A combinatorial approach towards the synthesis of non-hydrolysable triazole-iduronic acid hybrid inhibitors of human α-l-iduronidase: Discovery of enzyme stabilizers for the potential treatment of MPSI

Article abstract of DOI:10.1039/c7cc09642a

Preparation of substituent-diverse, triazole-iduronic acid hybrid molecules by click reaction of an azido iduronic acid derivative with randomly chosen alkynes is described. Library members were screened for their ability to inhibit α-l-iduronidase, and hit molecules and analogues were then investigated for their ability to stabilize rh-α-IDUA in a thermal denaturation study. This work resulted in the discovery of the first small molecules that can be used to stabilize exogenous rh-α-IDUA protein in vitro.

Full text of DOI:10.1039/c7cc09642a

ChemComm
View Article Online
View Journal
COMMUNICATION
A combinatorial approach towards the synthesis
of non-hydrolysable triazole–iduronic acid hybrid
inhibitors of human a-L-iduronidase: discovery
of enzyme stabilizers for the potential treatment
of MPSI†
Cite this:DOI: 10.1039/c7cc09642a
Received 19th December 2017,
Accepted 15th February 2018
DOI: 10.1039/c7cc09642a
ab
Wei-Chieh Cheng,
*
Cheng-Kun Lin,a Huang-Yi Li,c Yu-Chien Chang,b
Sheng-Jhih Lu,a Yu-Shin Chena and Shih-Ying Changa
rsc.li/chemcomm
Preparation of substituent-diverse, triazole–iduronic acid hybrid unstable.1–3 Accordingly, methods to enhance the stability of this
molecules by click reaction of an azido iduronic acid derivative protein drug are urgently needed. One means of enzyme stabili-
with randomly chosen alkynes is described. Library members were zation is the use of reversible inhibitors, which can protect a
screened for their ability to inhibit a-L-iduronidase, and hit mole- protein drug from denaturation by binding to it.4–6
cules and analogues were then investigated for their ability to
Because preparation of IdoA-typed molecules such as
stabilize rh-a-IDUA in a thermal denaturation study. This work 4-methylumbelliferyl a-L-iduronide (4-MUI, 1), the fluorogenic
resulted in the discovery of the first small molecules that can be substrate of IDUA, is tedious,7 most IDUA inhibitors except
used to stabilize exogenous rh-a-IDUA protein in vitro.
5F-IdoAF (2, see in Fig. 1) possess other structurally diverse
skeletons, and display only weak to moderate inhibitory
a-L-Iduronidase (IDUA; EC 3.2.1.76) is a human lysosomal enzyme activity.8,9 Interestingly, 2 acts as an enzyme inactivator because
that catalyzes the hydrolysis of the terminal a-L-iduronic acid it forms a covalent glycosyl-enzyme intermediate, followed by
(IdoA) residue in glycosaminoglycans, such as dermatan sulfate regeneration of free IDUA.9
and heparan sulfate (Fig. 1).1 Deficiencies in the activity of this
In previous work, we used natural product-inspired, combi-
enzyme result in the accumulation of glycosaminoglycans in the natorial chemistry to discover novel inhibitors or chemical
lysosomes, leading to mucopolysaccharidosis type I (MPS I). MPSI chaperones of sugar-processing enzymes. A straightforward
disease can be treated by enzyme replacement therapy (ERT),2 coupling reaction such as amide bond formation (with randomly
wherein recombinant human a-IDUA (rh-a-IDUA) is injected into selected carboxylic acids) was employed to prepare diverse
the patient to reduce accumulated substrates, and alleviate clinical molecules.10,11 However, our attempts to use the same strategy
symptoms.3 But rh-a-IDUA is highly expensive and inherently in this work were thwarted at an early stage when our proposed
amine synthesis (reduction of the protected azido iduronic acid
intermediate) did not work well (see the ESI†).12 Accordingly,
conjugation of the azido intermediate directly via a click reaction,
a type of 1,3-dipolar cycloaddition, was applied.13 Triazole-
containing, sugar-based molecules have been previously used to
generate bioactive molecules for the study of enzyme inhibitors
and receptor recognition.13 However, to the best of our knowledge,
they have not been applied to discover new, non-hydrolysable, and
reversible IDUA stabilizers.
Fig. 1 Natural and fluorogenic substrates of IDUA.
The initial designed molecules (types I and II), inspired by
IdoA and 4-MUI (1), are depicted in Fig. 2. Structurally, they
possess the conformation and configuration of IdoA. The sugar
and triazole moieties are linked by an alpha linkage in type I,
and a beta linkage in type II; and their corresponding pre-
cursors are iduronyl azides 3 and 4, respectively.
