A unique asymmetric bow-tie poly(amidoamine) (PAMAM) dendrimer (ABTD) scaffold was designed and developed as a well-defined macromolecular carrier for tumor-targeted drug delivery. The ABTD scaffold in this study consists of a G3-half-dendron (G3-HD) unit and a G1-half-dendron (G1-HD) unit, bearing thiol moiety in each unit and a bis(maleimide) linker unit, which undergo sequential thiol-maleimide coupling to assemble the scaffold. This assembly methodology is applicable to all other combinations of different generations of PAMAM dendrimers. In the prototype ABTD in this study, 16 biotin moieties were tethered to the G3-HD unit and 4 payloads (new-generation taxoid) to the G1-HD via a self-immolative linker to form an ABTD-tumor-targeting conjugate (ABTD-TTC-1). Two other ABTD-TTCs were synthesized, wherein the G1-HD unit was tethered to a fluorescence-labeled taxoid or to a fluorescent probe. These three ABTD-TTCs were constructed by using a common key ABTD 6 bearing a terminal acetylene group in the G1-HD unit, which was fully characterized as a single molecule by high-resolution mass spectrometry and NMR despite its high molecular weight (Mw: 12 876). Then, the click reaction was employed to couple ABTD 6 with a small-molecule payload or fluorescence probe unit bearing a terminal azide moiety. ABTD-TTC-3, as a surrogate of ABTD-TTC-2, showed substantially enhanced internalization into two cancer cell lines via receptor-mediated endocytosis, attributed to multibinding effect. ABTD-TTC-1 exhibited a remarkable selectivity to cancer cells (1400-7500 times) compared to human normal cells, which demonstrates the salient feature and bright prospect of the ABTD-based tumor-targeted drug-delivery system.
A unique asymmetric bow-tie <span class="Chemical">poly(amidoamine) (PAMAM) dendrimer (<span class="Chemical">ABTD) scaffold was designed and developed as a well-defined macromolecular carrier for tumor-targeted drug delivery. The ABTD scaffold in this study consists of a G3-half-dendron (G3-HD) unit and a G1-half-dendron (G1-HD) unit, bearing thiol moiety in each unit and a bis(maleimide) linker unit, which undergo sequential thiol-maleimide coupling to assemble the scaffold. This assembly methodology is applicable to all other combinations of different generations of PAMAM dendrimers. In the prototype ABTD in this study, 16 biotin moieties were tethered to the G3-HD unit and 4 payloads (new-generation taxoid) to the G1-HD via a self-immolative linker to form an ABTD-tumor-targeting conjugate (ABTD-TTC-1). Two other ABTD-TTCs were synthesized, wherein the G1-HD unit was tethered to a fluorescence-labeled taxoid or to a fluorescent probe. These three ABTD-TTCs were constructed by using a common key ABTD 6 bearing a terminal acetylene group in the G1-HD unit, which was fully characterized as a single molecule by high-resolution mass spectrometry and NMR despite its high molecular weight (Mw: 12 876). Then, the click reaction was employed to couple ABTD 6 with a small-molecule payload or fluorescence probe unit bearing a terminal azide moiety. ABTD-TTC-3, as a surrogate of ABTD-TTC-2, showed substantially enhanced internalization into two cancer cell lines via receptor-mediated endocytosis, attributed to multibinding effect. ABTD-TTC-1 exhibited a remarkable selectivity to cancer cells (1400-7500 times) compared to human normal cells, which demonstrates the salient feature and bright prospect of the ABTD-based tumor-targeted drug-delivery system.
<span class="Disease">Cancer is the second
leading cause of <span class="Disease">death, as well as the number
1 cause of death under the age of 85, in the United States.[1] It has been estimated that approximately 1.7
million new cancer cases will occur and over half a million deaths
will be attributed to cancer.[2] Despite
the significant progress in the development of cancer detection, prevention,
surgery, chemotherapy, and radiation therapy, no common cure has been
fully established for patients with this malignant disease. Traditional
chemotherapy relies on the premise that the rapidly proliferating
tumor cells are more likely to be killed by cytotoxic agents than
normal cells. In reality, however, due to the little or no specificity
of these cytotoxic agents, severe systemic toxicity and dose-limiting
side effects may occur and result in anemia, hair loss, and damage
to the liver, kidney, and bone marrow. Accordingly, extensive efforts
have been made for developing tumor-specific delivery systems to overcome
the shortcomings of conventional chemotherapy in the last few decades.[3,4]
In general, a tumor-targeted drug conjugate consists of a
tumor
recognition moiety and a cytotoxic agent connected directly or through
a suitable linker. The tumor-targeted drug conjugate should be systemically
nontoxic, which means the linker must be stable during the blood circulation,
but readily cleaved to release the cytotoxic agent efficiently upon
internalization into the tumor cells. In addition, the linker should
be bifunctional so that both cytotoxic agent and tumor-targeting moiety
can be connected to the linker.Rapidly growing <span class="Disease">cancer cells
require various nutrients and vitamins
so that certain <span class="Disease">cancer-specific receptors are overexpressed to enhance
the uptake. These receptors can be used as targets to deliver the
cytotoxic drugs specifically to the cancer cells via receptor-mediated
endocytosis (RME). Various ligands, including mAbs,[5−9] PUFAs,[10,11] folic acid,[12,13] transferrin,[14] oligopeptides,[15] aptamers,[16] and hyaluronic
acid,[17] have been studied as TTMs to construct
the tumor-targeted drug conjugates.
Biotin, well known as vitamin
H or B7, is essential for cell growth,
production of fatty acids, and metabolism of fats and amino acids.
Considerable evidence has shown that various cancer cells overexpress
biotin receptors (BRs) significantly more than normal cells.[18] Accordingly, we selected biotin receptors on
cancer cell surface as the molecular target and employed biotin as
the primary tumor-targeting moiety to construct our drug conjugate
with cytotoxic agents.<span class="Chemical">Paclitaxel and <span class="Chemical">docetaxel have made significant
impacts on cancer
chemotherapy.[19] However, these drugs seriously
suffer from the lack of tumor specificity and MDR. For solid tumors,
such as breast, ovarian, and lung cancers, paclitaxel and docetaxel
show efficacy, but these taxane drugs are not effective against colon,
pancreatic, melanoma, and renal cancers. For example, human colon
carcinoma is inherently multidrug-resistant because of the overexpression
of P-glycoprotein (P-gp), which is an effective ABC transporter. Overexpression
of P-gp can lead to multidrug resistance as the result of pumping
hydrophobic anticancer agents, such as paclitaxel and docetaxel, out
of cancer cells.[20] On the basis of our
extensive SAR study of taxoids, we have developed a series of highly
potent second- and third-generation taxoids.[21−29] Most of these new-generation taxoids exhibited 2–3 orders
of magnitude higher potency than paclitaxel and docetaxel against
MDR cell lines expressing P-gp.[30] Accordingly,
we chose a highly potent second-generation taxoid, SB-T-1214, as the
cytotoxic payload.
<span class="Disease">Tumor-targeted drug-delivery strategies are
based on “active”
or “passive” targeting. Active targeting relies on the
difference in the expression level of cell surface biomarkers, such
as antigen or receptor, between <span class="Disease">cancer and normal tissue. Extensive
preclinical and clinical studies have been performed on antibody–drug
conjugates (ADCs) for tumor-targeted drug delivery in the last two
decades, wherein monoclonal antibodies (mAbs) targeting cancer surface-specific
antigens, e.g., EGFR, Her-2, CD22, CD30, CD33, etc., are used for “active”
targeting.[4,9,31] Small-molecule
drug conjugates (SMDCs), especially targeting vitamin receptors in
tumor and cancer cells, have also been actively studied.[30,32−34] Another strategy termed as “passive targeting”
aims at employing macromolecules, including polymers or nanoparticles
as vehicles. These macromolecules do not directly interact with tumor
cells, but strongly influence the accumulation, transportation, and
biodistribution of their drug conjugates in tumor tissues due to the
EPR effect. The EPR effect was first elucidated by Maeda and co-workers
in the 1980s.[35] In principle, tumor cell
clusters induce angiogenesis for their increasing demands of oxygen
and nutrition. The new blood vessels generated by the tumor often
have irregular shape and large size pore compared to the healthy tissue.
The leaky and defective vessels make the vasculature of tumor tissue
permeable to macromolecules, which hardly pass the endothelial barrier
of the blood vessels of healthy tissue. Besides, the enhanced uptake
of macromolecules in tumor tissue is also related to its increased
retention in the tumor. If the molecular weight of a macromolecule
reaches 40 kDa, it is difficult to be cleared from the tumor due to
the lack of its lymphatic drainage system.[36]
Dendrimer is a class of <span class="Chemical">polymers that are also often called
starburst
<span class="Chemical">polymers because of their shapes. In general, a dendrimer has a treelike
molecular architecture with an interior core, interior layers (also
called generations), which consist of repeating units regularly attached
to the core, as well as an exterior surface of functional terminal
groups attached to the outermost generation. Generation number (usually
abbreviated to G#) is determined by the number of focal points going
from the core toward the surface.[37] The
molecular weight of dendrimer itself and the number of terminal functional
group can be easily determined by the generation. Dendrimer could
be divided into various classes, such as poly(amidoamine-organosilicon),
poly(propyleneimine), amphiphilic, chiral, multilingual, and micellar
dendrimers, as well as Tecto-dendrimers and Frechet-type dendrimers.[38] One of the commonly used dendrimers as carriers
in medicinal chemistry is poly(amidoamine) (PAMAM) dendrimer. PAMAM
dendrimers are commercially available, usually in a methanol solution,
which can reach generation 10 with molecular weight over 930 000
g/mol.[39]
Dendrimer has been used
in drug delivery as a macromolecular carrier
to target and deliver a wide range of therapeutic agents.[40−48] Compared to the small molecular drug conjugates, which only have
limited number of targeting moiety in each conjugate, dendrimer-based
drug conjugates can largely increase the loading capacity of the targeting
moiety to achieve high concentration at <span class="Disease">tumor sites. Recently, Baker
and his co-workers have used <span class="Chemical">PAMAM dendrimers as the platform for
functionalizing it with different types of biological molecules for
targeted drug delivery and have shown efficient receptor-mediated
cellular internalization in vitro and improved chemotherapeutic activity
in vivo.[49−53] For example, the surface amino groups of a G5 PAMAM dendrimer were
partially acetylated (up to 85%), and the remaining amino groups were
sequentially modified with different functionalities, including folic
acid, paclitaxel, and fluorescein isothiocyanate (FITC).[49]
However, it is impossible to modify the
surface of a G dendrimer with two different
modifiers, e.g., <span class="Disease">tumor-targeting
module and payload module, with an exact ratio, because these modifications
will produce modified G dendrimers with
statistical distributions of the two modifiers. Thus, this type of
traditional approach cannot produce reproducible materials with exact
loading of payloads or <span class="Disease">tumor-targeting molecules, which may seriously
limit its clinical applications.[47]
To solve this highly challenging problem, we came up with an idea
of coupling two fully functionalized half-<span class="Chemical">dendrons, modified with
two different modifiers, through a bifunctional linker (Figure ). A fully functionalized G <span class="Chemical">PAMAM dendrimer can be readily prepared by
modifying all amino terminals of the dendrimer with a cystamine core,
PAMAM(cystamine), using excess modifying reagents, e.g., an activated
ester of a tumor-targeting module or that of a cytotoxic drug module.
