Enzymic synthesis and biological evaluation of injectable glutathione-everolimus

Article abstract of DOI:10.1007/s00044-017-2084-6

An enzymic synthesis of glutathione-everolimus is reported. This process has been optimized and scaled up with high reproducibility and yields, which will facilitate the development of such conjugate. The stability of the conjugate supported that this prodrug can be prepared into lyophilized solid, which is to be reconstituted with 0.9% sodium chloride for injection before intravenous infusion. And the results of species-related drug release experiment displayed that the performance of the conjugate in human plasma, rat and monkey was similar. Moreover, the in vivo efficacy of glutathione-everolimus in the treatment of renal cell carcinoma was investigated in detail. The conjugate was proved to be an effective, safe and well-tolerated injectable prodrug in the treatment of renal cell carcinoma. The results indicated that three times injection with a high dosage in 1 week can achieve much better in vivo efficacy, and no obvious toxic response was observed.

Full text of DOI:10.1007/s00044-017-2084-6

MEDICINAL
CHEMISTRY
RESEARCH
Med Chem Res (2018) 27:583–591
DOI 10.1007/s00044-017-2084-6
ORIGINAL RESEARCH
Enzymic synthesis and biological evaluation of injectable
glutathione-everolimus
1,2
Xiaohe Zheng2 Lifei Mao2 Liang Qin2 Tianmin Zhu2
●
●
●
●
Haibo Wang
Received: 6 June 2017 / Accepted: 21 September 2017 / Published online: 23 November 2017
© Springer Science+Business Media, LLC 2017
●
●
●
Abstract An enzymic synthesis of glutathione-everolimus
is reported. This process has been optimized and scaled up
with high reproducibility and yields, which will facilitate
the development of such conjugate. The stability of the
conjugate supported that this prodrug can be prepared into
lyophilized solid, which is to be reconstituted with 0.9%
sodium chloride for injection before intravenous infusion.
And the results of species-related drug release experiment
displayed that the performance of the conjugate in human
plasma, rat and monkey was similar. Moreover, the in vivo
efficacy of glutathione-everolimus in the treatment of renal
cell carcinoma was investigated in detail. The conjugate
was proved to be an effective, safe and well-tolerated
injectable prodrug in the treatment of renal cell carcinoma.
The results indicated that three times injection with a high
dosage in 1 week can achieve much better in vivo efficacy,
and no obvious toxic response was observed.
Keywords Everolimus Glutathione Renal cell carcinoma
●
●
Prodrug Drug delivery mTOR
Introduction
Renal cell carcinoma (RCC) is the most common primary
malignant tumor of the kidney (Eble et al. 2004), account-
ing for approximately 85% of kidney carcinomas (Arai and
Kanai 2010). And the incidence of RCC is still increasing at
a rate of 2–3% per decade in most countries (Shirotake et al.
2016). Everolimus and temsirolimus (Fig. 1), two potent
mammalian target of rapamycin (mTOR) inhibitors, were
approved by the FDA for the treatment of advanced RCC
(Sherman et al. 2015). Observational retrospective data
suggested that temsirolimus, as an weekly and intrave-
nously administered agent, exhibited lots of advantages
compared to oral everolimus, including financial con-
siderations, assurance of patient compliance given its
intravenous administration, its toxicity profile, patient per-
formance status, and patient or physician preference
(Stenner-Liewen et al. 2013; Mackenzie et al. 2011; Wei-
kert et al. 2013; Alasker et al. 2013; Patel et al. 2012). For
those reasons, injectable temsirolimus was widely used as a
succedaneum of oral everolimus in the clinical therapy for
advanced cancer patients. However, more than eight species
of excipient for assisting solubilization, including ethanol
and castor oil, are used in the formulation of temsirolimus,
which lead to much adverse reactions (Coiffier 2013; Farag
et al. 2009).
Electronic supplementary material The online version of this article
(https://doi.org/10.1007/s00044-017-2084-6) contains supplementary
material, which is available to authorized users.
Haibo Wang and Xiaohe Zheng contributed equally to this work and
are co-first authors.
* Tianmin Zhu
tzhu@hisunpharm.com
Haibo Wang
wanghb616@163.com
Up to now, no injectable everolimus-based drug has been
developed in the market because of its poor water solubility.
