases and are presently the most
active inhibitors of bone degradation. Their chemical structure
is derived from pyrophosphate (Figure 1) in which the ester
group is replaced by an enzymatically stable CR2 group.
Figure 1. Structure of pyrophosphate and bisphosphonate.
To date, eight bisphosphonates (Figure 2) are clinically
established which vary in their side groups and binding
constants to calcium hydroxyapatite.
Although bisphosphonates are one of the best options to
date for alleviating bone pain, reducing bone destruction, and
inhibiting tumor growth, new therapeutic approaches are
urgently required in order to improve the outcome of treating
bone metastases that are at present merely palliative.
As bone-seeking agents, bisphosphonates demonstrate an
uptake in the bone of 20−80% of the administered dose.5 After
intravenous administration of a bisphosphonate, approximately
50−75% of the injected dose binds to exposed bone mineral,
with half-lives in the blood circulation in the range of 0.5−6 h
in humans depending on the bisphosphonate and administered
dose.5−7 An important feature of bisphosphonates is that in
general the uptake in bone metastases is 10−20-fold higher
than in healthy bone tissue.8−10
Received: April 7, 2012
Published: August 13, 2012
© 2012 American Chemical Society
7502
dx.doi.org/10.1021/jm300493m | J. Med. Chem. 2012, 55, 7502−7515

Journal of Medicinal Chemistry
Article
Figure 2. Structure of clinically established bisphosphonates.
Figure 3. Structures of bisphosphonate drug conjugates.
Surprisingly, only studies to a limited extent have exploited
the bone-seeking properties of low-molecular-weight bis-
phosphonates by conjugating them with an anticancer agent.
Figure 3 depicts synthetically conjugates or complexes of
bisphosphonates.
and 1,2,4-triglycidylurazol, was first reported in 1986 by
Wingen et al..11 This compound demonstrated a reduction of
tumor osteolysis but was inferior to pamidronate.11 Con-
jugation of pamidronate with melphalan yielded 4-[4-[bis(2-
chloroethyl)amino]phenyl]-1-hydroxybutane-1,1-bisphos-
phonic acid (BAD) with an improved survival time in a bone
metastasis model Walker 256 carcinosarcoma. However, the
Diglycidyl-[3-(3,3-diphosphonate-3-hydroxy-propylamino)-
2-hydroxy-propyl]-urazol (DDU), a conjugate of pamidronate
7503
dx.doi.org/10.1021/jm300493m | J. Med. Chem. 2012, 55, 7502−7515

Journal of Medicinal Chemistry
Article
cytostatic efficacy of BAD was similar compared to
melphalan.12 Although it showed a higher anticancer activity
than monotherapy with either melphalan or pamidronate in N-
methyl-N-nitrosourea induced mammary carcinomas in
Sprague−Dawley rats, a combination therapy of pamidronate
and melphalan showed the best therapeutic efficacy in this
tumor model.13 3-[Bis-(2-chloroethyl)-amino]-4-methylphenyl-
hydroxymethane-1,1-bisphosphonic acid (BCMP), a derivative
of BAD, had a lower cytostatic activity than BAD alone against
rat osteosarcoma, and as a result it was not further
investigated.14
Keppler et al. prepared a bisphosphonate cisplatin-analogue
in which a phosphonate is complexed with diaminoplatinum
(cis-diammine[nitrilotris(methylphosphonato)(2)-O1,N1]-
platinum(II), AMDP).15 The therapeutic efficacy of this
complex was better than pamidronate but not superior to
conventional cisplatin in a transplantable osteosarcoma model
of the rat, even at a 28-fold higher dose.16
A bisphosphonate derivative of methotrexate (MTX-BP)
demonstrated a higher therapeutic activity than conventional
methotrexate against human osteosarcoma implanted in nude
mice.17,18 A biodistribution study of a 99 mTc-labeled MTX-BP
conjugate was comparable to 99 mTc-labeled bisphosphonate,
demonstrating an accumulation of ∼20% of the administered
dose in skeletal tissue of mice.19
No antitumor effect at all was obtained when using a
conjugate incorporating a bisphosphonate bound to the amino
group of doxorubicin when evaluated against human tumor
xenografts.20
of bone metabolism that are especially predominant in the
formation and growth of bone metastases are: (a) the acidic
environment in the lacunae that is necessary for bone
desorption29−34 and (b) the expression of several proteases
such as matrix metalloproteinase, urokinase plasminogen
activator, and cathepsins (cathepsins K, L, and B)35−40 that
are responsible for osteolysis, bone remodeling, and the growth
(osteoblasts) or degradation (osteoclasts) of the bone matrix
and bone metastases. Thus we set out to develop novel
bisphosphonate prodrugs that exploit the profound bone-
seeking properties of the bisphosphonate moiety as well as the
microenvironment of bone metastases to ensure a selective
release and accumulation of anticancer agent at the desired site
of action.
In this work, we describe the synthesis of novel bi-
sphosphonate doxorubicin prodrugs that are effectively cleaved
in a pH-dependent manner or by cathepsin B and demonstrate
pronounced binding properties to hydroxyapatite and native
bone matrix.
RESULTS AND DISCUSSION
■
Rationale and Chemistry. The aim of this work was to
develop prodrugs with bone targeting properties that selectively
release an anticancer agent in the skeletal metastases. The
general design of our new bisphosphonate prodrugs is shown in
Figure 4.
El-Mabhouh et al. prepared a compound in which a
bisphosphonate was conjugated to the anticancer agent
gemcitabine (Gem/BP, see Figure 3). This conjugate was
examined in in vitro and in vivo studies, and approximately
∼67% of retained whole-body activity was bound to the bone at
8 h after administration of labeled 188Re-Gem/BP. In addition,
in a nude mice model using the human breast cancer cell line
MDA-MB-231BO, unlabeled Gem/BP reduced the number
and size of bone metastases relative to gemcitabine-treated and
untreated control groups. Histological examinations confirmed
the superior therapeutic efficacy of Gem/BP in comparison
with commercial gemcitabine.21−23
Recently, Shabat et al. developed bisphosphonate conjugates
with the anticancer drug camptothecin in which the hydroxyl
group of camptothecin was bound through an ester bond to
tetraethyl bisphosphonate carboxylic acid (camdronate) and
tryptophan as a model compound in which the amino group
was bound through a carbonate bond to a bisphosphonate
(trypdronate, see Figure 3). Both bisphosphonate conjugates
were shown to bind to hydroxyapatite in vitro.24 Besides low-
molecular-weight bisphosphonate compounds, HPMA copoly-
mer drug conjugates with alendronate and either paclitaxel or
the antiangiogenic drug TNP-470 have recently been described,
showing antiangiogenic activity in bone metastases models in
vivo.25−28
Figure 4. General design of our new bisphosphonate prodrugs.
The prodrugs comprise a thiol-bearing bisphosphonate that
has reacted with a maleimide-bearing doxorubicin prodrug.
This design is based on our long-term experience in the
development of thiol-binding doxorubicin prodrugs.41−46 Such
maleimide-based prodrugs were initially developed to bind
rapidly and selectively to the cysteine-34 position of circulating
albumin due to tumor uptake of albumin uptake in malignant
tissue.47 In situ binding of prodrugs to endogenous albumin is
meanwhile a preclinically and clinically validated technology of
increasing the therapeutic index of anticancer agents. An acid-
sensitive albumin-binding prodrug of doxorubicin 9, the (6-
maleimidohexanoyl)hydrazone derivative of doxorubicin
(INNO-206, previously known as DOXO-EMCH) developed
in our group42 has been studied clinically46 and is undergoing a
phase II study in soft tissue sarcoma and pancreatic cancer (see
http://www.cytrx.com).
However, proof-of-concepts for straightforward low-molec-
ular-weight anticancer bisphosphonate drug conjugates have,
on the whole, been disappointing. Although a bone targeting
effect for most bisphosphonate drug conjugates was obtained,
the major drawback in the design of these conjugates is to our
mind a lack of a predetermined breaking point that allows the
bound drug to be specifically released in bone metastases.