Azide 3 was prepared starting from D-glucuronolactone.
Methanolysis of 3 followed by peracetylation gave a mixture
of a- and b-anomers of 5 (Scheme 1).14 Bromination of these at
a Genomics Research Center, Academia Sinica, 128, Section 2, Academia Road,
Taipei, 11529, Taiwan. E-mail: wcheng@gate.sinica.edu.tw
b Department of Chemistry, National Cheng Kung University, 1, University Road,
Tainan City, 701, Taiwan
c Department of Life Sciences and Institute of Genome Sciences,
National Yang-Ming University, 155, Section 2, Linong Street, Taipei, 112, Taiwan
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c7cc09642a
This journal is ©The Royal Society of Chemistry 2018
Chem. Commun.

View Article Online
Communication
ChemComm
Fig. 2 Structures of desired iduronyl triazoles (types I and II) and their
corresponding precursors 3 and 4.
Scheme 2 Preparation of iduronyl triazoles 10 and 11 as model reactions.
the C5 position under photo-induced radical conditions using
N-bromosuccinimide gave 6 (70%).15 Next, C5 brominated 6
was treated with tributyltin hydride and azobisisobutyronitrile
(AIBN) in boiling THF, which gave a separable mixture of 7 and
5 (7/5 = 1/2) in 90% yield.16 Based on NMR analysis and
literature reports, 7 was confirmed to be a 1C4 conformer
because of the smaller coupling constants of J4,5 and J4,3
(2.4 and 3.6 Hz, respectively) observed.17 Treatment of 7 with
TMSN3/SnCl4 gave the a-azide intermediate 3 in 75% yield.18
Notably, the configuration at the anomeric center was con-
firmed by 2D (NOESY) NMR analysis (see the ESI†).
Additionally, we had planned to prepare azide 9 (bearing a
fluoride at the C5 position) for comparison purposes (3 vs. 9).
Although 6 could be converted to 8 through a nucleophilic
halogen exchange (AgF) in good yield (75%), its subsequent
conversion to 9 was not successful.19 Presumably, the electron-
withdrawing nature of the fluorine atom at C5 suppresses the
anomeric effect of the proximal oxygen atom such that nucleo-
philic substitution at C1 cannot take place.
With azide 3 now in hand, the click reaction could be
investigated. As shown in Scheme 2, treatment of azide 3 with
phenylacetylene in the presence of sodium ascorbate (NaAsc)
gave rise to a click reaction, which was followed by deprotection
to afford the iduronyl triazole 10 (type I, R = Ph). Likewise, 11
(b-form) could also be prepared from 7. Compound 7 was con-
verted to an unstable a-anomeric bromide intermediate, followed
by reaction with sodium azide to form the b-isomeric azide 4. The
anomeric configuration of 4 was confirmed by 2D (NOESY) NMR
analysis (see the ESI†). Triazole 11 (type II, R = Ph) was obtained
Table 1 Inhibitory study of iduronyl triazoles 10 and 11 against IDUA
%Inhibition
Compound
1 mM
100 mM
10
11
83
42
55
15
Apparently, triazole 10 (a-linked) exhibited better activity and
was a three-fold more potent inhibitor of IDUA than triazole 11
(b-linked), at 100 mM. This result led us choose 10 as the
structural template and 3 as the starting point for the genera-
tion of a structurally diverse, triazole-based library.
Next, to compare the influence on the enzyme activity of the
sugar conformation and the orientation of the C5 carboxylic acid,
a-L-IdoA configured 12 and b-D-GlcA configured 13 were synthe-
sized by hydrolysis of the corresponding glycosyl azides 3 and 4,
respectively (see the ESI†). In our test reactions, 12 and 13 reacted
with accessible terminal alkynes to give the corresponding 14a–d
and 15a–b, respectively (Table 2). As expected, 14a–d were found to
be more potent inhibitors of IDUA than 15a–b since the enzyme
can recognize more substrate like molecules. It can also be seen
from Table 2 that 14b, bearing a substituted aryl moiety, displayed
better activity than 14a, 14c and 14d (with bulky or linear chains).
Our results showed that both the sugar moiety and a glycan part
form certain interactions with the enzyme.
These encouraging results prompted us to generate more
structurally distinct iduronyl triazole-based molecules (type I).
´
from 4 via click chemistry and Zemplen deacetylation.
