N -acetylgalactosamino dendrons as clearing agents to enhance liver targeting of model antibody-fusion protein

Article abstract of DOI:10.1021/bc400333m

Dendrimer clearing agents represent a unique class of compounds for use in multistep targeting (MST) in radioimmunotherapy and imaging. These compounds were developed to facilitate the removal of excess tumor-targeting monoclonal antibody (mAb) prior to a

Full text of DOI:10.1021/bc400333m

Technical Note
pubs.acs.org/bc
N‑Acetylgalactosamino Dendrons as Clearing Agents to Enhance
Liver Targeting of Model Antibody-Fusion Protein
Barney Yoo,* Sarah M. Cheal, Geralda Torchon, Anna Dilhas, Guangbin Yang,†,‡ Jun Pu,†,‡
,
†,‡
§
†,‡
†,‡
§
‡,§,∥
,†,‡
Blesida Punzalan, Steven M. Larson,
and Ouathek Ouerfelli*
†
‡
§
∥
Organic Synthesis Core Facility, Molecular Pharmacology and Chemistry Program, Department of Radiology, and Molecular
Pharmacology and Therapy Service, Memorial Sloan-Kettering Cancer Center, New York, New York, United States
*
S Supporting Information
ABSTRACT: Dendrimer clearing agents represent a unique class of
compounds for use in multistep targeting (MST) in radioimmuno-
therapy and imaging. These compounds were developed to facilitate
the removal of excess tumor-targeting monoclonal antibody (mAb)
prior to administration of the radionuclide to minimize exposure of
normal tissue to radiation. Clearing agents are designed to capture the
circulating mAb, and target it to the liver for metabolism.
Glycodendrons are ideally suited for MST applications as these highly
branched compounds are chemically well-defined, thus advantageous
over heterogeneous macromolecules. Previous studies have described
glycodendron 3 as a clearing agent for use in three-step MST protocols,
and early in vivo assessment of 3 showed promise. However, synthetic
challenges have hampered its availability for further development. In this report we describe a new sequence of chemical steps
which enables the straightforward synthesis and analytical characterization of this class of dendrons. With accessibility and
analytical identification solved, we sought to evaluate both lower and higher generation dendrons for hepatocyte targeting as well
as clearance of a model protein. We prepared a series of clearing agents where a single biotin is connected to glycodendrons
displaying four, eight, sixteen or thirty-two α-thio-N-acetylgalactosamine (α-SGalNAc) units, resulting in compounds with
molecular weights ranging from 2 to 17 kDa, respectively. These compounds were fully characterized by LCMS and NMR. We
then evaluated the capacity of these agents to clear a model 1 I-labeled single chain variable fragment antibody-streptavidin ( I-
scFv-SAv) fusion protein from blood and tissue in mice, and compared their clearing efficiencies to that of a 500 kDa dextran-
biotin conjugate. Glycodendrons and dextran−biotin exhibited enhanced blood clearance of the scFv-SAv construct.
Biodistribution analysis showed liver targeting/uptake of the scFv-SAv construct to be 2-fold higher for compounds 1 to 4, as
well as for the 500 kDa dextran, over saline. Additionally, the data suggest the glycodendrons clear through the liver, whereas the
dextran through reticuloendothelial system (RES) metabolism.
31
131
INTRODUCTION
Dendrimers provide an ideal platform to fulfill the molecular
requirements of clearing agents for clinical MST applications.
They can allow for precise construction of a carbohydrate array
with optimum density and geometry for efficient lectin
recognition, while incorporating an orthogonal functional
group for binding of the tumor-targeting mAb. Additionally,
the polarity and size of dendrimers can be tuned to restrict
diffusion into extravascular compartments, such as tumors, thus
restricting access to the prelocalized mAb bound to the tumor.
While conjugates of macromolecules − such as glycoconjugates
of albumin with biotin ligands, and derivatized dextrans −
have been demonstrated to be effective clearing agents, they
possess a high degree of compositional heterogeneity. This
poses a barrier to reproducible synthesis and, eventually,
dosing. From a biological standpoint, use of these molecules
■
Dendrimers have broad application in the fields of biology and
medicine. For example, in cancer therapy, these highly
branched polymers have been utilized as multivalent carriers
for antimetastatic activity, the stimulation of immune responses,
tumor targeting, as well as delivery vehicles for chemo-
1
,2
therapeutics. One specific application where carbohydrate
containing dendrimers, or glycodendrimers, have shown
promise is as clearing agents in multistep targeting (MST)
strategies for radioimmunotherapy (RIT) and imaging. MST
clearing agents are bifunctional compounds designed to
facilitate the rapid clearance of an unbound tumor-targeting
protein, such as a recombinant monoclonal antibody (mAb),
from the vasculature and off-target tissue prior to admin-
5
6
3
istration of the imaging tracer or drug. This step is often
essential to achieving a high therapeutic index. The clearance
mechanisms exploited by these agents are through recognition
by the reticuloendothelial system (RES) and/or active targeting
Received: July 17, 2013
Revised: October 22, 2013
Published: October 22, 2013
4
to the liver through hepatocyte lectins.
©
2013 American Chemical Society
2088
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
Figure 1. Glycodendron clearing agents (1−4), α-thio-GalNAc end group (5), and biotin core.
1
1
can lead to interference with subsequently administered
imaging tracer or drug at target tissue, since processing of the
macromolecules by the RES can lead to the introduction of
metabolites back into circulation which can saturate binding
sites (for the imaging tracer or drug) on the targeting mAb.
The N-acetylgalactoseamine functionalized dendron 3
galactosamine (GalNAc) units. In addition, other studies
have shown that significant binding enhancements to the ASGP
receptor were obtained with branched polyvalent presentation
1
1
of the carbohydrate groups, further underscoring the
advantage of a dendron scaffold. On the glycodendron,
monosaccharides are S-linked, as opposed to O-linked, which
imparts greater resistance to hydrolysis. The single biotin is
attached to an N-methyl amide, which has been shown to
3
,7
(
Figure 1) has proven effective in MST, and has been utilized
in preclinical and clinical studies using the biotin/streptavidin
8
based three-step pretargeting approach. This glycodendron
12
provide stability against biotinases in vivo. While compound 3
has been utilized in MST studies with promising results,
contains a single biotin attached to a dendron scaffold
composed of aminohexanoic aliphatic arms connected through
tertiary amides, with sixteen α-configured thio-N-acetylgalactos-
amine end groups 5. These monosaccharides are used for liver
targeting, as they serve as ligands for the asialylglycoprotein
5
,8,13−15
synthesis of this glycodendron has remained a challenge, and
analytical characterization has not been previously described. In
addition, we had particular interest in evaluating lower
generation variants in order to further optimize structures
that may be less complex and more readily obtained for clinical
use.
9,10
(
ASGP) receptors expressed on hepatocytes.
The ASGP
receptor is a lectin that binds desialylated glycoproteins with
particular preference for terminal galactose and N-acetyl
2
089
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
Absent clear answers as well as analytical data, and present
the urge to gain insights into the clearing mechanism and
properties of these glycodendrons, we sought to synthesize
compounds 1 to 4 and subsequently test their clearing abilities.
Most prior work was limited to patents. Be that as it may, the
lack of a clear synthesis path, analytical data, and clearing
profiles dictated that we revisit all the above problems. With
clinical development and its need for large amounts of the best
clearing agent in mind, we embarked on the search for a better
clearing agent by solving all the above problems. From the
synthesis point of view, one could clearly discern an advantage
in more converging strategies that couple advanced inter-
mediates of the dendron or branches thereof bearing the
GalNac units in late stages of the synthesis to lead to the final
compound. However, despite previously reported methods to
obtain 3 we have found most convergent routes to be much
more problematic at several levels. At the level of chemistry,
there needed to be longer reaction times to ensure completion,
which, due the large number of protecting groups, led to partial
loss of some under different coupling conditions. At the
purification level, numerous purification steps ensued which
proved to be unpredictable, time-consuming, and material
wasting.
(DMAP), Lawesson’s reagent, sodium bromide (NaBr),
tetrabutylammonium bromide (Bu NBr), acetonitrile, diethyl
4
ether, dimethylformamide (DMF), ethyl acetate, hexanes,
methylene chloride (DCM), diisopropylethylamine (DIEA),
and triethylamine (TEA) were obtained from Sigma-Aldrich.
Concentrated HCl was obtained from Fisher Scientific. D-Biotin
was obtained from Fluka. 2-(7-Aza-1H-benzotriazole-1-yl)-
1,1,3,3-tetramethyluronium hexafluorophosphate (HATU)
was obtained from Genescript. 2-Acetomido-2-deoxy-3,4,6-tri-
O-acetyl-1-thio-α-D-galactopyranose and tert-butyl (5-
hydroxypentyl)carbamate were obtained from BOC Sciences.
Deuterium oxide and CDCl were obtained from Cambridge
3
Isotopes. Solid-phase synthesis was conducted in polypropylene
fritted syringes from Torviq. All resins were obtained from
131
Novabiochem. I was purchased in the form of carrier-free
131
Na I in NaOH containing 0.02 M Na SO from Nordion
2
4
(Canada) Inc. The CC49 single chain variable fragment scFv-
streptavidin (scFv-CC49-SAv) fusion protein was provided by
16
the NeoRx Corporation. Pierce precoated iodination tubes
(50 μg of Iodogen/tube) were purchased from Thermo
Scientific. L-Tyrosine, glycerol, and xylene cyanol, were
purchased from Sigma-Aldrich. Silica gel IB-F reverse-phase
thin-layer chromatography (TLC) sheets were obtained from
J.T. Baker. All buffers and solutions were prepared using
ultrapure water (18 MΩ-cm resistivity). Outbred female
athymic nude mice (4−6 weeks of age) were obtained from
Harlan Sprague−Dawley Inc. Biotin-deficient diet (5836) was
obtained from Purina Mills. The human cell carcinoma line
LS174T was obtained from ATCC (CL-188).
In this report, we devised a new synthetic approach to readily
obtain the aliphatic dendrons 11 to 14 using a semiconvergent
methodology where the building block 7 is successively used to
increase dendron generations, and later at dendron completion,
appropriately protected aminoalkylthiosugar moieties were
introduced prior to global deprotection. The incorporation of
an orthogonal protecting groups strategy provided the
versatility to site-specifically attach biotin as well as 4 to 32
α-thio-GalNAc units, and led to the desired family of
glycodendrons (1 to 4) with molecular weights ranging from
approximately 2 to 17 kDa, respectively (Figure 1). Addition-
ally, we have explored solid phase synthesis as a faster means
for development of such molecules. However, solid phase
synthesis has proven incompatible with the growing dendrimer
skeleton, and coupling became inefficient by the second
generation. Overall the homogeneous synthesis approach is
highly reproducible and efficient, requiring a significantly
reduced number of purification steps. Compounds 1 to 4 and
intermediates 11 to 14 were fully characterized by LCMS and
NMR to ensure product integrity and purity. To verify
biological efficacy and utility for MST in vivo, we examined
the blood clearance kinetics of a model antibody−streptavidin
fusion protein following administration of glycodendrons 1 to 4
in normal biotin-starved mice by serial-blood sampling. In
addition, we conducted biodistribution studies to determine
tracer uptake and retention in the liver, as well as other RES-
associated organs. The model antibody−streptavidin protein
LCMS Analysis and Purification. All liquid chromatog-
raphy mass spectrometry (LCMS) data was obtained with a
Waters Autopure system comprising the following: 2767
Sample Manager, 2545 Binary Gradient Module, System
Fluidics Organizer, 2424 Evaporative Light Scattering Detector,
2
998 Photodiode Array Detector, 3100 Mass Detector. Binary
solvent system: solvent A, 0.05% TFA in water; solvent B, 0.05%
TFA in acetonitrile. Analytical method (unless otherwise noted):
5
−95% solvent B in 10 min, 1.2 mL/min flow rate. Analytical
columns: Waters XBridge, BEH300, C4, 3.5 μm, 4.6 × 50 mm
and C18, 5 μm, 4.6 × 50 mm. Preparative method: 5−95%
solvent B in 30 min, 20 mL/min flow rate. Preparative Columns:
Waters XBridge Prep C18, 5 μm, OBD, 19 × 150 mm.
Silica Gel Purification. Silica gel purifications were
performed on the Teledyne ISCO CombiFlash Companion
Purification System. RediSep normal phase flash columns from
Teledyne ISCO were used for purifications.
NMR Characterization. All NMR data were obtained with
either a Bruker AV 500 or AV 600 instruments. The following
abbreviations are used: singlet (s), broad singlet (bs), doublet
(
(
d), triplet (t), quartet (q), pentet (p), doublet of a doublet
dd), multiplet (m).