Although well-tolerated profile, efficacy and safety of
everolimus in treating advanced RCC have been proved in
clinical trials (Motzer et al. 2010), there are limited for oral
1
Zhejiang Hongyuan pharmaceutical Co., Ltd., Linhai 317016
Zhejiang, China
2
Zhejiang Hisun pharmaceutical Co., Ltd., Taizhou 318000
Zhejiang, China

584
Med Chem Res (2018) 27:583–591
administration when treating patients with dysphagia
induced by dental ulcer and mucous irritation, which are the
most common side effects of long-term intake of everolimus
(Motzer et al. 2010). In addition, daily treatment is thought
to limit day-to-day variability in exposure attributable to its
low bioavailability (Crowe et al. 1999) and variable
absorption (Kirchner et al. 2004; Kovarik et al. 2002),
which lead to treatment failure and short survival.
glutathione-everolimus, showed good water solubility and
better in vivo efficacy (Wang et al. 2017). As a part of our
ongoing researches towards the development of this con-
jugate, in this paper, we report an enzymic synthesis of
glutathione-everolimus and evaluation of the injectable
glutathione-everolimus in the treatment of human renal cell
carcinoma in detail.
Thus, an injectable everolimus-based drug is required for
enhancement of advanced patient compliance and
improvement in therapeutic efficacy.
Results and discussion
Synthesis
Some approaches to circumvent the problem of water
solubility of water insoluble agents have been to conjugate
the drug directly to a water-soluble polymer such as
hydroxypropyl methacrylate (Tomalova et al. 2016), poly-
ethyleneglycol (PEG) (Zhao et al. 2008), and cyclodextrins
polymer (CDP) (Ryan 2013). For example, PEGylated
everolimus had been successful in solubilizing the ther-
apeutic agent (Zhu et al. 2002). An alternative solution to
drug delivery has been offered in our previous work, in
which an endogenous tripeptide, glutathione, was used
(Wang et al. 2016, 2017). The reported conjugate,
In our previous work, iodoacetate of everolimus was used
as an intermediate in the synthesis of glutathione-
everolimus. As we known, α-iodoaliphatic acid and its
ester are instable and easily decomposed, which led to low
yield in the second step of the reported synthetic route
(Wang et al. 2017). As an attempt to improve the synthetic
method, vinyl chloroacetate was used in the synthesis of
intermediate 1 by enzymic synthesis with immobilized
lipase as a catalyst (Scheme 1). Unfortunately, glutathione
Fig. 1 Structures of everolimus and tesirolimus
Scheme 1 The enzymic synthesis of glutathione-everolimus with vinyl choroacetate

Med Chem Res (2018) 27:583–591
585
Scheme 2 The enzymic synthesis of glutathione-everolimus with vinyl bromoacetate
Table 1 Stability of glutathione-everolimus at room temperature
100
50
0
Solution
Time of
Remaining of
detection (h)
the conjugate (%)
pH = 7.4
pH = 6.0
pH = 8.0
pH = 2.0
Saline (0.9% NaCl)
Water for injection
0
99.1
99.1
99.1
99.1
24
0
24
0
5
10
15
20
25
Time (h)
was hardly reacted with the resulting intermediate 1 because
of the weak leaving activity of the chlorine atom.
Fig. 2 The stability of glutathione-everolimus in different pH PBS
solutions
As an alternative, vinyl bromoacetate was used in the
synthesis of intermediate 2 by enzymic synthesis (Scheme
2). This step was almost quantitative without any purifica-
tion by filtering the immobilized lipase and evaporating the
excessive low-boiling vinyl bromoacetate and other by-
product. The more active bromine atom of intermediate 2
was easily reacted with the thiol group of glutathione to
give the target conjugate, which was purified by reversed
phase high performance liquid chromatography (RP-HPLC)
with 83% yield. By contrast, the chemical synthesis just
obtained about 42% total yield (Wang et al. 2017). This
enzymic synthesis approach has been successfully opti-
mized and scaled up to hundreds of grams per batch on a
repetitive base.
Table 2 pH stability studies for the conjugate at 37 °C
pH
2.0
6.0
7.4
8.0
8.1
t1/2 (h)
12.2
429.8
26.3
reconstituted with 0.9% sodium chloride for injection
before intravenous infusion.
Since the linkage between the acetyl spacer and ever-
olimus is an ester bond, besides, everolimus is a macrolide,
the conjugate would be easily hydrolyzed under basic
conditions. As predicted, the degradation of the conjugate
increased as the pH was increased above 6.0 in phosphate
buffers at 37 °C (Fig. 2; Table 2). But the conjugate was
relative stable at pH 6.0 with a half-life of more than 429 h.