Surprisingly, bisphosphonate prodrugs have not been reported
in the literature. Two physiological and biochemical properties
Furthermore, a series of albumin-binding prodrugs with
enzymatically cleavable linkers have been developed in the past
10 years demonstrating superior efficacy and tolerability in
preclinical tumor-bearing mice models.47 Maleimide-bearing
prodrugs were therefore an obvious choice for developing
bisphosphonate prodrugs.
For developing bone-seeking prodrugs, we chose doxorubicin
for the following reasons: (a) doxorubicin is commonly used in
those cancer indications where bone metastases frequently
7504
dx.doi.org/10.1021/jm300493m | J. Med. Chem. 2012, 55, 7502−7515

Journal of Medicinal Chemistry
Article
occur such as breast, prostate, and thyroid cancer as well as
against osteosarcoma,48 (b) in the past, we have developed
various thiol-binding prodrugs of doxorubicin with acid-
sensitive or enzymatically cleavable linkers.47,49,50
Scheme 1. Chemical Synthesis of the Thiol-Bearing
Bisphosphonate 8
a
A suitable thiol-bearing bisphosphonate is 2-(3-
mercaptopropylthio)ethane-1,1-diyl-diphosphonic acid that
has been described in the literature.51 The rationale for
selecting an appropriate predetermined breaking point when
designing bisphosphonate prodrugs was based on the character-
istic biochemical and physiological properties of bone
metastases. Osteoclasitc and osteoblastic bone metastases
exhibit an acidic environment in order to degrade the apatite
matrix during bone remodelling and the formation of bone
metastases. Bone metastases are characterized by a high
expression of several proteases including cathepsins, matrix
metalloproteases, plasmin, and uPA.35,37−40,52,53 Key proteases
which are overexpressed in bone metastasis are cathepsin K and
B.38−40
a
Reagents and conditions: (a) HNEt2, (HCHO)n, MeOH, reflux, 24
h; (b) toluene, p-TsOH, molecular sieve (4 Å), reflux, 24 h; (c)
BrSi(CH3)3, DCM, 0 °C → rt, MeOH; (d) 1,3-propanedithiol, 1-
butanol, reflux → rt, H2O.
Cathepsin B is a cysteine proteases which is activated in an
acidic microenvironment and can participate in the vicious
cycle of bone metastases.53 For example, it is expressed at low
levels in primary prostate tumors; however, bone metastatic
lesions express high levels of activated cathepsin B, suggesting
that protease activity is modulated by interactions between
tumor cells and the bone microenvironment.38,54
In summary, incorporating an acid-sensitive or cathepsin B
cleavable linker in a doxorubicin bisphosphonate prodrug
should be an effective way of releasing doxorubicin specifically
in areas of bone metastases after uptake of the prodrug in
skeletal metastases. As suitable maleimide-bearing prodrugs of
doxorubicin we selected (a) the acid-sensitive albumin-binding
prodrug, the (6-maleimidohexanoyl)hydrazone derivative of
doxorubicin (DOXO-EMCH), in which doxorubicin is
derivatized at its C-13 keto position with a thiol-binding linker
and (b) the cathepsin B cleavable doxorubicin prodrugs EMC-
Phe-Lys-PABC-DOXO or EMC-Val-Ala-PABC-DOXO (EMC
= 6-maleimidocaproic acid), which incorporate a 1,6-self-
immolative p-aminobenzyloxycarbonyl (PABC) linker between
the drug and the enzymatically cleavable dipeptide.50
of 1,3-propanedithiol in a high yield (88% over the last two
steps).
Synthesis of the Acid-Sensitive Bisphosphonate Prodrug 1.
The acid-sensitive bisphosphonate prodrug 1 was obtained in a
rapid Michael addition from the thiol-bearing moiety 8 when
coupled to the thiol-binding compound 9 (Scheme 2). Prodrug
1 was obtained as a water-soluble red solid after purification by
preparative reverse phase chromatography and subsequent
lyophilization.
Synthesis of Bisphosphonate Doxorubicin Prodrugs with
Cathepsin B Substrates (2 and 3). A suitable thiol-binding
doxorubicin derivative for synthesizing a cathepsin B cleavable
bisphosphonate prodrug from thiol-bearing aliphatic bis-
phosphonate 8 is EMC-Phe-Lys-PABC-DOXO 10, first
published by Dubowchik et al.,55 that we have previously
synthesized as an albumin-binding prodrug.50 10 incorporates a
self-immolative Phe-Lys-PABC-spacer that allows doxorubicin
to be released efficiently after cleavage of Phe-Lys as the
cathepsin B substrate. This compound bound to the cysteine-34
position of albumin was cleaved efficiently by cathepsin B
releasing the free drug and exhibited superior antitumor efficacy
in vivo compared to doxorubicin.50
Consequently, we coupled 10 to the bisphosphonate 8 as
depicted in Scheme 3 and obtained the doxorubicin bi-
sphosphonate 2.
Although 2 was cleaved efficiently by cathepsin B (see
Supporting Information), its low aqueous solubility (approx-
imately 1 mM) does not make it an optimal candidate and
limits its investigation for in vivo studies.
As an alternative to 2, we decided to use another cathepsin B
substrate, i.e., Val-Ala, described by Peterson et al.56 Following
our previous synthetic route for obtaining bisphosphonate
prodrugs, we synthesized a new thiol-binding prodrug 16
(EMC-Val-Ala-PABC-DOXO) as depicted in Scheme 4. The
self-immolative spacer of 12 was subsequently reacted with
bis(p-nitrophenyl) carbonate (bis-PNP carbonate) to yield the
activated carbonate 13. Afterward, doxorubicin was introduced
in the presence of N,N-diisopropylethylamine (DIEA) and
subsequently piperidine was added to remove the Fmoc-
protecting group. In a final step, the 6-maleimidocaproic-N-
hydroxysuccinimide ester (EMC-OSu) was reacted with the
amino-group of compound 15 to yield the cathepsin B
cleavable compound 16.
Synthesis of Bisphosphonate Doxorubicin Prodrugs That
Are Cleaved in a pH-Dependent Manner (1) and by
Cathepsin B (2 and 3). According to our synthetic approach,
the construction of the prodrugs is based on the condensation
of a thiol-bearing bisphosphonate moiety with a thiol-binding
group conjugated with the anticancer drug.
The initial step in the development of our bisphosphonate
prodrugs was the synthesis of a thiol-bearing aliphatic
bisphosphonate compound, which allows the attachment of
the thiol binding group of the doxorubicin prodrugs. The
synthesis of the bisphosphonate 8 (Scheme 1) was performed
according to published procedures,51 with modifications in
steps two and three (see Scheme 1 and Supporting
Information). 5 was obtained by reacting commercially
available tetraethyl methylene-bisphosphonate 4 with diethyl-
amine and paraformaldehyde in methanol. Subsequent acid-
catalyzed elimination furnished compound 6. In this step,
calcium hydride was substituted by powdered molecular sieve
(4 Å), and the product was lyophilized from the aqueous layer.
The bisphosphonate ester 6 was converted into the free acid 7
by using the dealkylating reagent bromotrimethyl silane
(BrSi(CH3)3). The modification in this step was the use of
methanol instead of 40% aqueous tetrabutylammonium
hydroxide. Isolation of compound 8 was achieved by addition
As shown in Scheme 5, the bisphosphonate prodrug 3 was
obtained by reacting the thiol-bearing bisphosphonate 8 with
7505
dx.doi.org/10.1021/jm300493m | J. Med. Chem. 2012, 55, 7502−7515

Journal of Medicinal Chemistry
Article
a
Scheme 2. Synthesis of the Acid-Sensitive Doxorubicin Bisphosphonate Prodrug 1
a
Reagents and conditions: 50 mM NH4HCO3 buffer (pH 7.4) including 25% tert-butanol, rt, 10 min; purification by C-18 RP chromatography.
a
Scheme 3. Synthesis of the Cathepsin B Cleavable Bisphosphonate Prodrug 2
a
Reagents and conditions: 50 mM sodium phosphate buffer (pH 7.4)/ethanol 1:1, 37 °C, 30 min.
