Synthesis of diacylated γ3-glutamyl-cysteamine prodrugs, and in vitro evaluation of their cytotoxicity and intracellular delivery of cysteamine

Article abstract of DOI:10.1016/j.ejmech.2015.12.027

To overcome the major disadvantages of cysteamine, the only registered treatment for the rare genetic disease cystinosis, nine prodrugs of γ3-glutamyl-cysteamine (4) were synthesized for evaluation. Esterification of the thiol conferred oxidative stabilit

Full text of DOI:10.1016/j.ejmech.2015.12.027

European Journal of Medicinal Chemistry 109 (2016) 206e215
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Synthesis of diacylated g-glutamyl-cysteamine prodrugs, and in vitro
evaluation of their cytotoxicity and intracellular delivery of
cysteamine
Lisa Frost, Pratap Suryadevara, Stephanie J. Cannell, Paul W. Groundwater 1,
Paul A. Hambleton, Rosaleen J. Anderson*
Sunderland Pharmacy School, University of Sunderland, Sunderland, SR1 3SD, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
To overcome the major disadvantages of cysteamine, the only registered treatment for the rare genetic
Received 2 September 2015
Received in revised form
28 November 2015
Accepted 14 December 2015
Available online 24 December 2015
disease cystinosis, nine prodrugs of
fication of the thiol conferred oxidative stability, while sufficient lipophilicity for oral bioavailability was
achieved by acylation of the -carboxyl group of -glutamyl-cysteamine (4). Low cytotoxicity was
observed in cultured HaCaT keratinocytes using the MTT assay, with EC50 values higher than or similar to
that of cysteamine. Successful uptake of the esterified prodrugs and the subsequent release of cyste-
amine into cultured human proximal tubule epithelial cells were demonstrated using CMQT derivati-
sation and HPLC with UV detection. These prodrugs show potential as novel delivery vehicles of
cysteamine to improve the treatment of the genetic disorder nephropathic cystinosis.
© 2015 Elsevier Masson SAS. All rights reserved.
g-glutamyl-cysteamine (4) were synthesized for evaluation. Esteri-
a
g
Keywords:
Nephropathic cystinosis
g-glutamyl transpeptidase
Cysteamine
Cystine
Drug targeting
Prodrug
1. Introduction
which reduces lysosomal cystine levels and therefore delays dis-
ease progression. However, there are several problems with its
Nephropathic cystinosis is an autosomal recessive disorder,
characterized by the accumulation of cystine crystals within the
lysosomes of all cells, as a result of deficiency of the cystine
transporter protein. The initial symptoms of cystinosis result from
the failure of renal tubules to reabsorb small molecules, causing
Fanconi syndrome and the associated polyuria, glucosuria, phos-
phaturia, and proteinuria [1]. While kidney involvement is one of
the classical features of cystinosis, it has been recognized more
recently as a multi-systemic disease, due to cystine accumulation
in, and damage to, non-renal organs and tissues [1,2]. Untreated,
this disease progresses, usually resulting in damage to all organs
and death by the age of 10 [3]. Treatment of cystinosis is typically by
administration of the aminothiol, cysteamine (as the bitartrate salt,
Cystagon®, or as its sustained release formulation, Procysbi®),
administration, which can lead to non-compliance in a large pro-
portion of those affected [4]. Even as the bitartrate salt, cysteamine
has an intensely unpleasant taste and smell, resulting in nausea and
vomiting [1], and frequently causes disturbance of the gastroin-
testinal (GI) mucosa; gastric or duodenal ulceration is a common
side effect [5]. Extensive first pass metabolism of cysteamine after
oral administration leads to urinary excretion of its conjugates, an
estimated bioavailability of 10e30% [2], and significant amounts of
dimethyl sulfide and methanethiol exhaled in the breath and
through the pores of the skin as body odour [6]. To maintain a
therapeutic plasma concentration, 60e90 mg/kg/day of cysteamine
is required in four divided doses [7], reaching 1 g (as either 50 or
150 mg tablets) per dose for many patients. The importance of a six
hourly intake of Cystagon®, despite disruption to sleep, was
demonstrated by a significant increase in polymorphonuclear
cystine levels after 9 hourly dosing, when compared to a 6 h dosing
interval [8]. The sustained release form Procysbi® allows dosing at
12 h intervals, alleviating the issue of sleep disturbance, but, as it
still releases cysteamine in the GI tract, the problems of first pass
* Corresponding author.
E-mail address: roz.anderson@sunderland.ac.uk (R.J. Anderson).
Present address: Faculty of Pharmacy, University of Sydney, NSW 2006,
1
Australia.
http://dx.doi.org/10.1016/j.ejmech.2015.12.027
0223-5234/© 2015 Elsevier Masson SAS. All rights reserved.

L. Frost et al. / European Journal of Medicinal Chemistry 109 (2016) 206e215
207
metabolism, resulting in low bioavailability, the release of strong
1.1. Prodrug design and rationale
smelling metabolites, and the gastric disturbance issues are not
significantly addressed [9]. Additionally, a cysteamine concentra-
tion effect on cell viability and cell proliferation was observed in
several cell lines, including cultured human fibroblasts, and
excessively high doses of cysteamine have been linked to the
appearance of bruise-like lesions on a number of cystinotic patients
[10].
Cysteamine is also of interest for its potential in treating other
diseases, such as Huntington's disease [11,12], cystic fibrosis
[13e15], chronic kidney disease [16], and non-alcoholic fatty liver
disease (NAFLD), including non-alcoholic steatohepatitis (NASH)
[17]. A current clinical trial is evaluating the potential of Pro-
cysbi® as a disease-modifying treatment for Huntington's dis-
ease. If the scope of cysteamine as a treatment increases to
include these other diseases, similar compliance problems due to
its side effects are likely and alternative delivery approaches
therefore have wider potential application than to cystinosis
alone.
Two further groups that are capable of bioactivation were
required to convert -glutamyl-cysteamine into a candidate for
clinical use: acylation of the thiol group to confer oxidative sta-
bility and an -carboxyl ester to address the likely limitation to
oral bioavailability caused by the zwitterionic nature of -glu-
tamyl-cysteamine 4 (Fig. 1). These modifications resulted in S-
thioester and -carboxyl ester -glutamyl-cysteamine derivatives
1a-c, 2a-c, and 3a-c (Fig. 1). The combined effects of varied
thioester and ester groups enable a limited investigation of suit-
able prodrug groups with desirable pharmacokinetic and phar-
macodynamic properties for the effective treatment of cystinosis
by oral administration. Esterification of carboxylic acid, alcohol
and thiol groups on pharmaceutical products is a successful
strategy to overcome chemical instability and limited oral
bioavailability, with convincing evidence from their extensive
clinical application [23e25]. Ester prodrugs are hydrolysed
rapidly after uptake from the GI tract by blood and liver esterases,
human carboxylesterases 1 and 2 (hCE1 and hCE2), with half-lives
typically 10e30 min [24].
g
a
g
a
g
Cysteamine exerts its therapeutic effect due to its ability to
undergo a thiol exchange with cystine, resulting in the formation of
cysteine and a cysteine-cysteamine mixed disulfide [18] (Scheme
1), which can efflux the lysosome via the transport protein
PQLC2. Evidence for the involvement of this transporter in the
cysteamine-driven depletion of cystine was obtained by its genetic
inactivation or silencing [19,20]. Molecular modelling has shown
There are three distinct enzyme cleavage steps to release
cysteamine from the prodrugs (Scheme 2), presenting two main
ways to deliver the active agent cysteamine into cells after ab-
sorption from the GI tract. Firstly, satisfactory lipophilicity of the
prodrugs permits rapid uptake into cells directly from the blood.
Once inside the cells, the prodrugs are hydrolysed to release
the structural and electronic similarities of the
L-cysteine-cyste-
amine mixed disulfide to
porter [21].