The degradation under basic condition, such as at pH 8.0,
took place quickly with a half-life of only 8.1 h. Meanwhile,
it was also hydrolized in highly acidic condition, and the
half-life of the conjugate at pH 2.0 was about 12.2 h. The
pH-dependent study clearly showed that everolimus could
not be released from the conjugate in the pH 7.4 PBS
solution at 37 °C with a half-life of about 26.3 h for
degradation. However, release of everolimus from the
conjugate in rat (Wang et al. 2017) indicated that enzymes
pH stability study
In the preformulation research, we investigated the stability
of glutathione-everolimus in different pH solutions. The
results showed that it was stable under neutral conditions in
solution. For example, the conjugate was very stable with-
out any degradation over 24 h at room temperature in
clinically relevant solutions (Table 1). Therefore,
glutathione-everolimus can be dissolved in WFI (water for
injection) and prepared into lyophilized solid, which is to be

586
Med Chem Res (2018) 27:583–591
in vivo played an important role in this process. These
enzymes could be either esterases or aminopeptidases (Yeo
et al. 1998).
and human, which indicated that the prodrug might create a
similar effect in all the three species.
Acute toxicity
Species-related drug release experiment
Glutathione-everolimus was given intravenously to mice at
different doses to evaluate its acute toxicity and the max-
imum tolerable doses. Mice administered glutathione-
everolimus at a single dose of 70 mg/kg (equivalent dose
of everolimus) were still alive after 1 week without any
toxic response. And no visible signs of toxicity and weight
loss were observed when glutathione-everolimus was given
at a dose of 45 mg/kg three times a week to the mice for
1 month. Therefore, glutathione-everolimus appeared to be
low toxic and good tolerance.
In our previous work, the release of everolimus from the
conjugate in rat was investigated, which showed that this
releasing process took about 2 h (Wang et al. 2017). To
evaluate this process in different species, we studied the
drug release of the conjugate in human plasma and cyno-
molgus monkey.
As Fig. 3. displayed, everolimus was gradually elimi-
nated in human plasma, which might be owing to in vitro
metabolism or instability. The conjugate can be converted
to everolimus in human plasma, and this process took more
than 5 h in vitro. Meanwhile, the released everolimus from
the conjugate was also continuously eliminated in human
plasma, which led to the curve declined at the end point.
The drug release experiment of the conjugate was also
studied in a cynomolgus monkey. As the results shown in
Fig. 4, the conjugate can quickly release everolimus in
cynomolgus monkey in a short time. For example, the
concentration of the released everolimus was about 4600
ng/mL at the fifth minute, which was slightly higher than
that of the remaining conjugate. And this slightly higher
phenomenon was persistent throughout the whole process.
We believe that this was owing to the in vivo metabolism of
the released everolimus is slower than the releasing of the
conjugate. Therefore, this releasing process only spent
approximately 4 h, but the released everolimus from the
conjugate can persist acting in monkey body for more than
8 h. The results also indicated that the release of everolimus
from the conjugate in monkey took place more slowly than
that in rat, but slightly faster than that in human plasma.
In brief, the results of the species-related drug release
experiment suggested that glutathione-everolimus can
release its parent compound, everolimus, in rat, monkey,
In vivo evaluation
Based on our previous work, a high-dose injection of
glutathione-everolimus once a week led to effective tumor
inhibition. To evaluate the influence of the administration
frequency and dosage to the treatment effect of glutathione-
everolimus, a more detailed experiment was carried out. A
10000
Glutathione-everolimus
1000
100
10
Everolimus released from
glutathione-everolimus
1
0
2
4
6
8
10
Time (h)
Fig. 4 The concentration–time curve of the injected glutathione-
everolimus (1 mg/kg) in cynomolgus monkey body
Fig. 3 Release of everolimus
from glutathione-everolimus in
human plasma
1000
Everolimus released from
everolimus-glutathione
Glutathione-everolimus
100
Everolimus
10
0
1
2
3
4
5
Time (h)

Med Chem Res (2018) 27:583–591
587
Fig. 5 a The tumor volume
changes over 29 days of
administration; b The body
weight changes of the mice over
29 days of administration; c The
representative photos of the
tumors stripped from the
sacrificed mice
(a)
1500
1000
500
0
G1 (Control)
G2 (Everolimus)
G3 (Glutathione-Everolimus)
G4 (Glutathione-Everolimus)
G5 (Glutathione-Everolimus)
G6 (Glutathione-Everolimus)
G7 (Glutathione-Everolimus)
0
5
10
15
20
25
30
Days after grouping
(b)
24
22
20
18
16
14
G1 (Control)
G2 (Everolimus)
G3 (Glutathione-Everolimus)
G4 (Glutathione-Everolimus)
G5 (Glutathione-Everolimus)
G6 (Glutathione-Everolimus)
G7 (Glutathione-Everolimus)
0
5
10
15
20
25
30
Days after grouping
(c)
G1
G2
G3
G4
G5 G6
G7
summary of the preliminary results is given in Fig. 5 and
Table 3.