16. 3 is formed within 15 min in a Michael addition as
confirmed by HPLC.
Purification of 3 was achieved with preparative C-18 reverse
phase chromatography and subsequent lyophilization yielded a
red solid. Water solubility of 3 was considerably better than for
2 exceeding 5 mM.
Metastases. Cleavage Studies of the Bisphosphonate
Prodrugs 1, 2, and 3. Following the synthesis of the prodrugs,
we studied their cleavage properties in analogy to our previous
studies.42,50 On the basis of our rationale that the new
bisphosphonate doxorubicin prodrugs should release doxo-
rubicin under conditions prevailing in the tumor environment
of bone metastases, we investigated the cleavage properties of
Cleavage and Binding Properties of Novel Bisphosph-
onate Doxorubicin Prodrugs for Targeting Bone
7506
dx.doi.org/10.1021/jm300493m | J. Med. Chem. 2012, 55, 7502−7515

Journal of Medicinal Chemistry
Article
a
Scheme 4. Synthesis of EMC-Val-Ala-PABC-DOXO 16
a
Reagents and conditions: (e) H-Ala-OH, NaHCO3, H2O, THF, rt; (f) PABOH, EEDQ, DCM, rt; (g) DIEA, bis(p-nitrophenyl) carbonate, DCM,
DMF, rt; (h) doxorubicin·HCl, DIEA, DMF, RT; (i) 20% piperidine in DMF, RT; (j) DCM, DMF, EMC-OSu, rt.
a
Scheme 5. Synthesis of Bisphosphonate Prodrug 3
a
Reagents and conditions: 50 mM NH4HCO3 buffer (pH 7.4)/ MeOH (1:1), rt; purification by C-18 RP chromatography.
the prodrugs according to the nature of the incorporated
cleavage point.
∼4% doxorubicin was observed with a time-dependent decrease
of the initial peak eluting at a retention time of ∼3 min of the
bisphosphonate prodrug 1. The peak intensity of doxorubicin
simultaneously increased. The half-life of 1 at pH 5.5 at room
temperature was approximately 2.5 h.
The acid-sensitive hydrazone bond in 1 should allow
doxorubicin to be released at the acidic pH values present in
the resorption lacuna. Hence, we incubated compound 1 at pH
5.5 and analyzed the release of doxorubicin over a period of 5 h
by HPLC (see Figure 5). Already after 5 min, an amount of
To confirm that the release of doxorubicin from 1 was acid-
promoted, we monitored the stability of 1 at pH 7.4 over 18 h.
7507
dx.doi.org/10.1021/jm300493m | J. Med. Chem. 2012, 55, 7502−7515

Journal of Medicinal Chemistry
Article
Figure 5. Chromatograms of an incubation study of compound 1 at pH 5.5 and room temperature over time. 1 was dissolved in 4 mM sodium
phosphate buffer containing 150 mM sodium chloride (pH 7.4), and the pH value was adjusted to pH 5.5 with 50 mM sodium acetate buffer to
obtain a final concentration of 2.4 mM of 1. For HPLC injection, the solution was diluted to 600 μM. A chromatogram of doxorubicin is included as
reference.
Figure 6. Chromatograms of a stability study of prodrug 1 (800 μM) at pH 7.4 (4 mM sodium phosphate buffer containing 150 mM sodium
chloride) at room temperature.
As noted in Figure 6, 1 exhibited a considerably higher stability
at this pH value with approximately 12% doxorubicin being
released after an 18 h incubation period.
cells. To determine the stability of prodrugs 1 and 3 in plasma,
these compounds were incubated in human plasma at 37 °C,
and subsequently samples were taken over an incubation period
of 6 h and analyzed by HPLC, as can be seen in Figure 8A,B.
While the cathepsin B cleavable prodrug 3 was stable under
these conditions with no doxorubicin or other anthracycline
derivatives being observed (Figure 8B), prodrug 1 showed a
slight degradation in human plasma with small amounts of
doxorubicin (retention time ∼7 min) and an unidentified
anthracycline derivative (∼ 2.5 min) appearing over the 6 h
incubation period (Figure 8A). Considering that the half-life in
human beings of circulating bisphosphonates is short, between
0.5 and 2 h due to rapid bone uptake as well as renal clearance,5
it can be surmised that the stability of both prodrugs is
sufficient to allow them to accumulate in bone and bone
metastases. In mice and rats, the half-life of bisphosphonates is
considerably shorter than in humans, in the order of only a few
minutes.5,6,59
To assess the cleavage properties of prodrugs 2 and 3, the
respective doxorubicin bisphosphonate derivative was incu-
bated with cathepsin B at pH 5.0 and pH 6.0 at 37 °C. Both
prodrugs were cleaved to doxorubicin (see Supporting
Information for 2 and Figure 7 for 3). Under the conditions
chosen (cathepsin B 0.4 mg/mL, 23.8 U/mg, buffer: 50 mM
sodium acetate, 100 mM NaCl, 4 mM EDTA, pH 5.0,
containing 8 mM L-cysteine), the half-life for 3 was ∼45 min,
with the sole cleavage product being doxorubicin and
completely released after 120 min (see Figure 7A). At pH 6.0
under the same conditions, cleavage was faster, with
doxorubicin already being completely liberated after 40 min
(see Figure 7C). Incubation of compound 3 without cathepsin
B at pH 5.0 at 37 °C served as a control experiment (see Figure
7B) and showed no cleavage over 3 h. The pH optimum of
cathepsin B is documented to vary in the range of pH 5.0−6.5
depending on the nature of the substrate.57,58
Hydroxyapatite Binding Assay for Both Bisphosphonate
Prodrugs at pH 6.0 and pH 7.4. A further aim of our work was
to assess the affinity between our new bisphosphonate prodrugs
and the bone mineral hydroxyapatite (HA). First, we incubated
prodrug 3 with different equivalents of hydroxyapatite (30, 40,
and 50 equiv) and monitored HA binding photometrically
Stability Studies of 1 and 3 in Human Plasma. The plasma
stability is an important criterion for the biological activity of a
synthesized prodrug. It should be stable long enough to enable
the transport of the active molecule to the tumor and tumor
7508
dx.doi.org/10.1021/jm300493m | J. Med. Chem. 2012, 55, 7502−7515

Journal of Medicinal Chemistry
Article
Figure 7. Chromatograms of an incubation study of compound 3 at pH 5.0 as well as at pH 6.0 (37 °C) with cathepsin B (A,C) and without
cathepsin B (B). (A) A 176 μL stock solution of compound 3 (1.5 mM) was diluted with 9 μL of cathepsin B (0.4 mg/mL, 23.8 U/mg) and 264 μL
of a buffer at pH 5.0 (50 mM sodium acetate, 100 mM NaCl, 4 mM EDTA, pH 5.0) or (C) of a buffer at pH 6.0 (50 mM sodium acetate, 100 mM
NaCl, 4 mM EDTA, pH 6.0) containing L-cysteine (8 mM). The cleavage product doxorubicin is included as reference.
determining the decrease of 3 in the supernatant over time. Not
unexpectedly, the kinetics of binding was dependent on the
excess of hydroxyapatite that was used (see Figure S16 in
Supporting Information). For further validation of the binding
assay, we opted to use 30 equiv of HA. Binding to
hydroxyapatite was carried out at pH 7.4 as well as pH 6.0 to
simulate the pH values prevalent in plasma as well as the acidic
bone lacunae. When doxorubicin was incubated with
hydroxyapatite powder at 37 °C at both pH 6.0 and pH 7.4,
no or only very weak binding of doxorubicin to hydroxyapatite
could be observed (see Figures 9 and 10). In contrast to
doxorubicin, the bisphosphonate prodrugs 1 and 3 were rapidly
bound to hydroxyapatite under these conditions. At pH 7.4,
approximately 50% of both prodrugs were bound to
hydroxyapatite after 1 h (see Figure 9) and a concomitant
red staining of the original white hydroxyapatite powder was
observed. The binding of prodrug 3 increases with time, and it
reached over 90% after 5 h and essentially 100% after 24 h. The
acid sensitive prodrug 1 demonstrated a similar curve shape but
saturation of HA binding occurred after 5 h with ∼80% of 1
being bound to HA. This phenomenon could be explained by
catalytic cleavage of the carboxylic hydrazone bond once 1 is
7509
dx.doi.org/10.1021/jm300493m | J. Med. Chem. 2012, 55, 7502−7515

Journal of Medicinal Chemistry
Article
Figure 8. Chromatograms of an incubation study of prodrugs 1 (A) and 3 (B) in human plasma at 37 °C over 6 h, recorded at 495 nm.
