A Novel Rapamycin-polymer conjugate based on a new poly(Ethylene Glycol) multiblock copolymer

Article abstract of DOI:10.1007/s11095-013-1192-3

Purpose: Rapamycin has demonstrated potent anti-tumor activity in preclinical and clinical studies. However, the clinical development of its formulations was hampered due to its poor solubility and undesirable distribution in vivo. Chemical modification of rapamycin presents an opportunity for overcoming the obstacles and improving its therapeutic index. The objective of this study is to develop a drug-polymer conjugate to increase the solubility and cellular uptake of rapamycin. Methods: We developed the rapamycin-polymer conjugate using a novel, linear, poly(ethylene glycol) (PEG) based multiblock copolymer. Cytotoxicity and cellular uptake of the rapamycin-polymer conjugate were evaluated in various cancer cells. Results: The rapamycin-polymer conjugate provides enhanced solubility in water compared with free rapamycin and shows profound activity against a panel of human cancer cell lines. The rapamycin-polymer conjugate also presents high drug loading capacity (wt% ~ 26%) when GlyGlyGly is used as a linker. Cellular uptake of the conjugate was confirmed by confocal microscopic examination of PC-3 cells that were cultured in the presence of FITC-labled polymer (FITC-polymer). Conclusion: This study suggests that the rapamycin-polymer conjugate is a novel anti-cancer agent that may provide an attractive strategy for treatment of a wide variety of tumors.

Full text of DOI:10.1007/s11095-013-1192-3

Pharm Res (2014) 31:706–719
DOI 10.1007/s11095-013-1192-3
RESEARCH PAPER
A Novel Rapamycin-Polymer Conjugate Based on a New
Poly(Ethylene Glycol) Multiblock Copolymer
Wanyi Tai & Zhijin Chen & Ashutosh Barve & Zhonghua Peng & Kun Cheng
Received: 13 June 2013 /Accepted: 9 August 2013 /Published online: 26 September 2013
#
Springer Science+Business Media New York 2013
ABSTRACT
ABBREVIATIONS
Purpose Rapamycin has demonstrated potent anti-tumor activity
in preclinical and clinical studies. However, the clinical develop-
ment of its formulations was hampered due to its poor solubility
and undesirable distribution in vivo. Chemical modification of
rapamycin presents an opportunity for overcoming the obstacles
and improving its therapeutic index. The objective of this study is
to develop a drug-polymer conjugate to increase the solubility and
cellular uptake of rapamycin.
FKBP12
mTOR
PEG
FK-binding protein 12
The mammalian target of rapamycin
Poly(ethylene glycol)
Rapalogues Rapamycin analogues
RBCs Red blood cells
Methods We developed the rapamycin-polymer conjugate
using a novel, linear, poly(ethylene glycol) (PEG) based multiblock
copolymer. Cytotoxicity and cellular uptake of the rapamycin-
polymer conjugate were evaluated in various cancer cells.
Results The rapamycin-polymer conjugate provides enhanced
solubility in water compared with free rapamycin and shows
profound activity against a panel of human cancer cell lines. The
rapamycin-polymer conjugate also presents high drug loading
capacity (wt% ~ 26%) when GlyGlyGly is used as a linker.
Cellular uptake of the conjugate was confirmed by confocal
microscopic examination of PC-3 cells that were cultured in the
presence of FITC-labled polymer (FITC-polymer).
INTRODUCTION
Rapamycin, also known as sirolimus, is a carboxylic macrolide
compound that was originally discovered as the product of a
bacteria from soil sample of island Rapa Nui in the South
Pacific (1). It was first developed as an antifungal drug by
Ayerst Research Laboratories, but the potent immunosup-
pressive and anti-proliferative effects were discovered later
(2). Rapamycin was approved as an immunosuppressive agent
by the US Food and Drug Administration (FDA) in 1999. Due
to its unique cytostatic effect, rapamycin has also entered into
clinical trials to treat various cancers, including breast cancer,
prostate cancer and glioblastoma (3–5). The molecular mech-
anism of rapamycin is to inhibit the mammalian target of
rapamycin (mTOR) signaling pathway by directly binding to
FK-binding protein 12 (FKBP12) and mTOR1. Because of
the importance of the mTOR signal pathway in cancer de-
velopment, rapamcyin was found to be active against almost
all types of solid tumors (2). Rapamycin exhibits an exquisite
selectivity to mTOR and demonstrates potent anti-tumor
activity in the nanomolar range in vitro. Besides inhibiting
cancer cell proliferation, rapamycin can also induce tumor
endothelial cell death and vessel thrombosis by acting as a
vascular AKT/mTOR inhibitor, which facilitates its anti-
tumor effect (6,7).
Conclusion This study suggests that the rapamycin-polymer
conjugate is a novel anti-cancer agent that may provide an attrac-
tive strategy for treatment of a wide variety of tumors.
.
.
KEY WORDS multiblock copolymer PEG polymer-drug
conjugate rapalogue rapamcyin
.
.
:
:
:
W. Tai Z. Chen A. Barve K. Cheng (*)
Division of Pharmaceutical Sciences, School of Pharmacy
University of Missouri-Kansas City, 2464 Charlotte Street
Kansas City, Missouri 64108, USA
e-mail: chengkun@umkc.edu
Although rapamycin has broad and potent anti-tumor
activity, it is insoluble in water and possesses poor distribution
in vivo, which dramatically limits its clinical application
Z. Peng
Department of Chemistry, University of Missouri-Kansas City
5100 Rockhill Road Kansas City, Missouri 64110, USA

A Novel Rapamycin-Polymer Conjugate
707
(8–10). Structurally, rapamycin does not contain any acidic or
basic group that can increase aqueous solubility by salt forma-
tion (8,11). As a result, rapamycin is very hydrophobic and the
solubility in water is only about 33 μg/mL (2.6 μg/mL in some
reports) (12). Moreover, it is very difficult to prepare injectable
rapamycin formulations for preclinical/clinical studies because
rapamycin is only slightly soluble in many cosolvents such as
ethanol, PEG 400 and polysorbate 80 (8). Rapamycin also
exhibits a poor distribution profile in vivo. Due to the ubiquitous
presence of the FK-binding proteins (high affinity to
rapamycin) in red blood cells (RBCs), rapamycin exhibits a
preferential distribution in RBCs. As reported by Yatscoff et al.,
95% of rapamycin in the blood is distributed in RBCs com-
pared to 3.1% in the plasma, 1.0% in lymphocytes and 1.0% in
granulocytes (9). Because RBCs can protect rapamycin from
hepatic metabolism and renal filtration, rapamycin exhibits a
relatively long terminal half-life (61–72 h in humans) compared
with other small molecules (13). However, low partition in the
plasma dramatically hinders the diffusion of rapamycin from
the blood into tumor cells. In addition, it was reported that
rapamycin was extensively distributed among many tissues in
rats (tissue to blood partition coefficient Kp>40 in some cases)
because of its lipophilic nature (14). The extensive distribution
of rapamcyin in tissues is believed to account for its high
apparent distribution volume (5.6–17.6 L/kg) and toxic effects
(15,16).
To overcome these limitations, rapamycin has been chem-
ically modified to hydrophilic analogues, such as CCI-779,
RAD001 and AP23573 (Fig. 1a) (17). CCI-779, also known as
Temsirolimus (Torisel, Wyeth), is an ester prodrug of
rapamycin for treatment of renal cell carcinoma. By introduc-
ing 2,2-bis(hydroxylmethyl) propionic ester at the 42-OH of
rapamycin, CCI-779 increases its hydrophilicity and makes
the analogue readily soluble for intravenous formulation (18).
RAD001, also known as Everolimus (Afinitor, Novartis), is a
hydroxyl-ethyl ether of rapamycin with improved aqueous
a
c
Poly[bis(ε-Lys)Glut-PEG]
Bis(ε-Lys)Glut
PEG
PEG
CCI-779
Rapamycin (Rapa)
GlyGlyGly
Rapamycin
AP23573
RAD001
b
Fig. 1 Engineering of the rapamycin-polymer conjugate. (a) Structural comparison of rapamycin and rapalogues (CCI-779, AP23573 and RAD001). All of them
are chemically modified at 42-OH position which is believed to minimally compromise their bioactivities. (b) The schematic graph illustrating the architecture of
rapamycin-polymer conjugate 8c. Instead of small molecular-weight moieties (highlighted in red in rapalogues structure), rapamycin in 8c is conjugated to high
molecular-weight polymer which can dramatically change its solubility and biodistribution profile. Rapamycin (Rapa) is uniformly distributed throughout the
multiblock copolymer. The drug molecules are separated from each other by PEG to avoid aggregation; (c) The representative chemical structure of one block of
the polymer-drug conjugate.