Diethyl N,N-6,6′-((4-Methoxybenzyl)amino)-
1
31
used for this study is the I-labeled scFv-CC49-SAv, a fusion
protein containing the single chain variable fragment (scFv) of
15−18
the murine mAb CC49 and streptavidin.
The CC49
dihexanoate (6). 4-Methoxybenzylamine (92 g, 87 mL,
fragment is specific for the tumor associated glycoprotein-72
0
.673 mol) and KI (143 g, 0.861 mol, 1.3 equiv) were mixed
in a 2 L flask. The reaction was heated in an oil bath to 75−80
C for 5 min, and DIEA (382 g, 514 mL, 2.959 mol, 4.4 equiv)
(
TAG72), which is known to be widely expressed on
18
adenocarcinomas.
°
was added. The resulting suspension was allowed to heat for
another 15 min, and then ethyl 6-bromohexanoate (300 g, 239
mL, 1.345 mol, 2.0 equiv) was introduced and reflux was
continued overnight. The mixture was cooled to ambient
temperature, filtered over a Buchner funnel, and the filtrate
̈
evaporated. The residue was resuspended in ethyl acetate and
EXPERIMENTAL PROCEDURES
General. Solvents and reagents purchased from commercial
sources were used without further purification. Ethyl 6-bromo-
hexanoate, N-methylcaprolactam, potassium iodide (KI), 4-
methoxybenzylamine (PBM), di-tert-butyl carbonate, p-tolue-
nesulfonyl chloride (TsCl), 4-(dimethylamino)pyridine
■
washed with water, 3×. The organic layer was dried with
2
090
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
sodium sulfate, evaporated, and purified on silica gel using a
(340 uL, 252 mg, 1.75 mmol, 3.5 equiv). HATU (374 mg,
0.902 mmol, 1.2 equiv) was dissolved in 1 mL DMF then added
to 9. After 20 min, solvent was partially removed under high
vacuum at room temperature, and the residue was transferred
to a separatory funnel containing diethyl ether. The organic
layer was washed with 0.1 N HCl (3×) and then with water
(1×). The ether and residual water was removed under reduced
pressure, then high vacuum. 253 mg (95%) of 10 was obtained
hexane/ethyl acetate gradient. The amine was isolated as a
1
slightly tan oil in typically 70−80% yield. H NMR (CDCl ,
3
5
00 MHz) δ = 7.21 (d, J = 8.5 Hz, 2H), 6.83 (d, J = 8.5 Hz,
H), 4.12 (q, J = 7 Hz, 4H), 3.80 (s, 3H), 3.46 (s, 2H), 2.36 (t,
2
J = 7.3 Hz, 4H), 2.25 (t, J = 7.5 Hz, 4H), 1.59 (p, J = 7.5 Hz,
H), 1.45 (p, J = 7.5 Hz, 4H), 1.27 (m, 10H). 13C NMR
CDCl , 500 MHz) δ = 173.8, 158.4, 132.0, 129.9, 113.5, 60.2,
4
(
3
5
7.9, 55.2, 53.4, 34.4, 27.0, 26.7, 24.9, 14.3. ESI-MS (m/z) for
as a light brown oil and was used without further purification.
+
1
C H NO (exact mass 421.28): [M+H] calc. 422.29; obs.
H NMR (CDCl , 600 MHz): δ = 4.13 (m, J = 7.1, 2.9 Hz,
24
39
5
3
4
22.3.
4H), 3.28 (t, J = 6.3 Hz, 2H), 3.20 (t, J = 6.5 Hz, 4H), 2.83 (s,
Aminodiethylhexanoate (7). Compound 6 (50 g, 119
3H), 2.29 (m, 6H), 1.65 (m, 6H), 1.54 (m, 6H), 1.45 (s, 9H),
13
mmol) was placed under a blanket of argon in a 1 L stainless
steel autoclave equipped with a magnetic stirrer. Ethanol (500
mL) was then introduced followed by ammonium formate (25
g, 392 mmol, 3.3 equiv). The vessel was then flooded with
argon such that the addition of palladium 10% on carbon (13 g,
1.31 (m, 6H), 1.26 (q, J = 7.0 Hz, 6H). C NMR (CDCl , 600
3
MHz): 173.7, 173.4, 155.8, 79.1, 60.3, 60.2, 47.8, 45.7, 34.2,
34.1, 33.1, 28.9, 27.5, 26.7, 26.6, 26.4, 25.2, 24.7, 24.7. ESI-MS
+
(m/z) for C H N O (exact mass 528.38): [M+H] calc.
28
52
2
7
529.38; obs. 529.4.
1
119 mmol, 1 equiv) does not spark. The autoclave was then
Generation 1 (G ) Dendron (11). Hydrolysis of Ethyl
0
sealed and the reaction was heated in a 110 °C oil bath for 5 h.
After cooling, the reaction vessel was depressurized and flushed
with argon before opening. The mixture was then filtered on a
thick pad of Celite, and evaporated in vacuo. The remaining
yellow oil was then purified on silica gel using ethyl acetate and
Esters of G Dendron. Compound 10 (253 mg, 0.5 mmol) was
dissolved in 2 mL ethanol and 2 mL of 2 N NaOH. The
hydrolysis was conducted in a 40 °C water bath for 1 h. The
solution was transferred to a separatory funnel and then
neutralized with 1 N HCl. Diethyl ether was added and the
organic layer was washed with 0.1 N HCl (3×). The organic
layer was collected and the solvent was removed under reduced
1
ethanol to provide 21 g (60%) of 7 as a brown oil. H NMR
(
CDCl , 500 MHz) δ = 4.12 (q, J = 7.1 Hz, 4H), 2.6 (t, J = 7.3
3
1
Hz, 4H), 2.3 (t, J = 7.5 Hz, 4H), 1.64 (p, J = 7.6 Hz, 4H), 1.50
pressure resulting in a tan oil. Coupling of 7 to form G dendron:
13
(
p, J = 7.5 Hz, 4H), 1.35 (m, 4H), 1.25 (t, J = 7.1 Hz, 6H). C
The oil was dissolved in 3 mL DMF. To this, compound 7 (440
mg, 1.3 mmol, 2.6 equiv) was added followed by DIEA (681
uL, 505 mg, 2 mmol, 4 equiv). HATU (456 mg, 1.2 mmol, 2.4
equiv) was dissolved in 2 mL DMF, then added. The reaction
was allowed to proceed for 20 min, at which point the solvent
was partially removed in vacuo, and transferred to a separatory
funnel with diethyl ether. The organic layer was washed with
water (1×), 0.1 N HCl (3×), and then water (1×). The organic
layer was collected and subjected to reduced pressure to
NMR (CDCl , 500 MHz, 298K) δ = 173.7, 60.2, 49.6, 34.4,
3
3
4.3, 29.4, 26.9, 25.0, 24.8, 14.3. ESI-MS (m/z) for C H NO
16 31 4
+
(exact mass 301.23): [M+H] calc. 302.23; obs. 302.3.
N-Methyl Caproic Acid·HCl (8). N-Methylcaprolactam (10
g, 78.8 mmol) was added to a 500 mL three-neck flask
equipped with a water condenser and pressure equalizing
addition funnel and cooled in an ice bath. Concentrated HCl
(
250 mL) was then carefully added dropwise, and the reaction
was allowed to reflux for two days. The reaction was evaporated
to dryness, and the product was crystallized in ethanol/ether to
remove solvents. 11 was obtained as a light brown oil (470 mg,
1
90%) and used without further purification. H NMR (CDCl ,
3
give 14.1 g (99%) of a white solid (HCl salt). Analytical data for
600 MHz) δ = 4.13 (m, J = 7.1, 3.6 Hz, 8H), 3.28 (q, J = 7.6,
6H), 3.2 (q, 8H), 2.83 (s, 3H), 2.29 (m, 14H), 1.65 (m, 14H),
1
free base: H NMR (D O, 600 MHz) δ = 2.96 (t, J = 7.7 Hz,
2
2
2
6
H), 2.63 (s, 1H), 2.32 (t, J = 7.4 Hz, 2H), 1.62 (p, J = 7.7 Hz,
1.54 (m, 14H), 1.45 (s, 9H), 1.31 (m, 14H), 1.26 (q, J = 7.1
H), 1.56 (p, J = 7.6 Hz, 2H), 1.33 (m, 2H). 1 C NMR (D O,
3
13
Hz, 12H). C NMR (CDCl , 600 MHz) δ = 173.68, 173.65,
2
3
00 MHz, 298K) δ = 178.82, 48.86, 33.43, 32.60, 25.03, 23.66.
173.44, 173.41, 172.39, 172.00, 155.83, 79.07, 60.35, 60.33,
60.22, 60.20, 47.80, 45.80, 45.72, 34.26, 34.24, 34.13, 34.10,
33.09, 32.89, 29.22, 28.93, 28.50, 27.78, 27.53, 27.51, 27.02,
26.87, 26.69, 26.59, 26.45, 25.28, 25.11, 24.73, 24.68, 24.65
14.26. ESI-MS (m/z) for C H N O (exact mass 1038.74):
+
ESI-MS (m/z) for C H NO (exact mass 145.11): [M+H]
calc. 146.12; obs. 146.3.
N-Methyl-(tert-Butyloxycarbonyl)-Carproic Acid (9).
N-Methyl caproic acid·HCl salt 8 (3.5 g, 0.019 mol) was
suspended in 100 mL DCM. tert-Butyloxy carbamate (4.5 g,
7
15
2
5
6
102
4
13
+
[M+H] calc. 1039.75, obs. 1040.06.
Generation 2 (G ) Dendron (12). Hydrolysis of Ethyl
Esters of G Dendron. Compound 11 (97 mg, 0.1 mmol) was
2
0.021 mol, 1.1 equiv) was then added, followed by TEA (5.6
1
mL, 0.04 mol, 2.1 equiv). The mixture was stirred for 60 min,
and the DCM/TEA was removed under reduced pressure. The
contents were resuspended in ethyl acetate, and washed with
dissolved in 2 mL ethanol, and 2 mL of 2.0 N NaOH were
added. The hydrolysis was conducted in a 40 °C water bath for
1 h. The solution was transferred to a separatory funnel and
then neutralized with 1.0 N HCl. A solution of diethyl ether
and DCM (10:1, v:v) was added and the organic layer was
washed with 0.1 N HCl (3×). The organic layer was collected
and subjected to reduced pressure to remove solvents and
0.2 N HCl. The organic layer was dried over MgSO , filtered,
4
and the volatiles were removed under reduced pressure to
afford 9 as a colorless oil (4.4 g, 95%). H NMR (CDCl , 500
MHz, 298K) δ = 3.20 (bs, 2H), 2.83 (bs, 3H), 2.35 (t, J = 7.4
Hz, 2H), 1.66 (p, J = 7.6 Hz, 2H), 1.53 (p, J = 7.5 Hz, 2H),
1
3
1
3
2
1
2
2
2
.45 (s, 9H), 1.33 (p, 2H). C NMR (CDCl , 600 MHz,
98K) δ = 179.1, 155.9, 79.3, 48.6, 48.2, 34.1, 33.9, 28.5, 27.6,
7.3, 26.1, 24.4. ESI-MS (m/z) for C H NO (exact mass
obtain a light brown oil. Coupling of 7 to form G dendron: The
3
oil was dissolved in 4 mL DMF. To this was added compound
7 (236 mg, 0.7 mmol, 7 equiv), followed by DIEA (280 uL, 207
mg, 1.6 mmol, 4 equiv) and HATU (228 mg, 0.6 mmol, 6
equiv) as dissolved solution in 1 mL DMF. The reaction was
allowed to proceed for 20 min, at which point the solvent was
partially removed under reduced pressure, and the mixture was
1
2
23
4
+
45.16): [M+H] calc. 246.17; obs. 246.3.
0
Generation 0 (G ) Dendron (10). Compound 9 (125 mg,
0
.5 mmol) was suspended in 1 mL DMF. Compound 7 (228
mg, 0.6 mmol, 1.2 equiv) was then added, followed by DIEA
2
091
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
transferred to a separatory funnel containing diethyl ether. The
organic layer was washed with 0.2 N HCl (3×), and then with
water (2×). The organic layer was collected and subjected to
reduced pressure to remove solvent. 176 mg (92%) of 12 was
lyophilized overnight. The dendron was dissolved in 1 mL
DMF, and DIEA (44 uL, 32 mg, 0.043 mmol, 10 equiv) was
added. Compound 9 (12 mg, 0.050 mmol, 2 equiv) was
premixed with HATU (16 mg, 0.043 mmol, 1.7 equiv) in 200
uL of DMF; to which DIEA (44 uL, 32 mg, 0.25 mmol, 10
equiv) was then added. This solution was then added to the
dendron and stirred at room temperature for 20 min. For
workup, DCM (10 mL) was added, and the organic solution
was washed with water (1×), 0.1 N HCl (3×), and then water
(1×). After evaporation, the residue was left to dry overnight
under high vacuum, to give an oil.