rate ≥ 60.0%) but with relative low inhibition rate, 60.8 and
67.4%, respectively, which was accordance with the results
in our previous work (Wang et al. 2017). Meanwhile, dose
escalation can not obviously improve the curative effect (G5
vs. G6). However, increasing administration frenquency
gave better results (G6 vs. G7). The tumor growth was
totally inhibited in group 7 (G7) with 86.6% inhibition rate,
when mice were administrated three times a week by
intravenous injection. Lower dosage with 10 mg/kg three
times a week also achieved good tumor inhibition, which
was 78.5%. Therefore, administration by intravenous
As the results indicated that tumors were effectively
inhibited in all the treated groups. The tumor inhibition of
oral glutathione-everolimus was 74.8%, which was very
close to that of oral everolimus (tumor inhibition: 71.7%),
meaning that the in vivo activity of the oral prodrug was
equivalent to that of oral everolimus. However, injected
glutathione-everolimus groups gave much different results.
The groups (G5 and G6) administrated intravenously once a
week exhibited effective inhibition to tumors (Inhibition

588
Med Chem Res (2018) 27:583–591
Table 3 In vivo activity of
everolimus and glutathione-
everolimus against renal cell
carcinoma OS-RC-2
d
Group Dose
Administration
Average tumor
Tumor
Body weight
changes (%)c
P value
(mg/kg)a frequency
weight (g) SEM inhibitionb
G1
G2
G3
G4
G5
G6
G7
a
–
–
0.795 0.127
0.225 0.057
0.200 0.071
0.171 0.066
0.312 0.087
0.258 0.130
0.106 0.073
–
−25.7
−7.7
−6.0
−1.7
−3.9
−3.2
−2.8
–
5
qd, ig
71.7%
74.8%
78.5%
60.8%
67.4%
86.6%
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
5
qd, ig
10
25
35
35
3 times/qw, iv
qw, iv
qw, iv
3 times/qw, iv
Equivalent dose of everolimus
b Tumor inhibition = (1 − average tumor weight of treated group/average tumor weight of negative control
group) × 100
c Body weight changes = (average body weight of mice at the twenty-ninth day/average body weight of mice
at first day − 1) × 100
d
Be relative to the negative control group for average tumor weight
Table 4 The liver coefficient of
the sacrificed mice
Group
G1
G2
G3
G4
G5
G6
G7
Liver coefficient
0.059
0.052
0.052
0.054
0.055
0.058
0.057
x
SD
0.006
0.003
0.0410
0.002
0.0269
0.003
0.2054
0.005
0.2582
0.004
0.7598
0.002
0.4626
P valuea
–
a
Be relative to the negative control for liver coefficient
injection once a week afforded improved therapy com-
pliance in clinical use for advanced cancer patients, while
administration by intravenous injection frequently, such as
three times a week, obtained excellent tumor growth
inhibition.
glutathione-everolimus, such as three times a week, might
decrease toxicity and increase therapeutic index.
There was no obvious body weight loss in all the treated
mice. But mice in the oral groups exhibited slightly higher
body weight loss, 7.7% for group 2 (G2) and 6.0% for
groups 3 (G3), respectively, when compared with those in
the injection groups. The body weight loss indirectly
showed that the relative more toxic reaction in oral groups.
As we previously reported, the average body weight of the
mice decreased remarkably by 25.7% in the negative group
(G1), which might be caused by the rapid growing tumor
and nutrition deficiency. No mice died throughout the study.