Then, the fully functionalized G PAMAM(cystamine)
dendrimer can be reduced to exactly the same half-dendrons, e.g.,
G-HD1-SH and G-HD2-SH. Subsequent stepwise coupling of G1-HD-SH and G2-HD-SH with a bis-maleimide
linker should give a novel asymmetric bow-tie PAMAM dendrimer (ABTD)
(Figure ). A salient
feature of this methodology is that it is applicable to the same-
or different-generation dendrimers; hence, it provides a highly robust
and flexible platform for a variety of applications.
Figure 1
Construction of a well-defined
asymmetric bow-tie dendrimer scaffold
using two differently functionalized half-dendrons.
Construction of a well-defined
asymmetric bow-tie dendrimer scaffold
using two differently functionalized half-<span class="Chemical">dendrons.
While our work on ABTDs was in progress,[54,55] Hartley and co-workers reported a G3 <span class="Chemical">PAMAM dendrimer-based platform
for vaccine delivery.[56] A peptide/protein
array, as well as a mannose array, was fully functionalized on two
different G3 PAMAM(cystamine) dendrimers. Then, these two G3 dendrimers
were reduced to the corresponding half-dendrons to construct G3 heterodendrimers
through orthogonal reattachment using a diblock linker bearing a fluorescein
probe for imaging purpose.[56] Although the
basic idea is similar to ours, their work was limited to the same-generation
(G3) dendrimers and not relevant for the construction of tumor-targeting
drug conjugates, in which only a small number of cytotoxic payloads
need to be introduced as compared to the tumor-targeting modules.
In contrast, our work deals with the construction of ABTDs bearing
two different generations of half-dendrons, which can accommodate
the loadings of functionally different modules. Although a number
of asymmetric bow-tie dendrimers consisting of two dendrimeric wedges
and terminated by two different functionalities, i.e., “Janus
dendrimers”, have been reported[57−59] and some of them have
shown promising results on the drug delivery in vivo,[60,61] our approach is unique and unprecedented.
We describe here
the design, synthesis, and biological evaluations
of a novel ABTD-based tumor-targeting drug conjugate (ABTD-TTC-1, 1), which exhibited extremely high selectivity (>1000 times)
to cancer cells compared to normal cells, and its fluorescent congeners
as mechanistic probes for tumor-targeting and efficient receptor-mediated
endocytosis (RME).
Results and Discussion
Design and Strategy for
Construction of Tumor-Targeted ABTD
Conjugates with Payloads
We have designed a novel <span class="Chemical">ABTD-based
<span class="Disease">tumor-targeted drug delivery system (DDS) platform, consisting of
16 tumor-targeting modules on a G3 half-dendron (G3-HD1) unit and
4 payloads (or probes) with self-immolative linkers on a G1 half-dendron
(G1-HD2) unit, which are connected by a bis-thiosuccinimide linker
unit (Figure ). To
increase the solubility of this macromolecule structure, a PEG short
chain (PEG3) is inserted into each dendrimer arm and the
linker unit. The payload/probe can be a cytotoxic drug, a fluorescent
dye, or a drug with a fluorescent dye.
Figure 2
Schematic representation
of asymmetric bow-tie dendrimer-based
conjugates.
Schematic representation
of asymmetric bow-tie dendrimer-based
conjugates.Building upon the <span class="Disease">ABTD-DDS
platform illustrated in Figure , we further designed three
novel <span class="Disease">tumor-targeting ABTD conjugates 1–3 with a payload or a probe, which are shown in Figure . Biotin (vitamin B7) was selected
as the tumor-targeting molecule since we have been successfully using
this vitamin in our SMDCs,[32,62−65] as well as single-walled carbon nanotube (SWNT)-based drug conjugates.[30,66,67] Also, a highly potent second-generation
taxoid, SB-T-1214 (4), was selected as the cytotoxic
payload. As fluorescent probes, fluorescein and SB-T-1214-fluorescein
(5) were selected. As Figure shows, all three conjugates (ABTD-TTCs)
share the same platform except for the payload or probe component.
ABTD-TTC-1 (1) bears taxoid 4 as the payload
with self-immolative disulfide linker unit, which has been successfully
used in our previous SMDCs and a SWNT-TTDDS.[32,62−66] ABTD-TTC-2 (2) has fluorescein isobutyl ester with
a short tether as a fluorescent probe and ABTD-TTC-3 (3) is a fluorescent probe of 1, wherein fluorescein isobutylester with a short tether is attached to 4. These three
ABTD-TTCs are planned to be constructed from the common key construct
bearing alkyne termini and three azide-bearing components using the
click reaction.
Synthesis of Biotin–ABTD–Alkyne
Construct 6, the
Key Intermediate for ABTD-TTCs
As Figure illustrates our retrosynthetic analysis,
<span class="Chemical">biotin–<span class="Chemical">ABTD–alkyne construct 6 was synthesized
following the strategy outlined in Figure . First, G3-PAMAM(cystamine) dendrimer was
fully functionalized by attaching biotin-NH-PEG3-(CH2)2CO units to give dendrimer 7. In
a similar manner, G1-PAMAM(cystamine) dendrimer was converted to fully
functionalized dendrimer 8, bearing 4-pentynoyl-NH-PEG3-(CH2)2CO units. Then, ω,ω′-bis-[6-maleimidohexanoyl-NH(CH2)2]2-PEG3 (9) was prepared as the linker unit. Next, both dendrimers, 7 and 8, were converted to the corresponding half-dendrons, 10 and 11, bearing sulfhydryl groups, through
reductive cleavage of the cystamine disulfide cores. Then, bismaleimide
linker 9 was reacted with a large excess of G1-half-dendron-SH 11 to afford the corresponding monoaddition product 12. Finally, G1-half-dendron-S-linker-maleimide 12 was reacted with G3-half-dendron-SH 10 to give the
key common intermediate, biotin-ABTD-alkyne intermediate 6. The syntheses of fully functionalized dendrimers 7 and 8 from G3- and G1-PAMAM(cystamine) dendrimers are
shown in Schemes and , respectively.
Figure 4
Retrosynthetic
analysis for the construction of key intermediate 6.
Scheme 1
Synthesis of Fully ω-Biotinylated
G3-PAMAM Dendrimer 7
Reaction
conditions: (i) N,N′-diisopropylcarbodiimide
(DIC)
(2 equiv), N-hydroxysuccinimide (NHS) (3 equiv),
dimethylformamide (DMF) (0.2 M), 45 °C, 18 h; (ii) CH2Cl2, 25 °C, 23 h; (iii) trifluoroacetic acid (TFA)
(1.0 equiv), CH2Cl2, 25 °C, 3 h; (iv) DIC
(2.0 equiv), NHS (3.0 equiv), CH2Cl2, 25 °C,
48 h; (v) MeOH, 25 °C, 24 h.
Scheme 2
Synthesis of Fully
Functionalized ω-Alkynoyl-G1-PAMAM Dendrimer 3
Reaction conditions: (i) 1-(3-Dimethylaminopropyl)-3-ethylcarbodiimide
hydrochloride (EDC·HCl) (1.1 equiv), 4-N,N-dimethylaminopyridine (DMAP) (0.5 equiv), CH2Cl2 (0.02 M), rt, 24 h; (ii) TFA (1.0 equiv), CH2Cl2, rt, 4 h (98%); (iii) NHS (2 equiv), EDC·HCl
(2.0 equiv), CH2Cl2, rt, 17 h (70%); (iv) MeOH,
rt, 48 h.
Retrosynthetic
analysis for the construction of key inte<span class="Chemical">rmediate 6.
Synthesis of Fully ω-Biotinylated
G3-PAMAM Dendrimer 7
Reaction
conditions: (i) N,N′-diisopropylcarbodiimide
(DIC)
(2 equiv), <span class="Chemical">N-hydroxysuccinimide (<span class="Chemical">NHS) (3 equiv),
dimethylformamide (DMF) (0.2 M), 45 °C, 18 h; (ii) CH2Cl2, 25 °C, 23 h; (iii) trifluoroacetic acid (TFA)
(1.0 equiv), CH2Cl2, 25 °C, 3 h; (iv) DIC
(2.0 equiv), NHS (3.0 equiv), CH2Cl2, 25 °C,
48 h; (v) MeOH, 25 °C, 24 h.
Synthesis of Fully
Functionalized ω-Alkynoyl-G1-PAMAM Dendrimer 3
Reaction conditions: (i) 1-(3-Dimethylaminopropyl)-3-ethylcarbodiimide
hydrochloride (<span class="Chemical">EDC·HCl) (1.1 equiv), <span class="Chemical">4-N,N-dimethylaminopyridine (DMAP) (0.5 equiv), CH2Cl2 (0.02 M), rt, 24 h; (ii) TFA (1.0 equiv), CH2Cl2, rt, 4 h (98%); (iii) NHS (2 equiv), EDC·HCl
(2.0 equiv), CH2Cl2, rt, 17 h (70%); (iv) MeOH,
rt, 48 h.
Fully ω-Biotinylated G3-PAMAM Dendrimer 7
<span class="Chemical">G3-PAMAM dendrimer bearing a <span class="Chemical">cystamine core (DNT-296)
was
fully functionalized with tumor-targeting modules (32 in total) through reaction with an N-hydroxysuccinimide
activated ester of the biotin-PEG module, biotin-NH-PEG3-(CH2)2-CO-NHS 14 (Scheme ). Activated ester 14 was readily prepared in good overall yield from biotin and tert-butyl 12-amino-4,7,10-trioxadodecanoate (13) through amide coupling, followed by deprotection of the tert-butyl moiety by TFA,[68] and
subsequent esterification with NHS.
Then, activated <span class="Chemical">ester 14 (2 equiv/amine terminus) was reacted with G3-dendrimer
(DNT-296) in methanol at ambient temperature for 48 h to give fully
ω-biotinylated G3-dendrimer 7. Due to limited solubility,
the purification of the product was performed by dialysis in methanol
using the molecular weight cut-off (MWCO) 6000–8000 tubing
membrane to remove unreacted excess 14. The structure
and purity of isolated G3-dendrimer 7 was confirmed by
matrix assisted laser desorption ionization-time of flight (MALDI-TOF)
using 2,5-dihydroxybenzoic acid (DHB) as matrix (Figure S35). The m/z peak
at 20 748.3 was consistent with the expected fully ω-biotinylated
G3-dendrimer 7. The highest m/z peak 10 378.6 corresponds to the half-dendron of 7, generated by interior disulfide bond cleavage under the
MALDI measurement conditions.