Increasing Dendron Generation. The sixteen ethyl esters of
the dendron were hydrolyzed in 1 mL of 1.0 N NaOH and 1
mL ethanol, then left to stir overnight at ambient temperature.
The volatiles were then removed under reduced pressure, and
residue was later transferred to a separatory funnel and then
neutralized with 1.0 N HCl. A solution of diethyl ether, tBuOH,
and DCM (6:3:1; v:v:v) was added and the organic layer was
washed with 0.1 N HCl (3×), then collected and evaporated
under vacuum. The residual brown oil was diluted in anhydrous
DMF (2 mL), and 7 (116 mg, 0.384 mmol, 16 equiv) was
added followed by DIEA (134 uL, 99 mg, 0.768 mmol, 32
equiv), then HATU (124 mg, 0.326 mmol, 0.041 equiv). The
reaction was allowed to proceed for 20 min under stirring, at
which point the mixture was transferred to a separatory funnel
with diethyl ether and DCM. The organic layer was washed
with 0.1 N HCl (3×), then water (2×), before it was collected
and evaporated under reduced pressure. 161 mg (79%) of 14
1
obtained as a tan oil, and used without further purification. H
NMR (CDCl , 600 MHz) δ = 4.12 (m, J = 7.1, 2.8 Hz, 16H),
3
3
3
3
.28 (q, J = 7.6 Hz, 14H), 3.21 (q, 16H), 2.83 (s, 3H), 2.29 (m,
0H), 1.65 (m, 30H), 1.54 (m, 30H), 1.45 (s, 9H), 1.31 (m,
1
3
0H), 1.26 (q, J = 6.9 Hz, 24H). C NMR (CDCl , 600 MHz)
3
δ = 173.68, 173.42, 172.52, 172.09, 79.07, 60.34, 60.21, 47.84,
4
2
2
5.97, 45.85, 45.78, 34.25, 34.12, 34.10, 33.07, 32.87, 29.22,
8.92, 28.50, 27.79, 27.74, 27.50, 27.04, 26.87, 26.59, 26.44,
5.34, 25.28, 25.16, 25.10, 24.73, 24.67, 14.27, 14.26. ESI-MS
+
(m/z) for C H N O (exact mass 2059.48): [M+H] calc.
12
202
8
25
2
+
2060.49; [M+2H] calc. 1030.75, obs. 1031.59.
3
Generation 3 (G ) Dendron (13). Compound 12 (90 mg,
0.043 mmol) was dissolved in 5 mL ethanol, and 4 mL of 2.0 N
NaOH was added. The hydrolysis was conducted in a 40 °C
water bath for 1 h. The solution was transferred to a separatory
funnel and then neutralized with 1.0 N HCl. A solution of
diethyl ether and DCM (10:1; v:v) was added and the organic
layer was washed with 0.1 N HCl (3×). The organic layer was
collected and subjected to reduced pressure to remove solvent.
The resulting oil was diluted in DMF (4 mL). To this, 7 (245
mg, 0.09 mmol, 16 equiv) was added followed by DIEA (280
uL, 207 mg, 1.6 mmol, 4 equiv). HATU (233 mg, 0.6 mmol, 6
equiv) was first dissolved in DMF (1 mL) then added. The
reaction was allowed to proceed for 20 min, at which point the
reaction was partially concentrated in vacuo, and the residue
was transferred to a separatory funnel preloaded with a diethyl
ether and DCM. The organic layer was washed with 0.1 N HCl
was obtained as a brown oil, which was used without further
1
purification. H NMR (CDCl , 600 MHz) δ = 4.12 (m, J = 7.1,
3
2.8 H, 64H), 3.28−3.21 (m, 140H), 2.97 (m, 3H), 2.90 (m,
3H), 2.83 (s, 3H), 2.29 (m, 130H), 1.65 (m, 130H), 1.55 (m,
(
3×) and with water (2×). Evaporation under reduced pressure
provided 168 mg (95% crude yield) of 13 was obtained as a
light brown oil, and was used without further purification. H
130H), 1.45 (s, 9H), 1.32 (m, 130H), 1.25 (q, J = 6.8 Hz,
1
13
96H). C NMR (CDCl , 600 MHz) δ = 173.66, 173.39,
3
NMR (CDCl , 600 MHz) δ = 4.12 (m, J = 7.1, 2.9 Hz, 32H),
172.41, 172.03, 60.31, 60.19, 47.88 47.80, 45.98, 45.81, 45.74,
34.25, 34.12, 34.10, 33.14, 33.09, 32.90, 29.23, 28.92, 28.50,
27.79, 27.51, 27.19, 27.07, 26.94, 26.87, 26.59, 26.43, 25.26,
25.14, 24.73, 24.67, 24.65, 14.29. ESI-MS (m/z) for
3
3
6
6
.28 (q, J = 7.5 Hz, 30H), 3.20 (q, 32H), 2.83 (s, 3H), 2.29 (m,
2H), 1.65 (m, 62H), 1.54 (m, 62H), 1.44 (s, 9H), 1.31 (m,
1
3
2H), 1.26 (q, J = 6.9 Hz, 48H). C NMR (CDCl , 600 MHz)
3
5
+
δ = 173.67, 173.43, 172.42, 172.08, 60.32, 60.21, 47.86, 47.82,
C462H828N O (exact mass: 8438.08): [M+5H] calc.
34 99
6+
4
2
2
5.82, 45.76, 34.25, 34.13, 34.10, 33.09, 32.90, 29.21, 28.92,
7.84, 27.51, 27.07, 26.88, 26.59, 26.44, 25.27, 25.15, 24.73,
4.68, 24.65, 14.28, 14.27. ESI-MS (m/z) for C224H402N O
1688.62, obs. 1689.85; [M+6H] calc. 1407.35, obs. 1408.37.
5-((tert-Butoxycarbonyl)amino)pentyl 4-Methylben-
zenesulfulfonate (15). tert-Butyl (5-hydroxypentyl)-
carbamate (28.6 g, 0.14 mmol) was dissolved in DCM (200
mL), and TEA (29 mL, 0.21 mol) at 0 °C, then TsCl (32.2 g,
0.169 mol) and a crystal DMAP were added to the solution.
After overnight stirring at RT, the reaction mixture was washed
by brine, briefly dried over Na SO , and concentrated under
16
3+
49
+
(exact mass 4100.95): [M+H] calc. 4101.95; [M+3H] calc.
4
+
1367.99, obs. 1368.40; [M+4H] calc. 1026.24, obs. 1026.89.
4
Generation 4 (G ) Dendron (14). Extension of Linker.
Compound 13 (101 mg, 0.025 mmol) was treated with neat
TFA (1 mL) for 10 min at room temperature to remove the
BOC protecting group. TFA was then removed by vacuum, and
the residual dendron was lyophilized overnight. The dendron
was resuspended in 1 mL anhydrous DMF to which DIEA (66
uL, 49 mg, 0.38 mmol, 15 equiv) was added. Compound 9 (12
mg, 0.050 mmol, 2 equiv) was premixed with HATU (16 mg,
2
4
reduced pressure. Silica gel chromatography using hexane/ethyl
acetate gradient from 9/1 to 3/1 provided the desired tosylate
1
40 g (80%). H NMR (500 MHz, CDCl ): δ = 7.78 (d, J = 8.2
3
Hz, 2H), 7.35 (d, J = 8.2 Hz, 2H), 4.56 (bs, 1H), 4.03 (t, J = 6.4
Hz, 2H), 3.05 (m, 2H), 2.45 (s, 3H), 1.65 (m, 2H), 1.40 (s,
9H), 1.36 (m, 2H), 1.32 (m, 2H). 13C NMR (125 MHz,
0.043 mmol, 1.7 equiv) in 500 uL of DMF and 44 uL DIEA (32
mg, 0.25 mmol, 10 equiv) was then added to the dendron and
stirred at room temperature for 20 min. DCM (10 mL) was
then added, and the reaction solution was washed with water
CDCl ): δ = 155.96, 144.76, 133.07, 129.86, 127.87, 79.10,
3
70.37, 40.20, 29.41, 28.41, 22.65, 21.64. ESI-MS (m/z) for
+
C H NO S (exact mass: 357.46): [M+H] calc. 358, obs. 358.
17
27
5
(
1×), 0.1 N HCl (3×), and water (1×). The DCM layer was
tert-Butyl (5-Bromoxypentyl)carbamate (16). The
tosylate 15 (39 g, 0.109 mmol) was dissolved in acetonitrile
evaporated, then left to dry overnight under high vacuum,
resulting in an oil. To further extend the linker one more
round, the procedure above was reiterated in its entirety where
the dendron was treated with neat TFA (∼1 mL) for 10 min at
room temperature. TFA was removed in vacuo, and the
dendron was dissolved in water:acetonitrile (1:1), frozen, and
(100 mL), then NaBr (45 g, 0.437 mmol) and Bu NBr (1.4 g,
4
0.0043 mmol) were added. The resulting mixture was heated to
reflux for an overnight period. Solvents were then removed
under vacuum, and the residue was dissolved in DCM (400
mL) and washed by brine. Drying over Na SO , filtration,
2
4
2
092
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
concentration, and separation by silica gel chromatography with
hexane/ethyl acetate gradient from 10/1 to 6/1 provided the
desired primary bromide 24 g (83%). H NMR (500 MHz,
and the organic layer was washed w/0.1 N HCl (3×). The
organic layer was collected and subjected to low then high
vacuum to remove volatiles and obtain a tan oil.
1
1
CDCl ): δ = 4.53 (bs, 1H), 3.41 (t, J = 6.7 Hz, 2H), 3.13 (m,
Coupling of αSGalNAc 19 to G Dendron. The oil was
3
13
2
(
H), 1.88 (m, 2H),1.59−1.47 (m, 4H), 1.44 (s, 9H). C NMR
125 MHz, CDCl ): δ = 155.97, 78.92, 40.24, 33.60, 32.29,
diluted in anhydrous DMF (1 mL) and treated with DIEA (210
uL, 155 mg, 1.2 mmol, 24 equiv). A solution of compound 10
(180 mg, 0.4 mmol, 8 equiv) in DMF (1 mL) was added,
followed by a solution of HATU (114 mg, 0.3 mmol, 6 equiv)
in DMF (1 mL). The resulting mixture was stirred for 30 min,
after which volatiles were partially removed under reduced
pressure, and the residue transferred to a separatory funnel.
DCM was added, and the DCM layer was washed with water
(1×) followed by 0.1 N HCl (3×), then water (1×). The DCM
layer was then collected, evaporated under reduced pressure
resulting in a brown oil. Coupling of D-biotin: The oil was treated
with neat TFA and stirred for 10 min, and the TFA was
removed in vacuo. The oil was dissolved in 5 mL of
water:acetonitrile (1:1), frozen, and lyophilized overnight.
The oil was suspended in DMF and a solution of preactivated
D-biotin (25 mg, 0.1 mmol, 2 equiv) was added and stirred for
30 min (see below). The solution was transferred to a
separatory funnel, and DCM was added, and washed with water
(4×). The DCM layer was evaporated resulting in a tan solid
(132 mg). 40 mg of the crude tan solid was dissolved in a 1:1
solution of methanol:0.2 N NaOH for 30 min, then neutralized
with 1.0 N HCl. The solution was evacuated to remove residual
methanol, and the remaining solution was purified by reversed-
3
2
9.20, 28.38, 25.29. ESI-MS (m/z) for C H BrNO (exact
10 20 2
+
mass: 265.07): [M+H] calc. 265, obs. 265.
-((tert-Butoxycarbonyl)amino)pentyl-2-acetamido-2-
deoxy-1-thio-3,4,6-tri-O-acetyl-α-D-galactopyranoside
18). To a suspension of 2-acetamido-2-deoxy-1,3,4,6-tetra-O-
5
(
acetyl-β-D-galactopyranose (50 g, 0.128 mol) in toluene (500
mL) was added Lawesson’s reagent (44.2 g, 0.109 mol, 0.85
equiv), and the resulting mixture was heated up to reflux for 2.5
h. LCMS indicated the reaction to be complete. After cooling
down to ambient temperature the crude mixture was directly
loaded onto a silica gel (500 g) column and eluted with ethyl
acetate. Collection and concentration of appropriate fractions
gave a brownish foam (17) which was dissolved in methanol
(
(
400 mL), and then 36 mL of a mixture of TFA and water
1:1) was added, and stirring was continued for 1.5 h at room
temperature. Solvents were then removed and 400 mL of DMF
was added followed by DIEA (35 mL) and 16 (40 g, 0.15 mol).