Figure 9. Binding of doxorubicin, of prodrugs 1 and 3 to hydroxyapatite (HA) at pH 7.4 and 37 °C.
bound to HA, releasing doxorubicin from the hydroxyapatite
matrix.
acid-sensitive prodrug 1 to HA did not increase, which can be
explained by a competing acid-promoted cleavage to
doxorubicin. In contrast, prodrug 3 showed a very high affinity
to hydroxyapatite at pH 6.0 with over 90% being bound to HA.
Bone Binding Assay for Both Bisphosphonate Prodrugs.
To confirm whether the new bisphosphonate prodrugs interact
with the bone matrix, we determined their binding property to
native bone. For this assay, bovine bone pieces were cut into
very small fragments, washed several times with distilled water
Because of acidic pH value in the Howship’s lacunae in
bones, we also investigated the binding properties of the
prodrugs and hydroxyapatite at pH 6.0 and at 37 °C (Figure
10). For both prodrugs, we observed a higher affinity for
hydroxyapatite after 1 h with 65−70% of the prodrug being
bound compared to ∼50% after incubation at pH 7.4. Not
unexpectedly, after this initial fast binding phase, binding of the
7510
dx.doi.org/10.1021/jm300493m | J. Med. Chem. 2012, 55, 7502−7515

Journal of Medicinal Chemistry
Article
Figure 10. Binding of doxorubicin, of prodrugs 1 and 3 to hydroxyapatite (HA) at pH 6.0 and 37 °C.
Figure 11. Incubation study of prodrugs 1 and 3 with native bone after 4 h at 37 °C and at pH 6.0 or pH 7.4.
and anhydrous ethanol, and dried at room temperature
overnight. Both prodrugs were incubated with the bone
matrices suspended in phosphate buffer at pH 7.4 and pH
6.0, respectively (at 37 °C). Figure 11 summarizes the data of
native bone incubated with both prodrugs at pH 7.4 and at pH
6.0. Because of the results obtained in the hydroxyapatite
binding assay, we incubated the prodrugs over a period of 4 h
and measured the absorbance of the supernatant at 495 nm.
Conventional doxorubicin was also incubated with the bone
fragments serving as the negative control. As expected, we
observed no binding of doxorubicin to the native bone. In
contrast, 80−90% of both prodrugs were bound to the bone
matrix after 4 h at pH 6.0 (Figure 11). At pH 7.4, binding to the
bone was lower, with 50−70% of 1 or 3 being bound after 4 h
incubation.
Orientating Toxicity Study of Prodrugs 1 and 3. For
initiating an in vivo trial in a bone metastases animal model, we
determined the systemic toxicity of 1 and 3 in female balb/C
nude mice. Initially, two groups of three mice (groups 1 and 4)
were treated intravenously once weekly with 8 mg/kg
doxorubicin equivalents of compounds 1 or 3 (see Supporting
Information) for three weeks. Within the first week after
therapy, no weight losses or other drug-related side effects were
observed (see Figure S17 of Supporting Information). Hence,
the administration of prodrug 1 and 3 at higher doses (16 and
24 mg/kg doxorubicin equivalents) was carried out. Shortly
after the initial treatment with both doses (16 and 24 mg/kg
doxorubicin equivalents) of prodrug 3, all animals died (group
2 and 3). Animals treated with 16 and 24 mg/kg doxorubicin
equivalents of prodrug 1, however, showed no acute side effects
(groups 5 and 6). All animals were observed for four weeks, and
behavior and body weight changes were documented. After 13
days after the last injection, all three animals of group 6 showed
signs of anemia and one animal had died. Body weights
remained constant or demonstrated the typical slight increase
characteristic for adolescent animals during the term of the
study. On the basis of this orientating toxicity study, the
maximum tolerated dose, dosed on a weekly schedule, was
estimated as 3 × 24 mg/kg for the acid-sensitive prodrug 1 and
3 × 8 mg/kg for 3, which is slightly above the MTD of 2 × 8
mg/kg for doxorubicin in nude mice.
CONCLUSION
■
In this work, we report on the development of tailor-made
prodrugs with bisphosphonates for treating bone metastases, a
synthetic approach that has not been described to date. Two
new water-soluble prodrugs which incorporated a bisphosph-
onate as the bone targeting moiety and doxorubicin as the
anticancer agent were prepared that release the parent
7511
dx.doi.org/10.1021/jm300493m | J. Med. Chem. 2012, 55, 7502−7515

Journal of Medicinal Chemistry
Article
phosphate buffer (pH 7.0); injection volume, 50 μL; gradient,
0−1.5 min 100% mobile phase A; 1.5−20 min increase to 100%
mobile phase B; 20−45 min 30% CH3CN, 70% 20 mM sodium
phosphate buffer; 45−55 min decrease to initial mobile phase;
55−58 min 100% mobile phase A.
anthracycline either at an acidic pH value or enzymatically by
cathepsin B. Sufficient stability in human plasma as well as high
affinity for hydroxyapatite and native bone was demonstrated,
making the novel prodrugs 1 and 3 suitable candidates for in
vivo evaluation in a bone metastases model.
(c) Waters System: Column, Waters, 100 Å, Symmetry C18 5 μm
[4.6 mm × 250 mm] with precolumn [3.9 mm × 20 mm].
Chromatographic conditions: isocratic flow, 1.0 mL/min;
mobile phase, 30% CH3CN, 70% 10 mM sodium phosphate
buffer (pH 7.8).
(d) Waters System: Column, Waters, 300 Å, Symmetry C18 5 μm
[4.6 mm × 250 mm] with precolumn [3.9 mm × 20 mm].
Chromatographic conditions: isocratic flow, 1.0 mL/min;
mobile phase, 30% CH3CN, 70% 20 mM sodium phosphate
buffer (pH 7.8).
EXPERIMENTAL SECTION
■
General. Doxorubicin ·HCl was purchased from Yick-Vic (Hong
Kong, China). The (6-maleimidohexanoyl)hydrazone derivative of
doxorubicin (DOXO-EMCH) was prepared as described previously.42
EMC-Phe-Lys-PABC-DOXO was prepared as described previously.60
All other organic solvents and chemicals used were at least reagent
grade and obtained from Merck (Damstadt, Germany), Sigma-Aldrich,
and Fluka (Munich, Germany) or Roth (Karlsruhe, Germany) and
used without further purification. The buffers used were vacuum-
filtered through a 0.2 μm membrane (Sartorius, Gottingen, Germany)
̈
Incubation Studies with Human Plasma and Prodrugs 1 and 3
at 37 °C. Human blood plasma (EDTA stabilized) was taken from
healthy volunteers. 1 or 3 was dissolved in 10 mM sodium phosphate
buffer (7.4) containing 0.15 M sodium chloride [3 mM]. The
respective sample was added to human plasma (EDTA as
anticoagulant) to obtain a final concentration of 600 μM, and the
samples incubated for 0, 0.5, 1, 3, and 6 h at 37 °C. Then 100 μL of
each sample was centrifuged in a 1.5 mL Eppendorf reaction tube at a
maximum of 13000 rpm (duration 60 s), and a 20 μL sample was
analyzed by HPLC using method (d) for compound 1 and method (a)
for compound 3.
pH-Dependent Studies of Compound 1 at pH 5.5. The stability of
1 was monitored by HPLC at pH 5.5 and 7.4. For studies at pH 7.4, 1
was dissolved in 4 mM sodium phosphate buffer containing 150 mM
sodium chloride (pH 7.4) to obtain a final concentration of 800 μM.