Fully Functionalized G1-PAMAM
Dendrimer with ω-Alkanoyl
Groups 8
In a manner similar to the synthesis
of <span class="Chemical">G3-dendrimer <span class="Chemical">7, 4-pentynoic acid (15)
was coupled with 13, followed by deprotection of tert-butyl group by TFA, and subsequent esterification with
NHS to give activated ester 16 in good overall yield.
Then, G1-PAMAM(cystamine) dendrimer (DNT-294) was reacted with 16 (3 equiv/amino terminus) in methanol at ambient temperature
to give the corresponding fully functionalized G1-dendrimer 8 as the product (Scheme ). The dendrimer 8 was isolated by preparative
high-performance liquid chromatography (HPLC) using a C18 column.
The structure and purity of ω-alkynoyl G1-dendrimer 8 was confirmed by MALDI analysis (Figure S36) and LC-UV-TOF (Figures S37 and S38).
The highest m/z peak 1895.205 corresponds
to the half-dendron of 8 via cleavage of the interior
disulfide linkage, generated under the MALDI-TOF measurement conditions.
The 3788.202 peak corresponds to the molecular ion. The LC-UV/TOF
analysis clearly showed a major peak with retention time at 21.8 min
(UV at 215 nm). The LC-TOF analysis showed 1895.0713 (M + 2H+)2+/2, 1263.7156 (M + 3H+)3+/3,
948.0384 (M + 4H+)4+/4, and 758.6320 (M + 5H+)5+/5 ion peaks (Figure S37), which were deconvoluted to the mass number of the molecular ion,
3788.125 (Figure S38).
Fluorescence-Labeled
Constructs, 19 and 20
In our previous
fluorescence imaging studies, we used <span class="Chemical">fluorescein
methyl <span class="Chemical">ester.[32,62,66] However, under the conditions of the present study, it was found
that fluorescein methyl ester was prone to be converted to the corresponding
nonfluorescent spirolactam.[69−71] Thus, we used more stable fluorescein
isobutylester as the fluorescent label for taxoid 4.
6-Hydroxycarbonylpropylfluorescein isobutyl ester 17 was
prepared using the method previously reported by our laboratory.[64] Amide coupling of 17 with 11-azido-3,6,9-trioxaundecan-1-amine
(18)[72,73] using EDC·HCl gave a click-ready
fluorescent label construct 6 in good yield (Scheme ). Amide coupling-ready
fluorescent label 20 was also prepared from 17 and NHS in excellent yield (Scheme ), which would be used to synthesize fluorescence-labeled
taxoid 5 (Scheme ).
Scheme 3
Preparation of Click-Ready and Coupling-Ready Fluorescent
Constructs
Reaction conditions: (i) 18 (1.0 equiv), DMAP (0.5 equiv), EDC·HCl (1.2 equiv),
DMF/CH2Cl2, rt, 21 h; (ii) NHS (1.5 equiv),
EDC (1.5 equiv), DMAP (0.5 equiv), CH2Cl2, rt,
17 h.
Scheme 4
Synthesis of a Fluorescein-Labeled Taxoid 5
Reaction conditions: (i) TBDMS-Cl
(5 equiv), imidazole (10 equiv), DMF, rt, 4 h; (ii) 20 (2.0 equiv), EDC·HCl (2.0 equiv), DMAP (1.0 equiv), CH2Cl2/DMF, rt, 9 d; (iii) HF/Py, CH3CN/Py,
0 °C to >rt, 24 h.
Preparation of Click-Ready and Coupling-Ready Fluorescent
Constructs
Reaction conditions: (i) 18 (1.0 equiv), DMAP (0.5 equiv), EDC·HCl (1.2 equiv),
DMF/CH2Cl2, rt, 21 h; (ii) NHS (1.5 equiv),
EDC (1.5 equiv), DMAP (0.5 equiv), CH2Cl2, rt,
17 h.
Synthesis of a Fluorescein-Labeled Taxoid 5
Reaction conditions: (i) TBDMS-Cl
(5 equiv), imidazole (10 equiv), DMF, rt, 4 h; (ii) 20 (2.0 equiv), EDC·HCl (2.0 equiv), DMAP (1.0 equiv), CH2Cl2/DMF, rt, 9 d; (iii) HF/Py, CH3CN/Py,
0 °C to >rt, 24 h.
Click-Ready Fluorescence-Labeled
Drug-Linker and Drug-Linker
Conjugates[62]
The C2′-hydroxy
group of <span class="Chemical">taxoid 4 was protected as a <span class="Chemical">tert-butyldimethylsilyl (TBDMS) ether, which was coupled with 20 under EDC·HCl/DMAP conditions, followed by deprotection with
HF/pyridine to afford 7-fluorescein-O-butanoyl-taxoid 5 (Scheme ).
<span class="Chemical">Fluorescein-labeled <span class="Chemical">taxoid 5 was coupled with
2-(triisopropylsiloxycarbonylbutan-2-yl)disulfanylphenylacetic acid
(21)[63−65] under EDC·HCl/DMAP conditions to give the corresponding
taxoid-2′-ester, followed by deprotection with HF/pyridine
and subsequent esterification with NHS/EDC·HCl to afford activated
ester 22 in good overall yield (Scheme ). Then, activated ester 22 was
coupled with 11-azido-3,6,9-trioxaundecan-1-amine (18) to give click-ready azido-SS-linker-taxoid-fluorescein conjugate 24 in 70% yield (Scheme ). The corresponding azido-SS-linker-taxoid conjugate 25 was also synthesized from activated ester 23 in a similar manner from taxoid 4, which was previously
reported by our laboratory (Scheme ).[64] It should be noted
that the disulfide linker unit incorporated into the conjugates 24 and 25 is a second-generation self-immolative
linker developed in our laboratory, which is proven to be stable in
blood plasma, but readily cleaved in cancer cells to release the taxoid
payload.[30,63−65,74]
Scheme 5
Synthesis of Click-Ready Fluorescence-Labeled Drug-Linker (24) and Drug-Linker (25) Components
Reaction conditions and yields:
(i) 21 (1.0 equiv), EDC·HCl (1.5 equiv), DMAP (0.25
equiv), CH2Cl2 (0.01 M), 0 °C to >rt,
16
h, 50–93%; (ii) HF/Py, CH3CN/Py, 0 °C to >rt,
25 h, 71%; (iii) NHS (1.2 equiv), EDC·HCl (1.2 equiv), CH2Cl2 (0.01 M), rt, 17 h, 86%; (iv) 18 (1.5 equiv),
CH2Cl2, rt, 17 h: 24, 70%; 25, 85%.
Synthesis of Click-Ready Fluorescence-Labeled Drug-Linker (24) and Drug-Linker (25) Components
Reaction conditions and yields:
(i) 21 (1.0 equiv), EDC·HCl (1.5 equiv), DMAP (0.25
equiv), CH2Cl2 (0.01 M), 0 °C to >rt,
16
h, 50–93%; (ii) HF/Py, CH3CN/Py, 0 °C to >rt,
25 h, 71%; (iii) NHS (1.2 equiv), EDC·HCl (1.2 equiv), CH2Cl2 (0.01 M), rt, 17 h, 86%; (iv) 18 (1.5 equiv),
CH2Cl2, rt, 17 h: 24, 70%; 25, 85%.
PEGylated Bis(maleimide)
Linker 9
Two
equivalents of 6-maleimidohexanoic acid (26) were coupled
with 4,7,10-trioxa-1,13-tridecanediamine (27) through
activated ester method using isobutyl chloroformate and N-methylmorphorine to give PEGylated bis(maleimide) linker 9 in moderate yield (Scheme ).
Scheme 6
Synthesis of Bis(maleimide)-PEG3-Linker 9
Reaction conditions and yields:
(i) ClCO2Bu-i (1.0 equiv), N-methylmorpholine (NMM) (1.0 equiv), THF, 0 to >25 °C, 3
h;
(ii) 4,7,10-trioxa-1,13-tridecanediamine (27) (0.5 equiv),
THF (0.1 M), 0 to >25 °C, 22 h; 50% for two steps.
Synthesis of Bis(maleimide)-PEG3-Linker 9
Reaction conditions and yields:
(i) <span class="Chemical">ClCO2Bu-i (1.0 equiv), <span class="Chemical">N-methylmorpholine (NMM) (1.0 equiv), THF, 0 to >25 °C, 3
h;
(ii) 4,7,10-trioxa-1,13-tridecanediamine (27) (0.5 equiv),
THF (0.1 M), 0 to >25 °C, 22 h; 50% for two steps.
“Click-Ready” ABTD Construct 6
As illustrated in the retrosynthetic analysis (Figure ), to construct the
“click-ready”
ABTD construct 6 bearing 4 terminal alkyne arms and 16
terminal biotin arms, fully functionalized G3 dendrimer 7 and G1 dendrimer 8 should be cleaved to form the corresponding
G3- and G1-half dendrons, 10 and 11, respectively,
which should be cross-coupled through bis(maleimido) linker 9 stepwise. We selected G1 dendrimer 8 over G3
dendrimer 7 to first couple with 9 because
(i) 8 has better solubility than 7, hence
purification with preparative HPLC is feasible and (ii) difference
in molecular weight is very large in the transformation of G1-half
dendron-linker 12 to ABTD construct 6 through
cross-coupling of 12 with 10, due to the
dominant contribution of 10 to the molecule size so that
dialysis can be used as the most relevant purification method after
the cross-coupling (Scheme ).
Scheme 7
Click-Ready ABTD Construct 6
Reaction conditions and yields:
(i) 9 (10 equiv), tris(2-carboxyethyl)phosphine (TCEP)
(3 equiv), MeOH, 25 °C, 24 h, 49%; (ii) 10 (0.5
equiv), TCEP (3 equiv), MeOH/H2O (1/1), 25 °C, 24
h, 49%.
Click-Ready ABTD Construct 6
Reaction conditions and yields:
(i) 9 (10 equiv), <span class="Chemical">tris(2-carboxyethyl)phosphine (<span class="Chemical">TCEP)
(3 equiv), MeOH, 25 °C, 24 h, 49%; (ii) 10 (0.5
equiv), TCEP (3 equiv), MeOH/H2O (1/1), 25 °C, 24
h, 49%.