708
Tai et al.
solubility and oral absorption (19). AP23573 (Deforolimus,
Merck/Ariad) is a novel non-prodrug rapamycin analogue
in which 42-OH is conjugated with a phosphate group to
increase aqueous solubility and oral bioavailability (20). How-
ever, these rapamycin analogues (rapalogues) still have limited
aqueous solubility (CCI-779: 0.12 mg/mL; RAD001:
0.1 mg/mL; AP23573: < 1 mg/mL) (15). Cosolvent such as
ethanol is always required for intravenous formulation (21).
Moreover, it was reported that the rapalogue CCI-779 can be
quickly hydrolyzed into rapamycin by plasma esterase and
subsequently redistributed into RBCs. Although other two
rapalogues, RAD001 and AP23573, cannot be converted into
rapamycin in vivo, moderate binding to RBCs was still ob-
served in preclinical studies (22). Therefore, it is necessary to
explore other rapamycin conjugates to improve aqueous sol-
ubility and tissue distribution profile, especially to minimize
the distribution to RBCs.
(Wood Dale, IL). TO-PRO-3 stain was obtained from Life
Technologies (Grand Island, NY). All solvents, including anhy-
drous solvents, HPLC grade solvents and other common organ-
ic solvents were purchased from commercial suppliers and used
without further purification.
Cell Culture
All the cell lines used in this study were obtained from ATCC.
LNCaP, PC-3, MCF-7, MCF-7/HER2 and SK-BR-3 cells
were grown in RPMI-1640 medium containing 10% Fetal Bo-
vine Serum (FBS), 100 units/mL penicillin, and 100 μg/mL
streptomycin. HeLa and DU145 cells were cultured in DMEM
medium supplemented with 10% FBS and 100 units/mL
penicillin-100 μg/mL streptomycin mixtures. All the cells were
grown at 37°C in a humidified atmosphere with 5% CO2.
Media were changed every other day, and the cells were pas-
saged when they reached 80–90% confluency.
Compared to the chemical modification, direct conjuga-
tion of rapamycin to a polymer may offer a better approach to
overcome its limitations. Many hydrophobic drugs, such as
paclitaxel and camptothecin, have demonstrated excellent
aqueous solubility and pharamcokinetics after being conjugat-
ed to hydrophilic polymers (23–26). However, rapamycin is
sensitive to light, temperature and pH, which make it hard to
tolerate many chemical reactions. Although rapamycin has
been successfully conjugated with wortmanin and other small
molecules under mild reaction conditions, little attempt has
been made to conjugate it with macromolecules, such as
polymers and antibodies (27–29). In this study, we developed
a novel, water-soluble, poly(ethylene glycol) (PEG) based
multiblock copolymer for rapamycin conjugation. To conju-
gate rapamycin with the polymer, a tri-peptide linker
GlyGlyGly was conjugated with rapamycin at 42-OH and
then grafted onto the pendant carboxylic groups in polymer
chains. The rapamycin-polymer conjugate exhibits excellent
solubility in water and potent anti-tumor activity, suggesting
its promising potential for clinical development.
Synthesis of Bis(ε-amino-L-Lysine Benzyl ester)
Glutaric amide, bis(ε-Lys-OBn)Glut, 5
The synthesis of comonomer bis(ε-Lys-OBn)Glut 5 was
started from Fmoc-Lys(Boc)-OH. Fmoc-Lys(Boc)-OH was
first esterified with benzyl alcohol, followed by removing
Fmoc group by 20% piperidine-DCM according to
Agnihotri’s report (30). One gram of NH2-Lys(Boc)-OBn 3
was dissolved in 50 mL of dry DCM, and then Glutaric acid
(0.16 g) and EDCI (0.57 g) were added to the solution. The
coupling reaction was initialized by adding 0.52 mL of
triethylamine (Et3N) to the solution. After stirring overnight
at room temperature, the reaction was stopped with 50 mL of
saturated NH4Cl aqueous solution. The organic layer was
separated, dried with Na2SO4, and concentrated under vac-
uum. The compound 4 was purified from the pale white
residue by silica gel chromatography (CHCl3/MeOH=50/1,
v/v). To remove the Boc groups, compound 4 (0.5 g) was
dissolved in 30 mL of TFA/DCM (v/v, 1/1) and stirred at
room temperature for 1 h. The solvent was evaporated under
vacuum, and ice-cold diethyl ether was added to precipitate the
product. The white precipitate was collected by filtration and
MATERIALS AND METHODS
Materials
1
dried under vacuum to get the pure product 5 (~ 60%). H
NMR (400 MHz, D2O) δ 7.27 (s, 5H), 5.05 (m, 2H), 4.25
(t, 1H), 2.79 (t, 2H), 1.95 (m, 2H), 1.65 (m, 3H), 1.53 (m, 2H),
1.26 (m, 2H). ESI-MS calcd. for C31H44N4O6 568.7; found
569.3 (M + H)1+.
All reagents listed below were obtained from commer-
cial sources and used without further purification.
Rapamycin was purchased from LC laboratories (Wo-
burn, MA). NHS-PEG3400-NHS was purchased from
Nanocs Inc. (New York, NJ). 2-(7-Aza-1H-benzotriazole-1-yl)-
1,1,3,3-tetramethyluronium hexafluorophosphate (HATU) and
N-Hydroxybenzotriazole (HOBT) were ordered from
Genscript Inc. (Piscataway, NJ). Thiazolyl Blue was obtained
from RPI Corp. (Prospect, IL). Fmoc-Lys(Boc)-OH and Trt-
Gly-OH were purchased from Chem-Impex International Inc.
Synthesis of Poly[bis(ε-Lys-OBn)Glut-PEG], 6a-d
Before the reaction, bis(ε-Lys-OBn)Glut 5 was dried in a
vacuum desiccator for 2 days. Bis(ε-Lys-OBn)Glut 5 (19 mg)
and NHS-PEG3400-NHS (100 mg) were transferred to a dry
5-mL flask, followed by adding 0.5 mL of extra dry DMF