L
-lysine, facilitating its use of this trans-
cysteamine, by (thio)esterase activity and hydrolysis of the
tamyl bond, probably by -glutamyl cyclotransferase. Alternatively,
after uptake from the GI tract into the hepatic portal vein, the
prodrugs undergo rapid esterase hydrolysis of the -carboxyl ester
(R1) and thioester (R2) groups in the blood and/or liver to release
the key intermediate 4, targeted to -glutamyl transpeptidase
(GGT) on the surface of cells. Interaction of -glutamyl-cysteamine
4 with membrane-bound GGT results in the hydrolysis of -glu-
g-glu-
g
The thiol and amino groups of cysteamine are therefore vital
for its role in the treatment of cystinosis, but the unpleasant taste
and smell contribute to low levels of patient compliance. Our
initial work demonstrated the feasibility of amino acid-cysteamine
conjugates as prodrugs, the depletion of cystine providing evi-
dence of their delivery of cysteamine to cystinotic cells, with a
delayed response consistent with the requirement for hydrolysis
of the amino acid-cysteamine amide bond to release the active
agent cysteamine [21]. A major drawback to these simple amino
acid-cysteamine derivatives was their lack of clinical relevance; in
particular, the thiol group confers nucleophilic and reducing
a
g
g
g
tamyl-cysteamine 4 and local release of cysteamine on the surface
of the cell, allowing its rapid internalisation. Our previous studies
provide evidence for the success of this approach: incubation of
cystinotic cells in medium containing
g-glutamyl-cysteamine 4
resulted in significant depletion of cystine [21].
properties, with disulfide bond forming potential, while
a
-amino
Although a low level of serum GGT is normal in humans and is
increased in various disease states, such as alcoholic liver disease,
the normal activity of serum GGT is around 30 times lower than
that of membrane-bound GGT [26] and is likely to make only a
small contribution to the release of cysteamine in the circulation.
acid prodrugs are known to be readily hydrolysed in the blood
[22], which would release cysteamine and fail to prevent its
metabolism. However, the initial studies also showed good
depletion of accumulated cystine in cystinotic cells by
cysteamine, providing evidence of hydrolysis of the
amide bond and release of cysteamine, while the
g
g
-glutamyl-
-glutamyl
The greater resistance of
amide isomers to proteolytic degradation by serum proteases, due
to low levels of serum -glutamyl hydrolysis [27], minimizes pre-
mature proteolysis of the -glutamyl-cysteamine bond in serum.
GGT is known to recognize a variety of small molecules linked to
glutamate via an amide bond at the -carboxyl group [28,29]. It is
g-glutamyl amide derivatives than their a-
a-glutamyl-
cysteamine analogue performed poorly in comparison [21]. Initial
viability experiments using -glutamyl-cysteamine in cells with a
high expression of -glutamyl transpeptidase (GGT), an external
cell surface enzyme that recognises and hydrolyses -glutamyl-
derivatives, indicated low cytotoxicity [21], providing encourage-
g
g
g
g
g
g
expressed on the surface of many cells across various tissues,
including the proximal convoluted tubule cells of the kidney [30],
the intestinal epithelium [31], and the luminal surface of the blood
ment for the further development of
clinical candidate.
g-glutamyl-cysteamine into a
Scheme 1. Intralysosomal thiol exchange of cysteamine with cystine forms a cysteine-cysteamine dimer and cysteine, both of which can efflux the lysosome.

208
L. Frost et al. / European Journal of Medicinal Chemistry 109 (2016) 206e215
Fig. 1.
g-Glutamyl-cysteamine prodrugs 1a-c, 2a-c and 3a-c and key intermediate g-glutamyl-cysteamine 4.
2. Results
2.1. Prodrug synthesis
The synthesis of esterified prodrugs 1a-c, 2a-c and 3a-c was
achieved by solution phase peptide coupling (Scheme 3).
a
-Esterification of commercially sourced benzyl ester 5 with the
appropriate alkyl bromide under mildly basic conditions gave es-
ters 6a-c in good yield. Palladium-catalysed reductive deprotection
yielded the
g-carboxylic acids 7a-c, which were coupled with
cysteamine using diisopropylcarbodiimide (DIC) and 1-
hydroxybenzotriazole (HOBt) to give
g-glutamyl-cysteamine de-
rivatives 8a-c.
Thioesters 9a-c were formed using acetic anhydride and trie-
thylamine, while thioesters 10a-c and 11a-c were prepared using
benzoyl chloride or pivaloyl chloride, as appropriate, and pyridine.
t-Boc deprotection of compounds 9e11 using hydrochloric acid
(2 M) in diethyl ether gave the nine target esterified prodrugs 1a-c,
2a-c and 3a-c, respectively.
Scheme 2. Sequential enzymatic hydrolysis of
a-c to release cysteamine.
g-glutamyl-cysteamine prodrugs 1-3
brain barrier [32]; the accumulation of cystine appears to correlate
with those cells expressing GGT. The high expression of GGT on
renal cells enables increased delivery of cysteamine to these cells,
which are the first to display pathology related to cystine accu-
mulation, leading to poor reuptake of essential electrolytes and
Fanconi syndrome. In cystinosis, however, cystine accumulates in
the majority of cells and tissues throughout the body; the wide
expression of GGT on cells and tissues therefore allows delivery of
cysteamine to all cells, providing an appropriate multi-systemic
treatment for cystinosis. Targeted delivery and GGT-mediated
release of cysteamine on the surface of cells is expected to result
in its rapid uptake into cells in the near vicinity, reducing the
amount of free cysteamine released into the circulation and mini-
mizing the opportunity for liver metabolism of cysteamine, maxi-
mizing its bioavailability. Benefits would include a less frequent and
reduced therapeutic dose, minimizing issues associated with high
peak serum cysteamine concentration, including the production of
noxious smelling metabolites.
2.2. Cytotoxicity studies
Despite its limitations, the MTT assay allows early identification
of cytotoxic or other adverse cellular effects. Initial cytotoxicity
studies on esterified prodrugs 1a-c, 2a-c and 3a-c, compared to
cysteamine hydrochloride, were carried out using spontaneously
transformed keratinocytes from histologically normal skin (HaCaT)
using a standard MTT assay, in which yellow MTT (3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide)
is
reduced to purple formazan in metabolically active cells [35]. Per-
centage survival value at each concentration was derived from the
data of 12 replicates, which were averaged and normalised to
calculate the percentage survival. In this work, the results of the
in vitro tests are expressed as EC50, which represents the mean dose
required to kill half of the members of a test in vitro population. The
EC50 values (calculated using GraphPad Prism®) for
cysteamine prodrugs and cysteamine hydrochloride (Table 1)
represent the dose required to kill 50% of the cells.
g-glutamyl-
The success of a similar GGT-targeting strategy for the delivery
of a cytotoxic glutathione analogue to pancreatic tumours was
recently demonstrated [26]. The low serum activity of GGT nor-
mally observed in humans resulted in only minimal activation of
the antitumour agent systemically. Targeted delivery was also
The range of EC50 values across the prodrugs underlines the
importance of the correct choice of thioester and ester groups. In
each thioester series, the ethyl esters (1a, 2a and 3a) showed lower
cytotoxicity than their propyl and butyl analogues. It is unlikely that
these differences are caused by differential uptake of the prodrugs
due to the lipophilicity of the ester groups, as the efficient uptake of
all members of the 1st series (1a, 1b and 1c) was confirmed in
subsequent experiments, when the intracellular cysteamine levels
achieved by ethyl ester 1a were comparable to those achieved by
propyl ester 1b and butyl ester 1c (Fig. 2). The relatively low
cytotoxicity of acetyl thioester prodrugs 1a-c and ethyl esters 2a
and 3a led to selection of these prodrugs for progression to further
studies designed to evaluate the uptake of these ester/thioester
achieved to the kidney with N-acyl-
methoxazole [33] and administration of the double prodrug
glutamyl- -DOPA resulted in a five-fold greater accumulation of
renal dopamine than the corresponding dose of the single prodrug
g-glutamyl prodrugs of sulfa-
g-
L
L-DOPA [34].
We present here our studies on nine S- and O- diacylated g-
glutamyl-cysteamine prodrugs d their synthesis and the in vitro
evaluation of their cytotoxicity, lipophilicity, cellular uptake and
hydrolysis to release cysteamine.
prodrugs of g-glutamyl-cysteamine 4 into cultured cell lines and

L. Frost et al. / European Journal of Medicinal Chemistry 109 (2016) 206e215
209
Compound
R2
CH3
Ph
Yields
1
2
1a (81 %), 1b (79 %), 1c (67 %)
2a (67%), 2b (52%), 2c (94%)
3a (71 %), 3b (88 %), 3c (71 %)
9a (34 %), 9b (62 %), 9c (56 %)
10a (25 %), 10b (24 %), 10c (47 %)
11a (56 %), 11b (43 %), 11c (40 %)
3
(CH3)3
CH3
Ph
9
10
11
(CH3)3
Scheme 3. General synthetic pathway to esterified prodrugs 1a-c, 2a-c and 3a-c. Reagents and conditions: (i) MeOH/H2O, Cs2CO3, RT, 24 h; (ii), ReBr, DMF, RT, 24 h; (iii), PdeC, H2,
MeOH, RT, 3 h; (iv) Diisopropylcarbodiimide, HOBt, cysteamine hydrochloride, Et3N, DCM, 0 ꢁC-RT, 24 h; (v) acetic anhydride, Et3N, THF, 40 ꢁC, 24 h; (vi) benzoyl chloride, pyridine,
DCM, RT, 24 h; (vii) pivaloyl chloride, pyridine, DCM, RT, 24 h; (viii) HCl (2M) in diethyl ether, RT.