The samples were incubated at room temperature and measured by
HPLC at t = 0, 0.5, 1, 3, 6, and 18 h (method c). For studies at pH 5.5,
the pH value of the acid-sensitive bisphosphonate prodrug 1, dissolved
in 4 mM sodium phosphate, 150 mM NaCl, pH 7.4, at a final
concentration of 2.4 mM, was adjusted to pH 5.5 with 50 mM sodium
acetate (pH 4.5) and incubated at room temperature; samples were
diluted to 600 μM and analyzed by HPLC (method c, injection
volume: 50 μL) at t = 5, 60, 120, 180, 240, and 300 min.
Cleavage Studies of Compounds 2 and 3 with Cathepsin B. First,
100 μL of prodrug 2 (300 μM in 50 mM sodium phosphate buffer, pH
7.0) was added to a solution of a 50 mM sodium acetate buffer
containing 100 mM NaCl, 4 mM EDTA containing L-cysteine (8 mM)
(pH 5.0). Afterward, 6.7 μL of cathepsin B (0.4 mg/mL, 23.8 U/mg)
were added and the solution was incubated for 1 h at 37 °C, and
samples were analyzed by HPLC (method b) after 0 and 60 min. For
3, a 176 μL stock solution of compound 3 (1.5 mM) was diluted with
9 μL of cathepsin B (0.4 mg/mL, 23.8 U/mg) and 264 μL of buffer
(50 mM sodium acetate, 100 mM NaCl, 4 mM EDTA, pH 5.0)
containing L-cysteine (8 mM). The mixture was incubated at 37 °C,
and aliquots (65 μL) were taken after 5, 20, 40, 60, and 120 min and
analyzed by HPLC (method a). As control, the incubation study was
performed with 3 in the absence of cathepsin B under the same
conditions as described above.
and thoroughly degassed prior to use. Enzymatically active cathepsin B
(human liver) was obtained from Calbiochem (Schwalbach am
Taunus, Germany). Hydroxyapatite (HA) was purchased from
Sigma-Aldrich (Munich, Germany).
1H NMR, 13C NMR, and 31P NMR were recorded on a Bruker
Avance DRX spectrometer 400 MHz (400, 100, and 161 MHz for 1H,
13C , and 31P, respectively). Analytical HPLC were performed either
with a Gilson 321 pump, a Kontron 535 detector (at 495 nm), and a
Bischoff Lambda 1010 detector (at 254 nm) or with a Waters system
(pump, Waters 600E+; detector, Waters 996 photodiode array
detector; controller, Waters 600E; auto sampler, Waters 717 plus).
For peak integration, Geminyx software (version 1.91 by Goebel
Instrumentelle Analytik GmbH, Germany) was used or Empower
2002 (version 1). UV/vis-spectrophotometry was carried out with a
double-beam UV/vis-spectrophotometer U-2000 from Hitachi. Mass
spectrometry (ESI-TOF-MS) was performed by A&M Labor fur
̈
Analytik, Bergheim, Germany, using a LTQ-Orbitrap-XL (Thermo
́
Fisher Scientific, San Jose, CA, USA), spray voltage 4.5 kV. Elementary
analyses were performed by vario EL (Elementar Analysensysteme
GmbH, Haunau, Germany). Melting points were determined using a
Buchi 530 (Buchi Labortechnik GmbH, Essen, Germany). Purity of all
̈
̈
compounds (≥95%) was determined by HPLC using one of the
methods specified below.
Chromatography. Preparative separation of crude products was
carried out by column chromatography using silica gel 60 (230−400
mesh, Roth, Karlsruhe, Germany) or Lichroprep RP-18 (Merck,
Darmstadt, Germany). Analytical TLC was performed either on
aluminum-coated plates using TLC silica gel 60 UV254 (Alugram Sil G,
Roth, Karlsruhe, Germany) or on Merck glass plates using silica gel 60
RP-18 F254S. Visualization of TLC plates was realized by suitable
detection methods; spots were identified by UV light absorption λ =
254 nm and/or by treatment with a solution containing
phosphormolybdic acid (25 mL), Cer(IV)-sulfate (10 g), water (940
mL), and concentrated sulfuric acid (60 mL) and subsequent heating
at 150 °C.
Methods. HPLC Studies. Four different types of HPLC methods
were used:
Hydroxyapatite Binding Assay of Compounds 1 and 3 at
Different pH Values. General procedure: In a falcon tube, the test
compound was diluted in phosphate buffer (4 mM sodium phosphate
containing 150 mM NaCl, pH 6.0 and pH 7.4) to obtain a solution
with a concentration of 300 μM, confirmed photometrically (λ = 495
nm, ε = 10650 M−1 cm−1). Hydroxyapatite (30 equiv) was added, and
the suspension was stirred well and incubated at 37 °C. Another
solution of the samples without hydroxyapatite was used as control. In
each case, after 0, 1, 2, 3, 4, 5, 6, and 20 h the samples were centrifuged
(4000 rpm, duration 60 s, Heraeus Megafuge 1.0) and the absorbance
of the supernatant was measured with a spectrophotometer at a
wavelength of 495 nm (ε495(doxorubicin) = 10650 M−1 cm−1). Each
result represents the arithmetic average of three measurements.
Percent HA binding was calculated as following: [(sample
concentration without HA − sample concentration with HA)]/
(sample concentration without HA) × 100%.
(a) Kontron System: Column, Waters, 300 Å, Symmetry C18 5 μm
[4.6 mm × 250 mm] with precolumn [3.9 mm × 20 mm].
Chromatographic conditions: flow, 1.2 mL/min; mobile phase
A, 20% CH3CN, 80% 20 mM sodium phosphate buffer (pH
7.0); mobile phase B, 70% CH3CN, 30% 20 mM sodium
phosphate buffer (pH 7.0); injection volume, 20 μL; gradient,
0−5 min 100% mobile phase A; 5−40 min increase to 70%
CH3CN, 30% 20 mM sodium phosphate buffer; 40−50 min
100% mobile phase B; 50−60 min decrease to initial mobile
phase; 60−65 min 100% mobile phase A.
(b) Waters System: Column, Waters, 300 Å, Symmetry C18 5 μm
[4.6 mm × 250 mm] with precolumn [3.9 mm × 20 mm].
Chromatographic conditions: flow, 1.0 mL/min; mobile phase
A, 15% CH3CN, 85% 20 mM sodium phosphate buffer (pH
7.0); mobile phase B, 30% CH3CN, 70% 20 mM sodium
7512
dx.doi.org/10.1021/jm300493m | J. Med. Chem. 2012, 55, 7502−7515

Journal of Medicinal Chemistry
Article
Bone Assay of Prodrugs 1 and 3 at Different pH Values. A bone
fragment was isolated from the backbone of a cow, and the cortical
bone was cut into small pieces (6 mm ×4 mm ×1 mm). The bone
pieces were left untreated, washed with water (bidest) and absolute
ethanol, and dried. For evaluating binding, the prodrugs were dissolved
in a falcon tube in phosphate buffer (4 mM Na2HPO4 containing 150
mM NaCl, pH 6.0 and pH 7.4) at a concentration of 300 μM, which
was monitored with a spectrophotometer (Hitachi U2000). Approx-
imately 50 mg of an untreated bone piece was added to the clear-red
solution, and the samples were incubated under stirring at 37 °C. As
control, a solution without a bone piece was prepared. Every hour,
over a period of 6 h and after 20 h, the samples were centrifuged (4000
rpm, duration 60 s, Heraeus Megafuge 1.0), and the absorbance of the
supernatant was measured with a spectrophotometer at a wavelength
of 495 nm (ε495(doxorubicin) = 10650 M−1 cm−1). Percent bone
binding was calculated as following: [(sample concentration without
bone − sample concentration with bone)]/(sample concentration
without bone) × 100%.
