Design, synthesis and evaluation of protein disulfide isomerase inhibitors with nitric oxide releasing activity

Article abstract of DOI:10.1016/j.bmcl.2019.126898

Protein disulfide isomerase (PDI), a chaperone protein mostly in endoplasmic reticulum, catalyzes disulfide bond breakage, formation, and rearrangement to promote protein folding. PDI is regarded as a new target for treatment of several disorders. Here, b

Full text of DOI:10.1016/j.bmcl.2019.126898

Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design, synthesis and evaluation of protein disulfide isomerase inhibitors
with nitric oxide releasing activity
⁎
Lin Lia, ,d, Jian Liub,d, Yaqi Dingc,d, Zhenxiong Shia, Bo Penga, Naidi Yangc, Danqi Hongb,
⁎
Chengwu Zhangc, Chuanhao Yaoa, Jingyan Geb, , Wei Huanga,c
a Shaanxi Institute of Flexible Electronics (SIFE), Northwestern Polytechnical University (NPU), 127 West Youyi Road, Xi’an 710072, PR China
b Key Laboratory of Bioorganic Synthesis of Zhejiang Province, College of Biotechnology and Bioengineering, Zhejiang University of Technology, Hangzhou 310014, PR
China
c Key Laboratory of Flexible Electronics (KLOFE) & Institute of Advanced Materials (IAM), Nanjing Tech University (Nanjing Tech), 30 South Puzhu Road, Nanjing
211800, PR China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Protein disulfide isomerase (PDI), a chaperone protein mostly in endoplasmic reticulum, catalyzes disulfide bond
breakage, formation, and rearrangement to promote protein folding. PDI is regarded as a new target for treat-
ment of several disorders. Here, based on the combination principle, we report a new PDI reversible modulator
16F16A-NO by replacing the reactive group in a known PDI inhibitor 16F16 with nitric oxide (NO) donor. Using
molecular docking experiment, 16F16A-NO could embed into the active cavity of PDI. From newly developed
fluorescent assay, 16F16A-NO showed rapid NO release. Furthermore, it is capable to moderately inhibit activity
of PDI and S-nitrosylate the protein, indicating by insulin aggregation assay and biotin-switch technique. Finally,
it displayed a dose-dependent antiproliferative activity against SH-SY5Y and HeLa tumor cells. Our designed
hybrid compound 16F16A-NO showed a reasonable activity and might offer a promising avenue to develop
novel PDI inhibitors for disease treatments.
Protein disulfide isomerase
S-nitrosylation
Nitric oxide
Biotin-switch technique
Protein disulfide isomerase (PDI, also known as PDIA1) is a 57-kDa
thiol oxidoreductase which catalyzes the thiol-based reactions with
their interested substrate proteins, including formation (oxidation),
breakage (reduction) and rearrangement (isomerization) through a
disulfide bond interchange.1–3 It mainly localizes at endoplasmic re-
ticulum (ER) through a C-terminal KDEL signal sequence and acts as a
chaperone for proper protein folding (Fig. 1A) to inhibit protein ag-
gregation and correct protein misfolding. The catalytic domains (a and
a’), containing redox catalytic Cys-Gly-His-Cys (CGHC, Cys53/56 and
Cys397/400, respectively) motifs, as shown in Fig. 1B, are responsible for
the oxidoreductase activity. Whereas, the other two noncatalytic do-
mains (b and b’), with a large hydrophobic pocket are essential for the
recognition and binding of incompletely folded protein substrates.4–6
Recently, as PDI is a crucial enzyme participating in protein folding,
the investigations of its functions have been revealed that it is linked
with a large number of disease states, including cancer, neurodegen-