A Novel Rapamycin-Polymer Conjugate
709
(water < 50 ppm). The solution was stirred at room temper-
ature for 10 min, and then anhydrous Et3N was added into
the reaction in nitrogen atmosphere. After stirring at room
temperature or 50°C for 2–24 h, the mixture was diluted with
50 mL of water and dialyzed against distilled water using a
dialysis tubing with MWCO 14,000. The polymer solution
was lyophilized, and pure polymer 6a-d was collected as
white powder (60–80%). The molecular weight of polymer
was determined on a Tosoh EcoSec HLC 8320GPC system
equipped with a differential refractometer. The gel perme-
ation column (Styragel®, 8 μm 300×7.5 mm, Waters Cor-
poration) was calibrated using polystyrene standards and elut-
ed with THF at a flow rate of 0.7 mL/min. As for 6b, Mw=
31.5 kDa, Mn=21.9 kDa, Mw/Mn=1.43. 1H NMR
(400 MHz, D2O) δ 7.30 (s, 5H), 5.08 (m, 2H), 4.26 (t, 1H),
4.07 (m, 2H), 3.61 (m, PEG-CH2), 3.43 (m, 2H), 2.95 (m, 2H),
2.16 (m, 2H), 2.15–2.18 (m, 5H), 1.35 (m, 2H).
4 mL of ice-cold THF, followed by adding 2 mL of 2 N
H2SO4 solution. The reaction was warmed to room temper-
ature and stirred for 30 min. After the reaction was complete,
the mixture was poured into 20 g of ice and neutralized by
adding excess amount of NaHCO3. The neutralized solution
was extracted with AcOEt. The organic phase was washed
with brine, dried and concentrated. The residue was applied
on a silica gel column and purified using petroleum ether/
acetone (v/v=3/1) as the elution solution (yield, ~ 80%). H
NMR (400 MHz, CDCl3) δ 7.46 (d, 6H), 7.28 (m, 6H), 7.26 (t,
3H), 6.36 (m, 2H), 6.13 (q, 1H), 5.95 (d, 1H), 5.55(q, 1H), 5.42
(d, 1H), 5.29 (d, 1H), 5.15(q, 1H), 4.79 (s, 1H), 4.61 (m, 1H),
4.17 (d, 1H), 3.87 (m, 1H), 3.75 (d, 1H), 3.66 (m, 1H), 3.55 (d,
1H), 3.38 (m, 1H), 3.33 (s, 3H), 3.30 (s, 3H), 3.27 (m, 1H), 3.14
(s, 3H), 3.08 (m, 1H), 2.69 (m, 2H), 2.58 (m, 1H), 2.32 (m, 2H),
2.17 (s, 2H), 2.05 (m, 1H), 1.93 (m, 3H), 0.75–1.80 (m, 41H).
ESI-MS calcd. for C71H98N2O13 1213.5; found 1213.7
[M + H]+.
Synthesis of Linker-Drug, Gly-Rapa and GlyGlyGly-Rapa
Gly-Rapa 13. Protecting group Trt was removed using 0.1 M
HOBT in trifluoroethanol (TFE) as reported (33). Briefly, com-
pound 12 (50 mg) was dissolved in 3 mL of freshly prepared
0.1 M anhydrous HOBT in TFE. The reaction was stirred at
room temperature for 10 min. The process of the reaction was
monitored by TLC. After the Trt group was completely re-
moved, 100 mL of water was added to stop the reaction. The
white suspension was then extracted with 100 mL of AcOEt. The
organic phase was washed with brine, dried by Na2SO4 and
concentrated under vacuum. Gly-Rapa 13 was purified from the
residue using silica gel chromatography (CHCl3/MeOH, v/v=
10/1) and yielded as white powder (~70%). H NMR (400 MHz,
CDCl3) δ 6.36 (m, 2H), 6.14 (q, 1H), 5.95 (d, 1H), 5.56(q, 1H),
5.43 (d, 1H), 5.28 (d, 1H), 5.16(q, 1H), 4.97 (s, 1H), 4.73 (m, 1H),
4.18 (d, 1H), 3.86 (m, 1H), 3.75 (d, 1H), 3.66 (m, 1H), 3.55
(d, 1H), 3.43 (m, 2H), 3.37 (s, 3H), 3.33 (s, 3H), 3.14 (s, 3H), 2.81
(m, 1H), 2.69 (m, 2H), 2.59 (m, 1H), 2.32 (m, 2H), 2.12 (s, 1H),
2.04 (m, 2H), 1.99 (m, 3H), 0.75–1.80 (m, 45H). ESI-MS calcd.
for C53H82N2O14 971.2; found 971.9 [M + H]+.
31-O-TMS Rapamycin 10. 31-Trimethylsilyl Rapamycin was
synthesized according to the procedure described by Shaw
and Nan (31,32). Briefly, Rapamycin (200 mg) and imidazole
(150 mg) were dissolved in 25 mL of DCM. After cooling to
0°C, trimethylsilyl chloride (TMSCl) solution in DCM (1 M,
1.3 mL) was slowly added into the reaction. The reaction was
stirred for around 10 min at 0°C. After the reaction was
complete, the mixture was directly applied on silica gel col-
umn and eluted with petroleum ether/acetone (v/v, 5/1). The
purified intermediate 31,42-bis-O-TMS Rapamycin 9
(210 mg) was then dissolved in 20 mL of DCM, followed by
adding 100 mg of imidazole/imidazole·HCl (mole ratio, 1/1).
After stirring at room temperature for 3 h, the reaction was
stopped and 31-O-TMS Rapamycin 10 was purified by flash
chromatography on silica gel (petroleum ether/acetone, v/v=
5/1). The overall yield for the two-step reaction is ~ 50%. 1H
NMR (400 MHz, CDCl3) δ 6.41 (m, 2H), 6.17 (q, 1H), 6.05 (d,
1H), 5.59 (q, 1H), 5.33 (d, 1H), 5.22 (d, 1H), 5.05 (q, 1H), 4.73
(s, 1H), 4.07 (d, 1H), 3.85 (m, 2H), 3.72 (m, 1H), 3.66 (d, 1H),
3.41 (s, 3H), 3.38 (m, 3H), 3.27 (s, 3H), 3.14 (s, 3H), 2.93
(m, 1H), 2.68 (m, 3H), 2.31 (m, 2H), 2.11 (m, 1H), 1.98 (m,
3H), 0.60–1.80 (m, 41H), 0.04 (s, 9H). ESI-MS calcd. for
C54H87NO13Si 986.4; found 1008.7 [M + Na]+.
GlyGlyGly-Rapa 15. Compound 13 (20 mg) and Trt-GlyGly-
OH (20 mg) were dissolved in 10 mL of dry DCM. After
stirring at room temperature for 5 min, EDCI (10 mg) and
Et3N (8 μL) were added to the mixture. After 2 h, the reaction
was stopped by adding 10 mL of saturated NH4Cl aqueous
solution. The organic phase was then separated, washed with
brine, dried by Na2SO4 and concentrated under vacuum.
The intermediate Trt-GlyGlyGly-Rapa 14 was purified on
silica gel column (CHCl3/MeOH, v/v=50/2.5). The Trt
group of 14 was deprotected using the same method as
described above. The final product GlyGlyGly-Rapa 15
was purified by silica gel chromatography (CHCl3/MeOH,
v/v=5/1, then 2/1). The yield for the two steps was 50 ~
60%. 1H NMR (400 MHz, CDCl3) δ 6.35 (m, 2H), 6.13 (q, 1H),
Trt-Gly-Rapa 12. 31-O-TMS Rapamycin 10 (100 mg) and
Trt-Gly-OH (64 mg) were dissolved in 5 mL of DCM. After
stirring at room temperature for 5 min, EDCI (40 mg) and
DMAP (2 mg) were added to the reaction. The reaction
mixture was continuously stirred for 2 h at room temperature.
Trt-Gly-Rapa(31-O-TMS) 11 was purified from the reaction
using silica gel chromatography (petroleum ether/acetone,
v/v=5/1). The TMS group at the 31 site of 11 was removed
using inorganic acid. Briefly, compound 11 was dissolved in