In order to further investigate the hepatic toxicity of
glutathione-everolimus, the liver coefficients of the mice
with long-term intake of glutathione-everolimus or ever-
olimus were calculated after mice humanely sacrificed. As
the results suggested (Table 4), compared with the negative
group (G1), the liver coefficients showed no significant
difference (P > 0.05) for intravenous injection groups (G4,
G5, G6, and G7), whereas, they exhibited significant dif-
ference (P < 0.05) for oral groups (G2 and G3). This toxic
response for oral groups might be partially attributed to
frequent administration and drug metabolism in liver, while
the less frequence by injection led to no obvious hepatic
toxicity for the mice. Thus, less frequent injection of
Conclusion
Glutathione is well-known for its multiple function in lots
of important physiological process. Using the endogenous
tripeptide as a new method for drug delivery, we success-
fully developed an injectable novel glutathione-drug con-
jugate with good water solubility. Glutathione-everolimus is
proved to be an effective, safe and well-tolerated injectable
prodrug in the treatment of renal cell carcinoma. The results
indicated that three times injection with a high dosage in
1 week can achieve much better in vivo efficacy, and no
obvious toxic response was observed. The results of
species-related drug release experiment displayed that the
performance of the conjugate in human plasma, rat, and
monkey was similar. Moreover, one synthetic process has
been optimized and scaled up with high reproducibility and
yields, which will facilitate the development of such con-
jugate. The stability of the conjugate supported that this
prodrug can be prepared into lyophilized solid, which is to
be reconstituted with 0.9% sodium chloride for injection
before intravenous infusion. As our continuous preclinical
researches, study of the in vivo efficacy of glutathione-
everolimus on other tumor models are also ongoing.

Med Chem Res (2018) 27:583–591
589
Experimental
16.0, 15.1, 13.8, 13.8, 10.9; HRESIMS m/z (pos) 1056.5414
[M + Na]+ (calcd. for C55H84ClNO15 1033.5529).
General procedures
Everolimus (1.0 g, 1.04 mmol) and vinyl bromoacetate
(0.257 g, 1.56 mmol) were dissolved in dichloromethane
(10 mL). Then the immobilized lipase (0.5 g) was added to
the mixture, which was stirred for about 4 h at room tem-
perature. After reaction finished, the immobilized lipase was
filtered with a funnel and the filtrate was concentrated in
vacuo to obtain compound 2 (1.12 g, 1.04 mmol) as a white
fluffy solid. Mp 64.6–66.3 °C; [α]2D0 −121.5° (c 1.0,
MeOH); IR (KBr) νmax 3443, 2931, 2870, 1722, 1643
Everolimus, obtained by semisynthesis, was supplied by
Hisunpharm. Other reagents and solvents were obtained
from commercial suppliers without further purification. All
the reactions were monitored by HPLC, which was per-
formed on Angilent apparatus. Nuclear magnetic resonance
(NMR) spectra were collected on a Bruker DPX 400 NMR
spectrometer with tetramethylsilane as an internal reference.
High-resolution mass spectra were obtained on a Waters
Micromass Q-Tof MicroTM instrument using the electro-
spray ionization (ESI) technique. All mice were purchased
from Weitong Lihua (Beijing, China). Mice were housed in
standard laboratory conditions (25 2 °C, 40–70% relative
humidity, and a 12-h light-dark cycle) in a barrier facility
with laminar flow cabinets. A cynomolgus monkey was
purchased from Academy of military medical sciences
(Beijing, China).
1
cm−1; H NMR (DMSO-d6, 400 MHz): δ = 6.48 (1H, s),
6.37–6.44 (1H, m), 6.19–6.25 (1H, m), 6.13 (2H, d, J =
10.8 Hz), 5.49 (1H, q, J = 24.1 Hz), 5.25–5.27 (1H, m),
5.10 (1H, d, J = 10.2 Hz), 4.93–4.96 (2H, m), 4.27 (2H, br
s), 4.15 (2H, s), 4.01 (2H, br s), 3.96 (1H, d, J = 4.12 Hz),
3.70–3.72 (2H, m), 3.64 (1H, d, J = 12.2 Hz), 3.42–3.45
(3H, m), 3.24–3.28 (1H, m), 3.15 (3H, s), 3.09–3.12 (1H,
m), 3.05 (3H, s), 2.98 (1H, s), 2.75 (1H, d, J = 17.1 Hz),
2.34–2.40 (2H, m), 1.98–2.22 (1H, m), 2.12 (1H, d, J =
12.7 Hz), 2.00–2.02 (1H, m), 1.85–1.92 (1H, m), 1.82–1.85
(2H, m), 1.74 (2H, s), 1.68–1.70 (2H, m), 1.60–1.63 (4H,
m), 1.46–1.54 (4H, m), 1.37–1.43 (2H, m), 1.24–1.29 (6H,
m), 1.14–1.18 (2H, m), 1.01–1.06 (3H, m), 0.94–0.99 (4H,
m), 0.86–0.87 (5H, m), 0.81–0.84 (3H, m), 0.78 (2H, d, J =
6.4 Hz), 0.74 (3H, d, J = 6.4 Hz); 13C NMR (DMSO-d6,
100 MHz): δ = 210.9, 207.9, 199.3, 169.6, 167.7, 167.6,
167.4, 139.7, 138.3, 137.6, 132.8, 130.9, 127.4, 125.1,
99.4, 86.0, 83.0, 82.7, 68.9, 76.2, 74.0, 67.4, 66.6, 65.9,
57.5, 57.5, 57.4, 56.3, 51.2, 45.6, 43.9, 36.4, 35.6, 35.2,
33.8, 32.7, 31.7, 31.3, 30.2, 30.1, 30.0, 28.8, 27.5, 26.9,
26.7, 24.9, 22.5, 22.1, 20.8, 16.0, 16.0, 15.1, 14.4, 13.8,
10.9; HRESIMS m/z (pos) 1100.4927 [M + Na]+ (calcd. for
C55H84BrNO15 1077.5024).