erative diseases and cardiovascular diseases.7–9 The growth of cancer
requires continued protein synthesis, giving rise to ER stress by acti-
vating unfolded protein response (UPR).10,11 Consequently, PDI has
been upregulated and its evaluated expression have been observed in a
wide range of human cancers.12–14 Besides, resistance of tumor cells to
chemotherapeutic agents was also reported to be associated with the
elevated PDI expression levels. Furthermore, dysfunction of PDI is also
involved with neurological disorders. Upregulated PDI and mutations in
the gene encoding PDI was observed in amyotrophic lateral sclerosis
(ALS).15 A wide range of synthetic and natural product-based PDI in-
hibitors have been explored in recent years, including 16F16,9 KSC-
34,16 RB-11-ca,17 PACMA-31,18 P1,19 S-CW355420 (structures shown
in Fig. S1) and so on. All these compounds contain a cysteine-reactive
electrophile group and exhibit elegant irreversible inhibition against
PDI. However, it was also reported that irreversible genetic silencing of
PDI is cytotoxic to cells and animal models because of its essential
chaperone role.21 Moreover, knockout of PDIA3, with similar features
as PDI, implicated the embryonic death in mice.22 Therefore, irrever-
sible inhibitors of PDI may exhibit cytotoxicity in vivo and reversible,
non-covalent inhibitors might bring new opportunities for improved
therapeutic purposes. In 2015, Stockwell group identified a reversible
inhibitor LOC14 (structure shown in Fig. S1) from screening of
⁎ Corresponding authors.
E-mail addresses: iamlli@nwpu.edu.cn (L. Li), gejy@zjut.edu.cn (J. Ge).
d These authors contributed equally to this work.
https://doi.org/10.1016/j.bmcl.2019.126898
Received 15 October 2019; Received in revised form 27 November 2019; Accepted 7 December 2019
0960-894X/©2019ElsevierLtd.Allrightsreserved.
Pleasecitethisarticleas:LinLi,etal.,Bioorganic&MedicinalChemistryLetters,https://doi.org/10.1016/j.bmcl.2019.126898

L. Li, et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxxx
Fig. 1. (A) Domain overview of protein disulfide isomerase, with a and a’ catalytic domain, b and b’ domains and C-terminal ER signal sequence. (B) PDI-catalytic
reaction. (C) Our strategy and the hybrid compound. (D) Docking analysis of 16F16A-NO and 16F16A against PDI. (E). Synthetic scheme of 16F16A-NO. (a) Fmoc-
Cl, 10% Na2CO3/H2O, 1,4-dioxane, 0 °C to r. t.; (b) 20% TFA/DCM, 0 °C to r. t.; (c) (COCl)2, DMF, DCM, 0 °C to r. t.; prop-2-yn-1-yl 1-methyl-2,3,4,9-tetrahydro-1H-
pyrido[3,4-b]indole-1-carboxylate, DMF, 0 °C to r. t.; (d) 20% piperidine/CHCl3, 0 °C to r. t.; (f) NaNO2, 20% AcOH/H2O, 0 °C to r. t.
~10,000 compounds.23 LOC14 has promising inhibition (Kd = 62 nM)
and protects neuron-like PC12 cells expressing toxic huntingtin pro-
teins.
stronger ligand binding affinity. The output was analyzed for the pre-
dicted binding interactions. The best scoring pose and results are shown
in Fig. 1D. It showed that 16F16A-NO had similar affinity value com-
pared with 16F16A. The docking pose confirmed 16F16A was in the
“a” catalytic domain and the aromatic ring of it was π-π interaction
with residues W52. Furthermore, 16F16A-NO, also bind in the C-
terminal active site pocket, forming similar interactions. It is important
to note that the NO donor groups was embedded deeper into pocket
with the additional interactions with other amino acids and closer to
the cysteine in the active site. Therefore, the result indicated that
16F16A-NO could be recognized by PDI.