710
Tai et al.
6.00 (d, 1H), 5.51(q, 1H), 5.41 (d, 1H), 5.27 (d, 1H), 5.14(q, 1H),
4.66 (m, 2H), 4.23 (d, 1H), 4.01 (m, 4H), 3.89 (m, 2H),
3.65 (m, 2H), 3.57 (d, 1H), 3.31–3.35 (m, 4H), 3.14 (s, 3H),
2.96 (s, 3H), 2.89 (s, 3H), 2.62 (m, 1H), 2.31 (m, 2H), 0.75–
2.00 (m, 53H). ESI-MS calcd. for C57H88N4O16 1085.3;
found 1085.8 [M + H]+.
Determination of Rapamycin Loading Efficacy (wt%)
The percentage of rapamycin conjugated to polymers (wt%)
was determined by Ultraviolet–visible (UV) spectrophotome-
try. The standard curve was generated with a series of
rapamycin solution in water (λmax=280 nm). The polymer-
drug conjugates were diluted in water, and the UV absor-
bance at 280 nm was measured. Rapamycin concentration
was calculated from the standard curve. The percentage of
rapamycin conjugated to polymer (wt%) was calculated by
dividing the amount of rapamycin by the total weight of the
polymer-drug conjugate.
Synthesis of Rapamycin-Polymer Conjugate
Synthesis of Poly[bis(ε-Lys)Glut–PEG] 7. Polymer 7 was synthe-
sized by removing the benzyl groups from carboxylic acids of
polymer 6b via hydrogenation. Briefly, polymer 6b (150 mg)
was dissolved in 50 mL of methanol in nitrogen atmosphere.
Twenty milligrams of Pd/C (palladium loading 10%) was
transferred into the reaction flask under nitrogen atmosphere.
After all the Pd/C was merged into methanol, nitrogen was
changed into hydrogen, and the reaction was stirred at room
temperature overnight. The reaction mixture was then filtered
to remove Pd/C. The filtrate was concentrated and dried
under vacuum overnight to get polymer 7 in quantitative yield.
1H NMR (400 MHz, D2O) δ 4.20 (m, 1H), 4.10 (m, 2H), 3.61
(m, PEG-CH2), 3.43 (m, 2H), 3.01 (t, 2H), 2.21 (t, 2H), 1.82 (m,
1H), 1.69 (m, 1H), 1.55 (m, 2H), 1.41 9 m, 2H), 1.25 (m, 2H).
Rapamycin Release from the Polymer Conjugates
Aliquots of GlyGlyGly-Rapa and rapamycin-polymer conju-
gate 8c stock solution (10 mM in DMSO) were diluted to
200 μM in water. The solutions were immediately mixed with
100% human/rat/mouse sera, which gave a final concentra-
tion of 100 μM rapamycin in 50% sera. Forty microliters of
the solution were divided into 0.2 mL Eppendorf tubes and
incubated at 37°C. At selected time intervals, serum proteins
were precipitated with 120 μL of acetonitrile plus 0.1% TFA.
After centrifugingation at 12,000 g for 10 min, 100 μL of the
supernatant was collected and analyzed with HPLC for quan-
tification of the released rapamycin.
The HPLC system was equipped with a Shimadzu LC-
20AT pump, a SIL-10AF auto-sampler and a SPD-10A UV
detector. The separation was achieved on a Waters C18 col-
umn (250 mm×4.6 mm, 5 μm) at a flow rate of 1.0 mL/min.
The mobile phases were consisted of phase A (water, 0.1%
formic acid) and B (acetonitrile, 0.1% formic acid), with a
gradual increase of phase B from 50–90% over 20 min. The
absorbance of rapamcyin was monitored at 280 nm.
Synthesis of Rapamycin-Polymer Conjugate 8a-c. Conjugate 8a-
c was synthesized by coupling linker-drugs (rapamycin, Gly-
Rapa and GlyGlyGly-Rapa) with polymer 7 using EDCI/
Et3N/NHS. Briefly, polymer 7 (5 mg, 0.0015 mmol of repeat
unit) and linker-drug (0.006 mmol) were dissolved in 0.5 mL
of dry DMF. EDCI (10 mg), DMAP (2 mg) and NHS (1 mg)
were added into the reaction and stirred for 24 h. The poly-
mer was then precipitated with diethyl ether (20 mL). After
centrifugation, ether was poured out and the precipitate was
dissolved in 5 mL of water. The insoluble solid was removed
by filtering through 0.2 μm filters. The solution was dialyzed
against distilled water by 14 kDa MWCO membrane at room
temperature for 48 h. Dialysis water was changed every 8 h.
The solution was free-dried to yield as a white solid (yield, 41–
88%).
In Vitro Cytotoxicity
In vitro cytotoxcity of rapamycin, the polymer and rapamycin-
polymer conjugates was determined according to the method
we reported before (34). Briefly, cells were seeded into 96-well
plates at a density of 5,000 cells/well. Twenty-four hours after
incubation, a series of drug dilutions ranging from 0.001 to
1000 nM were added to the wells and cultured for another
72 h. The relative cell numbers were measured using MTT
assay. To enhance the assay sensitivity, twenty-five microliters
of Sorensen’s glycine buffer (0.1 M glycine, 0.1 M NaCl, adjust
pH to 10.5 with 0.1 N NaOH) was added into wells according
to a previous report (35). The absorbance was measured using a
DTX 880 Multimode Detector (Beckman Coulter,Inc., Fuller-
ton, CA) at the wavelength of 570 nm. IC50 was calculated by
fitting a concentration – absorbance curve using Graphpad
Prism 5 (Graphpad software Inc, La Jolla, CA).
Synthesis of Rapamycin-Polymer Conjugate 8d. Conjugate 8d
was synthesized using the same method as described above
but the coupling reagent was changed to HATU/DIPEA.
Briefly, polymer 7 (5 mg) and GlyGlyGly-Rapa 15 (6 mg)
were added into 0.5 mL of dry DMF, followed by HATU
(20 mg) and DIPEA (16 μL). The reaction was continuously
stirred at room temperature for 24 h. Twenty milliliter of
diethyl ether was poured into the reaction to precipitate the
product. The precipitate was collected by centrifugation
(4,000 g, 5 min) and re-dissolved in 5 mL of water. After
dialysis and free drying, the pure conjugate 8d yielded as a
light brown powder (yield ~ 50%).

A Novel Rapamycin-Polymer Conjugate
711
Cellular Uptake Analysis by Confocal Laser Scanning
Microscopy
grafted by one or two rapamycin (Rapa) molecules and separat-
ed from other bis(ε-Lys)Glut unit by PEG (Fig. 1b and c).
Therefore, drugs are evenly distributed along the polymer chain
and all the drugs are well separated from each other by PEG.
Compared to the drug position patterns in other polymer-drug
conjugates (polyaspartice acid and polymerglutamic acid - drug
conjugate), this unique architecture can dramatically reduce the
heterogeneity and avoid aggregation (37). Moreover, this novel
alternating, multiblock copolymer contains two drug conjuga-
tion sites in each block, which enables high drug loading capac-
ity (wt%) (32.1% for rapamycin). Rapamycin is attached onto
the polymer by tri-peptide GlyGlyGly linker which can dramat-
ically increase drug loading efficiency. Rapamycin is linked with
GlyGlyGly via a potentially cleavable ester linkage, which can be
gradually released by enzymatic/chemical hydrolysis (Fig. 1c).
Cellular uptake of the FITC-labeled polymer was performed
and analyzed by confocal laser scanning microscopy as report-
ed (36). Briefly, PC-3 cells were seeded into 8-well Lab-TekTM
chamber slides (Nunc Inc. Rochester, NY) and cultured for
24 h. The culture medium was discarded, and monolayers were
washed once with RPMI-1640 medium. Polymer-FITC was
diluted to 10 μM in RPMI-1640 and added into the chambers.
After incubation for 24 h, the cells were washed six times with
DPBS and fixed in 4% buffered formalin at room temperature
for 10 min. The cells were washed three times to remove
remaining formalin, and the nucleus was stained with TO-
PRO-3 (Excitation/Emission: 642/661 nm) according to man-
ufacturer’s protocol (Invitrogen, Grand Island, NY). After
washing, cell monolayers were mounted onto glass microscope
slides using fluorescence mounting medium (Vector Lab, Bur-
lingame, CA). The cellular uptake and distribution was ana-
lyzed using Nikon C1 laser scanning confocal microscope
(Nikon Instrument Inc., Melville, NY).
Synthesis of multiblock copolymers Poly
[bis(ε-Lys-OBn)Glut-PEG], 6a-d, and Poly[bis(ε-Lys)
Glut-PEG], 7
The multiblock copolymer Poly[bis(ε-Lys-OBn)Glut-PEG]
was synthesized by condensation of monomer NHS-
PEG3400-NHS with comonomer bis(ε-Lys-OBn)Glut 5 in
the presence of organic base. Bis(ε-Lys-OBn)Glut 5 is com-
posed of one glutaric acid and two L-lysine linked at α-amino
groups (Fig. 2). The ε-amino group of L-Lysine was less
sterically hindered and more nucleophilic than the α-amino
group, therefore making it more reactive to monomer NHS-
PEG3400-NHS. Comonomer bis(ε-Lys-OBn)Glut 5 was syn-
thesized according to the scheme in Fig. 2. Compound 3 was
synthesized from 1 according to Agnihotri’s report (30).
Glutaric acid was coupled with at least 2 equivalents of com-
pound 3 using EDCI/DMAP. Compound 4 was carefully
purified by silica gel flash chromatography and treated with
TFA/DCM to get comonomer bis(ε-Lys-OBn)Glut 5. Poly-
merization was carried out in anhydrous DMF in nitrogen
atmosphere. After the polymerization was initialized by or-
ganic base Et3N, the viscosity of the solution increased dra-
matically in less than 1 h. The reaction mixture became very
viscous after 24 h, which indicated successful elongation of the
polymer chain. As shown in Table I, polymerization at 50°C
can produce higher Mw than that at 25°C. It might be
because that high temperature can reduce the viscosity of
the polymerization solution and therefore increase the con-
densation efficiency (38). Polymer Mw reduced to around
20 kDa when inorganic base NaHCO3 was used. Although
both Et3N and NaHCO3 can deprotonate the ε-amino groups
of bis(ε-Lys-OBn)Glut 5, NaHCO3 is partially insoluble in
DMF, making it less efficient in polymerization. Polymer 6a-
d are highly soluble in water, THF and chloroform, but
insoluble in diethyl ether and petroleum ether.
Aqueous Solubility Assay
The aqueous solubilities of rapamycin and its conjugates were
assayed by the thermodynamic method as reported before
(34). Briefly, excess amount of rapamycin or its conjugates
were suspended in 100 μL of distilled water. The suspension
was shaken at room temperature for 24 h. The suspension was
then centrifuged at 12,000 g for 15 min to remove undissolved
drug. The supernatant was collected, and the concentration
was quantified by UV spectrometry at 280 nm.
Statistical Analysis
Data were expressed as the mean standard deviation (SD).
Difference between any two groups was determined by
ANOVA. P <0.05 was considered statistically significant.
RESULTS
Novel Engineering of Rapamycin-Polymer Conjugate
In prior studies, rapamycin was specifically conjugated with
small molecular-weight moieties to increase its hydrophilicity
and specificity (CCI-779, APA23573 and RAD001, Fig. 1a).
Although the oral bioavailability was enhanced, the solubility
and biodistribution profile were only slightly improved. In this
paper, a rapamycin-polymer conjugate was engineered by con-
jugating rapamycin with a novel, PEG based, multiblock copol-
ymer. The architecture of the polymer is composed of alternat-
ing PEG and bis(ε-Lys)Glut units. Each bis(ε-Lys)Glut unit is
To graft rapamycin onto polymer, polymer 6b was
deprotected by hydrogenation to liberate the carboxyl groups