Purification of 12 was
performed by preparative HPLC
using Jupiter C18 column. Solvents were <span class="Chemical">water with 0.1% TFA (solvent
A) and acetonitrile with 0.1% TFA (solvent B). We were able to remove
the impurities and defect derivatives to solely isolate the desired
product as single component. The purity of 12 was confirmed
by MALDI-TOF (Figure S39) and LC-UV-TOF
(Figures S40 and S41). The formation of 6 (Mw: 12 876) was confirmed
by the LC-UV-TOF analysis. Construct 6 was purified by
dialysis with MWCO 6000–8000 dialysis tubing membrane, and
the structure of 6 was confirmed by MALDI-TOF (Figure S8) and LC-UV-TOF analyses (Figures S9–S11).
Click-ready <span class="Chemical">ABTD 6 was
reacted with <span class="Chemical">azido-SS-linker-taxoid conjugate 25(75) in the presence of a catalytic amount of copper
sulfate pentahydrate and sodium ascorbate in methanol/distilled water
(4/1) at room temperature for 24 h. The progress of the reaction was
monitored by FIA-MS. Upon completion, the reaction mixture was subjected
to dialysis using an MWCO 6000–8000 tubing membrane for 48
h against methanol. The resulting solution was filtrated to remove
insoluble solid and then lyophilized to give tumor-targeting asymmetric
bow-tie dendrimer 1 (ABTD-TTC-1) in 72% yield (Scheme ). In the same manner,
click-ready ABTD 6 was reacted with azido-SS-linker-taxoid(fluorescein)
probe 24 and azido-linker-fluorescein 19 to give 2 (ABTD-TTC-2) and 3 (ABTD-TTC-3)
in 69 and 76% yields, respectively (Scheme ). Attempted characterizations of these ABTD-TTCs 1–3 by MALDI-TOF or LC-TOF were unfortunately
not successful, due to the lack of sufficient ionization. Thus, these
compounds were characterized by 1H and 13C NMR,
as well as gel permeation chromatography (GPC) analyses, which showed
>98% purity for all compounds (Figures S1–S6, S45–S47 in the Supporting Information).
Scheme 8
Synthesis
of ABTD-TTCs 1–3
See Figure for
the structures of payload or probe moieties.
Synthesis
of ABTD-TTCs 1–3
See Figure for
the structures of payload or probe moieties.
Biological Evaluations
Internalization of ABTD-TTC-2 (2) and ABTD-TTC-3
(3) Fluorescent Probes into Cancer Cells via Receptor-Mediated
Endocytosis (RME)
Internalization of 2 via <span class="Chemical">RME
was confi<span class="Chemical">rmed by confocal fluorescence microscopy (CFM) and flow cytometry
(see Figure S48 in the Supporting Information).
Internalization of 3 was monitored by CFM, and the fluorescence
was quantified by flow cytometry, using ca. 10 000 treated
live ID-8 (ovarian) and MX-1 (breast) cancer cells, overexpressing
biotin receptors (Figure ). CFM images in Figure clearly show the efficient internalization of 3 via RME. Flow cytometry also indicates that at 20 μM
concentration, the fluorescein signal intensity increases from 1 to
3 h incubation period.
Figure 5
CFM images and flow cytometry analysis of different types
of cells
after incubation with 3 at the final concentration of
20 μM at 37 °C for different periods. (A) Flow cytometry
analysis of 3 in ID-8 at 0 h (red, control), 1 h (green),
and 3 h (blue). (B, C) CFM images and flow cytometry analysis in ID8
at 1 and 3 h, respectively. (D) Flow cytometry analysis of 3 in MX-1 at 0 h (red, control), 1 h (green), and 3 h (blue). (E,
F) CFM images and flow cytometry analysis in MX-1 at 1 and 3 h, respectively.
(For full internalization study results using ID8 and MX-1 cancer
cells, as well as WI-38 normal human lung fibroblast cells, see Figure S49 in the Supporting Information.).
CFM images and flow cytometry analysis of different types
of cells
after incubation with 3 at the final concentration of
20 μM at 37 °C for different periods. (A) Flow cytometry
analysis of 3 in ID-8 at 0 h (red, control), 1 h (green),
and 3 h (blue). (B, C) CFM images and flow cytometry analysis in ID8
at 1 and 3 h, respectively. (D) Flow cytometry analysis of 3 in <span class="Gene">MX-1 at 0 h (red, control), 1 h (green), and 3 h (blue). (E,
F) CFM images and flow cytometry analysis in <span class="Gene">MX-1 at 1 and 3 h, respectively.
(For full internalization study results using ID8 and MX-1cancer
cells, as well as WI-38 normal human lung fibroblast cells, see Figure S49 in the Supporting Information.).
Multibinding Effect
The multiple <span class="Gene">TTMs (16 <span class="Chemical">biotin modules
in this case) on a single ABTD macromolecular drug conjugate are expected
to increase its targeting efficiency to get internalized through RME
via multibinding to the biotin receptors that are overexpressed on
the surface of cancer cells. To prove this hypothesis, we synthesized
a small-molecule fluorescent probe 28 (Figure ), which only has one biotin
moiety for one fluorescent moiety (TTM/fluorescein ratio = 1), for
comparison with 3, bearing 16 biotin moieties for 4 fluorescent
moiety (TTM/fluorescein ratio = 4).
Figure 6
Biotin-PEG3-fluorescein probe 28 and biotin-SS-linker-taxoid
conjugate 29 (BLT-S).
Biotin-PEG3-fluorescein probe 28 and biotin-SS-linker-taxoid
conjugate 29 (BLT-S).Figure clearly
shows that at 10 μM concentration of fluorescent probes, with
increasing incubation time, both 28 and 3 were internalized via <span class="Chemical">RME into two different <span class="Disease">cancer cell lines (ID-8
and MX-1) overexpressing biotin receptors. It also reveals that at
the initial stage the RME rate of 3 is higher than 28, which indicates that the multiple TTM can increase the
targeting efficiency to the cancer cells. In addition, for each time
point, in terms of the different number of inherited fluorescein of 3 and 28, the fluorescein signal intensity of 3 is more than 4 times of 28 against ID-8 cell
line, suggesting that the multiple TTM can enhance the cellular uptake
of the whole conjugate via RME, resulting in more probes internalized
at the end.
Figure 7
Time-dependent RME rate comparison between a small fluorescent
probe 28 (blue) and 3 (red). Curves were
drawn based on the flow cytometry results by incubation of 3 at the concentration of 10 μM at 37 °C for 0 h (control),
1, 2, 4, 6, and 16 h, in ID-8 (left) and MX-1 (right) (for the flow
cytometry data to produce these graphs, see Figures S50 and S51).
Time-dependent <span class="Chemical">RME rate comparison between a small fluorescent
probe 28 (blue) and 3 (red). Curves were
drawn based on the flow cytometry results by incubation of 3 at the concentration of 10 μM at 37 °C for 0 h (control),
1, 2, 4, 6, and 16 h, in ID-8 (left) and <span class="Gene">MX-1 (right) (for the flow
cytometry data to produce these graphs, see Figures S50 and S51).
Cancer Cell Selective Cytotoxicity
<span class="Disease">Cytotoxicity assessment
of <span class="Chemical">ABTD conjugates 1–3 using MTT
assay against ID-8 (murineovarian cancer), MX-1 (humanbreast cancer),
and WI-38 (human lung fibroblast, normal) cell lines were performed.
For comparison purpose, paclitaxel, free taxoid 4, its
fluorescein conjugate 5, and small-molecule biotin-SS-linker-taxoid
conjugate, BLT-S 29 (Figure ),[64] were assayed
as well. The results are summarized in Table .
Table 1
Cytotoxicity (IC50, nM)
of Paclitaxel, 4, 5, 29, and
ABTD-TTCs 1–3b
entry
compound
ID-8
MX-1
WI-38
1
paclitaxel
21.2 ± 4.30
3.83 ± 0.59
175 ± 51.6
2
4 (SB-T-1214)
1.89 ± 0.30
2.90 ± 0.47
4.14 ± 0.82
3
5 (SB-T-1214-fluorescein)
846 ± 112
N.D.
519 ± 156
4
29 (BLT-S)
7.84 ± 1.85
26.7 ± 3.44
519 ± 90.3
5
29 + GSH-OEta
5.91 ± 0.32
1.52 ± 0.34
N.D.
6
1 (ABTD-TTC-1)
0.66 ± 0.03
2.05 ± 0.91
582 ± 48.8
7
1 + GSH-OEta
0.62 ± 0.07
0.12 ± 0.05
903 ± 15
8
2 (ABTD-TTC-2)
>5000
3630 ± 1300
N.D.
9
2 + GSH-OEta
651 ± 169
411 ± 80.9
N.D.
10
3 (ABTD-TTC-3)
>5000
>5000
N.D.
The concentration
of cytotoxic compounds
that inhibit 50% (IC50, nM) of different types of cells
after 72 h of drug exposure at 37 °C under 5% CO2 was
examined. N.D. = not determined.
24 h of drug exposure, followed
by thorough washing with Dulbecco's phosphate-buffered saline
(DPBS),
and 48 h incubation with 6 equiv of glutathione ethyl ester (GSH-OEt)
at 37 °C under 5% CO2.
The concentration
of <span class="Disease">cytotoxic compounds
that inhibit 50% (IC50, nM) of different types of cells
after 72 h of drug exposure at 37 °C under 5% CO2 was
examined. N.D. = not determined.
24 h of drug exposure, followed
by thorough washing with <span class="Chemical">Dulbecco's phosphate-buffered saline
(<span class="Chemical">DPBS),
and 48 h incubation with 6 equiv of glutathione ethyl ester (GSH-OEt)
at 37 °C under 5% CO2.