The reaction was stirred at room temperature overnight. The
volatiles were removed (DIEA and DMF) under low, then high
vacuum, and the residue was dissolved in ethyl acetate (700
mL), washed with brine (3×), and dried briefly over Na SO .
2
4
After filtration over Celite, the filtrate was concentrated and
purified by column chromatography using a hexane/ethyl
phase HPLC. 25 mg of pure compound 1 was recovered as a
1
white powder. H NMR (D O, 600 MHz) δ = 5.42 (d, J = 5.52,
2
acetate gradient − from 4/1 to 1/1 as the eluant to provide
4H), 4.54 (m, 1H), 4.35 (m, 1H), 4.26 (dd, J = 5.9, 5.5 Hz,
4H), 4.18 (t, J = 6.1 Hz, 4H), 3.91 (d, J = 3.1 Hz, 4H), 3.75
(dd, J = 8.2, 3.2 Hz, 4H), 3.69 (d, J = 6.1 Hz, 8H), 3.25 (m,
15H), 3.09 (t, 8H), 2.98−2.83 (m, 4H), 2.70 (d, J = 13.0 Hz,
1H), 2.58 (m, 4H), 2.50 (m, 4H), 2.32 (m, 8H), 2.16 (q, J = 7.3
1
5
3.9 g (89%) of the desired BOC-protected amine. H NMR
(
500 MHz, CDCl ) δ = 5.67 (d, J = 8.8 Hz, 1H), 5.49 (d, J =
3
5
1
1
.0 Hz, 1H), 5.38 (d, J = 2.5 Hz, 1H), 5.05 (dd, J = 2.5 Hz, J =
1.7 Hz, 1H), 4.77 (m, 1H), 4.61 (bs, 1H), 4.55 (t, J = 6.5 Hz,
H), 4.13 (dd, J = 6.5 Hz, J = 11.4 Hz, 1H), 4.10 (dd, J = 6.5
13
Hz, 8H), 1.96 (s, 12H), 1.69−1.22 (m, 72H). C NMR (D O,
2
Hz, J = 11.4 Hz, 1H), 3.11 (m, 2H), 2.60 (m, 2H), 2.16, 2.05,
600 MHz, 298K) δ = 176.38, 176.19, 174.38, 71.59, 68.44,
67.70, 62.14, 61.08, 60.26, 55.51, 50.21, 48.34, 45.84, 39.81,
39.11, 35.70, 35.66, 32.52, 30.19, 28.58, 28.03, 28.00, 26.68,
25.97, 25.80, 25.65, 25.52, 25.34, 25.25, 25.22, 25.17, 21.97.
ESI-MS (m/z) for C105H188N O S (exact mass 2269.23): [M
2
1
1
6
2
.01, 1.98 (4 s, 12H), 1.64 (m, 2H), 1.49 (m, 2H), 1.44 (s, 9H),
.38 (m, 2H). 1 C NMR (125 MHz, CDCl ): δ = 171.01,
3
3
70.41, 170.29, 170.16, 155.98, 84.92, 79.11, 68.53, 67.34,
7.23, 61.83, 48.32, 40.38, 30.97, 29.62, 29.28, 28.43, 25.94,
3.35, 20.76, 20.72. ESI-MS (m/z) for C H N O S (exact
14
29 5
+2H]2+ calc. 1135.62, obs. 1136.36; [M+3H] calc. 757.42, obs.
3+
2
4
40
2
10
+
mass: 548.24): [M+H] calc. 549, obs. 549.
758.07.
5
-Amino-pentyl-2-Acetamido-2-deoxy-1-thio-3,4,6-
Preactivated D-Biotin. For a typical reaction, D-biotin (2
equiv relative to amine) was dissolved in DMF (5 mg of D-
biotin per 1 mL DMF) with DIEA (52 uL, 39 mg, 0.3 mmol, 6
equiv), and HATU (38 mg, 0.1 mmol, 1.8 equiv) in DMF was
added and allowed to mix for 10 min.
tri-O-acetyl-α-D-galactopyranoside Trifluoroacetic Acid
Salt (19). A solution of the BOC-protected compound above
700 mg, 1.28 mmol) in TFA/DCM (7 mL, 2:5 (v:v)) was
stirred at room temperature for 1 h. The solvents were
removed, and the residue was dried over vacuum for 2 h, and
directly used for next step without purification. H NMR (500
MHz, CDCl ): δ = 7.52 (bs, 2H), 6.26 (d, J = 8.0 Hz, 1H), 5.59
d, J = 5.3 Hz, 1H), 5.38 (d, J = 2.2 Hz, 1H), 5.04 (dd, J = 2.2
Hz, J = 11.6 Hz, 1H), 4.67 (m, 1H), 4.53 (t, J = 6.5 Hz, 1H),
.19 (dd, J = 6.5 Hz, J = 11.6 Hz, 1H), 4.05 (dd, J = 6.5 Hz, J =
1.6 Hz, 1H), 3.10 (m, 2H), 2.69, 2.61 (2 m, 2H), 2.18, 2.06,
.05, 2.03 (4 s, 12H), 1.72 (m, 2H), 1.68 (m, 2H), 1.51 (m,
H). ESI-MS (m/z) for C H N O S (exact mass: 448.19): [M
(
2
Clearing Agent 2. Hydrolysis of Ethyl Esters of G
1
Dendron. Compound 12 (44 mg, 0.02 mmol, 0.16 mmol of
ethyl ester) was dissolved in 2 mL ethanol and 2 mL of 2 N
NaOH. The hydrolysis was conducted in a 40 °C water bath for
2 h. The solution was transferred to a separatory funnel and
then neutralized with 1.0 N HCl. Ether was added and the
organic layer was washed with 0.1 N HCl (3×). The organic
layer was collected and subjected to reduced pressure to
remove solvent resulting in a tan oil.
3
(
4
1
2
2
+
1
9
32
2
8
+
2
H] calc. 449, obs. 449.
Coupling of αSGalNAc 19 to G Dendron. The tan oil
Clearing Agent 1. Hydrolysis of Ethyl Esters of G1
(assumed to be 0.02 mmol dendron, and 0.16 mmol of
carboxylic acid as from the previous step) was suspended in
DMF (1 mL) and DIEA (167 uL, 124 mg, 0.96 mmol, 48
equiv). A solution of compound 19 (175 mg, 0.32 mmol, 16
equiv) in DMF (1 mL) was added, followed by a solution of
HATU (92 mg, 0.24 mmol, 12 equiv) in DMF (1 mL). The
Dendron. Compound 11 (50 mg, 0.05 mmol, 0.2 mmol of
ethyl ester) was dissolved in 2 mL ethanol and 2 mL of 2.0 N
NaOH was added. The hydrolysis was conducted in a 40 °C
water bath for 2 h. The solution was transferred to a separatory
funnel and then neutralized with 1.0 N HCl. Ether was added
2
093
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
reaction was allowed to proceeded for 30 min, then subject to
high vacuum for 30 min to partially remove solvent. The
remaining volume was transferred to a separatory funnel, DCM
was added, and the DCM layer was washed with water (1×),
then with 0.1 N HCl (3×), and finally with water (1×). The
DCM layer was collected and evaporated resulting in a brown
oil.
transferred to a separatory funnel, DCM was added, and
washed with water, 4×. The DCM layer was evaporated
resulting in a tan solid (150 mg). Final deprotection and
purification: 20 mg of the crude tan solid was dissolved in 2 mL
of a 1:1 solution of methanol:0.2 N NaOH for 30 min, then
neutralized with 1.0 N HCl. The solution was evacuated to
remove residual methanol, and the remaining solution was
Coupling of D-Biotin. The brown oil above was treated with
neat TFA and stirred for 10 min, and TFA was removed in
vacuo for 60 min. The oil was resuspended in DMF and a
solution of preactivated D-biotin (10 mg, 0.04 mmol, 2 equiv)
was added and stirred for 30 min. The solution was transferred
to a separatory funnel, DCM was added, and washed with water
purified by reversed-phase HPLC. 10 mg of pure compound 3
1
was isolated as a white powder. H NMR (D O, 600 MHz) δ =
2
5.43 (d, J = 5.4 Hz, 16H), 4.54 (m, 1H), 4.35 (m, 1H), 4.26
(dd, J = 11.3, 5.5 Hz, 16H), 4.16 (t, J = 6.0 Hz, 16H), 3.92 (d, J
= 3.2 Hz, 16H), 3.75 (dd, J = 3.1, 11.3 Hz, 16H), 3.68 (d, J =
6.1 Hz, 32H), 3.24 (m, 63H), 3.08 (t, J = 6.7 Hz, 32H), 2.97−
2.83 (m, 4H), 2.69 (d, 1H), 2.53 (m, 32H), 2.31 (m, 32H),
(4×). The DCM layer was evaporated resulting in a tan solid
(102 mg). 20 mg of the crude tan solid was dissolved in a 1:1
1
3
1.96 (s, 48H), 1.70−1.23 (m, 288H). C NMR (D O, 600
2
solution of methanol:0.2 N NaOH for 30 min, then the
mixtrure was neutralized with 1.0 N HCl. The solution was
evacuated to remove residual methanol, and the remaining
solution was purified by reversed-phase HPLC. 12 mg of pure
MHz, 298 K) δ = 174.3, 83.8, 71.6, 68.4, 67.7, 61.1, 50.2, 39.2,
35.7, 30.2, 28.7, 28.2, 25.5, 25.3, 22.0. ESI-MS (m/z) for
C405H728N O S : [M+H] calc. 8565.11; [M+5H] calc.
1713.83, obs. ; [M+6H] calc. 1428.36, obs.; [M+7H] calc.
1224.45, obs.
+
5+
50
113 17
6+
7+
compound 2 was obtained as a white powder (73% recovery
1
Clearing Agent 4. Hydrolysis of Ethyl Esters of G4
Dendron. Compound 14 (19 mg, 2.25 umol,) was dissolved
in 0.5 mL ethanol and 0.5 mL of 1.0 N NaOH and left
overnight at room temperature. The solution was transferred to
a separatory funnel with 10 mL of water, then neutralized with
1.0 N HCl. A volumetric mixture composed of diethyl ether,
tBuOH, and DCM (6:3:1) was added and the organic layer was
washed with 0.1 N HCl (3×), then water (3×). The organic
layer was recovered and solvents were removed under vacuum,
resulting in a brown oil. For an alternative workup after
overnight hydrolysis, ethanol was partially removed from the
reaction under reduced pressure. The dendron solution was
neutralized with 1.0 N HCl, then acidified. This resulted in the
precipitation of the dendron. After centrifugation (3000 rpm), a
brown viscous pellet formed and was washed with water (3×).
Residual water was removed under vacuum resulting a brown
oil.
from HPLC purification). H NMR (D O, 600 MHz) δ = 5.42
2
(
pd, J = 5.4 Hz, 8H), 4.53 (m, 1H), 4.34 (m, 1H), 4.26 (dd, J =
1
8
1
4
1
1
6
2
1.3, 5.5 Hz, 8H), 4.16 (t, J = 5.9 Hz, 8H), 3.91 (d, J = 3.2 Hz,
H), 3.75 (dd, J = 11.3, 3.2 Hz, 16H), 3.68 (d, J = 6.1 Hz,
6H), 3.24 (m, 31H), 3.08 (t, J = 6.8 Hz, 16H), 2.97−2.82 (m,
H), 2.69 (d, J = 12.8, 1H), 2.57−2.50 (m, m, 16H), 2.31 (m,
6H), 2.15 (q, J = 7.4 Hz, 16H), 1.96 (s, 24H), 1.53−1.22 (m,
44H). 13C NMR (D O, 600 MHz) δ = 174.33, 83.77, 71.59,
2
8.44, 67.72, 61.07, 50.24, 45.83, 39.15, 35.69, 30.21, 28.66,
8.13, 26.79, 25.66, 25.41, 25.28, 22.01. ESI-MS (m/z) for
3
+
C205H368N O S (exact mass 4394.42): [M+3H] calc.
26
57 9
4
+
1465.8, obs. 1466.9; [M+4H] calc. 1099.6, obs. 1100.5; [M
+
+
5H]5 calc. 879.9, obs. 880.8.