Compound 2 (0.90 g, 0.83 mmol) and glutathione (0.51
g, 1.66 mmol) were added to a mixed solution of DMF/
EtOH/H2O (1:2:2, 10 mL) to obtain a clear solution. K2CO3
(0.23 g, 1.66 mmol) was added subsequently to the solution,
and the obtained mixture was stirred for about 24 h. Then,
the reacted solution was concentrated in vacuo. The
obtained residue was purified by RP-HPLC to give
glutathione-everolimus (0.713 g, 0.69 mmol) as a white
powder.
Preparation of compounds
Everolimus (1.0 g, 1.04 mmol) and vinyl chloroacetate
(0.188 g, 1.56 mmol) were dissolved in dichloromethane
(10 mL). Then the immobilized lipase (0.5 g) was added to
the mixture, which was stirred for about 4 h at room tem-
perature. After reaction finished, the immobilized lipase was
filtered with a funnel and the filtrate was concentrated in
vacuo. The residue was purified by silica gel column
chromatography (n-hexane: acetone = 5:1 to 3:1) to obtain
compound 1 (0.958 g, 0.93 mmol) as a white powder. Mp
70.8–71.3 °C; [α]2D0 −129.6° (c 1.0, MeOH); infrared (IR)
(KBr) νmax 3425, 2932, 2871, 1721, 1643 cm−1; H NMR
1
(dimethyl sulphoxide (DMSO)-d6, 400 MHz): δ = 6.46 (1H,
d, J = 13.6 Hz), 6.40 (1H, d, J = 11.2 Hz), 6.23 (1H, t, J =
13.4 Hz), 6.12 (1H, d, J = 10.6 Hz), 5.49–5.43 (1H, m),
5.10 (1H, d, J = 9.2 Hz), 4.94 (1H, br s), 4.37 (2H, br s),
4.26–4.20 (2H, m), 4.01–3.96 (2H, m), 3.72 (2H, br s), 3.64
(1H, d, J = 11.0 Hz), 3.42 (2H, d, J = 11.0 Hz), 3.15 (3H,
br s), 3.05 (3H, br s), 3.01–3.05 (2H, m), 2.71–2.75 (2H,
m), 2.49–2.48 (1H, m), 2.35–2.38 (2H, m), 2.22–2.25 (2H,
m), 2.11 (2H, br s), 1.99–2.02 (2H, m), 1.85–1.91 (3H, m),
1.74 (2H, br s), 1.63 (6H, br s), 1.55–1.60 (6H, m),
1.35–1.40 (2H, m), 1.20–1.25 (4H, m), 1.14 (4H, br s),
0.97–1.04 (6H, m), 0.83–0.86 (6H, m), 0.73–0.78 (6H, m);
13C NMR (DMSO-d6, 100 MHz): δ = 210.9, 208.7, 207.9,
198.9, 169.6, 169.0, 167.7, 167.4, 139.7, 138.3, 137.6,
132.8, 130.9, 127.4, 125.4, 99.4, 86.0, 83.0, 82.9, 82.7,
76.2, 74.0, 68.9, 67.4, 66.6, 65.8, 57.5, 57.4, 55.9, 55.3,
51.2, 45.6, 43.9, 41.9, 41.5, 38.7, 36.4, 35.6, 35.2, 33.8,
32.7, 32.5, 31.3, 30.2, 30.0, 26.9, 26.7, 24.9, 22.1, 20.8,
pH stability study
To evaluate the stability of the conjugate in different pH
conditions, solutions of glutathione-everolimus (2 mg/mL)
were diluted into phosphate buffered saline (PBS), which
were adjusted to pH 2.0, 6.0, 7.4, and 8.0, respectively, and
incubated at 37 °C. The solutions were removed at different
time points and analyzed by HPLC, detecting disappearance
of the conjugate. Stability profile graph was generated by

590
Med Chem Res (2018) 27:583–591
plotting the percentage of remaining starting material over a
time course. The percentage was calculated on the basis of
the ratio of the peak area of the sample at 0, 1, 4, 8, 12, and
24 h vs. the initial area peak.