Table 1
Mean (n ¼ 12) EC50 values ( standard error) of esterified
g-glutamyl-cysteamine prodrugs 1a-c, 2a-c and 3a-c and cysteamine hydrochloride in cultured HaCaT keratinocytes.
Compound
EC50 M)
1a
1b
1c
2a
2b
2c
3a
3b
3c
Cysteamine. HCl
(
m
1732
548
563
331
75
97
228
152
102
309
(3.74)
(3.21)
(3.22)
(2.74)
(2.84)
(3.19)
(3.03)
(10.00)
(10.31)
(9.82)
their subsequent intracellular hydrolysis to release cysteamine.
Ethyl ester prodrug 3a was included for testing, even though it
showed greater cytotoxicity than cysteamine hydrochloride, to
provide an example from each series.
and variation of the thioester group. Although not optimal, these
characteristics are comparable to many orally administered agents
in clinical use and are consistent with the aim of designing pro-
drugs with clinical potential for oral administration.
2.4. In vitro studies
2.3. Physicochemical properties of esterified prodrugs
Provided cysteamine is not significantly released from the pro-
drugs during incubation in the medium, successful cellular uptake
and hydrolysis of the prodrugs can be demonstrated by measuring
the increased levels of intracellular cysteamine. The possibility of
prodrug hydrolysis in fetal bovine serum-containing medium was
The oral bioavailability of pharmaceuticals is heavily influenced
by the balance of aqueous solubility and lipophilicity, which are
affected by the pKa of any ionisable groups and the consequent log P
of the molecule at physiological pH values. Early investigation of
these properties is therefore desirable.
investigated in separate experiments using prodrug 1a and
tamyl-cysteamine 4 with LC-MS analysis. Slow hydrolysis of the
thioester and, to a lesser extent, the -carboxyl ester was observed,
but the -glutamyl-cysteamine bond was not cleaved over 24 h
g-glu-
Experimentally measured pKa and log P values (MlogP) were
obtained using a Sirius T3 and compared with the calculated log P
values (ClogP; ChemDraw®) and the log D values (derived from the
Sirius T3 experimental data (Equation (1)) at pH 7.4 (Table 2).
a
g
(data not shown). It was concluded that hydrolysis of the prodrug
to release cysteamine by the action of serum esterase and GGT in
the culture medium was unlikely.
Exposure of cultured cells to the prodrugs, followed by mea-
surement of the resulting intracellular cysteamine concentrations,
allowed the investigation of three indicators: the uptake of pro-
drugs into the cells, the successful ester/thioester hydrolysis of the
ꢀ
ꢁ
Log D ¼ Log P ꢀ Log 1 þ 10ðpKaꢀpHÞ
(1)
The measured pKa values indicate that the prodrugs are likely to
remain largely ionized throughout the GI tract after oral adminis-
tration, with up to 20% unionized amine present in the small in-
testine to facilitate absorption. Both the measured and calculated
log P values increased in parallel with homologation of the
a
-ester
prodrugs, and the release of cysteamine from g-glutamyl-

210
L. Frost et al. / European Journal of Medicinal Chemistry 109 (2016) 206e215
Fig. 2. Cysteamine concentration in PT37 cells treated with 20 mM prodrugs 1a-c, 2a, 3a or cystamine dihydrochloride, each in comparison to cysteamine hydrochloride (dashed
line). Each point represents mean concentration (n ¼ 3) and error bars represent standard error.
Table 2
Measured pKa
cysteamine prodrugs 1a-c, 2a and 3a, compared to cysteamine hydrochloride and cystamine dihydrochloride.
( g-glutamyl
standard error), calculated log P (ClogP, ChemDraw®), measured log P values (MlogP) and log D values (triplicate measurements) of ester/thioester
Prodrug
R1
R2
pKa (experimental)
ClogP
MlogP
Log D at pH7.4
1a
1b
1c
2a
CH2CH3
(CH2)2CH3
(CH2)3CH3
CH2CH3
CH2CH3
e
CH3
CH3
CH3
Ph
(CH3)3
e
8.94 ( 0.29)
8.80 ( 0.20)
8.69 ( 0.18)
8.85 ( 0.38)
8.90 ( 0.36)
8.15 ( 0.15 thiol), 10.52 ( 0.40 amine)
9.11 ( 0.12)
ꢀ0.98
ꢀ0.52
ꢀ0.07
þ0.31
þ0.48
ꢀ0.10
ꢀ0.74
þ0.11
þ0.21
þ0.43
þ1.09
þ1.22
þ0.06
ꢀ0.46
ꢀ1.44
ꢀ1.21
ꢀ0.88
ꢀ0.37
ꢀ0.29
ꢀ2.28
ꢀ2.18
3a
Cysteamine HCl
Cystamine dihydrochloride
e
e
cysteamine 4. Following treatment of cultured proximal tubule
epithelial cells with test agents, the intracellular cysteamine con-
centration was quantified using 2-chloro-1-methylquinolinium
tetrafluoroborate (CMQT) as a pre-column derivatising agent and
HPLC. CMQT reacts instantaneously and highly specifically with
thiols, under mild conditions, to form a thiol-CMQT derivative that
can be quantified using HPLC with ultraviolet or fluorescent
detection [34]. HPLC conditions, equipment and validation data are
included in the Supporting Information.
2.4.1. Prodrug uptake and cysteamine release in proximal tubular
cells
A normal cultured human proximal tubular (kidney) cell line,
PT37 [37], was used to assess the cellular uptake and release of
cysteamine from esterified prodrugs 1a-c, 2a and 3a, using cyste-
amine and its disulfide cystamine as controls. The concentration of
20
of patients treated with Cystagon® and is in line with the estab-
lished EC50 of 12.5 M [7]. Following prodrug exposure, the
mM prodrug was chosen to mimic cysteamine levels in the blood
m

L. Frost et al. / European Journal of Medicinal Chemistry 109 (2016) 206e215
211
intracellular concentration of cysteamine was measured after cell
lysis, centrifugation and Bradford assay of the pellet protein con-
tent, supernatant thiol derivatization with the tagging agent CMQT
[36], followed by HPLC analysis using simultaneous UV and fluo-
rescence detection, Fig. 2.
Analysis of untreated PT37 cells showed no cysteamine-CMQT
peak, indicating an initial intracellular cysteamine concentration
below the level of detection (<160 nM). Cysteamine uptake into
cells following treatment with cysteamine hydrochloride was
confirmed by the detection of cysteamine-CMQT. Cysteamine up-
take was rapid, peaking after 2 h; the level was maintained at
agreement with the measured log P values (Table 2), the S-acetyl
prodrugs 1a-c (with lower MlogP values) were associated with a
slower increase in cysteamine concentration than the S-benzoyl 2a
and S-pivaloyl 3a prodrugs (with higher MlogP values and greater
lipophilicity). In this instance, due to the minimal hydrolysis of
esters in the culture medium, it is probable that the ester/thioester
prodrugs were internalised by passive diffusion, with no involve-
ment of GGT, and broken down intracellularly to release
cysteamine.