Clearing Agent 3. Hydrolysis of Ethyl Esters of G
3
Dendron. Compound 13 (48 mg, 0.012 mmol) was dissolved
in 1 mL ethanol, and 1 mL of 1.0 N NaOH was added. The
hydrolysis was conducted overnight, at room temperature. The
solution was transferred to a separatory funnel and then
neutralized with 1.0 N HCl. A solution of diethyl ether, tBuOH,
and DCM (6:3:1; v:v:v) was added and the organic layer was
washed with 0.1 N HCl (3×), and water (3×). The organic
layer was transferred into a round bottomed flask and
evaporated under reduced pressure, then high vacuum, to
produce a brown oil.
Coupling of GalNAc Groups. The dendron was dissolved in
500 uL of DMF and DIEA (75 uL, 56 mg, 0.432 mmol, 192
equiv) was added followed by 19 (65 mg, 0.144 mmol, 64
equiv) in 500 uL DMF, and HATU (114 mg, 0.3 mmol, equiv)
also in 500 uL of DMF. The reaction was allowed to proceed
for 1 h, at which point solvent was partially removed under
reduced pressure for 30 min. DCM and ether (2:1) were added
and the solution was transferred to a separatory funnel and
decanted. The organic layer was then washed with 0.1 N HCl
(3×), then with water (3×), before evaporation under reduced
pressure.
Coupling with D-Biotin. The collected oil above was treated
with neat TFA (1 mL) for 10 min at room temperature to
remove the BOC protecting group. TFA was removed by
evacuation and the dendron was diluted in a solution of
water:acetonitrile (1:1), frozen, and lyophilized overnight. The
resulting oil was dissolved in DMF and a solution of
preactivated D-biotin (5 mg, 0.02 mmol) was added and
mixed for 1 h (see above). The solution was transferred to a
separatory funnel, DCM was added, and the organic layer
washed with water (4×). The DCM layer was evaporated,
resulting in a off-white solid. The material was dissolved in a
solution of methanol:0.2 N NaOH (1:1) for 30 min, monitored
by LCMS, then neutralized with 1 N HCl. The solution was
evacuated to remove residual methanol, and the remaining
solution was purified by reversed-phase HPLC. 13 mg of pure
Coupling of GalNAc Groups. The brown oil above was
suspended in 2 mL anhydrous DMF. DIEA (210 uL, 155 mg,
1.2 mmol, 96 equiv) was added followed by a solution of 19
(
(
215 mg, 0.384 mmol, 32 equiv) as a solution in DMF. HATU
114 mg, 0.3 mmol, 24 equiv) dissolved in DMF was then
added. The reaction was allowed to proceed for 30 min, at
which point the solvent was partially removed under reduced
pressure, and the mixture was transferred to a separatory
funnel. A solution of diethyl ether, tBuOH, and DCM (3:3:1;
v:v:v) was added and the organic layer was washed with 0.1 N
HCl (3×), and with water (2×). The organic layer was
collected and subjected to reduced pressure to remove volatiles.
150 mg of a tan oil was obtained.
Coupling with D-Biotin. The oil was treated with neat TFA
for 10 min at room temperature, to remove the BOC protecting
group. TFA was then removed by under high vacuum. The
resulting oil was diluted with DMF and a solution of
preactivated D-biotin (6 mg, 0.024 mmol, 2 equiv) was added
and stirred for 30 min (see above). The solution was
2
094
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
a
Scheme 1. Synthesis of Amino-Diethyl-Hexanoate
a
(
a) Δ, KI, DIEA, overnight (70−80%); (b) PdC, 110 °C, 5 h (60%).
compound 4 was recovered as a white powder (33% yield
to be ≥98.5% using instant thin-layer chromatography. The
functional integrity of the tracer-antibody was evaluated using a
binding assay with a TAG-72(+) cell-line (LS174T) according
1
starting from 14). H NMR (D O, 600 MHz) δ = 5.44 (pd, J =
2
5
5
.16 Hz, 32H), 4.55 (m, 1H), 4.36 (m, 1H), 4.28 (dd, J = 5.9,
.4 Hz, 32H), 4.17 (t, 32H), 3.93 (d, J = 2.9 Hz, 32H), 3.56
20
to the method by Lindmo et al. (59.3 ± 12.9%; n = 3). No
(
dd, J = 11.3, 2.7 Hz, 32H), 3.69 (d, J = 5.8 Hz, 64H), 3.25 (m,
efforts were made to determine biotin-binding capacity.
1
31
1
2
1
1
2
35H), 3.09 (t, 64H), 2.99−2.84 (m, 10H), 2.58 (m, 32H),
Blood Clearance of
I-scFv-CC49-SAv. All animal
.51 (m, 32H), 2.31 (m, 68H), 2.17 (m, 64H), 1.97 (s, 96H),
experiments were approved by the Institutional Animal Care
and Use Committee of Memorial Sloan-Kettering Cancer
Center, and institutional guidelines for the proper and humane
use of animals in research were followed. All mice were fed a
biotin-free diet for one week prior to study. The antibody was
prepared for injection by mixing the radioactive antibody with
additional nonlabeled antibody to achieve 100 μg per dose. All
blood sampling and injections were performed via the tail vein.
Mice (n ≥ 3 per group) were intravenously administered 15−
30 uCi/100 μg of 1 I-scFv-CC49-SAv (0.57 nmol). After 24 h,
a blood sample was collected to determine activity concen-
tration of the tracer (“baseline”, t = 0)) and within 1−2 h the
mice were injected with 50 ug of 1 (22 nmol) or 100 μg of 2, 3,
and 4 (5.7−23 nmol) formulated in 100 μL of saline, or vehicle.
Serial blood samples at various times 5−120 min postinjection
1
3
.55−1.24 (m, 584H). C NMR (D O, 600 MHz) δ = 174.2,
2
30.8, 117.4, 83.8, 71.6, 68.4, 67.7, 61.0, 50.3, 39.2, 35.7, 30.2,
8.8, 28.3 25.5 25.4, 22.1. ESI-MS (m/z) for
9+
C819H1474N100O227S33 (exact mass 17399.77): [M+9H] calc.
1
0+
1
934.3, obs. 1934.3; [M+10H] calc. 1741.0, obs. 1741.1.
General Procedure for Solid-Phase Synthesis of
Dendron Core. For a typical synthesis of 20 on solid support,
00 mg of Wang resin (0.5 mmol/g) was allowed to swell in
1
31
DCM (2 mL) for 10 min. The DCM was removed and carbon
tetrabromide (83 mg, 0.25 mmol, 5 equiv) and triphenylphos-
phine (66 mg, 0.25, 5 equiv) in DCM (2 mL) were added to
the resin, and allowed to shake for 3 h at room temperature.
The resin was washed 3× with DCM, and 3× with DMF.
Compound 8 (73 mg, 0.5 mmol, 10 equiv) in DMF (4 mL)
with DIEA (350 uL, 260 mg, 2 mmol, 40 equiv) was prepared
and added to the resin, and the resulting mixture shaken
overnight. The liquids were removed by suction filtration and
the resin was washed 6× with DMF. Coupling step: Compound
(
p.i.) of clearing agent (t = 5−120 min) were collected in
preweighed microcapillary tubes and counted in the gamma
counter for quantification of radioactivity. Differences between
cohorts were statistically analyzed with Student’s t test for
paired data. Two-sided significance levels were calculated and P
7
(45 mg, 0.15 mmol, 3 equiv) in DMF (2 mL) was mixed with
DIEA (52 uL, 39 mg, 0.3 mmol, 6 equiv) and added to the
resin. A solution of HATU (57 mg, 0.15 mmol, 3 equiv) in
DMF (1 mL) was then added and the reaction shaken for 60
min. The liquids were removed, the resin was washed 6× with
DMF.
<
0.05 was considered statistically significant.
Biodistribution Analysis of 131I-scFv-CC49-SAv with
Clearing Agent Step. For all groups, animals were sacrificed
via CO asphyxiation 22−24 h post injection of the clearing
2
agent (t = 22−24 h). Activities in blood, heart, liver, spleen, and
kidney were determined. Tissues and organs were excised,
rinsed with water and allowed to air-dry, weighed, and then
counted in an automatic well counter (Perkin-Elmer Wallac
Wizard 3″ Automatic Gamma Counter). The radioactive
uptake was determined by decay-correcting the net count rate
to the time of injection and expressing as the percentage of the
injected dose per gram of the tissue (%ID/g).
Deprotection Step. Sodium silanolate (10 equiv) was
prepared in DMF/DCM (1:1, 5 mL). Due to incomplete
solubility, the mixture was either filtered or centrifuged, and
filtrate or supernatant was added to the resin, then allowed to
shake for 2 h. The resin was washed 3× with DCM, 3× with
DMF, 3× with 5% acetic acid in DMF (5 min/wash), and
finally 6× with DMF. Coupling step: Similar to above, 8 (90 mg,
0.3 mmol, 6 equiv) in DMF (2 mL) with DIEA (105 uL, 78
mg, 0.6 mmol, 12 equiv) and added to the resin. HATU (114
mg, 0.3 mmol, 6 equiv) in DMF (1 mL) was added last, and the
mixture was shaken for 60 min. After suction filtration, the resin
was washed 6× with DMF. Cleavage from resin: Resin was
treated with a 10% solution of chloroethylchloroformate in
DCM for 60 min. The cleaved resin was then washed 3× with
DCM and filtered. Methanol is added to the filtrate, and then
subject to vacuum and residual products were analyzed by
LCMS.
RESULTS
■
Glycodendrons 1−4 were constructed in a stepwise manner
where dendrons 11−14 were synthesized first, followed by the
addition of the biotin to the dendron core, and then attachment
of the prefunctionalized α-thio-N-acetylgalactosamine units 19.
Orthogonal protecting groups (ethyl esters and BOC) were
utilized to site-specifically incorporate the desired functional
groups, i.e., biotin and carbohydrates, with a final single
deprotection step leading to the desired compounds 1−4.
Radioiodination of scFv-CC49-SAv. The scFv-CC49-SAv
1
31
protein was labeled with I using the Iodogen method and
Synthesis of AB Monomer 7 and Dendron Core. For
2
previously described methods to a specific activity of 3.0 ± 2.1
the preparation of AB type monomer 7, a number of synthetic
routes were initially considered. However, the route depicted in
2
1
6,19
μCi/μg (n = 3).
The radiochemical purity was determined
2
095
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
0
a
Scheme 2. Synthesis of Generation 0 (G ) Dendron
a
(
a) Conc. HCl, reflux, 48 h (99%); (b) CH Cl , di-tert-butylcarbonate, 60 min (95%); (c) DMF, 7, HATU, TEA, 20 min (Crude 95%).
2 2
0
1
2
3
4
Figure 2. HPLC Analysis of Dendrons 10 (G ), 11 (G ), 12 (G ), 13 (G ), and 14 (G ). Analysis conditions used: gradient of 45−95% acetonitrile
in water, 0.05% TFA) in 25 min; reversed phase C4 column (300 Å, 4.6 × 50 mm).
(
Scheme 1 was selected due to its high reproducibility, good
product yields, and low cost of reagents. Monomer 7 was
obtained by reacting 4-methoxybenzylamine (PMBA) with an
excess of ethyl 6-bromo-hexanoate in the presence of base
provide an orthogonal protection strategy against the base
sensitive ethyl esters in monomer 7. The product 9 was
obtained in excellent yield and purity, which was confirmed by
1
LCMS and H NMR.
(
(
DIEA) and NaI, while stirred overnight at room temperature
Scheme 1). After an aqueous workup and silica gel
Synthesis of Dendrons 11−14. Dendrons 11−14 were
constructed where the AB type monomer 7 was reiteratively
2
purification, the tertiary amine 6 was isolated in high yield.
Removal of the PMB group was achieved through catalytic
hydrogenolysis resulting in the desired product 7. Purity and
introduced through a series of coupling-deprotection steps.
This route provided the flexibility to obtain dendrons of
successive generations, in contrast to more convergent methods
1
21
identity of the monomer were assessed by LCMS and H
that have been previously described. In order to evaluate the
NMR.
molecular identity, purity, and structural integrity of the
It should be emphasized that the AB type monomer 7 is the
dendron products, while monitoring for impurities such as
truncated molecules, LCMS and H NMR were extensively
2
1
most important component in the synthesis of these dendrons.