Orientating Toxicity Study of Prodrugs 1 and 3. All experiments
were performed using five-week-old female Balb/C mice (Charles
River GmbH, Sulzfeld, Germany) with an approximate weight of 20 g.
Mice were maintained in individual ventilated cages (three mice/cage)
at constant temperature and humidity. Experimental protocols had
been approved by the Ethics Committee for Animal Experimentation,
University Freiburg. 1 and 3 were dissolved in a sterile buffer
containing 10 mM sodium phosphate and 0.15 mM sodium chloride
(pH 7.4) at a concentration of 3.4 mM. Bisphosphonate prodrugs 1
and 3 were intravenously administered to six groups of three mice
each. Doses of 8, 16, and 24 mg/kg doxorubicin equivalents were
administered once weekly over a period of three weeks. During this
time, animal behavior and welfare was monitored daily and changes in
body weight were checked every two days.
Synthesis of the Bisphosphonate Prodrug 3. First, 35 mg of
compound 8 was dissolved in 29 mL of 50 mM ammonium
bicarbonate buffer (pH 7.4) to obtain a thiol concentration of 3.2
mM. Then a solution of 99 mg of EMC-Val-Ala-PABC-DOXO 16
dissolved in 40 mL of 10 mM ammonium bicarbonate buffer (pH 7.4,
3.2 mM) was added dropwise to the bisphosphonate solution over 5
min. The mixture was stirred at room temperature for 15 min and was
frozen with liquid nitrogen. After lyophilization, the red crude product
was purified by flash chromatography (RP 18) eluting with CH3CN/
H2O (HPLC grade) 1:1 to give 3 as a red solid. HPLC analysis
(method a): A495 nm ≥ 98% of peak area. Due to the fact that the
lyophilized product contained salts, the content of doxorubicin in the
sample was determined using the ε-value for doxorubicin (ε495 (pH
7.4) = 10650 M−1 cm−1). The HPLC chromatogram and mass
spectrum is included as Supporting Information. C58H75N5O24P2S2
HRMS ESI-TOF: calculated [M + H]+ 1352.3791, found 1352.3778.
ASSOCIATED CONTENT
* Supporting Information
■
S
Syntheses of compounds 8 and 16 including analytical data,
chromatographic methods, HPLC purities, and mass spectra for
compounds 1, 2, 3, 12, 13, 15, and 16. This material is available
free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*Phone: 0049-761-2062930. Fax: 0049-761-2062905. E-mail:
kratz@tumorbio.uni-freiburg.de. Website: http://www.
tumorbio.uni-freiburg.de/04_forschung/makromolekulare_
prodrugs.html.
Synthesis of Compound 8. Synthesis of compound 8 including
analytical data is described in the Supporting Information.
Synthesis of Compound 16. Synthesis of compound 16 including
analytical data is described in the Supporting Information.
Present Address
§Leibniz Institute for Molecular Pharmacology (FMP), Group
of Chemical Systems Biology, Robert-Rossle-Straße 10, 13125
̈
Berlin, Germany.
Synthesis of Bisphosphonate Prodrug 1. First, 50.7 mg of
compound 8 (171 μmol) was dissolved in 43 mL of 50 mM
ammonium bicarbonate buffer (pH 7.4) to obtain a final concentration
of 4 mM. This solution contained approximately 80% free thiol groups
as assessed with the Ellmann's test.61 Subsequntly, a solution of 82 mg
of DOXO-EMCH (1 equiv) dissolved in a mixture of 40 mL of 50
mM ammonium bicarbonate buffer (pH 7.4) and tert-butanol (1:1)
was added dropwise to the bisphosphonate solution over 3 min. The
mixture was stirred at room temperature for 10 min, frozen with liquid
nitrogen, and lyophilized, yielding 236 mg as a red solid. The solid was
dissolved in mobile phase which was acetonitrile/10 mM sodium
phosphate buffer (pH 7.8) 20:80 and purified by C-18 RP
chromatography, and the pure fractions were collected and lyophilized.
The purity of 1 as assessed by HPLC (method c) was A495 nm ≥ 99% of
peak area. Due to the fact that the lyophilized product contained salts,
the content of doxorubicin in the sample was determined using the ε-
value for doxorubicin (ε495 (pH 7.4) = 10650 M−1 cm−1).
C42H56N4O19P2S2 HRMS ESI-TOF: calculated [M + H]+ 1047.2528,
found 1047.2514. The HPLC chromatogram and mass spectrum is
included as Supporting Information.
Synthesis of the Bisphosphonate Prodrug 2. First, 60 mg (3.4
mmol, 1 equiv) of EMC-Phe-Lys-PABC-Doxo 10 dissolved in 14 mL
of ethanol was added dropwise to a solution of 16.7 mg of 7 (3.4
mmol, 1 equiv) dissolved in 14 mL of 50 mM sodium phosphate
buffer (pH 7.4) at 37 °C, and the sample was stirred for 30 min. Then
the reaction mixture was centrifuged, and the solvent removed in high
vacuo, yielding 140 mg of 2 as a red powder. The purity of 2 was
monitored by HPLC (method b), A495 nm ≥ 95% of peak area. Due to
the fact that the lyophilized product contained salts, the content of
doxorubicin in the sample was determined using the ε-value for
doxorubicin (ε495 (pH 7.4) = 10650 M−1 cm−1). C65H81N6O24P2S2
HRMS ESI-TOF: calculated [M + H]+ 1457.4370, found 1457.4377.
The HPLC chromatogram and mass spectrum is included as
Supporting Information.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The financial support of the Deutsche Forschungsgemeinschaft
is gratefully acknowledged.
■
ABBREVIATIONS USED
■
AMDP, cis-diammine[nitrilotris(methylphosphonato)(2)-
O1,N1]-platinum(II); BAD, 4-[4-[bis(2-chloroethyl)amino]-
phenyl]-1-hydroxybutane-1,1-bisphosphonic acid; BCMP, 3-
[bis-(2-chloroethyl)-amino]-4-methylphenyl-hydroxymethane-
1,1-bisphosphonic acid; BP, bisphosphonate; DDU, diglycidyl-
[3-(3,3-diphosphonate-3-hydroxy-propylamino)-2-hydroxy-
propyl]-urazol; DIEA, N,N-diisopropylethylamine; DOXO,
doxorubicin; DOXO-EMCH, 6-maleimidohexanoyl)hydrazone
derivative of doxorubicin; EEDQ, 2-ethoxy-1-ethoxycarbonyl-
1,2-dihydroquinoline; EMC, 6-maleimidocaproic acid; EMC-
OSu, 6-maleimidocaproic-N-hydroxysuccinimide ester; Gem/
BP, bisphosphonate conjugated to gemcitabine; HA, hydrox-
yapatite; MTX-BP, a bisphosphonate derivative of methotrex-
ate; PABC, p-aminobenzyloxycarbonyl; PNP, p-nitrophenyl; p-
TsOH, p-toluenesulfonic acid; uPA, urokinase-type plasmino-
gen activator
REFERENCES
■
(1) Hong, W. K.; Bast, R. C.; Hait, W.; Kufe, D. W.; Holland, J. F.
et al. Holland-Frei Cancer Medicine; 8th ed.; BC Decker: Shelton, CT,
2010.