3. Conclusion
0.84
0.06 nmol/mg protein until around 6 h, dropping to
0.22 0.04 nmol/mg protein after 24 h, suggesting that cysteamine
is metabolized after internalization. The intracellular level of
cysteamine did not continue to rise when its concentration was
Nine esterified
to overcome some of the undesirable properties of the pharma-
ceutical product cysteamine. Derivatization as the -glutamyl-de-
g-glutamyl-cysteamine prodrugs were designed
g
maintained in the medium at 20
that cysteamine uptake may be saturable [38].
m
M, supporting the observation
rivative was successful, offering increased metabolic stability, and
physicochemical properties suitable for oral bioavailability were
When exposed to air, cysteamine is rapidly oxidized to its
disulfide form, cystamine [39], particularly when in dilute
aqueous solution. The rate of oxidation of cysteamine hydro-
chloride in aqueous solution was investigated by HPLC with UV
detection, using pre-analysis derivatisation by CMQT. Only the
reduced thiol form can react with CMQT and is selectively
detected, Fig. 3.
achieved by esterification of the glutamate a-carboxyl group, while
oxidation of the thiol was inhibited by thioester formation. In
addition, particularly low cytotoxicity of 1a, 1b and 1c in HaCaT
keratinocytes was observed using the MTT assay; a-ethyl esters 2a
and 3a showed similar cytotoxicity to that of cysteamine in this
system, while acetyl thioesters 1a-c exhibited lower cytotoxicity
than cysteamine. These ester/thioester
g-glutamyl-cysteamine
The oxidation of cysteamine hydrochloride in water (pH 4.66)
occurred rapidly, falling to 50% within 1 h; after approximately 3 h,
there was no detectable cysteamine in the solution. When
PT37 cells were treated directly with the disulfide, cystamine
dihydrochloride, the total intracellular levels of cysteamine were
considerably lower than those observed when the same cells were
treated with cysteamine hydrochloride; the dicationic hydrophilic
nature of cystamine is likely to hinder uptake into the cell and the
efficiency of its intracellular bioreduction has not been established,
thus cystamine has not found application for the treatment of
cystinosis. In addition to extensive first pass metabolism, oxidation
to the disulfide may be partly responsible for the poor bioavail-
ability of cysteamine in the therapy of cystinosis. A number of
thioesters of cysteamine were synthesized and showed signifi-
cantly increased stability to oxidation compared to cysteamine
(data not shown). The use of thioester groups in the prodrugs in-
hibits oxidation of the free thiol to the disulfide with the aim of
increasing cysteamine bioavailability.
prodrugs displayed uptake into cells, ester and thioester hydrolysis,
and release of cysteamine, with a delay to its release consistent
with the requirement for activation of the prodrug. Furthermore,
the prodrugs maintained the concentration of cysteamine at above
baseline levels for at least 24 h, comparative to or better than that
achieved by cysteamine hydrochloride and significantly better than
that achieved by cystamine dihydrochloride. These data support
the in vivo efficacy evaluation of cysteamine prodrugs in a validated
animal model for their potential as an alternative oral treatment for
cystinosis.
This targeted prodrug approach to the delivery of cysteamine
may also have application to the treatment of other diseases, such
as Huntington's disease, cystic fibrosis and non-alcoholic steato-
hepatitis (NASH), in which recent research has demonstrated a
beneficial effect of cysteamine.
4. Experimental
Cysteamine-CMQT was also observed when cells were treated
with any of prodrugs 1a-c, 2a and 3a, providing evidence for their
uptake and hydrolysis to release cysteamine. Peak levels of cyste-
amine were not reached until 6 h after exposure, consistent with
Supporting Information available: General methods and ex-
amples are shown in this section. Full experimental data are
available as Supporting Information.
The structures of products were assigned using IR, 1H NMR
and 13C NMR spectroscopy, including 2D techniques (COSY,
HMQC, HMBC), and by mass spectrometry. The purity of products
was confirmed as >95% using elemental analysis, HPLC with UV
detection, or LC-MS. Unless stated otherwise, all chemicals and
solvents were obtained from Sigma Aldrich (UK) and used
without further purification. Solvents were dried using the Pure
Solv™ purification system (Innovative Technology, Inc). TLC was
carried out using silica gel matrix on aluminium foil (Sigma
Aldrich). Column chromatography used silica gel, particle size
35e70 micron (Fisher Scientific). CHN analysis, NMR spectros-
copy and mass spectrometry were carried out using, respectively,
an Exeter Analytical CE 440 Elemental Analyser instrument,
AVANCE DPX 300 MHz Bruker NMR instrument at 300 MHz (1H)
or 75 MHz (13C) and an Esquire 3000 þ Ion Trap mass spec-
trometer. IR spectroscopy was carried out with a Perkin Elmer
Spectrum BX FT-IR system. Melting point determination was
carried out using a Stuart SMP 30 melting point apparatus. ESI
HRMS was performed courtesy of the EPSRC National Mass
Spectrometry Service Centre.
the requirement for enzymatic action on the
a-carboxyl ester,
thioester, and -glutamyl groups to release cysteamine. In
g
Fig. 3. Oxidation of cysteamine (0.01 M cysteamine hydrochloride in water, room
temperature, in triplicate); detected as the CMQT derivative using HPLC with UV
detection.

212
L. Frost et al. / European Journal of Medicinal Chemistry 109 (2016) 206e215
4.1. General method for preparation of compounds 6a-c
4.3. General method for preparation of compounds 8a-c
2-(S)-tert-Butoxycarbonylamino-pentanedioic acid 5-benzyl
ester 5 was dissolved in methanol with 10% water (10 mL). The
solution was neutralized to pH 7 by drop-wise addition of aqueous
Cs2CO3 (20%). Following evaporation under high vacuum, the
resulting syrup was dissolved in dry DMF (20 mL) and treated with
the appropriate alkyl bromide (2.88 molar equivalents). The reac-
tion was stirred at room temperature overnight. Ethyl acetate
(20 mL) was added and the organic layer washed with water
(3 ꢂ 20 mL), 10% K2CO3 (2 ꢂ 20 mL), and brine (20 mL), and then
dried over magnesium sulfate. The resulting solution was concen-
trated under reduced pressure to afford the product as a white
solid.
Compounds 7a-c were dissolved in dry DCM (1 g in 50 mL)
under nitrogen. To this was added triethylamine (1 mol equivalent),
HOBt (1 mol equivalent) and 1,3-diisopropylcarbodiimide (1 mol
equivalent) and the mixture was stirred, at 0 ꢁC for 15 min. Cyste-
amine hydrochloride (1 mol equivalent) was then added and the
mixture stirred at 0 ꢁC for 45 min, then at room temperature
overnight. The solution was filtered and the combined organic
layers were washed with 10% w/v potassium carbonate, 10% w/v
citric acid and water, and then dried over magnesium sulfate. The
resulting solution was concentrated under reduced pressure to
afford the crude product. Purification by column chromatography
(petrol/ethyl acetate as eluent) on silica was carried out to obtain
the corresponding pure compound as a white solid 8a-c.
4.1.1. 2-(S)-tert-Butoxycarbonylaminopentanedioic acid 5-benzyl
ester 1-propyl ester (6b)
4.3.1. 2-(S)-tert-Butoxycarbonylamino-4-(2-mercapto-
ethylcarbamoyl)butyric acid butyl ester (8c)
Following the procedure described in 4.1, the reaction of 2-tert-
butoxycarbonylaminopentanedioic acid 5-benzyl ester 5 (1.00 g,
2.96 mmol) with propyl bromide (1.05 mL, 8.52 mmol) gave
product 6b (1.01 g, 90%): Mp 51e52 ꢁC; 1H NMR (300 MHz, CDCl3)
dH 0.95 (t, J ¼ 7.1 Hz, 3H, 30-CH3), 1.43 (s, 9H, t-Boc CH3), 1.68 (sextet,
J ¼ 7.1 Hz, 2H, 20-CH2), 1.96 (m, 1H, 3-CHa), 2.21 (m, 1H, 3-CHb), 2.41
(dd, J ¼ 16.5 and 6.8 Hz,1H, 4-CHa), 2.50 (dd, J ¼ 16.5 and 6.8 Hz,1H,
4-CHb), 4.09 (t, J ¼ 7.1 Hz, 2H, 10-CH2), 4.33 (m, 1H, 2-CH), 5.08 (br d,
1H, NH), 5.12 (s, 2H, CH2Ph), 7.35 (m, 5H, CH2Ph); 13C NMR (75 MHz,
CDCl3) dC 10.3 (30-C), 21.9 (20-C), 28.0 (3-C), 28.3 (t-Boc CH3), 30.4
(4-C), 53.0 (2-C), 62.1 (10-C), 66.5 (CH2Ph), 80.0 (quat., t-Boc C),
128.3 (Ph CH),128.3 (Ph CH),128.6 (Ph CH),135.9 (quat., Ph C),155.4
(quat., NCO2), 172.2 (quat., 5-C), 172.5 (quat., 1-C); IR cmꢀ1 3361
(NeH), 1760 (C]O, ester), 1720 (C]O, ester), 1686 (C]O, carba-
mate), 1365 (CeN), 1169 (CeO), 750, 698 (Ar CeH); Anal. calculated
for C20H29NO6 (Mr 379.45) C 63.31, H 7.70, N 3.69%; Found C 63.51, H
7.68, N 3.73%; ESI MS 401.1 [MNaþ].