Both model iduronyl triazoles 10 and 11 were evaluated in
Table 2 Chemical structures of iduronyl and glucuronyl triazoles and
an IDUA inhibition study, and the results are shown in Table 1.
their inhibitory activities
Compound
%Inhibitiona
Compound
%Inhibitiona
12
46
58
81
37
14d
13
15a
15b
69
14a
14b
14c
NIb
NIb
17
a
b
Scheme 1 Synthetic route for the preparation of azide 3.
1 mM. NI refers to no inhibition (%inhibition o10%).
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2018

View Article Online
ChemComm
Communication
Table 3 Thermal shift study of small molecules toward rh-a-IDUA
Tm (1C)
Conditionsa
18b
19
20b
21
22
2c
0 mM
250 mM
500 mM
56.2
57.0
58.3
56.2
56.7
56.7
56.2
56.9
57.7
56.2
56.2
56.8
56.2
56.5
57.8
56.2
55.0
55.6
a
b
c
At pH 3.5. The curves are shown in the ESI. Reference compound.
thermal denaturation assay.5,6,22 The Tm of rh-a-IDUA on its own
was 56.2 1C. In contrast, the Tm of rh-a-IDUA increased to 58.3,
56.7, 57.7, 56.8 and 57.8 1C when incubated with 500 mM of 18–22,
respectively. The reference inhibitor 2 was also tested under the
same conditions, but no significant increase in Tm was observed.
Although a better inhibitor than 18–22,23 its covalent interaction
with IDUA might have no benefit for the protection of enzyme
during thermal denaturation. As shown in Table 3, these
concentration-dependent Tm enhancement effects suggest these
molecules protect rh-a-IDUA from denaturation. In contrast,
glycerol (up to 40 mM, see the ESI†) did not show any ability to
stabilize this enzyme.24
The ability of these molecules to protect the enzyme from
heat-induced inactivation was also evaluated.6,25 In a control
experiment without treating molecules, the rh-a-IDUA activity
was found to decrease below 5% of the initial value after
120 min of incubation at 48 1C. Upon treatment with 18 or
20, this enzyme activity still remained around 60% at 500 mM
(Fig. 5). These findings establish the utility of triazole–iduronic
acid hybrid molecules as rh-a-IDUA stabilizers, the first demon-
stration of the feasibility of using small molecules to protect
IDUA against denaturation.
Fig. 3 Combinatorial approach to prepare an iduronyl triazole-based
library (L16 + L17) and its IDUA inhibition screening.
In addition to the six alkynes available in hand, a diverse library
of alkynes was prepared by parallel conjugation of propargyl-
amine with a 42-member carboxylic acid library to generate a
structurally diverse N-acyl propargylamine-based library. As
shown in Fig. 3, a combinatorial approach toward parallel
synthesis of an iduronyl triazole-based library (L16 + L17) could
commence with a 48-membered alkyne library reacting with
azide 12 via click reaction (see the ESI†). To improve the
efficiency, all procedures were modified for compatibility with
semi-automated instruments including a synthesizer, a modern
multifunction liquid handler, and a multichannel concentrator.20
After proper purification (solid-phase extraction),21 a 48-membered
library of diverse iduronyl triazoles with high purity and excellent
regioselectivity was obtained (see the ESI†).
The IDUA inhibitory activities of all 48 members were tested
at 100 mM and the results are shown in Fig. S2, ESI†. The
activity is dependent on the structure of the N-substituent on
the triazole ring. Generally, most iduronyl triazoles that contain
an N-cinnamoyl substituent exhibit better inhibitory potency
than a saturated N-acyl moiety. Among them, the inhibitory
activity of L16-C3 (renamed 18, see in Fig. 4) bearing an
(E)-2-methyl-3-phenylacryloyl group was particularly high
(88% inhibition at 100 mM).
Based on the structure of 18, three similar analogues 20–22
were designed and synthesized with structural differences at
the substituted moiety. Besides, 19 (L16-C3), the saturated
form of 18 was also synthesized for comparison purposes.
The structures and inhibition results are shown in Fig. 4.
The ability of 18–22 to affect rh-a-IDUA stability was next
investigated. The enzyme melting temperature (Tm) of rh-a-IDUA
under various conditions was measured in a fluorescence-based
Fig. 5 A heat-induced denaturation study. (a and b) Treatment of 500 mM
of molecules with rh-a-IDUA at pH 3.5 and 48 1C for the indicated amount
of time. The enzyme was incubated at pH 3.5 with 4MU-a-iduronic acid.
(c) Treatment with varied concentrations of 18. (d) Treatment with varied
concentrations of 20.