7513
dx.doi.org/10.1021/jm300493m | J. Med. Chem. 2012, 55, 7502−7515

Journal of Medicinal Chemistry
Article
(2) Rubens, R. D. Bone metastasesthe clinical problem. Eur. J.
Cancer 1998, 34, 210−213.
to assess the potential for targeted chemotherapy of metastatic bone
cancer. Nucl. Med. Biol. 2006, 33, 715−722.
(22) El-Mabhouh, A. A.; Mercer, J. R. 188Re-labelled gemcitabine/
bisphosphonate (Gem/BP): a multifunctional, bone-specific agent as a
potential treatment for bone metastases. Eur. J. Nucl. Med. Mol.
Imaging 2008, 35, 1240−1248.
(23) El-Mabhouh, A. A.; Nation, P. N.; Abele, J. T.; Riauka, T.;
Postema, E.; et al. A conjugate of gemcitabine with bisphosphonate
(Gem/BP) shows potential as a targeted bone-specific therapeutic
agent in an animal model of human breast cancer bone metastases.
Oncol. Res. 2011, 19, 287−295.
(24) Erez, R.; Ebner, S.; Attali, B.; Shabat, D. Chemotherapeutic
bone-targeted bisphosphonate prodrugs with hydrolytic mode of
activation. Bioorg. Med. Chem. Lett. 2008, 18, 816−820.
(3) Coleman, R. E. Skeletal complications of malignancy. Cancer
1997, 80, 1588−1594.
(4) Rubens, R. D. Metastatic breast cancer. Curr. Opin. Oncol. 1995,
7, 523−526.
(5) Fleisch, H. Bisphosphonates in bone disease: from the laboratory
to the patient; 3rd ed.; The Parthenon Publishing Group: London,
1997.
(6) Lin, J. H. Bisphosphonates: a review of their pharmacokinetic
properties. Bone 1996, 18, 75−85.
(7) Bartl, R.; von Treschkow, E.; Bartl, C. Bisphosphonat-Manual:
Wirkungen-Indikationen-Strategien; Springer-Verlag: Berlin−Heidel-
berg, 2006.
(25) Miller, K.; Eldar-Boock, A.; Polyak, D.; Segal, E.; Benayoun, L.;
et al. Antiangiogenic antitumor activity of HPMA copolymer−
paclitaxel−alendronate conjugate on breast cancer bone metastasis
mouse model. Mol. Pharmaceutics 2011, 8, 1052−1062.
(26) Segal, E.; Pan, H.; Benayoun, L.; Kopeckova, P.; Shaked, Y.;
et al. Enhanced antitumor activity and safety profile of targeted
nanoscaled HPMA copolymer−alendronate−TNP-470 conjugate in
the treatment of bone malignances. Biomaterials 2011, 32, 4450−4463.
(27) Miller, K.; Erez, R.; Segal, E.; Shabat, D.; Satchi-Fainaro, R.
Targeting bone metastases with a bispecific anticancer and
antiangiogenic polymer−alendronate−taxane conjugate. Angew.
Chem., Int. Ed. Engl. 2009, 48, 2949−2954.
(28) Pan, H.; Sima, M.; Kopeckova, P.; Wu, K.; Gao, S.; et al.
Biodistribution and pharmacokinetic studies of bone-targeting N-(2-
hydroxypropyl)methacrylamide copolymer−alendronate conjugates.
Mol. Pharmaceutics 2008, 5, 548−558.
(29) Blair, H. C.; Teitelbaum, S. L.; Ghiselli, R.; Gluck, S.
Osteoclastic bone resorption by a polarized vacuolar proton pump.
Science 1989, 245, 855−857.
(30) Palokangas, H.; Mulari, M.; Vaananen, H. K. Endocytic pathway
from the basal plasma membrane to the ruffled border membrane in
bone-resorbing osteoclasts. J. Cell Sci. 1997, 110 (Pt 15), 1767−1780.
(31) Baron, R.; Neff, L.; Louvard, D.; Courtoy, P. J. Cell-mediated
extracellular acidification and bone resorption: evidence for a low pH
in resorbing lacunae and localization of a 100 kD lysosomal membrane
protein at the osteoclast ruffled border. J. Cell Biol. 1985, 101, 2210−
2222.
(8) Goeckeler, W. F.; Edwards, B.; Volkert, W. A.; Holmes, R. A.;
Simon, J.; et al. Skeletal localization of samarium-153 chelates:
potential therapeutic bone agents. J. Nucl. Med. 1987, 28, 495−504.
(9) Bayouth, J. E.; Macey, D. J.; Kasi, L. P.; Fossella, F. V. Dosimetry
and toxicity of samarium-153-EDTMP administered for bone pain due
to skeletal metastases. J. Nucl. Med. 1994, 35, 63−69.
(10) Hirabayashi, H.; Sawamoto, T.; Fujisaki, J.; Tokunaga, Y.;
Kimura, S.; et al. Relationship between physicochemical and
osteotropic properties of bisphosphonic derivatives: rational design
for osteotropic drug delivery system (ODDS). Pharm. Res. 2001, 18,
646−651.
(11) Wingen, F.; Eichmann, T.; Manegold, C.; Krempien, B. Effects
of new bisphosphonic acids on tumor-induced bone destruction in the
rat. J. Cancer Res. Clin. Oncol. 1986, 111, 35−41.
(12) Wingen, F.; Sterz, H.; Blum, H.; Moller, H.; Pittermann, W.;
et al. Synthesis, antitumor activity, distribution and toxicity of 4-[4-
[bis(2-chloroethyl)amino]phenyl]-1-hydroxybutane-1 1-bisphos-
phonic acid (BAD), a new lost derivative with increased accumulation
in rat osteosarcoma. J. Cancer Res. Clin. Oncol. 1986, 111, 209−219.
(13) Wingen, F.; Pool, B. L.; Klein, P.; Klenner, T.; Schmahl, D.
Anticancer activity of bisphosphonic acids in methylnitrosourea-
induced mammary carcinoma of the ratbenefit of combining
bisphosphonates with cytostatic agents. Invest. New Drugs 1988, 6,
155−167.
(14) Pool, B. L.; Berger, M.; Schlehofer, J. R.; Wingen, F. In vivo and
in vitro investigations on biological effects of aromatic bis-(2-
chloroethyl)amino-bisphosphonic acids, new agents proposed for
chemotherapy of bone tumors: cytostatic activity in rat osteosarcoma;
toxicity and genotoxicity in liver and bone marrow; mutagenicity in S.
typhimurium. Invest. New Drugs 1988, 6, 67−78.
(15) Klenner, T.; Valenzuela-Paz, P.; Keppler, B. K.; Angres, G.;
Scherf, H. R.; et al. Cisplatin-linked phosphonates in the treatment of
the transplantable osteosarcoma in vitro and in vivo. Cancer Treat. Rev.
1990, 17, 253−259.
(16) Klenner, T.; Wingen, F.; Keppler, B. K.; Krempien, B.; Schmahl,
D. Anticancer-agent-linked phosphonates with antiosteolytic and
antineoplastic properties: a promising perspective in the treatment
of bone-related malignancies? J. Cancer Res. Clin. Oncol. 1990, 116,
341−350.
(17) Sturtz, G. L.; Appere, G.; Breistol, K.; Fodstad, O.;
Schwartsmann, G.; et al. Eur. J. Med. Chem. 1992, 27, 825−833.
(18) Sturtz, G. L.; Couthon, H. M.; Fabulet, O.; Mian, M.; Rosini, S.
Synthesis of gem-bisphosphonic methotrexate conjugates and their
biological response towards Walker's osteosarcoma. Eur. J. Med. Chem.
1993, 28, 899−903.
(19) Hosain, F.; Spencer, R. P.; Couthon, H. M.; Sturtz, G. L.