One of promising strategies to achieve reversible inhibitors is based
on the use of nitric oxide (NO).24 NO is prone to make amines, aromatic
rings, alcohols and reductive thiols on protein nitrosylation modifica-
tion, thus changing protein conformation and affecting its activity and
function. Among them, NO-mediated S-nitrosylation of cysteine re-
sidues is one of typical post translational modifications (PTM), which is
a reversible modification.25 It is estimated that ~70% of the universal
proteome may be subjected to post translational regulation by S-ni-
trosylation, including transcription factors, kinases, channel proteins.26
In 2006, Lipton et al reported that PDI was S-nitrosylated in vitro and in
vivo to form an S-nitrosylated protein, causing decrement of its ac-
tivity.8
Synthesis of 16F16A-NO
The synthetic scheme was shown in Fig. 1E with introduction of
alkyne at the carboxylic acid site and replacement of chloroacetamide
with phenyl nitrous amide. A terminal propargyl alkyne modification,
biorthogonal and not causing to lose its biological activity, was de-
signed to validate whether there is any covalent bound to proteins.
First, Fmoc protected with phenylaminoacetic acid was prepared
through alkylation and protecting group exchanges. Next it was at-
tached with three-ring key core structure with propargyl ester under
basic condition. After deprotection with Fmoc group, sodium nitrite in
acidic condition was used to oxidate secondary amine to obtain the
final product 16F16A-NO. The product was kept in the dark and always
fresh prepared in DMSO. Detailed synthetic procedures and NMR
spectra were enclosed in electronic supporting information.
Herein, we designed a new reversible modulator (16F16A-NO,
shown in Figs. 1C and S1C), which is a hybrid compound between a
potent PDI inhibitor 16F16 and
a nitric oxide donor. 16F16
(IC50 = 1.5 µM), an irreversible inhibitor with a chloroacetyl electro-
phile (Fig. S1B), was discovered by Stockwell group through a small-
molecule screening approach.9 The electrophile on 16F16 was replaced
with phenyl nitrous amide and erased its covalent linkage properties.27
The attachment of alkyne group would not affect the proper but pro-
vides a tool to understand the binding type of the hybrid. 16F16A-NO
could recognize and embed into the active cavity of PDI. We also found
NO release from 16F16A-NO was rapid in vitro and the protein could be
S-nitrosylated by 16F16A-NO resulted from a biotin-switch assay. We
further found the inhibition of PDI activity and the proliferation of SH-
SY5Y cells and HeLa cells by 16F16A-NO. These findings would re-
present a new class of reversible inhibitors of PDI that can probe its
potential as a drug for treatment of human cancers.
Nitric oxide measurement via a fluorescent assay
Next, we analyzed whether 16F16A-NO could release NO in water
solution working as a NO donor. A fluorescence probe BT-NH devel-
oped by our group was used to detect NO production.29–31 Fluorescent
signal is turned on when BT-NH (λex = 520 nm) reacts with NO to
convert a fluorescent product (λem = 620 nm). (Basic mechanism
shown in Fig. 2). S-nitrosylglutathione (GSNO), a commercial stable NO
donor, known to be decomposed under light, metal ion catalysis and
aqueous solution, and subsequently released NO, has been the subject
as a control to compare. Hence, quantification of NO release was car-
ried out with 50 µM BT-NH in PBS at 37 °C. Different concentrations of
16F16A-NO and GSNO respectively, upon mixing with the reaction, led
to a fluorescent increase. From Fig. 2B, with 600 µM 16F16A-NO, there
Molecular docking studies
First, it is important to evaluate the influence of NO donor mod-
ification on the binding toward PDI protein. The new inhibitor was
converted from 16F16, which is the first PDI inhibitor generating from
high-throughput screening9 and investigated to localize in the “a”
catalytic domain of PDI.23 Hence, using together with the alkyne ver-
sion 16F16A as control, 16F16A-NO was examined for their ability to
enter the active activity of PDI (PDB code 4EKZ28) using Autodock
software. The docking protocol ranks the output values indicating a
2

L. Li, et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxxx
Fig. 4. Effect of PDI inhibition against different concentrations of 16F16A-NO,
GSNO and 16F16A using insulin aggregation assay. 50 ng PDI was treated with
inhibitors (0.1, 1, 10, 30 and 50 μM) in buffer (100 mM Na3PO4, 0.2 mM EDTA,
pH = 7.0) at 37 °C for 30 min followed by adding with 0.16 mM insulin, re-
acting at room temperature for 1 h. The absorbance was measured at 650 nm.