Following the procedure described in 4.3, the reaction of com-
pound 7c (0.74 g, 2.44 mmol) with cysteamine hydrochloride gave
product 8c (0.62 g, 70%): Mp 62e63 ꢁC; 1H NMR (300 MHz, CDCl3)
dH 0.94 (t, J ¼ 7.2 Hz, 3H, 40-CH3), 1.3 (t, J ¼ 6.5 Hz, 1H, SH), 1.31
(sextet, J ¼ 7.2 Hz, 2H, 30-CH2), 1.43 (s, 9H, t-Boc CH3), 1.65 (quintet,
J ¼ 7.2 Hz, 2H, 20-CH2), 1.97 (m, 1H, 3-CHa), 2.20 (m, 1H, 3-CHb), 2.41
(dd, J ¼ 16.5 and 6.8 Hz, 1H, 4-CHa), 2.50 (dd, J ¼ 16.5 and 6.8 Hz,1H,
4-CHb), 2.62 (q, J ¼ 6.5 Hz, 2H, 8-CH2), 3.36 (q, J ¼ 6.5 Hz, 2H, 7-
CH2), 4.07 (t, J ¼ 7.2 Hz, 2H, 10-CH2), 4.33 (m, 1H, 2-CH), 5.14 (br
d, 1H, 2-CNH), 6.22 (br s, 1H, 6-NH); 13C NMR (75 MHz, CDCl3) dC
10.35 (40-C), 21.9 (30-C), 24.6 (8-C), 28.3 (t-Boc CH3), 28.7 (3-C), 30.4
(8-C), 32.6 (4-C), 38.5 (20-C), 39.5(7-C), 53.3 (2-C), 67.1 (10-C), 80.2
(quat., t-Boc C), 155.9 (quat., NCO2), 172.4 (quat., 5-C), 172.6 (quat.,
1-C); IR cmꢀ1 3350 (NeH), 2422 (SH), 1721 (C]O, ester), 1683 (C]
O, carbamate), 1649 (C]O, amide), 1348 (CeN), 1162 (CeO); Anal.
calculated for C16H30N2O5S (Mr 362.48): C 53.28, H 8.34, N 7.73%;
Found: C 53.41, H 8.18, N 7.78%; ESI MS 385.2 [MNaþ].
4.2. General method for preparation of compounds 7a-c
4.4. General method for preparation of compounds 9a-c
Compounds 6a-c were dissolved in methanol (50 mL). Palla-
dium on carbon (10%, Alfa Aesar) was added (0.1 mass equivalent)
and the mixture was stirred for approximately 3 h, at room tem-
perature, under a hydrogen atmosphere at 3 bars of pressure. The
catalyst was removed by filtration over celite and the resulting
solution was concentrated under reduced pressure to afford a col-
ourless oil, which was dissolved in a minimum amount of ethyl
acetate, and then hexane was added until the solution became
turbid. The solution was stored at 0 ꢁC overnight and the resultant
precipitate was collected by filtration.
Compounds 8a-c and triethylamine (1 mol equivalent) were
protected from moisture and stirred in dry THF (1 g in 30 mL) at
room temperature. To this acetic anhydride (1 mol equivalent) was
added and the mixture stirred at 40 ꢁC overnight. The organic layer
was washed with 10% potassium carbonate, brine and water, and
then dried over magnesium sulfate. The resulting solution was
concentrated under reduced pressure to afford the crude product.
Purification by column chromatography (dichloromethane/meth-
anol as eluent) was required for obtaining the corresponding pure
products 9a-c as a white solid.
4.2.1. 2-(S)-tert-Butoxycarbonylamino-pentanedioic acid 1-butyl
ester (7c)
4.4.1. 4-(2-Acetylsulfanyl-ethylcarbamoyl)-2-(S)-tert-
butoxycarbonylaminobutyric acid butyl ester (9c)
Following the procedure described in 4.2, compound 6c (2.0 g,
5.08 mmol) was reduced to give product 7c (1.377 g, 89%): Mp
52e54 ꢁC; 1H NMR (300 MHz, CDCl3) dH 0.92 (t, J ¼ 7.1 Hz, 3H, 40-
CH3), 1.36 (sextet, J ¼ 7.1 Hz, 2H, 30-CH2), 1.43 (s, 9H, t-Boc CH3), 1.62
(quintet, J ¼ 6.5 Hz, 2H, 20-CH2), 1.95 (m, 1H, 3-CHa), 2.15 (m, 1H, m,
3-CHb), 2.41 (dd, J ¼ 16.7 and 6.9 Hz, 1H, 4-CHa), 2.51 (dd, J ¼ 16.7
and 6.9 Hz,1H, 4-CHb), 4.13 (t, J ¼ 6.5 Hz, 2H,10-CH2), 4.30 (m, 1H, 2-
CH), 5.24 (br d, 1H, NH); 13C NMR (75 MHz, CDCl3) dC 13.6 (40-C),
19.0 (30-C), 27.7 (3-C), 28.2 (t-Boc CH3), 30.0 (4-C), 30.5 (20-C), 52.9
(2-C), 65.4 (10-C), 80.1 (quat., t-Boc C), 155.5 (NCO2), 172.3 (quat., 1-
C), 177.0 (quat., 5-C); IR cmꢀ1 3357 (NeH), 2962 (broad, OeH), 1709
(broad, C]O),1367 (CeN), 1160 (CeO); Anal. calculated for
Following the procedure described in 4.4, the reaction of com-
pound 8c (1.00 g, 2.76 mmol)) with acetic anhydride gave product
9c (0.62 g, 55%): Mp 70e72 ꢁC; 1H NMR (300 MHz, CDCl3) dH 0.93 (t,
J ¼ 7.1 Hz, 3H, 40-CH3), 1.38 (sextet, J ¼ 7.1 Hz, 2H, 30-CH2), 1.45 (s,
9H, t-Boc CH3), 1.63 (quintet, J ¼ 7.1 Hz, 2H, 20-CH2), 1.92 (m, 1H, 3-
CHa), 2.15 (m, 1H, 3-CHb), 2.25 (dd, J ¼ 16.4 and 6.7 Hz, 1H, 4-CHa),
2.29 (dd, J ¼ 16.4 and 6.7 Hz,1H, 4-CHb), 2.35 (s, 3H,11-CH3), 3.04 (t,
J ¼ 6.5 Hz, 2H, 8-CH2), 3.44 (q, J ¼ 6.5 Hz, 2H, 7-CH2), 4.14 (t,
J ¼ 7.1 Hz, 2H, 10-CH2), 4.27 (m, 1H, 2-CH), 5.30 (br d, 1H, 2-CNH),
6.30 (br s,1H,m, 6-NH); 13C NMR (75 MHz, CDCl3) dC 13.7 (40-C),19.1
(30-C), 28.4 (t-Boc CH3), 28.9 (3-C), 29.1(8-C), 30.6 (11-C) 30.6 (20-C),
32.6 (4-C), 39.5 (7-C), 53.2 (2-C), 65.5 (10-C), 80.1 (quat., t-Boc C),
155.9 (quat., NCO2), 172.1 (quat., 5-C), 172.3 (quat., 1-C), 196.0 (quat.,
10-C); IR cmꢀ1 3320 (NeH), 1728 (C]O, ester), 1695 (C]O,
C
14H25NO6 (Mr 303.36) C 55.43, H 8.31, N 4.62%; Found C 55.57, H
8.19, N 4.78%; ESI MS 326.2 [MNaþ].

L. Frost et al. / European Journal of Medicinal Chemistry 109 (2016) 206e215
213
thioester) 1686 (C]O, carbamate), 1648 (C]O, amide), 1367 (CeN),
1164 (CeO); Anal. calculated for C18H32N2O5S (Mr 404.52) C 53.44,
H 7.97, N 6.93%; Found C 53.77, H 8.03, 6.95%; ESI MS 427.2 [MNaþ].