weight and tumor sizes were measured every 3 or 4 days
throughout the study. Finally, tumors were retrieved and
weighed after 29 days administration, then, the tumor
inhibition was calculated. The liver of the human sacrificed
mice was taken out and weighed. The liver coefficient was
calculated using the following formula: the liver coefficient
= the average weight of the liver/the average body weight
of the mice.
Drug release experiment
In human plasma
Ten microliter 0.1 mmol/L solution of glutathione-
everolimus was added to 990 μL human plasma. The two
solutions were mixed thoroughly, and then incubated at 37 °
C. Fifty microliter of the mixture was removed from the
mixed solution at predetermined time intervals of 0, 1, 15,
30 min, 1, 2, and 5 h, respectively. One hundered and fifty
microliter acetonitrile was added immediately to the
removed solution to quench the reaction, which was then
analyzed with HPLC to determine the remainder of the
conjugate and the released everolimus in the solution.
Acknowledgements This work was financially supported by the
Key Science and Technology Innovation Team of Zhejiang Province
(2013TD10).
Compliance with ethical standards
Conflict of interest The authors declare that they have no competing
interests.
In cynomolgus monkey
References
The conjugate, dissolved in normal saline, was given to a
5–6 years old monkey with a dose of 1 mg/kg by intrave-
nous injection. One milliliter blood was taken from the
forelimbs of the monkey at predetermined time intervals of
5, 10, 30 min, 1, 1.5, 2, 3, 6, 8, and 24 h, respectively.
Centrifugation was used to remove the red blood cells.
Then, 50 μL of the obtained liquid supernatant was sam-
pled, and 150 μL acetonitrile was subsequently added to the
sample. Centrifugation was used once again to remove the
denatured protein in the solution. Afterwards, the remainder
of the conjugate and the released everolimus in the solution
were analyzed by HPLC.
Alasker A, Meskawi M, Sun M, Ismail S, Hanna N, Hansen J, Tian Z,
Bianchi M, Perrotte P, Karakiewica PI (2013) A contemporary
update on rates and management of toxicities of targeted therapies
for metastatic renal cell carcinoma. Cancer Treat Rev 39:388–401
Arai E, Kanai Y (2010) Genetic and epigenetic alterations during renal
carcinogenesis. Int J Clin Exp Pathol 4:58–73
Coiffier B (2013) Clinical efficacy and management of temsirolimus in
patients with relapsed or refractory mantle cell lymphoma. Clin
Lymphoma Myeloma Leuk 13:351–359
Crowe A, Bruelisauer A, Duerr L, Guntz P, Lemaire M (1999)
Absorption and intestinal metabolism of SDZ-RAD and rapa-
mycin in rats. Drug Metab Dispos 27:627–632
Eble JN, Sauter G, Epstein JI, Sesterhenn IA (2004) World Health
Organization classification of tumors. IARC, Lyon
Farag SS, Zhang S, Jansak BS, Wang X, Kraut E, Chan K, Dancey JE,
Grever MR (2009) Phase II trial of temsirolimus in patients with
relapsed or refractory multiple myeloma. Leuk Res
33:1475–1480
Evaluation of in vivo efficacy
Kirchner GI, Meier-Wiedenbach I, Manns MP (2004) Clinical phar-
macokinetics of everolimus. Clin Pharmacokinet 43:83–95
Kovarik JM, Hartmann S, Figueiredo J, Rordorf C, Golor G, Lison A,
Budde K, Neumayer HH (2002) Effect of food on everolimus
absorption: quantification in healthy subjects and a confirmatory
screening in patients with renal transplants. Pharmacotherapy
22:154–159
Mackenzie MJ, Rini BI, Elson P, Schwandt A, Wood L, Trinkhaus M,
Bjarnason G, Knox J (2011) Temsirolimus in VEGF-refractory
metastatic renal cell carcinoma. Ann Oncol 22:145–148
Motzer RJ, Escudier B, Oudard S, Hutson TE, Porta C, Bracarda S,
Grünwald V, Thompson JA, Figlin RA, Hollaender N, Kay A,
Ravaud A (2010) Phase 3 trial of everolimus for metastatic renal
cell carcinoma: final results and analysis of prognostic factors.