Fig. 4 Structures of L16-C3 (18) and its analogues 20–22 and their IDUA
inhibitory activities.
This journal is ©The Royal Society of Chemistry 2018
Chem. Commun.

View Article Online
Communication
ChemComm
R. Soska, M. Frascella, L. J. Pellegrino, Y. Lun, D. Guillen, D. J. Lockhart
and K. J. Valenzano, PLoS One, 2012, 7, e40776; (d) R. E. Boyd, G. Lee,
P. Rybczynski, E. R. Benjamin, R. Khanna, B. A. Wustman and
K. J. Valenzano, J. Med. Chem., 2013, 56, 2705–2725.
6 W.-C. Cheng, J.-H. Wang, H.-Y. Li, S.-J. Lu, J.-M. Hu, W.-Y. Yun,
C.-H. Chiu, W.-B. Yang, Y.-H. Chien and W.-L. Hwu, Eur. J. Med.
Chem., 2016, 123, 14–20.
7 (a) N. Baggett, A. K. Samra and A. Smithson, Carbohydr. Res., 1983, 124,
63–74; (b) F.-C. Lu, L. S. Lico and S.-C. Hung, ARKIVOC, 2013, 13–21.
8 (a) J. Pabba, B. P. Rempel, S. G. Withers and A. Vasella, Helv. Chim.
Acta, 2006, 89, 635–666; (b) I. C. di Bello, P. Darling, L. Fellows and
B. Winchester, FEBS Lett., 1984, 176, 61–64.
9 C. E. Nieman, A. W. Wong, S. He, L. Clarke, J. J. Hopwood and
S. G. Withers, Biochemistry, 2003, 42, 8054–8065.
10 (a) T.-J. R. Cheng, T.-H. Chan, E.-L. Tsou, S.-Y. Chang, W.-Y. Yun,
P.-J. Yang, Y.-T. Wu and W.-C. Cheng, Chem. – Asian J., 2013, 8,
2600–2604; (b) H.-W. Shih, K.-T. Chen, S.-K. Chen, C.-Y. Huang, T.-J.
R. Cheng, C. Ma, C.-H. Wong and W.-C. Cheng, Org. Biomol. Chem.,
2010, 8, 2586–2593.
11 (a) W.-C. Cheng, C.-Y. Weng, W.-Y. Yun, S.-Y. Chang, Y.-C. Lin,
F.-J. Tsai, F.-Y. Huang and Y.-R. Chen, Bioorg. Med. Chem., 2013, 21,
5021–5028; (b) W.-C. Cheng, J.-H. Wang, W.-Y. Yun, H.-Y. Li and
J.-M. Hu, Eur. J. Med. Chem., 2017, 126, 1–6.
12 E. G. von Roedern, E. Lohof, G. Hessler, M. Hoffmann and
H. Kessler, J. Am. Chem. Soc., 1996, 118, 10156–10167.
13 V. K. Tiwari, B. B. Mishra, K. B. Mishra, N. Mishra, A. S. Singh and
X. Chen, Chem. Rev., 2016, 116, 3086–3240.
In summary, a library of iduronyl triazole-based analogues
has been efficiently synthesized using click reaction to conju-
gate an azido iduronic acid precursor with a wide range of
structurally diverse alkynes. Library members were screened for
their ability to inhibit IDUA, and hit molecules were then
investigated for their ability to stabilize rh-a-IDUA in a thermal
denaturation study. It was found that 18 and 20 could protect
the enzyme from heat-induced inactivation. This is the first
study to establish that small molecules can be used to stabilize
exogenous rh-a-IDUA protein in vitro, and could potentially
increase the lifetime of ERT when used in the treatment of
MPSI disease. Our search for a more potent stabilizer of rh-a-
IDUA for testing against MPSI patient cell lines by the synthesis
of more diverse libraries is ongoing, and the results will be
reported in due course.
We thank Academia Sinica and Ministry of Science and
Technology (MOST) for financial support. We gratefully
acknowledge the Taiwan Foundation for Rare Disorders for
graduate student research & fellowship support.
14 G. N. Bollenback, J. W. Long, D. G. Benjamin and J. A. Lindquist,
J. Am. Chem. Soc., 1955, 77, 3310–3315.
Conflicts of interest
15 R. Blattner, R. J. Ferrier and P. C. Tyler, J. Chem. Soc., Perkin Trans. 1,
1980, 1535–1539.
There are no conflicts to declare.