As Table shows,
the IC50 values of second-generation taxoid 4 against ID-8, MX-1, and WI-38 are 1.89, 2.90, and 4.14 nM, respectively,
which are 1–2 orders of magnitude more potent than those of
paclitaxel (entries 1 and 2). Taxoid 4 exhibited similar
IC50 values against these three cell lines (entry 2). Paclitaxel
showed small differences in their IC50 values against the
three cell lines (entry 1). Thus, paclitaxel possesses appreciable
selectivity difference between cancer cells and normal cells, but 4 shows little selectivity between cancer cells and normal
cells, as anticipated. Table also shows that fluorescence-labeled taxoid 5 with modification at the C7 position exhibits 2–3 orders
of magnitude less potency than 4 (entry 3). Similar results
were also observed from the comparison between 1 and 2 against ID-8 (IC50 0.66 nM, >5000 nM, respectively)
and MX-1 (IC50 2.05, 3630 nM, respectively) (entries 6–10).ABTD-TTC-1 1 exhibits remarkable potency (IC50 0.66 nM), 3 times higher than that of free taxoid 4 (IC50 1.89 nM) against ID-8 (entry 6), indicating excellent
efficiency of this ABTD-based drug-delivery system, which is consistent
with the CFM and flow cytometry results shown in Figures and 6. When 6 equiv of glutathione ethyl ester were added to fully cleave
the linker and release all taxoid payload, the IC50 did
not change much (IC50 0.62 nM), indicating that endogenous
glutathione level (and possibly some other thiols) was sufficient
to cleave all of the internalized ABTD conjugate in 72 h incubation
(entry 7). Drug conjugate 1 shows similar potency to
free taxoid 4 against MX-1 (IC50 2.05, 2.90
nM, respectively) without addition of glutathione ethyl ester (entry
6). However, as anticipated, 1 exhibited 1 order of magnitude
higher potency (IC50 0.12 nM) once 6 equiv of glutathione
ethyl ester were added after 24 h and incubated for another 48 h,
indicating the insufficient endogenous glutathione level in MX-1 cells
(entry 7).ABTD-TTC-1 (1) and small-molecule conjugate 29 exhibit a weak cytotoxicity (IC50 519 and 582
nM, respectively) against WI-38 normal cell line, while showing strong
cytotoxicity against cancer cells (entries 4 and 6). The results indicate
that excellent tumor-targeting efficiency was achieved by these conjugates
using the biotin receptor as tumor-specific target. Moreover, 1 exhibited 2–3 orders of magnitude higher cytotoxicity
against ID8 and MX-1 compared to normal cell line WI-38 (entry 6),
which indicates that 1 has more than 10 times higher
cancer cell selectivity than 29. Same trend was also
observed on addition of 6 equiv of glutathione ethyl ester (entry
7). Therefore, the results indicate that a higher intracellular concentration
of taxoid payload was achieved by utilizing the ABTD-DDS with multiple
payloads, resulting in a higher potency of drug conjugate than the
small-molecule conjugate 29 bearing single TTM with single
taxoid payload. To assess a possible cytotoxicity of dendrimer itself
as macromolecular vehicle, cytotoxicity of 3, which only
has TTM and fluorescein moiety without cytotoxic payload, was examined.
As Table clearly
shows, the IC50 values of 3 against ID-8 and
MX-1 are both over 5000 nM, indicating that the cytotoxicity of ABTD
scaffold itself is nontoxic and biocompatible (entry 10).
Conclusions
Asymmetric bow-tie <span class="Chemical">PAMAM dendrimer-based (<span class="Chemical">ABTD)
platform as a novel
tumor-targeting drug-delivery system has been designed, and three
tumor-targeting ABTD conjugates (ABTD-TTC-1∼3) have been successfully
synthesized. The ABTD-TTCs were obtained in good purity and biological
evaluations were reproducible. By means of flow cytometry and confocal
fluorescence microscopy (CFM), we confirmed that these macromolecular
drug conjugates were internalized into cancer cells through RME. We
also disclosed the multibinding effect of the tumor-targeting moieties
by comparing the RME internalization rate and the fluorescein signal
intensity between 3 and a small fluorescent probe 28. The IC50 values observed for 1 clearly showed the high efficiency of this dendrimer-based drug-delivery
system compared to the drug itself and a small drug conjugate 29. The IC50 values observed for 3 indicated the nontoxicity and biocompatibility of this drug-delivery
vehicle. Further studies along this line include different tumor-targeting
molecules as well as newer-generation taxoids with higher potency.
The present synthetic protocol is applicable to higher-generation
dendrimers. In place of attaching a fluorescent probe at the C7 position
of a taxoid, which resulted in significant loss of cytotoxicity, a
branched bis(maleimide) linker that can accommodate a fluorescent
probe or a radionuclide for PET, SPECT, or magnetic resonance imaging
for diagnostic and theranostic studies has been designed. Further
studies along this line are actively underway in our laboratory.
Experimental
Section
General Methods
1H NMR and <span class="Chemical">13C NMR spectra were measured on a Varian 300 or 500 MHz spectrometer
or a Bruker 400, 500, or 700 MHz NMR spectrometer. Melting points
were measured on a Thomas–Hoover capillary melting point apparatus
and are uncorrected. Thin-layer chromatography (TLC) analyses were
performed on aluminum-backed Silica G TLC plates (Sorbent Technologies,
200 μm, 20 cm × 20 cm), visualized with UV light, and stained
with sulfuric acid-EtOH, 10% PMA-EtOH, 10% vanillin-EtOH with 1% sulfuric
acid or ninhydrin-butanol with 10% AcOH. Column chromatography was
carried out on silica gel 60 (Merck, 230–400 mesh ASTM). Purification
were also performed using Jupiter C18 semipreparative HPLC column
(5 μm, 300 Å, 250 mm × 10 mm) or Jupiter C18 preparative
HPLC column (5 μm, 300 Å, 250 mm × 21.2 mm) on a Shimadzu
CBM-10AW VP communications bus module, Shimadzu SPD-10A VP UV–vis
detector, and Shimadzu LC-6AD liquid chromatography assembly. Chemical
purity was determined with a Shimadzu L-2010A HPLC HT series HPLC
assembly, using a Kinetex PFP column (4.6 mm × 100 mm, 2.6 μm),
employing CH3CN/water as the solvent system with a flow
rate of 1 mL/min. GPC analysis was performed with a Shimadzu L-2010A
HPLC HT series HPLC assembly, using a Ultrahydrogel 500 column (7.8
mm × 300 mm, 10 μm), employing CH3CN/water (20/80)
with a flow rate of 0.5 mL/min. High-resolution mass spectrometry
analysis was carried out on an Agilent LC-UV-TOF mass spectrometer
using Jupiter C18 analytical column with 2.6 μm, 100 Å,
2.1 mm × 100 mm, with 0.1% TFA in water (optima grade) as solvent
A and 0.1% TFA in CH3CN (optima grade) as solvent B, running
temperature 25 °C, and flow rate 0.5 mL/min, at the Institute
of Chemical Biology and Drug Discovery (ICB&DD), Stony Brook,
NY, or at the Mass Spectrometry Laboratory, University of Illinois
at Urbana–Champaign, Urbana, IL. MALDI-TOF analysis for the
determination of molecular weight was carried out using DHB as matrix
at the Institute of Chemical Biology and Drug Discovery (ICB&DD),
Stony Brook, NY.
Materials
The chemicals were purchased
from Sigma-Aldrich,
Fisher Scientific, or VWR International and used as received or purified
before use by standard methods. <span class="Chemical">Dichloromethane and <span class="Chemical">methanol were
dried before use by distillation over calcium hydride under nitrogen.
Ether and tetrahydrofuran were dried before use by distillation over
sodium benzophenone kept under nitrogen. 10-Deacetylbaccatin III was
obtained from Indena SpA, Italy. Reaction flasks were dried in a 100
°C oven and allowed to cool to room temperature in a desiccator
over calcium sulfate and assembled under an inert nitrogen gas atmosphere.
4-(9-(2-Isobutoxycarbonylphenyl)-3-oxo-3H-xanthen-6-yloxy)butanoic
acid (17),[64] 11-azido-3,6,9-trioxaundecan-1-amine
(18),[72] SB-T-1214 (4),[28] 2′-TBDMS-4,[62] 2-(5-oxo-triisopropylsiloxypent-2-yl)disulfanylphenylacetic
acid (21),[75] SB-T-1214-linker-SS-PEG3-azide 25,[75]tert-butyl 12-amino-4,7,10-trioxadodecanoate (13),[76,77] 12-biotynylamino-4,7,10-trioxadodecanoic
acid,[55,78] and biotin-PEG3-tether-SS-linker-SB-T-1214
(BLT-S) 29(64) were prepared
by the literature methods. The PAMAM dendrimers bearing cystamine
cores were purchased as methanol solution from Dendritic Nanotechnologies,
Inc. G3 PAPAM(cystamine) dendrimer, DNT-296 (MW: 7001 g/mol) (10.1% in methanol) and G1 PAPAM(cystamine)
dendrimer, and DNT-294 (MW: 1522 g/mol)
(20.1% in methanol) were used as received. Biochemical materials,
including RPMI-1640 cell culture media, DPBS buffer, fetal bovine
serum (FBS), PenStrep, and TrypLE, were obtained from Gibco and VWR
International and used as received for cell-based assays.