Cancer 116:4256–4265
Human renal cell carcinoma OS-RC-2 tumors were estab-
lished by implanting a 4–5 mm3 tissue fragment of OS-RC-
2 tumor collected from donor mice into the right axillary
flank of recipient nude mice. When tumors reached an
average volume of about 148 mm3, mice were divided into
seven groups. Normal saline was given to mice by means of
intragastric administration daily for negative control (G1).
Mice administered everolimus orally everyday at a dose of
5 mg/kg as the positive control (G2). As a parallel test, 5
mg/kg glutathione-everolimus was administered orally to
the mice daily (G3). To investigate the influence of different
administration dosage and frequency on the therapeutic
effect, mice were given glutathione-everolimus intrave-
nously at a dose of 10 mg/kg three times a week (G4), 25
mg/kg once a week (G5), 35 mg/kg once a week (G6), or
35 mg/kg three times a week (G7), respectively. Mouse
Patel SB, Stenehjem DD, Gill DM, Tantravahi SK, Agarwal AM, Hsu
J, Vuong W, Pal SK, Agarwal N (2012) Everolimus versus
temsirolimus in metastatic renal cell carcinoma after progression
with previous systemic therapies. Clin Genitourin Cancer
14:153–159

Med Chem Res (2018) 27:583–591
591
Ryan J, PA P (2013) Cyclodextrin-based polymers for therapeutics
delivery. US Patent 2012-0065161. Filled 12 Aug 2011, Issued
15 Mar 2012
Wang HB, Kuang H, Zhang W, Zhou X, Zhu T (2016) Synthesis and
evaluation of
a water-soluble epothilone B prodrug with
remarkably decreased toxicity. Lett Org Chem 13:619–624
Wang H, Zheng X, Cai Z, Yu O, Zheng S, Zhu T (2017) Synthesis and
evaluation of an injectable everolimus prodrug. Bioorg Med
Chem Lett 27:1175–1178
Weikert S, Kempkensteffen C, Busch J, Johannsen M, Grünwald V,
Zimmermann K, Flörcken A, Westermann J, Weinkauf L, Miller
K, Keilholz U (2013) Sequential mTOR inhibitor treatment with
temsirolimus in metastatic renal cell carcinoma following failure
of VEGF receptor tyrosine kinase inhibitors. World J Urol
31:805–809
Yeo SH, Nihira T, Yamada Y (1998) Screening and identification of a
novel lipase from Burkholderia sp. YY62 which hydrolyzes t-
butyl esters effectively. J Gen Appl Microbiol 44:147–152
Zhao H, Rubio B, Sapra P, Wu D, Reddy P, Sai P, Martinez A, Gao Y,
Lozanguiez Y, Longley C, Greenberger LM, Horak ID (2008)
Novel prodrugs of SN38 using multiarm poly(ethylene glycol)
linkers. Bioconjugate Chem 19:849–859
Zhu T, Shah SM, Saunders RW (2002) Water soluble SDZ-RAD
esters. EP Patent 2002-1210350. Filled 16 Aug 2000, Issued 06
May 2002
Sherman S, Amzal B, Calvo E, Wang X, Park J, Liu Z, Lin C, Cas-
ciano R (2015) An indirect comparison of everolimus versus
axitinib in US patients with advanced renal cell carcinoma in
whom prior sunitinib therapy failed. Clin Ther 37:2552–2559
Shirotake S, Yasumizu Y, Ito K, Masunaga A, Ito Y, Miyazaki Y,
Hagiwara M, Kanao K, Mikami S, Nakagawa K, Momma T,
Masuda T, Asano T, Oyama M, Tanaka N, Mizuno R, Oya M
(2016) Impact of second-line targeted therapy dose intensity on
patients with metastatic renal cell carcinoma. Clin Genitourin
Cancer 14:e575–e583
Stenner-Liewen F, Grünwald V, Greil R, Porta C (2013) The clinical
potential of temsirolimus in second or later lines of treatment for
metastatic renal cell carcinoma. Expert Rev Anticancer Ther
13:1021–1033
Tomalova B, Sirova M, Rossmann P, Pola R, Strohalm J, Chytil P,
Cerny V, Tomala J, Kabesova M, Rihova B, Ulbrich K, Etrych T,
Kovar M (2016) The structure-dependent toxicity, pharmacoki-
netics and anti-tumor activity of HPMA copolymer conjugates in
the treatment of solid tumors and leukemia. J Control Release
223:1–10