Targeted delivery of antineoplastic agent to bone: biodistribution
studies of technetium-99m-labeled gem-bisphosphonate conjugate of
methotrexate. J. Nucl. Med. 1996, 37, 105−107.
(32) Kingsley, L. A.; Fournier, P. G.; Chirgwin, J. M.; Guise, T. A.
Molecular biology of bone metastasis. Mol. Cancer Ther. 2007, 6,
2609−2617.
(33) Teitelbaum, S. L. Bone resorption by osteoclasts. Science 2000,
289, 1504−1508.
(34) Arnett, T. Regulation of bone cell function by acid−base
balance. Proc. Nutr. Soc. 2003, 62, 511−520.
(35) Miyake, H.; Hara, I.; Yamanaka, K.; Arakawa, S.; Kamidono, S.
Elevation of urokinase-type plasminogen activator and its receptor
densities as new predictors of disease progression and prognosis in
men with prostate cancer. Int. J. Oncol. 1999, 14, 535−541.
(36) Woodward, J. K. L.; Holen, I.; Coleman, R. E.; Buttle, D. J. The
roles of proteolytic enzymes in the development of tumour-induced
bone disease in breast and prostate cancer. Bone 2007, 41, 912−927.
(37) Cher, M. L.; Biliran, H. R., Jr.; Bhagat, S.; Meng, Y.; Che, M.;
et al. Maspin expression inhibits osteolysis, tumor growth, and
angiogenesis in a model of prostate cancer bone metastasis. Proc. Natl.
Acad. Sci. U.S.A. 2003, 100, 7847−7852.
(38) Podgorski, I.; Linebaugh, B. E.; Sameni, M.; Jedeszko, C.;
Bhagat, S.; et al. Bone microenvironment modulates expression and
activity of cathepsin B in prostate cancer. Neoplasia 2005, 7, 207−223.
(39) Littlewood-Evans, A. J.; Bilbe, G.; Bowler, W. B.; Farley, D.;
Wlodarski, B.; et al. The osteoclast-associated protease cathepsin K is
expressed in human breast carcinoma. Cancer Res. 1997, 57, 5386−
5390.
(20) Fabulet, O.; Sturtz, G. L. Synthesis of gem-bisphosphonic
doxorubicin conjugates. Phosphorus, Sulfur Silicon Relat. Elem. 1995,
101, 225−234.
(21) El-Mabhouh, A. A.; Angelov, C. A.; Cavell, R.; Mercer, J. R. A
99mTc-labeled gemcitabine bisphosphonate drug conjugate as a probe
(40) Brubaker, K. D.; Vessella, R. L.; True, L. D.; Thomas, R.; Corey,
E. Cathepsin K mRNA and protein expression in prostate cancer
progression. J. Bone Miner. Res. 2003, 18, 222−230.
7514
dx.doi.org/10.1021/jm300493m | J. Med. Chem. 2012, 55, 7502−7515

Journal of Medicinal Chemistry
Article
(41) Kratz, F.; Muller-Driver, R.; Hofmann, I.; Drevs, J.; Unger, C. A
novel macromolecular prodrug concept exploiting endogenous serum
albumin as a drug carrier for cancer chemotherapy. J. Med. Chem.
2000, 43, 1253−1256.
(61) Ellman, G. L. Tissue sulfhydryl groups. Arch. Biochem. Biophys.
1959, 82, 70−77.
(42) Kratz, F.; Warnecke, A.; Scheuermann, K.; Stockmar, C.;
Schwab, J.; et al. Probing the cysteine-34 position of endogenous
serum albumin with thiol-binding doxorubicin derivatives. Improved
efficacy of an acid-sensitive doxorubicin derivative with specific
albumin-binding properties compared to that of the parent compound.
J. Med. Chem. 2002, 45, 5523−5533.
(43) Kratz, F.; Mansour, A.; Soltau, J.; Warnecke, A.; Fichtner, I.;
et al. Development of albumin-binding doxorubicin prodrugs that are
cleaved by prostate-specific antigen. Arch. Pharm. (Weinheim) 2005,
338, 462−472.
(44) Chung, D.-E.; Kratz, F. Development of a novel albumin-
binding prodrug that is cleaved by urokinase-type plasminogen
activator (uPA). Bioorg. Med. Chem. Lett. 2006, 16, 5157−5163.
(45) Schmid, B.; Chung, D.-E.; Warnecke, A.; Fichtner, I.; Kratz, F.
Albumin-binding prodrugs of camptothecin and doxorubicin with an
Ala-Leu-Ala-Leu-linker that are cleaved by cathepsin B: synthesis and
antitumor efficacy. Bioconjugate Chem. 2007, 18, 702−716.
(46) Kratz, F. DOXO-EMCH (INNO-206): the first albumin-
binding prodrug of doxorubicin to enter clinical trials. Expert Opin.
Invest. Drugs 2007, 16, 855−866.
(47) Kratz, F. Albumin as a drug carrier: design of prodrugs, drug
conjugates and nanoparticles. J. Controlled Release 2008, 132, 171−
183.
(48) Gianni, L.; Grasselli, G.; Cresta, S.; Locatelli, A.; Vigano, L.;
et al. Anthracyclines. Cancer Chemother. Biol. Response Modif. 2002, 20,
59−69.
(49) Kratz, F.; Warnecke, A.; Schmid, B.; Chung, D. E.; Gitzel, M.
Prodrugs of anthracyclines in cancer chemotherapy. Curr. Med. Chem.
2006, 13, 477−523.
(50) Abu Ajaj, K.; Graeser, R.; Fichtner, I.; Kratz, F. In vitro and in
vivo study of an albumin-binding prodrug of doxorubicin that is
cleaved by cathepsin B. Cancer Chemother. Pharmacol. 2009, 64, 413−
418.
(51) Bansal, G.; Wright, J. E. I.; Zhang, S.; Zernicke, R. F.; Uludag, H.
Imparting mineral affinity to proteins with thiol-labile disulfide
linkages. J. Biomed. Mater. Res., Part A 2005, 74, 618−628.
(52) Woodward, J. K.; Holen, I.; Coleman, R. E.; Buttle, D. J. The
roles of proteolytic enzymes in the development of tumour-induced
bone disease in breast and prostate cancer. Bone 2007, 41, 912−927.
(53) Webb, S. D.; Sherratt, J. A.; Fish, R. G. Alterations in proteolytic
activity at low pH and its association with invasion: a theoretical
model. Clin. Exp. Metastasis 1999, 17, 397−407.
(54) Arkona, C.; Wiederanders, B. Expression, subcellular distribu-
tion and plasma membrane binding of cathepsin B and gelatinases in
bone metastatic tissue. Biol. Chem. 1996, 377, 695−702.
(55) Dubowchik, G. M.; Firestone, R. A.; Padilla, L.; Willner, D.;
Hofstead, S. J.; et al. 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.
(56) Peterson, J. J.; Meares, C. F. Cathepsin substrates as cleavable
peptide linkers in bioconjugates, selected from a fluorescence quench
combinatorial library. Bioconjugate Chem. 1998, 9, 618−626.
(57) Barrett, A. J.; Kirschke, H. Cathepsin B, cathepsin H, and
cathepsin L. Methods Enzymol. 1981, 80 (Pt C), 535−561.
(58) Swanson, A. A.; Martin, B. J.; Spicer, S. S. Human placental
cathepsin B1. Isolation and some physical properties. Biochem. J. 1974,
137, 223−228.
(59) Fleisch, H. Bisphosphonates. Pharmacology and use in the
treatment of tumour-induced hypercalcaemic and metastatic bone
disease. Drugs 1991, 42, 919−944.
(60) Abu Ajaj, K.; Biniossek, M. L.; Kratz, F. Development of
protein-binding bifunctional linkers for a new generation of dual-acting
prodrugs. Bioconjugate Chem. 2009, 20, 390−396.
7515
dx.doi.org/10.1021/jm300493m | J. Med. Chem. 2012, 55, 7502−7515