712
Tai et al.
a
c
b
d
1
2
3
4
e
bis(ε-Lys-OBn)Glut, 5
Poly[bis(ε-Lys-OBn)Glut-PEG], 6a-d
Fig. 2 Synthesis of Poly[bis(ε-Lys-OBn)Glut-PEG]. Reagents and conditions: (a) Benzyl alcohol, EDCI, DMAP; (b) 20% piperidine in DCM; (c) Glutaric acid,
EDCI, DMAP; (d) TFA/DCM, rt; (e) Et3N, DMF.
(Fig. 3). Polymer 7 is a novel, linear multiblock copolymer
containing two carboxyl groups for drug conjugation
in each block. This polymer is also soluble in water,
DMF, and methanol, slightly soluble in CHCl3, but
insoluble in THF and diethyl ether.
31-OH of rapamycin was selectively protected with TMS
group according to the procedure described by Shaw and
Nan (31,32). Trt-Gly-OH was then conjugated with 42-OH
using a conventional coupling method. After removing TMS
at 31-OH in diluted inorganic acid H2SO4, Trt-Gly-Rapa
was deprotected with 0.1 M HOBT in TFE to get Gly-Rapa
13. Routine Trt deprotection methods, such as 1% TFA in
DCM, were inapplicable because rapamycin is very sensitive
to TFA. Gly-Rapa 13 was further coupled with Trt-Gly-Gly-
OH using EDCI/Et3N to give compound 14. GlyGlyGly-
Rapa 15 was finally obtained by removing the Trt group from
compound 14.
Linker-drugs Gly-Rapa 13 and GlyGlyGly-Rapa 15 were
conjugated with polymer 7 using the same coupling method
as the synthesis of 8a. Conjugate 8b was obtained with a yield
of 65–88%, and the weight percentage (wt%) of rapamycin in
the polymer-drug conjugate was 19.5%. The highest weight
percentage of rapamycin in polymer-drug conjugates was
observed in conjugate 8c (wt% ~ 27.3%) where around
80% of carboxyl groups in polymer 7 were conjugated with
rapamycin. There was no significant difference in drug load-
ing capacity when other coupling methods, such as HATU/
DIPEA, were used (conjugate 8d, Table II). Therefore, the
conjugation efficiency is mainly determined by linkers rather
than coupling methods.
Synthesis of the Rapamycin-Polymer Conjugate
Rapamycin was first conjugated with polymer 7 by forming
an ester bond between its hydroxyl groups and polymer’s
carboxyl groups in the presence of EDCI, Et3N and NHS.
Conjugate 8a was purified by dialysis with a yield around
67%. However, conjugate 8a had very low drug loading
capacity (wt% ~ 0.48%). If all the carboxyl groups were
conjugated with rapamycin, the theoretical maximal
drug loading capacity (wt%) could be as high as
34.1%. Therefore, the hydroxyl group of rapamycin
only reacted with around 1.4% of all the carboxyl
groups in polymer 7. The low coupling efficiency could
be explained partially by the steric hindrance between
the polymer and rapamycin. To reduce the steric hin-
drance, two linkers, Gly and GlyGlyGly, were utilized
to link polymer 7 and rapamycin.
Linker-drugs were synthesized according to the scheme in
Fig. 4. In order to selectively conjugate the linkers at 42-OH,
Tab le I Polymerization of the
Polymer
Base
Temperature
Reaction Iime
Mw (kDa)
Mn (kDa)
Mw/Mn
Yield (%)
Monomer NHS-PEG3400-NHS
with Comonomer bis(ε-Lys-
OBn)Glut in Different Conditions
6a
6b
6c
6d
Et3N
50°C
50°C
25°C
50°C
2 h
20.9
31.5
22.8
19.9
15.2
21.9
16.1
14.1
1.37
1.43
1.41
1.42
71
59
66
79
Et3N
24 h
24 h
24 h
Et3N
NaHCO3

A Novel Rapamycin-Polymer Conjugate
713
a
b
6b
Poly[bis(ε-Lys)Glut-PEG], 7
Polymer-Rapamycin conjugate, 8a-d
R:
8a
8b
8c-d
Fig. 3 Synthesis of the rapamycin-polymer conjugate. Reagents and conditions: (a) H2, Pd/C in methanol; (b) EDCI, Et3N, NHS.
Characterization of Rapamycin-Polymer Conjugate
conjugate 8c also shows strong absorbance at near ultraviolet
region (200–230 nm), which is contributed mainly by polymer 7
and minimally by rapamycin (Fig. 5). The carboxyl groups in
polymer 7 and double bonds in rapamycin are believed to be
responsible for the UV absorbance at 200~230 nm.
The UV spectrum of conjugate 8c was scanned over the range
of 190 to 400 nm and compared with free rapamycin. As shown
in Fig. 5, The UV spectrum of rapamycin exhibits a maximum
absorbance peak at 280 nm and two small peaks on either side.
Similar peaks were also observed in the UV spectrum of purified
conjugate 8c, which indicates that rapamycin is conjugated to
the polymer chains. Besides three peaks around 280 nm,
Proton NMR (1H NMR) spectroscopy was also determined to
prove the presence of rapamycin in the rapamycin-polymer
1
conjugate 8c. Figure 6a shows the H NMR spectrum of
rapamycin. The seven groups of peaks (a - g) at low field
d
c
a
b
Rapamycin
9
10
11
e
f
e
12
Gly-Rapa, 13
14
GlyGlyGly-Rapa,15
Fig. 4 Synthesis of GlyGlyGly-Rapa. Reagents and conditions: (a) TMSCl, imidazole, DCM; (b) imidazole/imidazole·HCl, DCM; (c) Trt-Gly-OH, EDCI,
DMAP, DCM; (d) THF/2 N H2SO4 (v/v=2/1); (e) 0.1 M HOBT in TFE; (f) Trt-Gly-Gly-OH, EDCI, Et3N, DCM.