CH3), 29.0 (8-C), 32.6 (4-C), 39.6 (7-C), 46.5 (quat., 11-C), 53.1 (2-C),
67.1 (10-C), 80.0 (quat., t-Boc C), 155.8 (quat., NCO2), 172.0 (quat., 5-
C), 172.3 (quat., 1-C), 207.1 (quat., 10-C); IR cmꢀ1 3320 (NeH), 1678
(broad, C]O), 1365 (CeN), 1165 (CeO); Anal. calculated for
4.5. General method for preparation of compounds 10a-c
C20H36N2O6S (Mr 432.57) C 55.53, H 8.39, N 6.48%; Found C 55.30, H
8.65, N 6.55%; ESI MS 455.2 [MNaþ].
Compounds 8a-c were stirred in dry dichloromethane (1 g in
30 mL), under dry conditions and at room temperature. To this was
added pyridine (1 mol equivalent) and benzoyl chloride (1 mol
equivalent) and the mixture was stirred at room temperature
overnight. The organic layer was washed with 10% potassium car-
bonate, brine and water, and then dried over magnesium sulfate.
The resulting solution was concentrated under reduced pressure to
afford the crude product. Column chromatography (petrol/ethyl
acetate as eluent) on silica was required to obtain the corre-
sponding pure product 10a-c as a white solid.
4.7. General method for the preparation of prodrugs 1-3a-c
Individually, compounds 9a-c,10a-c and 11a-c were dissolved in
a solution of HCl (2 M) in diethyl ether (Alfa Aesar) and stirred at
room temperature. The reaction was monitored by TLC (petroleum
ether/ethyl acetate).
4.7.1. 4-{[2-(S)-(Acetylsulfanyl)ethyl]carbamoyl}-1-ethoxy-1-
oxobutan-2-aminium chloride (1a)
Following the general procedure described in 4.7, compound 9a
(3.00 g, 3.97 mmol) was stirred in HCl (2 M) in diethyl ether
overnight (100 mL). The white precipitate was collected, washed
with diethyl ether and dried under vacuum (2.01 g, 80%): Mp:
97e100 ꢁC; 1H NMR (300 MHz, d6-DMSO) dH 1.21 (t, J ¼ 7.0 Hz, 3H,
20-CH3), 2.10 (m, 1H, 3-CHa), 2.15 (m, 1H, 3-CHb), 2.25 (dd, J ¼ 16.6
and 7.2 Hz,1H, 4-CHa), 2.29 (dd, J ¼ 16.6 and 7.2 Hz, 1H, 4-CHb), 2.30
(s, 3H, 11-CH3), 2.74 (t, J ¼ 6.6 Hz, 2H, 8-CH2), 3.29 (q, J ¼ 6.6 Hz, 2H,
7-CH2), 4.15 (q, J ¼ 7.0 Hz, 2H, 10-CH2), 4.20 (m, 1H, 2-CH); 13C NMR
(75 MHz, d6-DMSO) dC 14.7 (20-C), 23.2 (3-C), 28.8 (11-C), 30.5 (8-C),
32.6 (4-C), 39.5 (7-C), 53.1 (2-C), 61.5 (10-C), 172.4 (quat., 5-C), 172.6
(quat., 1-C), 196.0 (quat., 10-C); IR cmꢀ1 3330 (Nþ-H3), 1656 (broad,
C]O), 1366 (CeN), 1151 (CeO); Anal. calculated for C11H21ClN2O4S
(Mr 312.81) C 42.24, H 6.77, N 8.96%; Found C 42.08, H 6.53, N 8.74%;
HRMS calculated for [C11H21N2O4S]þ 277.1217, found 277.1217.
4.5.1. 4-(2-Acetylsulfanyl-ethylcarbamoyl)-2-(S)-tert-
butoxycarbonylaminobutyric acid ethyl ester (10a)
Following the procedure described in 4.5, the reaction of com-
pound 8a (0.25 g, 0.75 mmol) with benzoyl chloride gave product
10a (0.08 g, 24%): Mp 78e79 ꢁC; 1H NMR (300 MHz, CDCl3) dH 1.26
(t, 3H, J ¼ 7.0 Hz, 20-CH3), 1.43 (s, 9H, t-Boc CH3), 1.93 (m, 1H, 3-CHa),
2.19 (m, 1H, 3-CHb), 2.26 (dd, J ¼ 16.6 and 6.9 Hz, 1H, 4-CHa), 2.29
(dt, J ¼ 16.6 and 6.9 Hz, 1H, 4-CHb), 3.25 (t, J ¼ 6.3 Hz, 2H, 8-CH2),
3.55 (q, J ¼ 6.3 Hz, 2H, 7-CH2), 4.18 (q, J ¼ 7.0 Hz, 2H, 10-CH2), 4.27
(m, 1H, 2-CH), 5.28, (br d, 1H, 2-CNH), 6.44 (br s, 1H, 6-NH), 7.44 (t,
J ¼ 7.2 Hz, 2H, 13-CH), 7.59 (tt, J ¼ 7.2 and 1.4 Hz, 1H, 14-CH), 7.96
(dd, J ¼ 7.2 and 1.4 Hz, 2H, 12-CH); 13C NMR (75 MHz, CDCl3) dC 14.2
(20-C), 28.3 (t-Boc CH3), 28.6 (3-C), 29.1(8-C), 32.6 (4-C), 39.6 (7-C),
53.1 (2-C), 61.5 (10-C), 80.1 (quat., t-Boc C), 127.3 (Ph CH), 128.7 (Ph
CH), 133.6 (Ph CH), 136.8 (quat., 11-C), 158.3 (quat., NCO2), 172.2
(quat., 5-C), 172.3 (quat., 1-C), 192.0 (quat., 10-C); IR cmꢀ1 3348
(NeH), 1734 (C]O, ester), 1678 (C]O, carbamate), 1662 (C]O,
thioester), 1644 (C]O, amide), 1367 (CeN), 1160 (CeO), 777 and
688 (Ar CeH); Anal. calculated for C21H30N2O6S (Mr 438.54) C 57.51,
H 6.90, N 6.39%; Found C 57.52, H 6.96, N 6.36%; ESI MS 461.2
[MNaþ].
4.8. Cell lines
HaCaT keratinocytes were commercially obtained from Cell Line
Services. PT37 cells were donated by the Leuven University Hos-
pital, Belgium. Cell lines were cultured in DMEM/F12 medium
(Invitrogen) supplemented with 2 mM L-glutamine (1%, Sigma
Aldrich), penicillin-streptomycin (1%, Sigma Aldrich) and heat
inactivated fetal calf serum (10%, Biosera). Cell monolayers were
maintained at 37 ꢁC under an atmosphere containing 5% carbon
dioxide.
4.6. General method for the preparation of compounds 11a-c
Compounds 8a-c were stirred in dry dichloromethane (1 g in
30 mL), under dry conditions and at room temperature. To this was
added pyridine (1 mol equivalent) and pivaloyl chloride (1 mol
equivalent) and the mixture was stirred at room temperature
overnight. The organic layer was washed with 10% potassium car-
bonate, brine and water, then dried over magnesium sulfate. The
resulting solution was concentrated under reduced pressure to
afford the crude product. Column chromatography (petrol/ethyl
acetate as eluent) on silica and recrystallisation (ethyl acetate/
hexane) was required to obtain the corresponding pure product
11a-c as a white solid.