Extreme care was taken to ensure the complete absence of
impurities in 7, especially amine byproducts which have proven
to adversely complicate purification and isolation of homoge-
neous dendron products later in the synthesis. LCMS analysis
was essential for identifying impurities not readily observed by
used during the course of the syntheses. High performance
liquid chromatography (HPLC) with three modes of detection
− mass, evaporative light scattering (ELS), and standard UV
absorption − was utilized for determining product identity and
1
purity. Dendron products 11−14 were also analyzed by H
1
H NMR. For example, in the early synthesis of 7, low levels,
NMR, which was used to assess product purity, as well provide
supporting structural information. For example, the integration
values for two specific resonance − a singlet (δ ≈ 2.83) from
the N-methyl group(s) of the dendron core and a multiplet (δ
≈ 4.13) from the ethyl ester (CH CH O−) end groups − were
typically less than 1% of ethyl-6-aminohexanoate were observed
1
by LCMS but difficult to detect by H NMR, reacted much
faster than higher order dendrons, and gave inseparable
contaminations in later steps (data not shown).
3
2
The dendron core 9, was obtained in two steps (Scheme 2)
where the N-methylcaprolactam was hydrolyzed in refluxing
concentrated HCl to quantitatively provide caproate 8. The
resulting amine was then protected as the respective tert-
butoxycarbamate 9. The acid-labile BOC group was utilized to
used to determine the relative number of N-methyl to ethyl
ester protons (Supporting Information). This served as an
additional means to verify the dendron products.
0
The generation zero (G ) dendron 10 was obtained by
coupling compounds 9 and 7 with HATU in the presence of
2
096
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
a
Scheme 3. Synthesis of Dendron Scaffold 11−13
a
(
a) Ethyl Ester Deprotection: 1 N NaOH, 40 °C, 1−2 h; (b) Coupling: 7, HATU, DIEA in DMF, 20−30 min (Crude yields are shown).
a
Scheme 4. Synthesis of Dendron Scaffold 14
a
(
a) BOC Deprotection: TFA, 10 min; (b) Coupling: 9, HATU, TEA in DMF, 20 min; (c) Ethyl Ester Deprotection: 1 N NaOH, overnight; (d)
Coupling: 7, HATU, TEA in DMF, 20 min.
DIEA (Scheme 2). This reaction was complete within 20 min,
and was followed by an acidic workup to remove unreacted 7 as
well as HATU byproducts. Analysis of the crude product by
LCMS revealed a single peak (Figure 2), with a molecular ion
at 529.36 m/z (calc. C H N O : 528.3775), indicative of a
LCMS analysis of 11 showed a single peak (Figure 2) with
+
molecular ions of [M+H] m/z = 1039.59 (calc. C H N O :
5
6
103
4
13
1
1039.7443) corresponding to the G tetraethyl ester product.
The ratio of N-methyl to ethyl ester (CH CH O−) protons
3
2
1
was 3:8 as determined by H NMR (Figure S5).
2
8
53
2
7
1
2
product of high purity. The H NMR spectrum of the product
was consistent with the desired compound 10. The resonances
for the N-methyl singlet and the CH CH O− multiplet for the
To obtain the G dendron 12, the deprotection−coupling
procedure above was repeated with 11 (Scheme 3). Ester
hydrolysis of 11 yielded four free carboxylic acids, which were
3
2
2
methylene in the ethyl esters were determined to be 3 and 4
respectively, as expected (Supporting Information, Figure S2).
Continuing with the deprotection-coupling cycle, dendrons
coupled with the monomer 7 resulting in the G octaethyl ester
dendron 12. LCMS analysis for the final product 12 showed a
single peak (Figure 2) with multiply charged ion peaks [M
+2H]2+ m/z = 1031.04 and [M+3H] m/z = 687.77 (calc.
3+
1
1−13 were synthesized as shown in Scheme 3. Starting with
1
compound 10, dendron 11 was obtained in two steps. First,
ester 10 was saponified with 1.0 N NaOH, followed by an
acidic workup, yielding two free carboxylic acids. Two
equivalents of monomer 7 were then coupled to the two
exposed carboxylic acids using HATU/DIEA in anhydrous
DMF. The reaction was complete within 30 min as determined
by LCMS. An acidic workup was used to remove any unreacted
monomer 7 as well as HATU related contaminants from 11.
C112H202N O : 2059.4781). H NMR integration data further
8
25
confirmed the presence of the desired product ratio of 3:15.4
(Figure S8).
3
Similarly, the G dendron was obtained by hydrolysis of the
ethyl esters of 12 resulting in eight carboxylic acids; and
HATU/DIEA mediated monomer addition yielded the
hexadeca-ethyl ester 13 (Scheme 3). After workup the final
product exhibited a single peak by HPLC (Figure 2), with the
2
097
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
a
Scheme 5. Synthesis of tert-Butyl-(5-Bromoxypentyl)carbamate 16
a
(
a) TEA, TsCl, DMAP (cat.), 0 °C; overnight at RT (80%); (b) ACN, NaBr, Bu NBr, reflux overnight (83%).
4
Scheme 6. Synthesis of N-tert-Butyloxycarbonyl-S-(2-Acetamido-2-deoxy-3,4,6-tri-O-acetyl-α-D-galactopyranosyl)-5-
a
Thiopentylamine Trifluoroacetic Acid Salt 19
a
(
a) Toluene, Lawesson’s reagent, reflux 2.5 h; (b) DMF, DIEA, 16 (two steps 89%); (c) TFA/DCM (2:5, v/v), 2 h.
expected masses (calc. C224H402N O : 4100.9457, obs. [M
+
workup, further helping to minimize impurities in the final
dendron products.
16
4+
49
3H]3 m/z = 1368.57, [M+4H] m/z = 1026.77). H NMR
+
1
spectrum was also consistent with the desired dendron−proton
ratios of 3 to 32.3 from the N-methyl and ethyl ester groups,
respectively, were observed (Figure S11).
Synthesis of Amino-alkyl Peracetylated α-Thio-Gal-
NAc 19. The prefunctionalized α-thio-N-acetylgalactosamine
19 was synthesized as shown in Schemes 5 and 6. N-BOC-
aminopentanol was tosylated (15), then converted into 16
(Scheme 5). To obtain the desired α-GalNAc mercaptan 17 (2-
acetomido-2-deoxy-3,4,6-tri-O-acetyl-1-thio-α-D-galactopyra-
nose), commercially available peracetylated β-galactosamine
was treated with Lawesson’s reagent, 2,4-bis(4-methoxyphen-
yl)-1,3-dithia-2,4-diphosphetane-2,4-disulfide, resulting in the
α-thioglycoside (Scheme 6). 16 and 17 were then combined
through a displacement reaction in the presence of base to yield
the BOC protected peracetylated α-sugar 18. Finally acid was
used to remove the BOC group, and the aminopentyl conjugate
4
In preparation of the G dotriaconta-ethyl ester dendron, the
core of 13 was further extended by two N-methyl-6-hexanoic
acid units (Scheme 4). Since three-dimensional structures of
these dendrons are not available, there were general concerns
4
that, for the G glycodendron, the central amine and the biotin
may be cloaked by 32 densely associated carbohydrate groups
thereby precluding binding to circulating/off-target antibody−
streptavidin fusion protein. To circumvent this possibility, we
extended the linker attachment to the biotin by 14 atoms.
Hence, 13 was elongated with two units of N-methyl-6-
hexanoic acid via the corresponding BOC analog 9. N-Methyl-
24
19 was obtained as a TFA salt. Purity and identity of the
intermediates and final thioglycoside were assessed by LCMS
6-hexanoic acid groups were sequentially added, i.e., after
1
and H NMR.
removal of the central BOC group with TFA, 9 was coupled
onto the exposed N-methyl amino terminus; and these steps
were repeated for the addition of another molecule of 9.
Following final BOC deprotection and the incorporation of D-
biotin, ethyl esters hydrolyses resulted in sixteen carboxylic acid
groups which were then extended by the addition of 7 to afford
Synthesis of Clearing Agents 1−4. To obtain clearing
agents 1−4, the end groups for dendrons 11−14 were first
modified with the carbohydrate moieties, followed by the
addition of D-biotin (Scheme 7). Starting with dendrons 11−14
the ethyl esters were hydrolyzed, exposing 4, 8, 16, or 32
carboxylic acid groups, respectively. This was followed by
HATU/DIEA mediated coupling of tetraacetyl-α-galactosami-
no-5-thiopentylamine groups 19 to the free acids (Scheme 7).
The lone BOC group on the glycodendrons was then removed
by brief treatment with TFA, and the resulting amine was
coupled with D-biotin. Finally, global deprotection of all the O-
acetyl groups with 0.2 N NaOH afforded the desired clearing
agents 1, 2, 3, and 4 as determined by LCMS (1, calc.
4
the G dendron 14. As shown in Figure 2, the final product
exhibited a single peak by LCMS (calc. C462H828N O :
34
99
5
+
6+
8
1
438.0802, obs. [M+5H] m/z = 1689.53, [M+6H] m/z =
407.99, [M+7H] m/z = 1207.06). The H NMR spectrum
7
+
1
revealed three well resolved N-methyl resonances (3H, δ =
.83; 3H, δ = 2.90; 3H, δ = 2.97) along with the CH CH O−
2
3
2
multiplet (64H, δ = 4.12) with integrations of 3.0, 3.0, 3.2, and
6
3.9, respectively (Figure S14).
2+
C105H184N O S : 2273.1631, obs. [M+2H] m/z = 1135.96,
10
3+
33 5
In contrast to prior work which used BOP or other
[
M+3H] m/z = 757.69; 2, calc. C205H368N O S : 4394.4183,
26 57 9
7,21
carboxylic acid activators,
we utilized 2-(7-aza-1H-benzo-
3+
4+
obs. [M+3H] m/z = 1466.42, [M+4H] m/z = 1100.08 [M
5H] m/z = 880.07; 3, calc. C405H728N O113S17: 8644.8009,
triazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate
HATU) for these coupling reactions. HATU has been
shown to be highly effective in the condensation of carboxylic
5+
+
50
6+
5+
(
obs. [M+5H] m/z = 1731.08, [M+6H] m/z = 1442.69, [M
+
1
7H]7+ m/z = 1236.61; 4, calc. C819H1474N100
O
227
10+
S :
33
2
2,23
9+
acids and secondary amines.
Indeed, for the reactions
7399.7654, obs. [M+9H] m/z = 1935.47, [M+10H] m/
described in this work, HATU proved to be efficient in forming
z = 1741.91). Crude products 1, 2, 3, and 4 were typically
obtained in >90% purity and required a single purification step
using reversed-phase HPLC of the final product (Figure 3).
Solid-Phase Synthesis of Dendrons. Given the sequen-
tial and repetitive nature of the dendron synthesis, we also
investigated the use of solid-phase methods as a possible
amide bonds between the AB monomer 7 and the free
2
carboxylic acid end groups. All reactions typically required only
a slight excess of monomer (1.2−3.0 equiv per free carboxylic
acid), and were complete within 30 min. Additionally, the
HATU byproducts could be removed with a standard acidic
2
098
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
2
7
Scheme 7. Synthesis of Clearing Agents 1−4
was used to cleave dendron products from the resin for the
purposes of efficacy analyses. Following this approach,
compounds 20, 21, and 22 were characterized by LCMS.
However, attempts to grow the dendron beyond generation 3
were not successful, and we could not isolate product 23 by this
method. Interestingly, upon coupling of 7 to the second
generation dendron-resin, the resin became visually distorted
and lost the capacity to swell in solvent. It is worth noting,
however, that some preliminary work suggests that the
synthesis of higher generation dendrons may be possible on
larger PEG-based resins and/or rigid macroporous nonswelling
28
beads (Synbeads).
Evaluation of Blood Clearance of 131I-scFv-CC49-SAv.
Experiments were conducted in vivo to evaluate the kinetics of
blood clearance of the scFv-CC49-SAv model protein following
a separate injection (24 h later) of an excess dose of clearing
agent to antibody (10:1 to ∼40:1 molar ratio). Additionally, to
provide further insights into the biology of these compounds
and for comparison, we also evaluated a high-molecular-weight
(
500 kDa) biotinylated-dextran. The scFv-CC49-SAv was
131
radioiodinated with I using standard Iodogen procedures to
20
directly measure levels of the fusion protein. Time activity
data for the 1 I-tracer in blood were calculated as %ID/g
relative to baseline collected prior to injection of compounds (t
= 0 min). As shown in Figure 5, clearing agents 1 to 4 produced
an almost immediate effect on blood activity, with radioactivity
levels dropping as much as 40% from baseline within 30 min
p.i. (post injection). During the same time interval, the
clearance of activity in animals given vehicle or biotinylated
dextran was ≥15% to a concentration of ≥85% of baseline. At
120 min all cohorts given either glycodendron or biotinylated-
dextran had blood activities significantly less than vehicle (P <
0.05 for biotinylated-dextran and 1, and P < 0.005 for 2, 3, and
4).