714
Tai et al.
Table II Properties of the
Rapamycin-Polymer Conjugates
Polymer-drug
conjugate
Conjugation
reagents
Linker-drug
Theoretical maximal
drug conjugated (wt%)
Drug conjugated
(wt%)
Yield (%)
8a
8b
8c
8d
EDCI/Et3N/NHS
EDCI/Et3N/NHS
EDCI/Et3N/NHS
HATU/DIPEA
Rapamycin
34.1
33.4
32.1
32.1
0.68 0.18
19.5 1.8
27.3 5.1
24.9 3.1
57~71
65~88
41~63
50~64
Gly-Rapa
GlyGlyGly-Rapa
GlyGlyGly-Rapa
corresponds very well to the protons of double bonds (a–e) and
carbonyl groups (f, g) in rapamycin. The single peaks (h–j),
between 3.0 and 3.5 ppm, corresponds to the presence of three
~20%, ~ 9% and ~3% of rapamycin were released from
conjugate 8c in mouse, rat and human serum, respectively.
The bulky polymer structure in polymer-rapamycin conjugate
dramatically increases the steric hindrance of the ester linker
between rapamycin 42-OH and GlyGlyGly, which makes ester
bond less accessible by esterase and therefore decreases the
hydrolysis rate (40). In addition, the release of rapamycin from
polymer-rapamycin reached its plateau at 1 h and 2 h in mouse
and rat serum, respectively. These premature plateau-release
profiles can be attributed to the instability of rapamycin in sera.
The half-life of rapamycin in rat plasma has been reported to be
only 2.2 h, and a shorter half-life would be expected in mouse
plasma due to its higher metabolic rate (39,41). Therefore, the
plateau was reached when the degradation rate of rapamycin
was equal to its release rate.
1
CH3O- groups in rapamycin. The H NMR spectrum of
rapamycin-polymer conjugate 8c (Fig. 6c) contains peaks
representing both rapamycin and the polymer chain. Peaks a -
j in Fig. 6c show the same chemical shifts (ppm) and splitting
patterns as that in Fig. 6a, indicating the presence of rapamycin
in the conjugate 8c. While the polymer chain in the conjugate 8c
corresponds to the peaks a’ - c’, in which peaks a’ - b’ represent
bis(ε-Lys)Glut and peak c’ represents PEG3400 (Fig. 6b).
Release of Rapamycin from Rapamycin Conjugates
The release of rapamycin from two conjugates GlyGlyGly-Rapa
and conjugate 8c was evaluated in mouse, rat and mouse serum.
Rapamycin conjugates were incubated with 50% of different
serum at 37°C for 4 h. The release profile of rapamycin was
monitored using HPLC. In mouse serum, the half-life of releas-
ing rapamycin from GlyGlyGly-Rapa is approximately 0.5 h. In
contrast, the half-life increases to 4 h in rat serum. A minimal
release was observed in human serum where only 10% of
rapamycin was released after 4 h (Fig. 7). The species-
dependent release profile is due to the metabolic differences
between species (39). The similar phenomenon was also observed
in Wortmannin-Rapamycin conjugate (27). The release of
rapamycin from rapamycin-polymer conjugate 8c is much
slower than that of GlyGlyGly-Rapa. After 4-h incubation, only
In Vitro Activity of Rapamycin and its Conjugates
The in vitro cytotoxicity of polymer 7, rapamycin, GlyGlyGly-
Rapa and conjugate 8c was determined in prostate, breast,
and cervical cancer cell lines using a MTT assay. No obvious
cytotoxicity (IC50>100 μM) was observed in polymer 7 treat-
ed cells, indicating that polymer 7 is not toxic to these cancer
cell lines. Rapamycin showed potent in vitro cytotoxicity to all
cancer cell lines (Table III). The IC50 values of rapamycin in
all seven cancer cell lines ranged from 0.1 to 5.2 nM. Com-
pared to breast cancer cell lines, prostate and cervical cancer
cell lines were much more sensitive to rapamycin treatment.
GlyGlyGly-Rapa was 2- to 36-fold less potent than rapamycin
with the exception of MCF-7/HER2, in which similar poten-
cy was observed. Given the fact that GlyGlyGly-Rapa is an
ester prodrug of rapamycin, and hydrolysis is required for its
activity, it is reasonable that GlyGlyGly-Rapa shows lower
potency than rapamycin. The IC50 values of conjugate 8c are
in a range of 25–50 nM. Although it is less potent than free
rapamycin, conjugate 8c still exhibits low nanomolar IC50
values against all cancer cell lines tested.
Polymer-Rapamycin conjugate 8c
Poly[bis(ε-Lys)Glut-PEG] 7
Rapamycin
Blank
Cellular Uptake and Distribution
of Rapamycin-Polymer Conjugates
FITC ethylenediamine (FITC-NH2) was conjugated to poly-
mer 7 using the same coupling method (EDCI/NEt3/NHS)
as conjugate 8a-c. The weight percentage (wt%) of FITC in
Wavelength (nm)
Fig. 5 UV spectra of rapamycin, rapamycin-polymer conjugate 8c and
polymer 7.

A Novel Rapamycin-Polymer Conjugate
715
1H NMR of rapamycin
a
1H NMR of Polymer 7
b
1H NMR of polymer-rapamycin conjugate 8c
c
Fig. 6 1H NMR spectra of rapamycin (a), polymer 7 (b) and rapamycin-polymer conjugate 8c (c) in CDCl3 solvent.
the labeled polymer 7 is 4.5% (Data not shown). To deter-
mine the cellular uptake and distribution, the FITC-labeled
polymer was incubated with PC-3 cells in serum-free medium
for 24 h. Punctuated staining pattern was observed in PC-3
cells after 24-h treatment (Fig. 8). Other fluorescently labeled
polymers, such as β-cyclodextrin polymer and PEG, were
Fig. 7 The release profiles of
rapamycin from GlyGlyGly-Rapa
(a) and conjugate 8c (b) in different
sera.

716
Tai et al.
Tab le I II In Vitro Cytotoxicity (IC50, nM). The Cancer Cells were Treated
Table IV Aqueous Solubility of Polymer 7, Rapamycin and Conjugate 8c
with the Compounds for 72 h Before MTTAssay. Values Shown are mean
SD
Compound
Aqueous solubility
(mg/mL)
Aqueous solubility
(Rapamycin equivalents)
Cancer
type
Cell line
Polymer
7a
Rapamycin GlyGlyGly- Conjugate
Rapa
8cb
mg/mL
mM
Prostate LNCaP
PC-3
>100,000 0.9 0.9
>100,000 0.8 0.7
>100,000 0.1 0.5
>100,000 4.4 0.3
4.1 0.3
51.2 0.6
29.9 0.7
30.6 0.6
40.7 0.7
26.3 0.4
25.9 0.5
Polymer 7
100.332 22.278
0.034 0.006
−/−
−/−
2.3 0.6
5.6 0.5
9.2 0.6
Rapamycin
Conjugate 8c
0.034 0.006
21.061 3.285
0.037 0.007
23.014 3.594
DU145
78.000 12.165
Breast
SK-BR-3
MCF-7
>100,000 5.2 0.3 10.3 0.5
>100,000 3.3 0.2 3.9 0.4
rapamycin. The aqueous solubility of the rapamycin-polymer
conjugate 8c is around 78 mg/mL.
MCF-7/
HER2
Cervical HeLa
>100,000 1.0 0.1 36.5 0.1
47.4 0.1
a IC50 is calculated based on molar concentration of repeated units
b IC50 of polymer rapamycin conjugate 8c is expressed in terms of rapamycin
equivalents
DISCUSSION
Rapamycin has been examined alone or in combination with
other drugs for treatment of various cancers in clinical studies
(3–5). Although it has shown promising therapeutic effects, its
clinical development was interrupted by poor aqueous solu-
bility and preferential distribution in RBCs. Various formula-
tions and conjugates, such as cosolvent system, polymeric
micellar formulation and some rapalogues, have been devel-
oped to overcome the limitations (8,15,17). All these ap-
proaches can only partially increase the solubility, but has
little effect on the blood distribution and pharmacokinetics.
Therefore, the objective of this study is to develop a
rapamycin-polymer conjugate that not only enhance the sol-
ubility but also improve the biodistribution profiles of
rapamycin.
reported to share the similar cellular uptake and distribution
pattern (36,42). Polymers are believed to enter cells via endo-
cytosis because the punctated vesicular distribution of fluores-
cence indicated a lysosomal localization. The uptake of FITC-
NH2 in PC-3 cells was also examined. Because FITC-NH2 is a
membrane non-permeable fluorescent dye, no fluorescence
was observed even after 24-h incubation (Fig. 8) (43).
Conjugation with Polymer Dramatically Enhanced
the Solubility of Rapamcyin
Due to the hydrophilic PEG component, polymer 7 is highly
soluble in water (Table IV). The aqueous solubility of
rapamycin is only around 0.034 mg/mL at room tempera-
ture. After conjugated to polymer 7, its solubility increases to
21.061 mg/mL which is more than 600 times higher than free
In this study, rapamycin was conjugated with a novel,
linear, PEG containing multiblock copolymer. The polymer
was synthesized via the condensation of a difunctionalized
PEG monomer (NHS-PEG3400-NHS) with a comonomer
Fig. 8 Cellular uptake of FITC
FITC
TO-PRO-3
Merge
labeled polymer (Polymer-FITC)
and FITC ethylenediamine (FITC-
NH2) by confocal microscopy. PC-
3 cells were treated with Polymer-
FITC and FITC-NH2 in serum free
RPMC-1640 media for 24 h. Cell
nucleus was stained with TO-PRO-
3 and imaged in two fluorescent
channels.