Synthetic
Procedures
7-FOB-Taxoid 5
A solution
of 2′-TBDMS-4[62] (184
mg, 0.19 mmol), DMAP (23
mg, 0.19 mmol), 20 (180 mg, 0.38 mmol), and EDC·HCl
(72.8 mg, 0.38 mmol) in DCM (10 mL) and DMF (1 mL) was stirred for
9 days at rt. The progress of the reaction was monitored by TLC. Upon
completion, the reaction mixture was concentrated in vacuo and the
crude product was purified on a silica gel column (DCM/MeOH with 99/1
to 9/1 gradient) to give 7-FOB-2′-TBDMS-taxoid (260 mg, 95%
yield) as an orange solid: 1H NMR (500 MHz, CDCl3) δ 0.07 (s, 3H), 0.10 (s, 3H), 0.67 (d, J = 6.7 Hz, 3H), 0.69 (d, J = 6.7 Hz, 3H), 0.87 (m,
2H), 0.93 (s, 9H), 0.98 (m, 2H), 1.05 (m, 1H), 1.16 (s, 3H), 1.23
(s, 3H), 1.34 (s, 9H), 1.59 (m, 1H), 1.67 (m, 1H), 1.74 (s, 3H), 1.77
(s, 3H), 1.79 (s, 3H), 1.93 (s, 3H), 2.03 (d, J =
4.8 Hz, 2H), 2.09 (m, 2H), 2.29 (m, 1H), 2.40 (s, 3H), 2.46 (m, 2H),
3.71 (m, 2H), 3.96 (d, J = 6.8 Hz, 1H), 4.05 (m,
2H), 4.17 (d, J = 8.4 Hz, 1H), 4.23 (d, J = 3.2 Hz, 1H), 4.30 (d, J = 8.4 Hz, 1H), 4.75 (m,
1H), 4.82 (d, J = 9.7 Hz, 1H), 4.94 (d, J = 8.8 Hz, 1H), 5.22 (d, J = 7.4 Hz, 1H), 5.59 (m,
1H), 5.66 (d, J = 7.4 Hz, 1H), 6.12 (t, J = 8.8 Hz, 1H), 6.29 (s, 1H), 6.46 (s, 1H), 6.54 (m, 1H), 6.71 (m,
1H), 6.86 (t, J = 9.7 Hz, 2H), 6.95 (m, 1H), 7.27
(d, J = 7.6 Hz, 1H), 7.45 (t, J =
7.4 Hz, 2H), 7.59 (t, J = 7.4 Hz, 1H), 7.64 (t, J = 7.6 Hz, 1H), 7.70 (t, J = 7.6 Hz, 1H),
8.00 (s, 1H), 8.08 (d, J = 7.4 Hz, 2H), 8.24 (d, J = 7.6 Hz, 1H). 13C NMR (125 MHz, CDCl3) δ −5.2, −4.8, 8.8, 10.9, 12.8, 13.7, 14.6,
18.4, 18.6, 18.8, 18.9, 19.1, 20.7, 21.4, 22.6, 23.7, 25.7, 25.8,
26.1, 27.5, 28.3, 30.4, 30.6, 31.4, 33.3, 35.4, 36.5, 43.2, 46.8,
56.0, 60.4, 64.4, 67.9, 71.2, 71.4, 71.9, 74.6, 74.9, 75.0, 78.8,
80.7, 84.0, 100.7, 105.7, 113.9, 114.7, 117.5, 121.8, 128.6, 128.9,
129.2, 129.6, 129.8, 130.1, 130.3, 130.5, 130.8, 131.3, 132.3, 132.5,
133.7, 134.2, 136.5, 141.6, 150.6, 154.4, 155.1, 159.0, 162.6, 163.6,
165.5, 166.8, 169.9, 171.7, 171.9, 172.9, 173.8, 185.7, 202.4. HRMS
(TOF) [M + H]+ calcd for C79H98NO21Si+ 1424.6395, found 1424.6398 (Δ = 0.2
ppm).To an ice-cooled solution of 7-FOB-2′-TBDMS-taxoid
(243 mg, 0.17 mmol) in a 1:1 mixture of acetonitrile/pyridine (9 mL,
0.02 M) was slowly added 2.4 mL of HF/pyridine under nitrogen. The
reaction mixture was stirred at rt for 24 h and the progress was monitored
by TLC. Upon completion, the reaction was quenched with 0.2 M aqueous
citric acid (5 mL) and extracted with EtOAc (30 mL × 3). The
organic layer was collected and washed with CuSO4 and distilled
water until the organic layer became clear. Then, the organic layer
was washed with brine (30 mL × 3), dried over anhydrous MgSO4, and concentrated in vacuo. The resulting crud product was
purified by column chromatography on silica gel (DCM/MeOH with 99/1
to 9/1 gradient) to give 5 (115 mg, 53% yield) as an
orange solid: mp 170–172 °C; 1H NMR (500 MHz,
CDCl3) δ 0.67 (d, J = 6.7 Hz, 3H),
0.69 (d, J = 6.7 Hz, 3H), 0.85 (m, 2H), 0.95 (m,
2H), 1.00 (m, 1H), 1.07 (m, 1H), 1.18 (s, 3H), 1.23 (s, 3H), 1.35
(s, 9H), 1.59 (m, 1H), 1.68 (m, 1H), 1.76 (s, 3H), 1.79 (s, 3H), 1.80
(s, 3H), 2.01 (s, 3H), 2.08 (m, 2H), 2.37 (s, 3H), 2.41 (m, 2H), 2.56
(m, 1H), 3.73 (m, 2H), 3.96 (d, J = 6.8 Hz, 1H),
4.13 (m, 2H), 4.17 (d, J = 8.5 Hz, 1H), 4.26 (s,
1H), 4.31 (d, J = 8.5 Hz, 1H), 4.75 (t, J = 8.5 Hz, 1H), 4.95 (m, 2H), 5.35 (d, J = 7.4 Hz,
1H), 5.65 (t, J = 7.4 Hz, 2H), 6.11 (t, J = 8.5 Hz, 1H), 6.31 (s, 1H), 6.43 (s, 1H), 6.53 (m, 1H), 6.72 (d, J = 8.5 Hz, 1H), 6.86 (t, J = 9.2 Hz, 2H),
7.11 (s, 1H), 7.28 (d, J = 7.6 Hz, 1H), 7.46 (t, J = 7.4 Hz, 2H), 7.59 (t, J = 7.4 Hz, 1H),
7.65 (t, J = 7.6 Hz, 1H), 7.70 (t, J = 7.6 Hz, 1H), 8.09 (d, J = 7.4 Hz, 2H), 8.24 (d, J = 7.6 Hz, 1H). 13C NMR (125 MHz, CDCl3) δ 8.8, 8.9, 10.8, 12.9, 14.1, 14.7, 18.6, 18.8, 18.9, 21.1,
22.3, 22.6, 23.4, 25.7, 26.2, 27.5, 28.2, 29.9, 30.0, 31.6, 33.4,
35.5, 43.2, 47.0, 51.5, 56.1, 67.9, 71.5, 71.9, 72.0, 73.6, 74.4,
75.1, 78.6, 79.7, 80.8, 83.9, 100.8, 105.7, 114.3, 114.7, 117.4, 120.9,
128.7, 128.8, 128.9, 129.1, 129.6, 129.8, 130.1, 130.3, 130.5, 130.8,
130.9, 131.3, 132.5, 133.8, 134.2, 137.5, 141.2, 150.8, 154.5, 155.4,
159.1, 163.7, 165.5, 166.8, 170.1, 172.1, 172.1, 172.9, 173.1, 185.7,
202.4. HRMS (TOF) [M + H]+ calcd for C73H84NO21+ 1310.5530, found 1310.5540 (Δ
= 0.7 ppm).
To an ice-cooled solution of <span class="Chemical">N-maleimidohexanoic acid (26) (1.1 g, 5.4 mmol) and <span class="Chemical">NMM
(546 mg, 5.4 mmol, 0.6 mL) in THF (20 mL, 0.25 M) was slowly added
isobutyl chloroformate (738 mg, 5.4 mmol) with stirring. A white precipitate
was immediately generated. The reaction mixture was allowed to warm
to rt for 3 h, and the progress of the reaction was monitored by TLC.
Upon completion, the precipitate was filtered out and the resulting
THF solution was immediately taken to the next step. A small portion
of the solution was concentrated in vacuo for 1H NMR analysis,
which confirmed the formation of the corresponding activate ester,
6-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)hexanoic
propionic anhydride: 1H NMR (500 MHz, CDCl3)
δ 0.96 (d, J = 6.7 Hz, 6H), 1.35 (m, 2H), 1.58
(m, 2H), 1.66 (m, 2H), 1.99 (m, 1H), 2.44 (t, J =
7.4 Hz, 2H), 3.50 (t, J = 7.4 Hz, 2H), 4.03 (d, J = 6.7 Hz, 2H), 6.68 (s, 2H). 13C NMR (125 MHz,
CDCl3) δ 18.8, 23.6, 25.9, 27.6, 28.1, 33.9, 37.5,
75.5, 134.0, 149.2, 167.7, 170.8.
To an ice-cooled solution
of the activated ester of 26 in THF was slowly added
4,7,10-trioxa-1,13-tridecanediamine (27) (595 mg, 2.7
mmol) with stirring. The resulting solution was allowed to warm to
rt for 22 h, and the progress of the reaction was monitored by TLC.
Upon completion, the solvent was removed in vacuo and the crude product
was purified by column chromatography on silica gel (DCM/MeOH with
gradient 99/1 to 9/1) to give 9 (814 mg, 50% yield for
two steps) as a white foam-like solid: 1H NMR (500 MHz,
CDCl3) δ 1.30 (quint, J = 7.4 Hz,
4H), 1.61 (quint, J = 7.4 Hz, 4H), 1.65 (quint, J = 7.4 Hz, 4H), 1.77 (quint, J = 7.4 Hz,
4H), 2.15 (t, J = 7.4 Hz, 4H), 3.35 (q, J = 6.4 Hz, 4H), 3.52 (t, J = 7.4 Hz, 4H), 3.57 (t, J = 6.4 Hz, 4H), 3.61 (m, 4H), 3.66 (m, 4H), 6.22 (m, 2H),
6.71 (s, 4H). 13C NMR (125 MHz, CDCl3) δ
25.2, 26.4, 28.3, 29.1, 36.5, 37.7, 37.8, 70.0, 70.1, 70.4, 134.0,
170.8, 172.7. HRMS (TOF) [M + H]+ calcd for C30H47N4O9+ 607.3338, found
607.3334 (Δ = −0.6 ppm).
To a solution of 8 (66 mg, 17 μmol) and bis(maleimido) linker 9 (103
mg, 170 μmol) in MeOH (3.0 mL) was slowly added 0.5 M TCEP (51
μmol, 0.1 mL) at rt with stirring. Upon addition, the solution
turned cloudy. The resulting mixture was stirred at rt for 6 h, and
the progress of the reaction was monitored by FIA-MS. Upon completion,
the reaction mixture was diluted with distilled water (1.0 mL) and
purified by semipreparative chromatography using a Jupiter C18 column
(5 μm, 300 Å, 250 mm × 21.2 mm; acidic solvent A:
H2O + 0.1% TFA; solvent B: acetonitrile + 0.1% TFA; gradient:
45% B to 55% B; flow rate: 10 mL/min). The fractions containing product
were combined and concentrated by lyophilization to give 12 (21 mg, 25% yield) as a yellow oil: 1H NMR (700 MHz,
CD3OD) δ 1.28 (m, 4H), 1.58 (m, 4H), 1.62 (m, 4H),
1.74 (m, 4H), 2.17 (q, J = 7.2 Hz, 4H), 2.31 (t, J = 2.6 Hz, 4H), 2.41 (t, J = 7.2 Hz, 8H),
2.47 (t, J = 5.7 Hz, 16H), 2.76 (t, J = 6.2 Hz, 8H), 2.94 (m, 4H), 3.26 (m, 6H), 3.30 (m, 8H), 3.35 (m,
8H), 3.37 (t, J = 5.7 Hz, 8H), 3.43 (t, J = 5.7 Hz, 4H), 3.50 (m, 8H), 3.54 (m, 16H), 3.60 (m, 6H), 3.65 (m,
36H), 3.73 (m, 4H), 3.75 (t, J = 6.2 Hz, 8H), 5.50
(s, 4H), 6.82 (s, 2H). 13C NMR (176 MHz, CD3OD) δ 14.3, 25.1, 25.9, 26.0, 26.9, 27.8, 28.0, 28.3, 29.0,
29.1, 34.2, 34.5, 35.2, 35.4, 35.5, 36.2, 36.4, 36.5, 37.0, 38.0,
38.4, 38.9, 39.0, 39.5, 50.0, 50.1, 52.3, 52.6, 53.4, 66.8, 68.5,
68.6, 69.0, 69.1, 69.8, 69.9, 70.0, 70.1, 82.2, 133.9, 171.1, 171.3,
172.5, 172.6, 172.9, 174.3, 174.4, 175.0, 177.6. MS (LC-TOF) [M +
H]+ calcd for C118H198N21O35S+ 2503.0535, found 657.3406, 835.4750, 1252.7086; deconvolution
value: 2503.40.