4.9. General method for MTT assay
Prodrugs were dissolved in medium to give a starting concen-
tration of 500
mg/mL. In a 96 well plate, serial dilutions were carried
out to give subsequent concentrations of 250, 125, 62.5, 31.25,
15.625, 7.81, 3.91 and 1.95 mg/mL. Cells were exposed to the pro-
drugs for a period of 48 h while incubated at 37 ꢁC under 5% carbon
dioxide. Following exposure, the prodrug containing medium was
removed and the cells in the 96 well plates were rinsed with PBS
4.6.1. 2-(S)-tert-Butoxycarbonylamino-4-[2-(2,2-dimethyl-
(phosphate buffered saline, pH 7.4). MTT (0.5 mg/mL in PBS, 200 mL)
propionylsulfanyl)-ethylcarbamoyl]butyric acid propyl ester (11b)
Following the procedure described in 4.6, the reaction of com-
pound 8b (0.20 g, 0.57 mmol) with pivaloyl chloride gave product
11b (0.10 g, 43%): Mp 38e40 ꢁC; 1H NMR (300 MHz, CDCl3) dH 0.94
(t, J ¼ 7.0 Hz, 3H, 30-CH3),1.24 (s, 9H,12-CH3), 1.44 (s, 9H, t-Boc CH3),
1.67 (sextet, J ¼ 7.0 Hz, 2H, 20-CH2), 1.91 (m, 1H, 3-CHa), 2.15 (m, 1H,
3-CHb), 2.24 (dd, J ¼ 16.4 and 6.8 Hz, 1H, 4-CHa), 2.26 (dd, J ¼ 16.4
and 6.8 Hz, 1H, 4-CHb), 3.00 (t, J ¼ 6.3 Hz, 2H, 8-CH2), 3.42 (q,
J ¼ 6.3 Hz, 2H, 7-CH2), 4.10 (t, J ¼ 7.0 Hz, 2H, 10-CH2), 4.27 (m, 1H, 2-
CH), 5.30 (br d, 1H, 2-CNH), 6.35 (br s, 1H, 6-NH); 13C NMR (75 MHz,
CDCl3) dC 10.3 (30-C), 21.9 (20-C), 27.4 (12-C), 28.1 (3-C), 28.3 (t-Boc
was added to each well and incubated for 2 h at 37 ꢁC under 5%
carbon dioxide. Upon removal of the dye, the cells were again
washed with PBS before the addition of a solution of 90% iso-
propanol and 10% DMSO (200 mL). The plates were incubated in the
dark for 10 min and then the absorbance was recorded at 595 nm
using a Teccan plate reader and Revelation software. EC50 values
were calculated from the dose response curves by non-linear
regression analysis using GraphPad Prism®. Cell survival was
expressed as a percentage compared to control wells. Percentage
survival value at each concentration represents 12 replicates, which
were averaged and normalised to calculate the percentage survival.

214
L. Frost et al. / European Journal of Medicinal Chemistry 109 (2016) 206e215
multisystemic disease, Pediatr. Nephrol. 23 (2008) 863e878.
[3] F. Manz, N. Gretz, Progression of chronic renal faluire in a historical group of
patients with nephropathic cystinosis. European collaborative study on cys-
tinosis, Pediatr. Nephrol. 8 (1994) 466e471.
[4] W.A. Gahl, J.Z. Balog, R. Kleta, Nephropathoc cystinosis in adults: natural
history and effects of oral cysteamine therapy, Ann. Intern. Med. 147 (2007)
242e250.
[5] R. Dohil, R.O. Newbury, Z.M. Sellers, R. Deutsch, J.A. Schneider, The evaluation
and treatment of gastrointestinal disease in children with cystinosis receiving
cysteamine, J. Pediatr. 143 (2003) 224e230.
[6] M. Besouw, H. Blom, A. Tangerman, A. de-Graff-Hess, E. Levtchenko, The origin
of halitosis in cystinotic patients due to cysteamine treatment, Mol. Gen.
Metab. 91 (2007) 228e233.
4.10. General method for Sirius T3 measurement of pKa and Log P
To measure the pKa, between 1 and 2 mg of test compound was
weighed accurately into a vessel and placed into the Sirius T3 in-
strument for pH metric analysis. The pKa was computationally
determined following the automated potentiometric acid-base
titration over a pH range of 2e12 (mean of 3 titrations). For the
automated pH-metric measurement of Log P, an accurately
weighed sample was dissolved in a two-phase water-octanol sys-
tem (1:2), and titrated over a pH range (2e12). Log P results were
obtained by computational process using the Sirius T3 software.
[7] N. Bouazza, J.M. Treluyer, C. Ottolenghi, S. Urien, G. Deschenes, D. Ricquier,
P. Niaudet, B. Chadefaux-Vekemans, Population pharmacokinetics and phar-
macodynamics of cysteamine in nephropathic cystinosis patients, Orphanet J.
Rare. Dis. 6 (2011) 86e94.
4.11. General method for measurement of total cellular cysteamine
content
[8] E.N. Levtchenko, C.M. van Dael, A.C. de Graaf-Hess, M.J.G. Wilmer, L.P. van den
Heuvel, L.A. Monnens, H.J. Blom, Strict cysteamine dose regimen is required to
prevent nocturnal cystine accumulation in cystinosis, Pediatr. Nephrol. 21
(2006) 110e113.
Test prodrugs were dissolved in culture medium to obtain the
required concentration (20 mM). Confluent cell monolayers were
ꢀ
[9] C.B. Langman, L.A. Greenbaum, M. Sarwal, P. Grimm, P. Niaudet, G. Deschenes,
E. Cornelissen, D. Morin, P. Cochat, D. Matossian, S. Gaillard, M.J. Bagger,
incubated with the prodrug-containing medium at 37 ꢁC for a
defined time period (0, 2, 6, 16 or 24 h). Monolayers were harvested
using trypsin-EDTA (0.25%), re-suspended in growth medium and
centrifuged for 5 min at 1000 rpm. The resulting cell pellets were
washed with cold PBS with centrifugation at 1000 rpm for a further
P. Rioux,
A randomized controlled crossover trial with delayed-release
cysteamine bitartrate in nephropathic cystinosis: effectiveness on white
blood cell cysteine levels and comparison of safety, Clin. J. Am. Soc. Nephrol. 7
(2012) 1112e1120.
[10] M.T. Besouw, R. Bowker, J.P. Duterte, F. Emma, W.A. Gahl, M. Greco,
M.R. Lilien, J. McKiernan, F. Nobili, J.A. Schneider, F. Skovby, L.P. van den
Heuvel, W.G. Van't Hoff, E.N. Levtchenko, Cysteamine toxicity in patients with
cystinosis, J. Pediatr. 159 (2011) 1004e1011.
[11] M. Borrell-Pages, J.M. Canals, F.P. Cordelieres, J.A. Parker, J.R. Pineda,
G. Grange, E.A. Bryson, M. Guillermier, E. Hirsch, P. Hantraye, M.E. Cheetham,
C. Neri, J. Alberch, E. Brouillet, F. Saudou, S. Humbert, Cystamine and cyste-
amine increase brain levels of BDNF in Huntington disease via HSJ1b and
transglutaminase, J. Clin. Invest. 116 (2006) 1410e1424.
[12] A. Dedeoglu, J.K. Kubilus, T.M. Jeitner, S.A. Matson, M. Bogdanov, N.W. Kowall,
W.R. Matson, A.J.L. Cooper, R.R. Ratan, M.F. Beal, S.M. Hersch, R.J. Ferrante,
Therapeutic effects of cystamine in a murine model of Huntington's disease,
J. Neurosci. 22 (2002) 8942e8950.
[13] C. Charrier, C. Rodger, J. Robertson, A. Kowalczuk, N. Shand, D. Fraser-Pitt,
D. Mercer, D. O'Neil, Cysteamine (Lynovex®), a novel mucoactive antimicro-
bial & antibiofilm agent for the treatment of cystic fibrosis, Orphanet J. Rare
Dis. 9 (2014) 189e200.
[14] D. De Stefano, V.R. Villella, S. Esposito, A. Tosco, A. Sepe, F. De Gregorio,
L. Salvadori, R. Grassia, C.A. Leone, G. De Rosa, M.C. Maiuri, M.M. Pettoello,
S. Guido, A. Bossi, A. Zoli, A. Venerando, L.A. Pinna, A. Mehta, G. Bona,
G. Kroemer, L. Maiuri, V. Raia, Restoration of CFTR function in patients with
cystic fibrosis carrying the F508del-CFTR mutation, Autophagy 10 (2014)
2053e2074.
[15] G. Devereux, D. Fraser-Pitt, J. Roberston, E. Devlin, D. Mercer, D. O'Neil,
Cysteamine as a future intervention in cystic fibrosis against current and
emerging pathogens: a patient-based ex vivo study confirming its antimi-
crobial and mucoactive potential in sputum, EBioMed 2 (2015) 1507e1512.
[16] R. Dohil, S. Schmeltzer, B.L. Cabrera, T. Wang, J. Durelle, K.B. Duke,
J.B. Schwimmer, J.E. lavine, Enteric-coated cysteamine for the treatment of
paediatric non-alcoholic fatty liver disease, Ailm. Pharmacol. Ther. 33 (2011),
1036e1034.