A Novel Rapamycin-Polymer Conjugate
717
bis(ε-Lys-OBn)Glut 5 containing two pendant carboxyl
groups for drug conjugation. Comonomer bis (ε-Lys-
OBn)Glut 5 is small in size and symmetric in structure, which
is expected to enhance the condensation efficiency. The purity
of the comonomer has a significant impact on achieving high
Table II, directly conjugating rapamycin to the polymer can
only load a small amount of drug (wt%~0.68). In contrast, the
drug loading capacities increased to 19.5% and 27.3%, re-
spectively, by inserting Gly and GlyGlyGly between
rapamcyin and the polymer. The use of longer linker, such
as GlyGlyGly, resulted in a higher drug loading capacity
because the long and flexible linker may reduce the steric
hindrance between the polymer and rapamycin during cou-
pling reaction. The resulting polymer rapamycin conjugate
8c dissolves freely in water. Its aqueous solution is stable and
does not precipitate/change color in the absence of strong
ionic strength and light.
The most striking observation of this study is the great
solubility of rapamycin after conjugating with the polymer.
Rapamycin is practically insoluble in water (0.034 mg/mL),
but its solubility increases to 21.061 mg/mL after conjugating
with the polymer. This solubility meets the requirement for i.v.
formulations in animal studies, in which the minimal dose
concentration is 3~6 mg/mL, and clinical studies, in which
the minimal dose concentration is 1 mg/mL (8). The back-
bone of polymer 7 contains repeating units of two compo-
nents, PEG3400 and bis(ε-Lys)Glut, in which PEG3400 attri-
butes to the aqueous solubility. Moreover, PEG3400 is also
expected to reduce the toxicity and immunogenicity of the
polymer (44). As shown in Table I, polymer 7 does not show
obvious toxicity in various cultured tumor cells.
The release of rapamycin from the polymer-drug conju-
gate in the presence of serum was observed (Fig. 7). The
polymer-rapamycin conjugate exhibits different release rates
in the serum of different species, which indicates that the
release of rapamycin is mainly mediated by enzymatic hydro-
lysis other than pure chemical hydrolysis. Bulk structure of the
polymer dramatically hinders the enzymatic hydrolysis and
therefore slowdowns the rapamycin release in the serum,
which can avoid the redistribution of rapamycin into RBCs.
Same to other polymer-drug conjugates, the rapamycin-
polymer conjugate is taken up by tumor cells via the endocy-
totic pathway (Fig. 8) (36,42). In the intracellular compart-
ment endosomes and lysosomes, rapamycin-polymer would
be attacked by peptidase/esterase and release the drug
rapamycin.
In conclusion, a novel, linear, PEG-based multiblock
copolymer was synthesized and utilized as a carrier for
hydrophobic drug rapamycin. The rapamycin-polymer
conjugate shows significantly increased solubility in wa-
ter and potent cytotoxicity against a panel of cancer cell
lines. It also demonstrates that the polymer-drug conju-
gate can be taken up by tumor cells. The result of the
uptake study indicates that the lysosome is the main site
of the intracellular localization of the polymer-drug
conjugate. Thus, in this study we reported a novel
rapamycin conjugate which holds great promising for
treatment of a wide variety of tumors.
MW polymers. The presence of impurities can terminate the
chain elongation, leading to low MW polymers (26). After
precipitation in diethyl ether twice, bis(ε-Lys-OBn)Glut was
obtained with a purity of ~ 97%. Beside impurity, moisture
can also dramatically decrease the MW of the polymer. Poly-
merization in non-drying solvent or with non-drying comono-
mer would only generate short polymer chains with MW of
~18 kDa (data not shown).
This study represents the first attempt to conjugate
rapamycin with a polymer. Rapamycin is a structurally com-
plicated natural product which contains three hydroxyl
groups. Although the tertiary alcohol at position 13 is highly
hindered and unreactive, rapamycin contains two secondary
hydroxyl groups at position 31 and 42 which are accessible for
esterification and etherification. Directly reacting with
acrylating agents would produce a mixture of 31,42-bis-
acrylated product and 31- or 42-monoacrylated products,
which results in inevitable heterogeneity of the rapamycin-
polymer conjugate. In this study, we achieved the
regioselective conjugation of rapamycin at 42-OH by first
protecting the 31-OH with TMS and then synthesizing 42-
monoacrylated rapamycin via esterification. Another chal-
lenge of rapamycin conjugation is its instability in both acidic
and basic conditions which make it intolerable to many chem-
ical reactions. Gly-Rapa was firstly attempted to synthesize
from Fmoc-Gly-Rapa, but rapamycin was quickly degraded
when deprotecting Fmoc in 10% piperidine in DCM. Re-
moving Boc from Boc-Gly-Rapa was also problematic due to
its sensitivity to TFA (even 1% TFA in DCM). Gly-Rapa was
finally obtained with a very low yield (~ 10%) from Trt-Gly-
Rapa by deprotecting Trt in THF/2 N H2SO4 (v/v=2/1,
overnight). However, the yield was dramatically increased to
60~80% when the deprotection method was switched to an
uncommonly used, but mild condition (0.1 N HOBT in TFE,
10 min).
Conjugation of rapamycin to the polymer was achieved by
coupling a linker to rapamycin at position 42-OH, and sub-
sequently grafting the linker-Rapa at the pendant carboxyl
groups of this multiblock copolymer. The expected mecha-
nism of drug release from the rapamycin-polymer conjugate is
that tumor esterase would hydrolyze the 42-ester bond and
release free rapamycin from the polymer backbone. Insertion
of linkers, such as Gly or GlyGlyGly, between rapamycin and
the polymer backbone is believed to accelerate the drug
release (25,26,37). Moreover, these amino acids or peptide
linkers can transform the hydroxyl group of rapamycin at
position 42 to a more reactive amino end group, which could
dramatically increase the drug loading capacity. As shown in