´
16 (a) D. Medakovic, Carbohydr. Res., 1994, 253, 299–300; (b) H. N. Yu,
J. Furukawa, T. Ikeda and C.-H. Wong, Org. Lett., 2004, 6, 723–726;
(c) S. Mohamed, E. H. Krenske and V. Ferro, Org. Biomol. Chem.,
2016, 14, 2950–2960.
Notes and references
1 H. Bie, J. Yin, Xu He, A. R. Kermode, E. D. Goddard-Borger,
S. G. Withers and M. N. G. James, Nat. Chem. Biol., 2013, 9, 739–745. 17 (a) G. Speciale, A. J. Thompson, G. J. Davies and S. J. Williams, Curr.
2 R. Giugliani, A. Federhen, A. A. Silva, C. M. Bittar, C. F. M. de Souza,
C. B. O. Netto, F. Q. Mayer, G. Baldo and U. Matte, Res. Rep. Endocr.
Disord., 2012, 2, 53–64.
Opin. Struct. Biol., 2014, 28, 1–13; (b) D. R. Ferro, A. Provasoli,
M. Ragazzi, B. Casu, G. Torri, V. Bossennec, B. Perly, P. Sina¨y,
M. Petitou and J. Choay, Carbohydr. Res., 1990, 195, 157–167.
3 (a) V. Valayannopoulos and F. A. Wijburg, Rheumatology, 2011, 50, 18 (a) M. Tosin and P. V. Murphy, Org. Lett., 2002, 4, 3675–3678; (b) The
v49–v59; (b) M. A. Saif, B. W. Bigger, K. E. Brookes, J. Mercer, conformation analysis of 3 is discussed in ESI†.
K. L. Tylee, H. J. Church, D. K. Bonney, S. Jones, J. E. Wraith and 19 P. Florio, R. J. Thomson and M. von Itzstein, Carbohydr. Res., 2000,
R. F. Wynn, Haematologica, 2012, 97, 1320–1328; (c) the cost of ERT 328, 445–448.
is estimated at US$150 000–300 000 per patient per annum in MPSI. 20 (a) H.-W. Shih, C.-W. Guo, K.-H. Lo, M.-Y. Huang and Wei-Chieh
4 (a) I. M. Slaymaker, M. Bracey, M. Mileni, J. Garfunkle, B. F. Cravatt,
D. L. Boger and R. C. Stevens, Bioorg. Med. Chem. Lett., 2008, 18,
5847–5850; (b) G. Barbato, D. O. Cicero, F. Cordier, F. Narjes,
Cheng, J. Comb. Chem., 2009, 11, 281–287; (b) Y.-W. Pan, C.-W.
Guo, H.-Y. Tu, C.-W. Tsai and W.-C. Cheng, ACS Comb. Sci., 2013, 15,
425–434.
B. Gerlach, S. Sambucini, S. Grzesiek, V. G. Matassa, R. De Francesco 21 E.-L. Tsou, S.-Y. Chen, M.-H. Yang, S.-C. Wang, T.-R. R. Cheng and
and R. Bazzo, EMBO J., 2000, 19, 1195–1206. W.-C. Cheng, Bioorg. Med. Chem., 2008, 16, 10198–10204.
5 (a) E. R. Benjamin, R. Khanna, A. Schilling, J. J. Flanagan, L. J. 22 M.-C. Lo, A. Aulabaugh, G. Jin, R. Cowling, J. Bard, M. Malamas and
Pellegrino, N. Brignol, Y. Lun, D. Guillen, B. E. Ranes, M. Frascella, G. Ellestad, Anal. Biochem., 2004, 332, 153–159.
R. Soska, J. Feng, L. Dungan, B. Young, D. J. Lockhart and K. J. 23 In our study, the IC50 values of 2 and 18 against IDUA are 12 mM and
Valenzano, Mol. Ther., 2012, 20, 717–726; (b) Y. Zhu, J.-L. Jiang, 41 mM, respectively.
N. K. Gumlaw, J. Zhang, S. D. Bercury, R. J. Ziegler, K. Lee, M. Kudo, 24 K. Gekko and S. N. Timasheff, Biochemistry, 1981, 20, 4667–4676.
W. M. Canfield, T. Edmunds, C. Jiang, R. J. Mattaliano and S. H. Cheng, 25 A. R. Sawkar, W.-C. Cheng, E. Beutler, C.-H. Wong, W. E. Balch and
Mol. Ther., 2009, 17, 954–963; (c) R. Khanna, J. J. Flanagan, J. Feng,
J. W. Kelly, Proc. Natl. Acad. Sci. U. S. A., 2002, 99, 15428–15433.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2018