[17] D.M. Okamura, N.M. Bahrami, S. Ren, K. Pasichnyk, J.M. Williams, J.A. Gangoiti,
J.M. Lopez-Guisa, I. Yamaguchi, B.A. Barshop, J.S. Duffield, A.A. Eddy, Cyste-
amine modulates oxidative stress and blocks myofibroblast activity in CKD,
J. Am. Soc. Nephrol. 25 (2014) 43e54.
[18] R.L. Pisoni, G.Y. Park, V.Q. Velilla, J.G. Thoene, Detection and characterization
of a transport system mediating cysteamine entry into human fibroblast ly-
sosomes. Specificity for aminoethylthiol and aminoethylsulfide derivatives,
J. Biol. Chem. 270 (1995) 1179e1184.
[19] A. Jezegou, E. Llinares, C. Anne, S. Kieffer-Jaquinod, S. O'Regan, J. Aupetit,
A. Chabli, C. Sagne, C. Debacker, B. Chadefaux-Vekemans, A. Journet, B. Andre,
B. Gasnier, Heptahelical protein PQLC2 is a lysosomal cationic amino acid
exporter underlying the action of cysteamine in cystinosis therapy, Proc. Nat.
Aacd. Sci. 109 (2012) E3434eE3443.
[20] B. Liu, H. Du, R. Rutkowski, A. Gartner, X. Wang, LAAT-1 is the lysosomal
lysine/arginine transporter that maintains amino acid homeostasis, Science
337 (2012) 351e354.
[21] R.J. Anderson, D. Cairns, W.A. Cardwell, M. Case, P.W. Groundwater, A.G. Hall,
L. Hogarth, A.L. Jones, O. Meth-Cohn, P. Suryadevara, A. Tindall, J.G. Thoene,
Design, synthesis and initial in vitro evaluation of novel prodrugs for the
treatment of cystinosis, Lett. Drug Des. Discov. 3 (2006) 336e345.
[22] B. Testa, J.M. Mayer, Hydrolysis in Drug and Prodrug Metabolism, Wiley,
Zurich, 2003, pp. 236e365.
[23] S.C. Laizure, V. Herring, Z. Hu, K. Witbrodt, R.B. Parker, The role of human
carboxylesterases in drug metabolism: have we overlooked their importance?
Pharmacotherapy 33 (2013) 210e222.
5 min. The pellets were then re-suspended in sucrose (400
0.25 M) and sonicated before the suspension was centrifuged at
13,000 rpm for 10 min. Following pellet suspension, 5 L of the cell
supernatant was added to a microplate (in triplicate). To this, was
mL,
m
added 240 mL Bradford reagent (Sigma Aldrich, Poole, UK) and the
plate incubated for 5 min at 37 ꢁC. The absorbance was then
measured at 595 nm. The protein concentration was determined by
comparison to a calibration curve prepared using bovine serum
albumin standards (0.1e1.5 mg/mL protein).
The resultant supernatant was stored at ꢀ80 ꢁC until HPLC
analysis. For HPLC analysis, cell supernatant (150
mL), 1-octanol
(30
NaOH/DMSO (300 mL/L)) and HCl (30
were added to a reaction vial and vortex mixed for 3 min. This was
followed by the addition of N-ethylmorpholine (100 L, 1.6 M),
phosphate buffer (100 L, 0.2 M) and CMQT (20 L, 0.1 M), before
L, 0.5 M)
mL), sodium borohydride (40
mL, 4 M in a solution of 0.1 M
mL, 3 M added dropwise)
m
m
m
mixing for a further 3 min. Lastly, glacial acetic acid (50
was added before HPLC analysis.
m
Acknowledgements
This work was made possible by funding provided by The Cys-
tinosis Foundation UK, who had no involvement in the experi-
mental design, analysis and interpretation of results, writing of the
manuscript or decision to publish. Gratitude is expressed to Pro-
fessor Elena Levtchenko and Dr Martine Besouw (Leuven University
Hospital, Belgium), and Dr Martijn Wilmers (Nijmegen University,
Netherlands), for the provision of cells, to Dr Stephen Waldek for
his helpful suggestions on the manuscript, and to Barrie Thynne, Joy
Otun, and Phillip Rolfe for their technical contributions. We also
thank the EPSRC UK National Mass Spectrometry Facility at Swan-
sea University for providing accurate mass measurements.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2015.12.027.
References
[1] W.A. Gahl, J.G. Thoene, J.A. Schneider, Cystinosis, New Eng. J. Med. 347 (2000)
[24] M.R. Redinbo, P.M. Potter, Mammalian carboxylesterases: from drug targets to
protein therapeutics, Drug Discov. Today 10 (2005) 313e325.
111e121.
[2] G. Nesterova, W. Gahl, Nephropatic cystinosis: late complications of
a

L. Frost et al. / European Journal of Medicinal Chemistry 109 (2016) 206e215
215
[25] V.R. Guarino, The molecular design of prodrugs by functional groups, in:
J. Rautio, R. Mannhold, H. Kubinyi, G. Folkers (Eds.), Prodrugs and Targeted
Delivery: Towards Better ADME Properties;, Wiley, Weinheim, 2011, pp.
31e111.
blood-brain barrier, J. Biol. Chem. 270 (1995) 14907e14912.
[33] S. Wilk, H. Mizoguchi, M. Orlowski, N-acyl-gamma-glutamyl derivatives of
sulfamethoxazole as models of kidney-selective prodrugs, J. Pharmacol. Exp.
Ther. 212 (1980) 167e172.
[26] E.E. Ramsay, S. Decollogne, S. Joshi, A. Corn, M. Apte, A. Pompella, P. Hogg,
[34] S. Wilk, H. Mizoguchi, M. Orlowski, gamma-Glutamyl dopa: a kidney-specific
dopamine precursor, J. Pharmacol. Exp. Ther. 206 (1978) 227e232.
[35] T. Mossman, Rapid colorimetric assay for cellular growth and survival:
application to proliferation and cytotoxicity assays, J. Immunol. Methods. 65
(1983) 55e63.
[36] E. Bald, R. Glowacki, 2-Chloro-1-methylquinoliniumtetrafluoroborate as an
effective and thiol specific UV-tagging reagent for liquid chromatography,
J. Liq. Chrom. Rel. Technol. 24 (2001) 1323e1339.
[37] M.J. Wilmer, A. de Graaf-Hees, H.J. Blom, H.B. Dijkman, L.A. Monnens, L.P. van
den Heuvel, E.N. Levtchenko, Elevated oxidized glutathione in cystinotic
proximal tubular epithelial cells, Biochem. Biophys. Res. Commun. 337 (2005)
610e614.
[38] T. Khomenko, J. Kolodney, J.T. Pinto, G.D. McLaren, X. Deng, L. Chen,
G. Tolstanova, B. Paunovic, B.F. Krasnikov, N. Hoa, A.J.L. Cooper, S. Szabo, New
mechanistic explanation for the localization of ulcers in the rat duodenum:
role of iron and selective uptake of cysteamine, Arch. Biochem. Biophys. 525
(2012) 60e70.
P.J. Dilda, Employing pancreatic tumor geglutamyltransferase for therapeutic
delivery, Mol. Pharmacol. 11 (2014) 1500e1511.
[27] S.B. Rosalki, Gamma-glutamyl transpeptidase, in: O. Bodanski, A.L. Latner
(Eds.), Advances in Clinical Chemistry, 17, Academic Press, New York, London,
1975, pp. 53e107.
[28] S.D.J. Magnan, F.N. Shirota, H.T. Nagasawa, Drug latentiation by gamma-
glutamyl transpeptidase, J. Med. Chem. 25 (1982) 1018e1021.
[29] S. Wickham, M.B. West, P.F. Cook, M.H. Hanigan, Gamma-glutamyl com-
pounds: substrate specificity of gamma-glutamyl transpeptidase enzymes,
Anal. Biochem. 414 (2011) 208e214.
[30] K.A. Curto, W.E. Sweeney, E.D. Avner, N.P. Piesco, N.P. Curthoys, Immunocy-
tochemical localization of gamma-glutamyltranspeptidase during fetal
development of mouse kidney, J. Histochem. Cytochem. 36 (1988) 159e166.
[31] T.Q. Garvey, P.E. Hyman, K.H. Isselbacher, Gamma-glutamyl transpeptidase of
rat intestine: localization and possible role in amino acid transport, Gastro-
enterology 71 (1976) 778e785.
[32] M.M. Sanchez del Pino, R.A. Hawkins, D.R. Peterson, Biochemical discrimina-
tion between luminal and abluminal enzyme and transport activities of the
[39] B.E. Svensson, S. Lindvall, Myeloperoxidase-oxidase oxidation of cysteamine,
Biochem. J. 249 (1988) 521e530.