718
Tai et al.
19. Goudar RK et al. Combination therapy of inhibitors of epidermal
growth factor receptor/vascular endothelial growth factor receptor 2
(AEE788) and the mammalian target of rapamycin (RAD001) offers
improved glioblastoma tumor growth inhibition. Mol Cancer Ther.
2005;4(1):101–12.
20. Fetterly GJ, Mita MM, Britten CD, Poplin E, Tap WD, Carmona A,
et al. Pharmacokinetics of oral deforolimus (AP23573, MK-8669).
2008 ASCO Annual Meeting, 2008.
ACKNOWLEDGMENTS AND DISCLOSURES
This project has been supported by two awards
(R21CA143683 and R01AA021510) from the National Insti-
tute of Health (NIH). The authors gratefully acknowledge Lu
Jin, Bin Qin and Ravi S. Shukla for helpful discussion and
techical assistance.
21. Rubino JTS, Siskavich V, Harrison MM, Gandhi P. Parenteral CCI-
779 formulations containing cosolvents, an antioxidant, and a sur-
factant. US Patent. 2011;8026276.
22. Laplanche R, Meno-Tetang GM, Kawai R. Physiologically based
pharmacokinetic (PBPK) modeling of everolimus (RAD001) in rats
involving non-linear tissue uptake. J Pharmacokinet Pharmacodyn.
2007;34(3):373–400.
REFERENCES
23. Li C et al. Complete regression of well-established tumors using a
novel water-soluble poly(L-glutamic acid)-paclitaxel conjugate. Can-
cer Res. 1998;58(11):2404–9.
24. Li C et al. Biodistribution of paclitaxel and poly(L-glutamic acid)-
paclitaxel conjugate in mice with ovarian OCa-1 tumor. Cancer
Chemother Pharmacol. 2000;46(5):416–22.
1. Vezina C, Kudelski A, Sehgal SN. Rapamycin (AY-22,989), a new
antifungal antibiotic. I. Taxonomy of the producing streptomycete
and isolation of the active principle. J Antibiot (Tokyo).
1975;28(10):721–6.
2. Garber K. Rapamycin's resurrection: a new way to target the cancer
cell cycle. J Natl Cancer Inst. 2001;93(20):1517–9.
25. Sapra P et al. Novel delivery of SN38 markedly inhibits tumor growth
in xenografts, including a camptothecin-11-refractory model. Clin
Cancer Res. 2008;14(6):1888–96.
3. Zhao H, Cui K, Nie F, Jin G, Li F, Wu L, et al. Effects of rapamycin
on Breast Cancer cell migration through the cross-talk of MAPK
pathway. Cancer Res. 2009;69(24 suppl).
26. Cheng J et al. Synthesis of linear, beta-cyclodextrin-based
polymers and their camptothecin conjugates. Bioconjug
Chem. 2003;14(5):1007–17.
4. Armstrong AJ et al. A pharmacodynamic study of rapamycin in men
with intermediate- to high-risk localized prostate cancer. Clin Cancer
Res. 2010;16(11):3057–66.
27. Ayral-Kaloustian S et al. Hybrid inhibitors of phosphatidylinositol 3-
kinase (PI3K) and the mammalian target of rapamycin (mTOR): design,
synthesis, and superior antitumor activity of novel wortmannin-
rapamycin conjugates. J Med Chem. 2010;53(1):452–9.
28. Umeda N et al. A photocleavable rapamycin conjugate for spatiotempo-
ral control of small GTPase activity. J Am Chem Soc. 2011;133(1):12–4.
29. Naicker S, Sharma SK,Thomas W. Rapamycin peptides conjugates:
synthesis and uses thereof. Patent. 2004. US Patent: 7659244.
30. Agnihotri G et al. Structure-activity relationships in nucleotide oligomer-
ization domain 1 (Nod1) agonistic gamma-glutamyldiaminopimelic acid
derivatives. J Med Chem. 2011;54(5):1490–510.
5. Cloughesy TF et al. Antitumor activity of rapamycin in a Phase I trial
for patients with recurrent PTEN-deficient glioblastoma. PLoS Med.
2008;5(1):e8.
6. Phung TL et al. Pathological angiogenesis is induced by sustained Akt
signaling and inhibited by rapamycin. Cancer Cell. 2006;10(2):159–
70.
7. Bruns CJ et al. Rapamycin-induced endothelial cell death and tumor
vessel thrombosis potentiate cytotoxic therapy against pancreatic
cancer. Clin Cancer Res. 2004;10(6):2109–19.
8. Simamora P, Alvarez JM, Yalkowsky SH. Solubilization of
rapamycin. Int J Pharm. 2001;213(1–2):25–9.
31. Shaw C-C, Sellstedt J, Noureldin R, Cheal GK, Fortier G.
Regioselective synthesis of rapamycin derivatives. 2001. US Patent
6,277,983.
32. Fajun Nan JD, Zuo J, Yu L, Meng L, Zhang Y, Yang N, et al.
Rapamycin carbonic ester analogues, pharmaceutical compositions,
preparations and uses thereof. 2011. USPTO Applicaton No.
20110166172.
33. Bodanszky M, Bednarek MA, Bodanszky A. Coupling in the ab-
sence of tertiary amines. Int J Pept Protein Res. 1982;20(4):387–95.
34. Tai W et al. Development of a peptide-drug conjugate for prostate
cancer therapy. Mol Pharm. 2011;8(3):901–12.
35. Plumb JA, Milroy R, Kaye SB. Effects of the pH dependence of 3-
(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide-
formazan absorption on chemosensitivity determined by a novel
tetrazolium-based assay. Cancer Res. 1989;49(16):4435–40.
36. Cheng J, Khin KT, Davis ME. Antitumor activity of beta-
cyclodextrin polymer-camptothecin conjugates. Mol Pharm.
2004;1(3):183–93.
9. Yatscoff R et al. Blood distribution of rapamycin. Transplantation.
1993;56(5):1202–6.
10. Ferron GM, Conway WD, Jusko WJ. Lipophilic benzamide and
anilide derivatives as high-performance liquid chromatography in-
ternal standards: application to sirolimus (rapamycin) determination.
J Chromatogr B Biomed Sci Appl. 1997;703(1–2):243–51.
11. Serajuddin AT. Salt formation to improve drug solubility. Adv Drug
Deliv Rev. 2007;59(7):603–16.
12. Sun M et al. The influence of co-solvents on the stability and
bioavailability of rapamycin formulated in self-microemulsifying
drug delivery systems. Drug Dev Ind Pharm. 2011;37(8):986–
94.
13. Gallant-Haidner HL et al. Pharmacokinetics and metabolism of
sirolimus. Ther Drug Monit. 2000;22(1):31–5.
14. Napoli KL et al. Distribution of sirolimus in rat tissue. Clin Biochem.
1997;30(2):135–42.
15. Yanez JA et al. Pharmacometrics and delivery of novel nanoformulated
PEG-b-poly(epsilon-caprolactone) micelles of rapamycin. Cancer
Chemother Pharmacol. 2008;61(1):133–44.
16. Serkova N et al. Assessment of the mechanism of astrocyte swelling
induced by the macrolide immunosuppressant sirolimus using multi-
nuclear nuclear magnetic resonance spectroscopy. Chem Res
Toxicol. 1997;10(12):1359–63.
37. Bhatt R et al. Synthesis and in vivo antitumor activity of poly(l-glutamic
acid) conjugates of 20S-camptothecin. J Med Chem. 2003;46(1):190–3.
38. Liu C-Y et al. New linearized relation for the universal viscosity −
temperature behavior of polymer melts. Macromolecules.
2006;39(25):8867–9.
39. Decarie A et al. Serum interspecies differences in metabolic pathways
of bradykinin and [des-Arg9]BK: influence of enalaprilat. Am J
Physiol. 1996;271(4 Pt 2):H1340–7.
40. Chandran SS et al. A prostate-specific antigen activated N-(2-
hydroxypropyl) methacrylamide copolymer prodrug as dual-
17. Benjamin D et al. Rapamycin passes the torch: a new generation of
mTOR inhibitors. Nat Rev Drug Discov. 2011;10(11):868–80.
18. Hidalgo M, Rowinsky EK. The rapamycin-sensitive signal
transduction pathway as a target for cancer therapy. Onco-
gene. 2000;19(56):6680–6.

A Novel Rapamycin-Polymer Conjugate
719
targeted therapy for prostate cancer. Mol Cancer Ther.
2007;6(11):2928–37.
delivery of oligonucleotides. J Control Release. 2003;89(3):365–
74.
41. Ferron GM, Jusko WJ. Species differences in sirolimus stabil-
ity in humans, rabbits, and rats. Drug Metab Dispos.
1998;26(1):83–4.
43. Sun Y et al. Intracellular labeling method for chip-based capillary
electrophoresis fluorimetric single cell analysis using liposomes. J
Chromatogr A. 2006;1135(1):109–14.
42. Murthy N et al. Design and synthesis of pH-responsive poly-
meric carriers that target uptake and enhance the intracellular
44. Veronese FM, Mero A. The impact of PEGylation on biological
therapies. BioDrugs. 2008;22(5):315–29.