All experiments were repeated three times.
Fig. 2. (A) Schematic representation of the fluorogenic probe BT-NH used in
detection of NO release. (B) Kinetic profile of NO level was evaluated after
adding 16F16A-NO (600 μM) in PBS buffer. Inset is end-point NO level for
incubation of different concentrations of GSNO and 16F16A-NO with BT-NH.
Experiments were performed in triplicate with data plotted as mean
SEM.
PDI inhibition assay
was a rapid rise of NO to concentrations in 10 min, which has a sta-
bilized NO release rate after 30 min of reaction. In comparison, GSNO
showed similar NO release quality at 100 µM, demonstrating superior
NO release capability than 16F16A-NO. Only one fifth of maximal re-
lative theoretical NO release was obtained. It might be too fast to
capture NO by o-diamine in the fluorophore. Nevertheless, NO was
clearly released from the hybrid to generate detectable NO, encoura-
ging us to test the compound in protein and cell-based assays.
To determine whether this NO releasing hybrid affects PDI function,
we further validated 16F16A-NO as a PDI inhibitor using insulin ag-
gregation assay at 50, 30, 10, 1, 0.1 and 0 μM.19 As shown in Fig. 4, it is
observed that the hybrid inhibited PDI reductase activity in a dose-
dependent manner. At 50 μM, it showed approximately 50% inhibitory.
The original 16F16A inhibited PDI activity more potently compared to
the derivative. Furthermore, the NO donor GSNO, as the control, did
not strongly affect PDI activity at 50 μM. It is indicated that although
the hybrid inhibitor 16F16A-NO is also without covalent affinity, it is
more effective and potent in combination of an inhibitor-like structure
and NO donor, in comparison with 16F16-DC which is a 16F16 analog
lacking with chloro-substituent (structure shown in Fig. S1).
S-nitrosylation of PDI
Next, we sought to determine whether NO release from 16F16A-NO
could lead S-nitrosylation on PDI protein (SNO-PDI). A well-known
assay, biotin-switch technique (BST), was proceeded.26 S-nitrosylation
was identified by anti-Biotin after substitution of SNO group with a
biotin group by chemical reduction with sodium ascorbate. Based on
results of nitric oxide release, in this assay, 600 μM 16F16A-NO and
50 μM GSNO were chosen to treat with PDI protein, respectively. From
Fig. 3, it is obviously found that exposure to both compounds suc-
cessfully led to generation of SNO-PDI, although 16F16A-NO appeared
weaker. Moreover, under dithiothreitol (DTT) reducing condition, S-
nitrosylation is abolished, indicating the reversible ability of this
modification under such reduction conditions. Besides, from results of
protein profiling using click chemistry,32–35 there is no significant la-
beling to any protein (data not shown), indicating there is no covalent
binding occurred.