1-Biotynyl-1-amino-3,6,9-trioxadodecanoyl-OSu 14
Activated ester 14 was prepared by
the condensation
of NHS with 12-biotynylamino-4,7,10-trioxadodecanoic acid,[55,78] which was readily prepared from biotin through coupling of biotin
with tert-butyl 12-amino-4,7,10-trioxadodecanoate
(13),[76,77] followed by deprotection of the tert-butyl ester moiety with HF/pyridine in accordance with
the patent literature method.[55,78] A solution of 12-biotynylamino-4,7,10-trioxadodecanoic
acid (950 mg, 2.12 mmol), EDC·HCl (487 mg, 2.54 mmol), and NHS
(293 mg, 2.54 mmol) in DCM (21 mL, 0.1 M) was stirred at rt for 48
h. The progress of the reaction was monitored by TLC. Upon completion,
the reaction mixture was concentrated in vacuo and the resulting crude
product was purified by column chromatography on silica gel (DCM/MeOH
with 99/1 to 9/1 gradient) to give 14 (977 mg, 86% yield)
as a white sticky solid: 1H NMR (500 MHz, CDCl3) δ 1.24 (m, 2H), 1.40 (m, 2H), 1.64 (m, 2H), 1.71 (m, 2H),
2.19 (t, J = 7.4 Hz, 2H), 2.56 (t, J = 6.4 Hz, 1H), 2.68 (s, 1H), 2.71 (d, J = 12.8
Hz, 1H), 2.82 (s, 4H), 2.86 (t, J = 6.3 Hz, 2H),
3.10 (m, 1H), 3.40 (m, 2H), 3.53 (t, J = 5.0 Hz,
2H), 3.60 (m, 8H), 3.66 (s, 1H), 3.81 (t, J = 6.3
Hz, 2H), 4.28 (m, 1H), 4.47 (m, 1H). 13C NMR (125 MHz,
CDCl3) δ 14.1, 22.6, 25.6, 28.2, 31.6, 32.2, 36.0,
39.1, 40.5, 55.6, 60.3, 61.8, 65.7, 66.6, 69.9, 70.1, 70.3, 70.4,
70.7, 164.3, 166.8, 169.1, 173.4. HRMS (TOF) [M + H]+ calcd
for C23H37N4O9S+ 545.2276, found 545.2276 (Δ = 0 ppm).
ω-Alkynyl acid-activated <span class="Chemical">ester 16 was
prepared from <span class="Chemical">4-pentynoic acid (15) and 13, followed by deprotection of the tert-butyl moiety
and esterification with NHS, as described below.
To a solution
of <span class="Chemical">4-pentynoic acid (15) (1.04 g, 5.35 mmol) in <span class="Chemical">DCM (27
mL, 0.2 M) was added tert-butyl 12-amino-4,7,10-trioxadodecanoate
(13)[76,77] (1.63 g, 5.88 mmol) and EDC·HCl
salt (1.12 g, 5.88 mmol), and the solution was stirred for 35 h under
rt. The progress of the reaction was monitored by TLC. Upon completion,
the reaction mixture was concentrated in vacuo and purified by column
chromatography on silica gel (hexanes/EtOAc with 10/90-70/30 gradient)
to give tert-butyl 12-(pent-4-ynoylamino)-4,7,10-trioxadodecanoate
(1.64 g, 86% yield) as a light yellow oil: 1H NMR (500
MHz, CDCl3) δ 1.44 (s, 9H), 2.02 (m, 1H), 2.41 (m,
2H), 2.50 (m, 4H), 3.45 (m, 2H), 3.56 (m, 2H), 3.63 (m, 8H), 3.72
(m, 2H), 6.34 (s, 1H). 13C NMR (125 MHz, CDCl3) δ 14.8, 28.0, 35.2, 36.2, 39.2, 66.9, 69.2, 69.8, 70.2, 70.3,
70.4, 70.5, 80.5, 83.0, 170.8, 170.9. HRMS (TOF) [M + H]+ calcd for C18H32NO6+ 358.2224, found 358.2231 (Δ = 1.9 ppm).
To a solution
of <span class="Chemical">tert-butyl 12-(pent-4-ynoylamino)-4,7,10-trioxadodecanoate
(1.56 g, 4.36 mmol) in <span class="Chemical">DCM (22 mL, 0.2 M) was slowly added TFA (4.4
mL) and the resulting solution was stirred for 4 h at rt. The progress
of the reaction was monitored by TLC. Upon completion, the reaction
mixture was concentrated in vacuo and purified by column chromatography
on silica gel (DCM/MeOH with 99/1 to 9/1 gradient) to afford 12-(pent-4-ynoylamino)-4,7,10-trioxadodecanoic
acid (950 mg, 96% yield.) as a yellow oil: 1H NMR (500
MHz, CDCl3) δ 2.05 (m, 1H), 2.47 (m, 2H), 2.53 (m,
2H), 2.64 (m, 2H), 3.49 (m, 2H), 3.59 (m, 2H), 3.65 (m, 8H), 3.78
(m, 2H), 6.79 (s, 1H), 10.78 (s, 1H). 13C NMR (125 MHz,
CDCl3) δ 14.9, 34.8, 35.0, 39.5, 66.4, 69.6, 69.7,
70.2, 70.3, 70.4, 70.5, 82.6, 172.3, 175.2. HRMS (TOF) [M –
H]− calcd for C14H22NO6– 300.1453, found 300.1465 (Δ = 4.0
ppm).
A solution of 12-(pent-4-ynoylamino)-4,7,10-trioxadodecanoic
acid
(1.35 g, 4.48 mmol), <span class="Chemical">NHS (618 mg, 5.37 mmol), and <span class="Chemical">EDC·HCl (1.03
g, 5.37 mmol) in DCM (22 mL, 0.2 M) was stirred for 19 h at rt. The
progress of the reaction was monitored by TLC. Upon completion, the
reaction mixture was concentrated in vacuo and purified by column
chromatography on silica gel (DCM/MeOH with 99/1 to 9/1 gradient)
to give activated ester 16 (1.22 g, 70% yield) as a yellow
oil: 1H NMR (500 MHz, CDCl3) δ 2.00 (m,
1H), 2.35 (m, 2H), 2.46 (m, 2H), 2.80 (s, 4H), 2.86 (m, 2H), 3.40
(m, 2H), 3.51 (m, 2H), 3.60 (m, 8H), 3.80 (m, 2H), 5.27 (s, 1H). 13C NMR (125 MHz, CDCl3) δ 14.8, 25.4, 25.5,
32.1, 35.1, 39.2, 65.6, 69.2, 69.7, 70.1, 70.3, 70.5, 70.6, 83.1,
166.8, 169.2, 171.0. HRMS (TOF) [M + H]+ calcd for C18H27N2O8+ 399.1762,
found 399.1783 (Δ = 5.2 ppm).
Fluorescein-O-butanoyl-NH-PEG3-(CH2)2-N319
To a
solution of 4-(9-(2-isobutoxycarbonylphenyl)-3-oxo-3H-xanthen-6-yloxy)butanoic acid (17)[64] (116 mg, 0.24 mmol), 1.2 equiv EDC·HCl salt (55 mg,
0.29 mmol), and 0.5 equiv DMAP (14.7 mg, 0.29 mmol) in DCM (12 mL)
was slowly added a solution of 11-azido-3,6,9-trioxaundecan-1-amine
(18)[72] (63 mg, 0.29 mmol)
in DCM (2 mL), and the resulting solution was stirred under rt for
24 h. The progress of the reaction was monitored by TLC. Upon completion,
the solvent was evaporated and the resulting crude product was purified
by column chromatography on silica gel (DCM/MeOH with 99/1 to 9/1
gradient) to give 19 (136 mg, 0.20 mmol, 82% yield) as
a sticky orange solid: 1H NMR (500 MHz, CD3OD)
δ 0.70 (d, J = 6.7 Hz, 3H), 0.71 (d, J = 6.7 Hz, 3H), 1.09 (m, J = 6.7 Hz, 1H),
2.14 (quint, J = 6.7 Hz, 2H), 2.43 (t, J = 6.7 Hz, 2H), 3.36 (t, J = 6.7 Hz, 2H), 3.38 (t, J = 6.7 Hz, 2H), 3.54 (t, J = 5.4 Hz, 2H),
3.60 (m, 2H), 3.64–3.67 (m, 8H), 3.73–3.82 (m, 2H),
4.21 (t, J = 6.3 Hz, 2H), 6.51 (s, 1H), 6.59 (d, J = 9.6 Hz, 1H), 6.96 (d, J = 9.6 Hz, 1H),
7.06 (d, J = 4.0 Hz, 1H), 7.08 (d, J = 4.0 Hz, 1H), 7.25 (s, 1H), 7.44 (d, J = 7.6 Hz,
1H), 7.80 (t, J = 7.6 Hz, 1H), 7.85 (t, J = 7.6 Hz, 1H), 8.32 (d, J = 7.6 Hz, 1H). 13C NMR (125 MHz, CD3OD) δ 17.8, 17.9, 24.8, 27.4,
31.8, 39.0, 50.3, 68.1, 69.1, 69.7, 69.8, 70.1, 70.2, 70.3, 71.4,
100.6, 104.1, 114.6, 114.8, 116.8, 128.0, 129.5, 129.9, 130.3, 130.5,
130.9, 131.2, 132.6, 133.6, 154.7, 155.0, 160.1, 164.8, 165.5, 173.8,
185.8. HRMS (TOF) [M + H]+ calcd for C36H43N4O9+ 675.3025, found 675.3064
(Δ = 5.7 ppm).
Fluorescein-O-butanoyl-OSu 20
A dark orange solution of 17 (100
mg, 0.21 mmol),
NHS (36.8 mg, 0.32 mmol), DMAP (12.8 mg, 0.11 mmol), and EDC·HCl
(61.3 mg, 0.32 mmol) in DCM (10 mL, 0.02 M) was stirred at rt for
17 h. The progress of the reaction was monitored by TLC. Upon completion,
the solvent was removed in vacuo and the resulting crude product was
directly purified by column chromatography on silica gel (DCM/MeOH
with 99/1 to 9/1 gradient) to give FOB-OSu 20 (114 mg,
95% yield) as an orange solid: 1H NMR (500 MHz, CDCl3) δ 0.64 (d, J = 6.7 Hz, 3H), 0.66
(d, J = 6.7 Hz, 3H), 1.54 (m, J =
6.7 Hz, 1H), 2.20 (quint, J = 6.7 Hz, 2H), 2.79 (s,
4H), 2.80 (m, 2H), 3.67 (m, 2H), 4.11 (t, J = 6.7
Hz, 2H), 6.37 (s, 1H), 6.46 (d, J = 7.8 Hz, 1H),
6.71 (d, J = 7.8 Hz, 1H), 6.82 (d, J = 4.0 Hz, 2H), 6.91 (d, J = 4.0 Hz, 1H), 7.24 (d, J = 7.6 Hz, 1H), 7.61 (t, J = 7.6 Hz, 1H),
7.66 (t, J = 7.6 Hz, 1H), 8.20 (d, J = 7.6 Hz, 1H). 13C NMR (125 MHz, CDCl3) δ
18.8, 18.9, 24.1, 25.6, 27.4, 27.6, 66.7, 71.8, 100.8, 105.7, 113.6,
115.0, 117.6, 129.0, 129.7, 129.8, 130.3, 130.5, 130.7, 131.2, 132.6,
134.1, 150.3, 154.2, 158.9, 163.0, 165.5, 168.1, 169.1, 185.6. HRMS
(TOF) [M + H]+ calcd for C32H30NO9+ 572.1915, found 572.2016 (Δ = 17 ppm).