31
expedient route to access limited quantities of the clearing
agents in order to accelerate development (Supporting
Information, Figure S28). Hydroxymethylphenyl resins of
various loading capacities (0.2−1.0 mmol/g) were converted
into the corresponding bromomethyl form upon treatment
with excess of carbon tetrabromide. The amine of the N-
methyl-6-aminohexanoic acid (Figure S28) was then reacted
with the bromo-functionalized resin in the presence of DIEA,
and dendron was grown from resulting carboxylic acid using a
repetitive coupling/deprotection cycles which were compatible
with solid phase synthesis. Hence, on-resin hydrolysis of the
ethyl esters was achieved by the use of sodium trimethylsila-
Biodistribution Analysis. At 24 h p.i. of compounds or
controls, biodistribution analysis was conducted to determine
2
5,26
nolate in DCM/DMF,
and 1-chloroethyl chloroformate
Figure 3. HPLC analysis of clearing agents 1−4. Analysis conditions used: gradient of 20−40% acetonitrile (in water, 0.05% TFA) in 20 min;
reversed phase C4 column (300 Å, 4.6 × 50 mm).
2
099
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
Figure 4. Blood clearance of 131I-scFv-CC49-SAv. Time activity curve for 131I-activity in blood determined by serial blood collection up to 120 min
following injection of 1 I-scFv-CC49-SAv (t = −22 h to −24 h) and compounds 1−4 (t = 0) in biotin-starved mice. Cohorts given vehicle or 100 μg
of biotinylated-dextran were included as controls. Data is shown as mean ± SEM, *P < 0.05, **P < 0.005. Note: scale on y-axis is 50−100%.
31
Figure 5. Glycodendrons enhance liver targeting/uptake of 131I-scFv-CC49-SAv. Tissue-to-blood ratios determined from biodistribution obtained 24
h p.i. of vehicle (Veh), glycodendrons (1, 2, 3, 4), or biotin-dextran (Dex) (Figure S29). Data is shown as mean ± SEM.
activities of 131I-tracer in the blood, kidneys, liver, and spleen
Figure S29). Animals treated with clearing agents 1 and 4
consistently showed lower levels of I in the blood, spleen,
Overall, the in vivo data suggests enhanced blood clearance of
the 1 I-scFv-CC49-SAv when a glycodendron compound or
the 500 kDa biotin−dextran conjugate is administered.
Additionally, the biodistribution results reveal increased liver
31
(
1
31
and kidneys, but notably higher levels (3.14 ± 0.24 and 3.22 ±
131
0.68%ID/g) in the liver, relative to saline (2.2 ± 0.31%ID/g).
retention of the I-tracer, suggesting that all clearing agents
facilitate 1 I-scFv-CC49-SAv targeting to the liver.
31
Compounds 2 and 3 showed radioactivity levels similar or
moderately higher than saline for all except the liver where
nearly twice as much I accumulation was observed (4.30 ±
1
31
DISCUSSION
■
0
.41 and 4.25 ± 0.07%ID/g). The 500 kDa dextran−biotin
The goal of this work is to evaluate and advance pretargeted
RIT approaches for cancer imaging and therapy. Our laboratory
has been interested in the synthesis of effective clearing agents
to be used in MST applications. The basic design strategy used
thus far has been to construct bimodal molecules that have the
capacity to target both the desired antibody and hepatocytes.
Early clearing agents utilized human serum albumin (HSA) as a
conjugate followed a similar trend to glycodendrons 1 and 4,
except for the spleen where notably higher radioactivity levels
were observed (3.17 ± 0.67% ID/g).
The tissue-to-blood ratios were calculated to evaluate tracer
retention (Figure 5). Mice treated with glycodendrons or the
5
>
00 kDa dextran−biotin compounds showed liver/blood ratios
29
1:1 compared with vehicle, suggesting retention of the tracer.
scaffold to which biotin and galactose were conjugated.
For spleen/blood, the biotinylated-dextran cohort was 1.5:1,
whereas for glycodendrons and control, the ratio was ≤1:1. In
the kidney, no differences were observed between groups, and
the kidney/blood ratio was ∼0.75:1.
Macromolecular dextrans with biotin, as shown here, have also
been used as clearing agents. The primary drawback with these
agents is the heterogeneity of the final product as well as the
challenges of synthetic reproducibility. While adequate for small
2
100
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
scale preclinical work, such issues become significant barriers to
clinical studies in humans. This lead to the design of
glycodendron 3, which has been used in several preclinical
the predominant clearing mechanism for dextran and its
complexes.
6
Kinetic analysis of blood clearance of 131I-scFv-CC49-SAv
following administration of compounds 1 to 4 indicate that all
can bind the protein in vivo and efficiently remove it from
circulation within 15 min following administration of an
intravenous bolus dose ≤100 μg with no adverse effects. For
this experiment the primary goal was to determine basic blood
clearing activity for compounds 1 to 4. Starting with 0.57 nmol
of the labeled fusion protein (i.e., 1 I-scFv-CC49-SAv), a
convenient dose of each CA was chosen to achieve a >10-fold
molar excess, relative to protein. Consequently, 50−100 μg of
CA were used (50 μg of 1, 100 μg of 2−4). This is in contrast
to previously reported work where 1.4 to 3.5 nmol protein and
5,8,13−15
and clinical PRIT related studies.
The use of a dendron
polymer offers the advantages of incorporating multiple
carbohydrate units and biotin in a predetermined and
controlled manner, resulting in a product that is chemically
well-defined and monodisperse. However, due to past
difficulties to synthesize compound 3, efforts to explore lower
molecular weight compounds using rational design approaches
31
30
with multiple biotin ligands have been reported.
In this study we describe an alternative solution to efficiently
obtain compound 3 as well as its lower and higher generation
homologues. Previously reported synthetic protocols for 3
followed a more convergent approach where more advanced
glycodendron intermediates were linked together to form the
31−33
5
0−100 μg of 3 were typically employed.
These amounts
were optimized for pretargeting protocols where the higher
fusion protein doses were necessary to allow for saturation of
the tumor target (by the fusion protein). Consequently, direct
comparisons between Figure 4 and analogous pharmacokinetic
profiles from previous pretargeting studies may not be possible.
Additionally, the presence of low levels of endogenous biotin in
mice may significantly impact the CA mediated clearance of
7
final product. However, efforts to repeat these steps resulted in
polydisperse products which were difficult to purify. To avoid
potential steric issues associated with joining high molecular
weight fragments and partial loss of protecting groups, we
focused on a more linear route where high purity low molecular
weight building blocks are used to first construct the dendron
scaffold, followed by carbohydrate and biotin conjugations. Our
approach relies upon a new route to synthesize high purity AB2
monomer 7, which was used as the basic dendron building
block. Dendrons 11 to 14 were then readily obtained by
implementing iterative deprotection/coupling cycles reminis-
cent of solid phase synthesis. To minimize purification steps,
coupling reactions were optimized to ensure that all reactions
were complete before proceeding. We were then able to take
advantage of the aliphatic nature of the dendron products and
use simple aqueous workup procedures to remove impurities.
This enabled us to forego several column chromatography
purification steps while affording the desired dendrons with
high purities and good yields as determined by LCMS and
NMR (Figure 2 and Supporting Information). Once dendrons
131
I-scFv-CC49-SAv at the lower dose used here. This may
explain why greater relative percentage drops in circulating
fusion protein (Figure 4) were not achieved in our experiments.
The pharmacokinetic profiles in Figure 4 provide little insight
as to which glycodendron is best suited for MST applications.
More rigorous studies are certainly needed, especially in the
context of a pretargeting protocol. The advantages of lower
molecular weight CAs, such as compounds 1 and 2, are clear:
they require fewer synthetic steps, and are more readily
obtained at lower cost. However, their small size can potentially
lead to unwanted complications, such as extravasation and
accumulation at the tumor site, leading to binding of the tumor
prelocalized mAb-SAv. Conversely, the larger CAs, such as
compounds 3 and 4, are synthetically more demanding, thus
more expensive. Yet their size may afford reduced tissue
penetration and increased residence time in the blood thereby
facilitating clearance of excess mAb-SAv.
11 to 14 were obtained, biotin and sugar 19 could be readily
conjugated to obtain glycodendrons 1 to 4. A single HPLC
purification step was performed resulting in materials of
excellent quality (Figure 3 and Supporting Information). An
analogous methodology which used a different protecting
group schema has been described for similar aliphatic dendrons.
In this example a reduced number of synthetic steps were used,
but potentially difficult purification steps were necessary to
Apart from the use of compounds 1 to 4 as clearing agents in
the context of MST, we also sought to evaluate these
glycondendrons as general liver targeting agents. To measure
CA dependent liver targeting/uptake of the fusion protein,
131
biodistribution analysis for I-scFv-CC49-SAv 24 h p.i. of
glycodendrons in select organs was performed (Figure S29).
21
obtain dendrons of good purity.
131
When I in the tissue are normalized to blood levels, we
We conducted biological testing of the glycodendrons to
verify hepatocyte targeting in vivo. For utility in MST, the
glycodendron must demonstrate the capacity to redirect the
tumor targeting mAb from circulation to the liver. We
administered 1 I- scFv-CC49-SAv, a familiar MST-antibody
tracer, to biotin-starved mice followed by an excess dose of
clearing agent relative to antibody dose. We conducted serial-
blood sampling for kinetic analysis of radioactivity, as well as
biodistribution studies to determine if animals given glycoden-
drons showed increased liver/blood ratios, indicating increased
uptake and retention (i.e., liver targeting). The inclusion of
biotinylated-dextran was twofold: a dextran-based clearing
agent was reported recently to be an effective clearing agent
for 3-step MST, and supporting biodistribution experiments
described here with tracer and dextran-clearing agent showed
significant uptake in the spleen and liver 4 h p.i. (30%ID/g and
observe an enhancement in liver targeting over the spleen and
kidneys when 1 to 4 are used (Figure 5). Additionally, the data
suggest liver uptake and retention to be the predominant
mechanism for the glycodendron dependent clearance of the
fusion protein. This study does not reveal a correlation between
sugar valency (or molecular weight) and liver targeting
efficiency, so further studies would be needed to address this
point.
The glycodendrons described here provide a highly versatile
molecular scaffold upon which to design further optimized
clearing agents as well as incorporate other ligand/receptor
pairs for MST applications. The dendron scaffold is ideal for
undertaking further structure−activity relationship (SAR)
studies to identify modifications that could possibly enhance
its function. For example, the chain lengths between the
carbohydrates and the dendron and/or the biotin and dendron
can be further extended to ensure optimal binding to the
31
∼10%ID/g, respectively), suggesting that RES metabolism is
2
101
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
respective receptor or binding partner. It would be of particular
benefit to evaluate such glycodendrons in protein, liver
membrane, or cell based binding assays and correlate the
(3) Goldenberg, D. M., Chang, C. H., Rossi, E. A., McBride, W. J.,
and Sharkey, R. M. (2012) Pretargeted molecular imaging and
radioimmunotherapy. Theranostics 2, 523−540.
34,35
(4) Axworthy, D. B., and Reno, J. M. (1997) Hexose derivatized
human serum albumin clearing agents. U.S. Patent 5,616,690.
results with in vivo clearance data.
In addition, the central amine functionality on the dendron
linker can be used to orthogonally introduce different targeting
moieties. Biotin can be replaced with other molecules to utilize
different ligand/receptor pairs. While biotin/streptavidin has
been shown to be effective, the presence of biotin in humans as
(5) Axworthy, D. B., Reno, J. M., Hylarides, M. D., Mallett, R. W.,
Theodore, L. J., Gustavson, L. M., Su, F.-M., Hobson, L. J., Beaumier,
P. L., and Fritzberg, A. R. (2000) Cure of human carcinoma xenografts
by a single dose of pretargeted yttrium-90 with negligible toxicity. Proc.
Natl. Acad. Sci. U.S.A. 97, 1802−1807.
18
well the immunogenicity of streptavidin may limit its use.
Consequently, different ligand/receptor pairs are being
(6) Orcutt, K. D., Rhoden, J. J., Ruiz-Yi, B., Frangioni, J. V., and
Wittrup, K. D. (2012) Effect of small-moleculebinding affinity on
tumor uptake in vivo: a systematic study using a pretargeted bispecific
antibody. Mol. Cancer Ther. 11, 1365−1372.
(7) Theodore, L. J. and Axworthy, D. B. (2001) Cluster clearing
agents. U.S. Patent 6,172,045.
1
8,36−40
investigated.
In particular, bispecific mAbs, which
recognize a tumor antigen and a small molecule such as
3
,6
metal bound DOTA, have made promising advances. We are
interested in utilizing these glycodendrons in the context of
such bispecific mAbs MST protocols, and our results will be
reported in due course.