A mixture
of 6 (11 mg, 0.85 μmol), fluorescein-tether-PEG3-azide 19 (2.3 mg, 3.4 μmol), copper sulfate
pentylhydrate
(1.7 mg, 6.8 μmol), and sodium ascorbic acid (1.2 mg, 6.8 μmol)
in MeOH/distilled water (4/1, 2.5 mL) was stirred for 24 h at rt,
and the progress of the reaction was monitored by FIA-MS. Upon completion,
the reaction mixture was subjected to dialysis using an MWCO 6000–8000
tubing membrane for 48 h against MeOH. The resulting solution was
filtrated to remove insoluble solid and then lyophilized to give ABTD-TTC-3
(3) (10 mg, 76% yield) as a yellow solid: 1H NMR (700 MHz, CD3OD) δ 0.70 (m, 8H), 0.89 (m,
32H), 1.19 (m, 16H), 1.30 (m, 32H), 1.39 (m, 12H), 1.46 (m, 48H),
1.62 (m, 48H), 1.67 (m, 32H), 1.76 (m, 28H), 1.94 (m, 8H), 2.19 (m,
24H), 2.19 (m, 32H), 2.31 (m, 8H), 2.42 (m, 24H), 2.48 (m, 120H),
2.74 (m, 24H), 2.95 (m, 36H), 3.19 (m, 16H), 3.23 (m, 32H), 3.26 (m,
16H), 3.36 (s, 16H), 3.38 (m, 32H), 3.42 (m, 8H), 3.55 (m, 68H), 3.65
(m, 188H), 3.75 (m, 40H), 4.34 (m, 16H), 4.53 (m, 16H), 4.62 (m, 8H),
4.93 (s, 16H), 4.99 (m, 8H), 6.66 (m, 4H), 6.94 (m, 4H), 7.30 (m,
4H), 7.49 (m, 4H), 7.62 (m, 4H), 7.74 (m, 4H), 8.15 (m, 4H), 8.21
(m, 4H). 13C NMR (176 MHz, CD3OD) δ 7.5,
9.9, 11.2, 11.3, 16.7, 17.8, 19.2, 21.3, 21.7, 22.3, 24.8, 25.0, 25.1,
25.2, 25.9, 27.2, 27.4, 28.1, 28.4, 28.8, 29.2, 29.3, 29.5,. 35.4,
35.8, 38.9, 39.7, 43.2. 50.3, 55.6, 55.8, 56.9, 60.2, 61.9, 66.9,
67.5, 68.4, 68.5, 69.0, 69.2, 69.7, 69.9, 70.1, 70.2, 100.7, 128.4,
129.4, 129.7, 130.2, 130.3, 130.4, 130.9, 132.0, 131.7, 132.6, 163.6,
164.9, 166.1, 166.8, 170.1, 172.6, 174.7, 174.6. GPC analysis (see General Methods for conditions): tR = 22.1 min; >99% purity (see Figure S14).
Biological Evaluations
Cell Culture
ID-8 (ovary), MX-1 (breast), and WI-38
(human lung fibroblast) cell lines were cultured as monolayers on
100 mm tissue culture dishes in the RPMI-1640 cell culture medium
(Gibco) supplemented with 10% (v/v) heat-inactivated fetal bovine
serum (FBS) as well as 1% (v/v) penicillin and streptomycin (P/S)
at 37 °C in a humidified atmosphere with 5% CO2. The
cells were harvested and collected by centrifugation at 950 rpm for
5 min and finally suspended in fresh cell culture medium containing
different cell densities for subsequent biological experiments and
analysis.
Incubation of Cells with Fluorescent Probes
The cell
suspension at 5 × 105 cells/well was split into each
well of a six-well plate with 3 mL of <span class="CellLine">RPMI-1640 cell culture medium
and allowed to grow for 24 h. The fluorescent probe 28 or 3 in dimethyl sulfoxide (<span class="Chemical">DMSO) was diluted in RPMI-1640
cell culture medium (3 mL) at a final concentration of 10 or 20 μM
for multibinding effect study and internalization through RME study,
respectively, and was used to replace the medium in each well followed
by incubation at 37 °C for different periods (16, 6, 4, 2, and
1 h for multibinding effect study, and 3 and 1 h for internalization
through RME study). One well with cells was incubated without any
fluorescent probes as control for different periods (16 h for multibinding
effect study, and 3 h for internalization through RME study). After
incubation, the medium was aspirated and the cells were washed twice
with PBS, collected by centrifugation, and resuspended in 300 μL
of PBS for flow cytometry fluorescein measurement and confocal microscopy
imaging.
Flow Cytometry Analysis of Treated Cells
Flow cytometry
analysis of the treated cells was perfo<span class="Chemical">rmed with a flow cytometer,
FACSCalibur, operating at a 488 nm excitation wavelength and detecting
emission wavelengths with a 530/30 nm bandpass filter. At least 10 000
cells were counted for each experiment using CellQuest 3.3 software
(Becton Dickinson), and the distribution of fluorescence was analyzed
using WinMDI 2.8 freeware (Joseph Trotter, Scripps Research Institute).
<span class="Chemical">Propidium iodide staining was used in all experiments to rule out
dead cell count in the analysis.
Confocal Microscopy Analysis
of Treated Cells
Cells
treated as described above were dropped onto an uncoated glass dish
(MatTek Corp.). Confocal fluorescence microscopy (CFM) experiments
were perfo<span class="Chemical">rmed using a Zeiss LSM 510 META NLO two-photon laser scanning
confocal microscope system, operating at a 488 nm excitation wavelength
and at 527 nm (23 nm detecting emission wavelength using a 505–550
nm bandpass filter). Images were captured using a C-Apochromat 63×/1.2
<span class="Chemical">water (corr.) objective or a Plan-Apochromat 100×/1.45 oil objective.
Acquired data were analyzed using LSM 510 META software.
Cytotoxicity
Assay
Cytotoxicities of paclitaxel, SB-T-1214
(4), SB-T-1214-fluorescein (5), biotin-PEG3-linker-SB-T-1214 (BLT-S) (29), ABTD-TCC-1 (1), ABTD-TTC-2 (2), and ABTD-TTC-3 (3) were evaluated in vitro against ID-8 (ovary, BR++) and MX-1 (breast,
BR+) cancer cell lines, wherein biotin receptor (BR) is overexpressed,
as well as WI-38 (human lung fibroblast cell line, normal), by means
of a quantitative colorimetric assay using a tetrazolium salt-based
analysis (“MTT assay”; MTT = 3-(4,5-dimeth-ylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide; Sigma Chemical Co.). The inhibitory activity of each compound
is represented by the IC50 value, which is defined as the
concentration required for inhibiting 50% of the cell growth. The
cells were harvested, collected, and resuspended in 200 μL of
medium at a concentration of ∼0.5 × 106 cells
(ID-8) or ∼1.5 × 106 (MX-1 and WI-38) per well
over a 96-well plate. The cells were allowed to descend to the bottom
of the plates overnight, and fresh medium was added to each well upon
removal of the old medium. A drug or drug conjugate in DMSO stock
solution was diluted to a series of concentrations in the cell culture
medium to prepare test solutions. These test solutions at different
concentrations ranging from 500 pM to 5 μM (100 μL each)
were added to the wells in the 96-well plate, and the cells were subsequently
cultured for 72 h. After removing the old medium by aspiration, 50
μL of DPBS buffer containing MTT (0.5 mg/mL) was added to each
well and incubated at 37 °C for 3 h. The resulting DPBS buffer
was then removed and the as-produced insoluble violet formazan crystals
were dissolved in 50 μL of 0.1 N HCl in isopropanol to give
a violet solution. The plate was allowed to shake for 8 min to fully
dissolve the violet formazan crystal and then the spectrophotometric
absorbance measurement of each well in the 96-well plate was run at
568 nm. The IC50 values and their standard errors were
calculated from the viability–concentration curve using the
Four Parameter Logistic Model of Sigmaplot. The concentration of DMSO
per well was ≤1% in all cases.
Authors: Jan Kovár; Marie Ehrlichová; Barbora Smejkalová; Ilaria Zanardi; Iwao Ojima; Ivan Gut Journal: Anticancer Res Date: 2009-08 Impact factor: 2.480
Authors: Larissa Kuznetsova; Jin Chen; Liang Sun; Xinyuan Wu; Antonella Pepe; Jean M Veith; Paula Pera; Ralph J Bernacki; Iwao Ojima Journal: Bioorg Med Chem Lett Date: 2005-11-18 Impact factor: 2.823
Authors: Baohua Huang; Jolanta F Kukowska-Latallo; Shengzhuang Tang; Hong Zong; Kali B Johnson; Ankur Desai; Chris L Gordon; Pascale R Leroueil; James R Baker Journal: Bioorg Med Chem Lett Date: 2012-03-21 Impact factor: 2.823
Authors: Iwao Ojima; Jin Chen; Liang Sun; Christopher P Borella; Tao Wang; Michael L Miller; Songnian Lin; Xudong Geng; Larisa Kuznetsova; Chuanxing Qu; David Gallager; Xianrui Zhao; Ilaria Zanardi; Shujun Xia; Susan B Horwitz; Jon Mallen-St Clair; Jennifer L Guerriero; Dafna Bar-Sagi; Jean M Veith; Paula Pera; Ralph J Bernacki Journal: J Med Chem Date: 2008-05-09 Impact factor: 7.446
Authors: Changwei Wang; Xin Wang; Yi Sun; Adam K Taouil; Su Yan; Galina I Botchkina; Iwao Ojima Journal: Bioorg Chem Date: 2019-12-20 Impact factor: 5.275
Figure 1
Figure 2
Figure 3
Figure 4
Scheme 1
Scheme 2
Scheme 3
Scheme 4
Scheme 5
Scheme 6
Scheme 7
Scheme 8
Figure 5
Figure 6