(8) Press, O. W., Corcoran, M., Subbiah, K., Hamlin, D. K., Wilbur,
D. S., Johnson, T., Theodore, L., Yau, E., Mallett, R., Meyer, D. L., and
Axworthy, D. (2001) A comparative evaluation of conventional and
pretargeted radioimmunotherapy of CD20-expressing lymphoma
xenografts. Blood 98, 2535−2543.
In conclusion, we describe the synthesis and characterization
of a class of glycodendron clearing agents. The synthesis
comprises high yielding steps, potentially amenable to
scalability, where crude products are obtained with good
purity. We then evaluated the capacity for these glycodendrons
compared to biotin−dextran conjugates to clear a circulating
(
9) Ashwell, G., and Harford, J. (1982) Carbohydrate-specific
receptors of the liver. Annu. Rev. Biochem. 51, 531−554.
10) Baenziger, J. U., and Maynard, Y. (1980) Human hepatic lectin.
Physiochemical properties and specificity. J. Biol. Chem. 255, 4607−
613.
(11) Lee, Y. C., Townsend, R. R., Hardy, M. R., Lo
(
1
31
radiolabeled mAb streptavidin fusion protein, I-scFv-CC49-
SAv, in mice. We find that all the glycodendrons and biotin−
dextran conjugate enhance blood clearance of the fusion
protein, and observe higher accumulation of the radiolabel in
the liver, suggesting appropriate targeting of the ASGP
receptor. Experiments seeking to evaluate these compounds
in tumor pretargeting experiments are underway and will be
reported in due time.
4
̈
nngren, J., Arnarp,
nn, H. (1983) Binding of synthetic
J., Haraldsson, M., and Lo
̈
oligosaccharides to the hepatic Gal/GalNAc lectin. Dependence on
fine structural features. J. Biol. Chem. 258, 199−202.
(12) Chauhan, J., and Dakshinamurti, K. (1986) Purification and
characterization of human serum biotinidase. J. Biol. Chem. 261, 4268−
275.
13) Cheung, N.-K. V., Modak, S., Lin, Y., Guo, H., Zanzonico, P.,
4
(
Chung, J., Zuo, Y., Sanderson, J., Wilbert, S., Theodore, L. J.,
Axworthy, D. B., and Larson, S. M. (2004) Single-chain Fv-streptavidin
substantially improved therapeutic index in multistep targeting
directed at disialoganglioside GD2. J. Nucl. Med. 45, 867−877.
(14) Sato, N., Hassan, R., Axworthy, D. B., Wong, K. J., Yu, S.,
Theodore, L. J., Lin, Y., Park, L., Brechbiel, M. W., Pastan, I., Paik, C.
H., and Carrasquillo, J. A. (2005) Pretargeted radioimmunotherapy of
mesothelin-expressing cancer using a tetravalent single-chain Fv-
streptavidin fusion protein. J. Nucl. Med. 46, 1201−1209.
ASSOCIATED CONTENT
■
*
S
Supporting Information
LCMS, H NMR, 13C NMR for dendrons and biodistribution
data. This material is available free of charge via the Internet at
http://pubs.acs.org.
1
AUTHOR INFORMATION
Corresponding Authors
Tel: 1 646 888 3272; fax: 1 646 888 3059. E-mail: yoob@
mskcc.org.
Tel: 1 212 639 7771; fax: 1 212 717 3654. E-mail: o-
ouerfelli@ski.mskcc.org.
Notes
■
*
(
15) Forster, G. J., Santos, E. B., Smith-Jones, P. M., Zanzonico, P.,
and Larson, S. M. (2006) Pretargeted radioimmunotherapy with a
single-chain antibody/streptavidin construct and radiolabeled DOTA-
biotin: Strategies for reduction of the renal dose. J. Nucl. Med. 47,
*
1
40−149.
(16) Schultz, J., Lin, Y., Sanderson, J., Zuo, Y., Stone, D., Mallett, R.,
The authors declare no competing financial interest.
Wilbert, S., and Axworthy, D. (2000) A tetravalent single-chain
antibody-streptavidin fusion protein for pretargeted lymphoma
therapy. Cancer Res. 60, 6663−6669.
ACKNOWLEDGMENTS
■
(
17) Graves, S. S., Dearstyne, E., Lin, Y., Zuo, Y., Sanderson, J.,
We would like to thank Dr. George Sukenick and Ms. Hui Fang
from the NMR Analytical Core Facility at MSKCC for
analytical data collection and interpretation. We thank Dr.
Cynthia Jung for her assistance in the preparation of this
manuscript. We also thank Don Axworthy for critical reading of
the manuscript. Work at the Organic Synthesis Core at
MSKCC is partially supported by NCI Grant P01 129243.
Schultz, J., Pantalias, A., Gray, D., Axworthy, D., Jones, H. M., and
Auditore-Hargreaves, K. (2003) Combination therapy with pretarget
CC49 radioimmunotherapy and gemcitabine prolongs tumor doubling
time in a murine xenograft model of colon cancer more effectively than
either monotherapy. Clin. Cancer Res. 9, 3712−3721.
(18) Lewis, M. R., Zhang, J., Jia, F., Owen, N. K., Cutler, C. S.,
Embree, M. F., Schultz, J., Theodore, L. J., Ketring, A. R., Jurisson, S.
S., and Axworthy, D. B. (2004) Biological comparison of 149Pm-,
REFERENCES
1
66Ho-, and 177Lu-DOTA-biotin pretargeted by CC49 scFv-
■
(
1) Heegaard, P. M. H., Boas, U., and Sorensen, N. S. (2009)
Dendrimers for vaccine and immunostimulatory uses. A review.
streptavidin fusion protein in xenograft-bearing nude mice. Nucl.
Med. Biol. 31, 213−223.
(19) Harlow, E., and Lane, D. (1988) Antibodies: A Laboratory
Manual, Cold Spring Harbor Laboratory.
Bioconjugate Chem. 21, 405−418.
(
2) Medina, S. H., and El-Sayed, M. E. H. (2009) Dendrimers as
carriers for delivery of chemotherapeutic agents. Chem. Rev. 109,
(20) Lindmo, T., Boven, E., Cuttitta, F., Fedorko, J., and Bunn, P. A.,
3141−3157.
Jr. (1984) Determination of the immunoreactive fraction of radio-
2
102
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103

Bioconjugate Chemistry
Technical Note
labeled monoclonal antibodies by linear extrapolation to binding at
antibody-oligonucleotide conjugates by radiolabeled antisense nucleo-
tides: a novel approach for two-step radioimmunotherapy of cancer.
Bioconjugate Chem. 4, 94−102.
(39) Le Doussal, J.-M., Chetanneau, A., Gruaz-Guyon, A., Martin, M.,
Gautherot, E., Lehur, P.-A., Chatal, J.-F. o., Delaage, M., and Barbet, J.
(1993) Bispecific monoclonal antibody-mediated targeting of an
indium-111-labeled DTPA dimer to primary colorectal tumors:
pharmacokinetics, biodistribution, scintigraphy and immune response.
J. Nucl. Med. 34, 1662−1671.
(40) Goodwin, D. A., Meares, C. F., McCall, M. J., McTigue, M., and
Chaovapong, W. (1988) Pre-targeted immunoscintigraphy of murine
tumors with indium-111-labeled bifunctional haptens. J. Nucl. Med. 29,
226−234.
infinite antigen excess. J. Immunol. Methods 72, 77−89.
(
21) Balaji, B. S., and Lewis, M. R. (2009) Double exponential
growth of aliphatic polyamide dendrimers via AB2 hypermonomer
strategy. Chem. Commun. 30, 4593−4595.
(
22) Angell, Y. M., Garcia-Echeverria, C., and Rich, D. H. (1994)
Comparative studies of the coupling of N-methylated, sterically
hindered amino-acids during solid-phase peptide-synthesis. Tetrahe-
dron Lett. 35, 5981−5984.
(
23) Carpino, L. A. (1993) 1-Hydroxy-7-azabenzotriazole. An
efficient peptide coupling additive. J. Am. Chem. Soc. 115, 4397−4398.
24) Knapp, S., and Myers, D. S. (2002) Synthesis of alpha-GalNAc
thioconjugates from an alpha-GalNAc mercaptan. J. Org. Chem. 67,
995−2999.
25) Laganis, E. D., and Chenard, B. L. (1984) Metal silanolates:
(
2
(
−2
organic soluble equivalents for O . Tetrahedron Lett. 25, 5831−5834.
(
26) Lovric, M., Cepanec, I., Litvic, M., Bartolincic, A., and Vinkovic,
V. (2007) Scope and limitations of sodium and potassium
trimethylsilanolate as reagents for conversion of esters to carboxylic
acids. Croat. Chem. Acta 80, 109−115.
(
27) Conti, P., Demont, D., Cals, J., Ottenheijm, H. C. J., and Leysen,
D. (1997) A new cleavage strategy for the solid-phase synthesis of
secondary amines. Tetrahedron Lett. 38, 2915−2918.
(
28) Basso, A., Braiuca, P., De Martin, L., Ebert, C., Gardossi, L.,
Linda, P., Verdelli, S., and Tam, A. (2004) Nonswelling macroporous
synbeads for improved efficiency of solid-phase biotransformations.
Chem.Eur. J. 10, 1007−1013.
(
29) Breitz, H. B., Weiden, P. L., Beaumier, P. L., Axworthy, D. B.,
Seiler, C., Su, F.-M., Graves, S., Bryan, K., and Reno, J. M. (2000)
Clinical optimization of pretargeted radioimmunotherapy with anti-
body-streptavidin conjugate and 90Y-DOTA-biotin. J. Nucl. Med. 41,
1
(
31−140.
30) Wilbur, D. S., Park, S. I., Chyan, M.-K., Wan, F., Hamlin, D. K.,
Shenoi, J., Lin, Y., Wilbur, S. M., Buchegger, F., Pantelias, A., Pagel, J.
M., and Press, O. W. (2010) Design and synthesis of bis-biotin-
containing reagents for applications utilizing monoclonal antibody-
based pretargeting systems with streptavidin mutants. Bioconjugate
Chem. 21, 1225−1238.
(
31) Goshorn, S., Sanderson, J., Axworthy, D., Lin, Y. K., Hylarides,
M., and Schultz, J. (2001) Preclinical evaluation of a humanized NR-
LU-10 antibody-streptavidin fusion protein for pretargeted cancer
therapy. Cancer Biother. Radiopharm. 16, 109−123.
(
32) Lin, Y. K., Pagel, L. M., Axworthy, D., Pantelias, A., Hedin, N.,
and Press, O. W. (2006) A genetically engineered anti-CD45 single-
chain antibody-streptavidin fusion protein for pretargeted Radio-
immunotherapy of hematologic malignancies. Cancer Res. 66, 3884−
3892.
(
33) Pantelias, A., Pagel, J. M., Hedin, N., Saganic, L., Wilbur, S.,
Hamlin, D. K., Wilbur, D. S., Lin, Y., Stone, D., Axworthy, D., Gopal,
A. K., and Press, O. W. (2007) Comparative biodistributions of
pretargeted radioimmunoconjugates targeting CD20, CD22, and DR
molecules on human B-cell lymphomas. Blood 109, 4980−4987.
(
34) Park, E. I., Mi, Y., Unverzagt, C., Gabius, H.-J., and Baenziger, J.
U. (2005) The asialoglycoprotein receptor clears glycoconjugates
terminating with sialic acid α2,6 GalNAc. Proc. Natl. Acad. Sci. U.S.A.
1
(
02, 17125−17129.
35) Kornilova, A. Y., Algayer, B., Breslin, M., and Uebele, V. (2012)
Development of a fluorescence polarization binding assay for
asialoglycoprotein receptor. Anal. Biochem. 425, 43−46.
(
36) Rossin, R., Renart Verkerk, P., van den Bosch, S. M., Vulders, R.
C. M., Verel, I., Lub, J., and Robillard, M. S. (2010) In vivo chemistry
for pretargeted tumor imaging in live mice. Angew. Chem., Int. Ed. 49,
3
(
375−3378.
37) Liu, G., Dou, S., Liu, Y., Wang, Y., Rusckowski, M., and
Hnatowich, D. J. (2011) 90Y labeled phosphorodiamidate morpholino
oligomer for pretargeting radiotherapy. Bioconjugate Chem. 22, 2539−
2545.
(
38) Kuijpers, W. H. A., Kaspersen, F. M., Veeneman, G. H., Van
Boeckel, C. A. A., and Bos, E. S. (1993) Specific recognition of
2
103
dx.doi.org/10.1021/bc400333m | Bioconjugate Chem. 2013, 24, 2088−2103