Antiproliferation activity of 16F16A-NO on cancer cells
Cancer cells require increased protein synthesis and thus respond to
oxidative stress which is mediated by PDI. PDI is responsible for the
isomerization, reduction, and oxidation of nonnative disulfide bonds in
unfolded proteins entering the ER, so PDI is usually up-regulated in
cancer cells. Previous study found that S-nitrosylation of PDI could
abrogate its activity and function. Having successfully confirmed that
16F16A-NO as NO donor and an in vitro PDI inhibitor, we further in-
vestigated its cellular activity by evaluating its growth inhibition effect
against two cancer cell lines (human neuroblastoma cell line SH-SY5Y
and human cervical cancer cell line HeLa) in the presence of DMSO
control or at different dosages of 16F16, 16F16A-NO and GSNO. The
results were shown in Figs. 5 and S2. All three of them showed no-
ticeable notable cell toxicity under high concentrations. At 600 μM,
16F16A-NO and GSNO showed 40% inhibition against both cancer cell
lines, although less potent than 16F16 (5 μM). SH-SY5Y cells were more
sensitive than HeLa cells to both NO releasing compounds 16F16A-NO
and GSNO treatment. While, in HeLa cells, 16F16A-NO cause more
significant potency at a lower dose than GSNO, indicating that core
structures of 16F16A moiety also enhanced the inhibition. Further-
more, we have tested them on a normal cell line (human embryo kidney
cell HEK293T). The inhibition of 16F16A-NO and GSNO to this cell line
is slightly weak. These results suggest that this hybrid analogs may be
effective and selective to cancer cells.
Fig. 3. In vitro S-nitrosylation (SNO) of recombinant PDI. Purified protein SNO-
PDI were detected by reacting 400 ng protein with in absence or presence of
concentrations of GSNO (1, 50 μM) or 16F16A-NO (300, 600 μM) for 30 min.
Purified protein was subjected to BST assay and were analyzed by western blot
with anti-biotin and anti-PDI.
3

L. Li, et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxxx
Acknowledgements
This work was financially supported by the National Natural Science
Foundation of China (21877100, 81672508 and 21708034), Natural
Science Foundation of Shaanxi Province (2019JM-016), Jiangsu
Provincial
Foundation
for
Distinguished
Young
Scholars
(BK20170041), China-Sweden Joint Mobility Project (51811530018),
Fundamental Research Funds for Central Universities and the
Provincial Universities of Zhejiang (RF-B2019003).
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmcl.2019.126898.
References
1. Hatahet F, Ruddock LW. Antioxid Redox Signal. 2009;11:2807.
2. Shergalis A, Neamati N. Protein disulfide isomerase. In: Choi S, ed. Encyclopedia of
Signaling Molecules. New York: Springer; 2016.
3. Benham AM. Antioxid Redox Signal. 2012;16:781.
4. Tian G, Xiang S, Noiva R, Lennarz WJ, Schindelin H. Cell. 2006;124:61.
5. Okumura M, Noi K, Kanemura S, et al. Nat Chem Biol. 2019;15:499.
6. Klappa P, Ruddock LW, Darby NJ, Freedman RB. EMBO J. 1998;17:927.
7. Xu S, Sankar S, Neamati N. Drug Discov Today. 2014;19:222.
8. Uehara T, Nakamura T, Yao D, et al. Nature. 2006;441:513.
9. Hoffstrom BG, Kaplan A, Letso R, et al. Nat Chem Biol. 2010;6:900.
10. Luo B, Lee AS. Oncogene. 2013;32:805.
11. Kim I, Xu W, Reed JC. Nat Rev Drug Discov. 2008;7:1013.
12. Eirich J, Braig S, Schyschka L, et al. Angew Chem Int Ed. 2014;53:12960.
13. Goplen D, Wang J, Enger PØ, et al. Cancer Res. 2006;66:9895.
14. Lovat PE, Corrazzari M, Armstrong JL, et al. Cancer Res. 2008;68:5363.
15. Woehlbier U, Colombo A, Saaranen MJ, et al. EMBO J. 2016;35:845.
16. Cole KS, Grandjean JMD, Chen K, et al. Biochemistry. 2018;57:2035.
17. Banerjee R, Pace NJ, Brown DR, Weerapana E. J Am Chem Soc. 2013;135:2497.
18. Xu S, Butkevich AN, Yamada R, et al. Proc Natl Acad Sci U S A. 2012;109:16348.
19. Ge J, Zhang C, Li L, et al. ACS Chem Biol. 2013;8:2577.
Fig. 5. Cell viability of SH-SY5Y and HeLa cells with different concentrations of
GSNO and 16F16A-NO assessed by CCK-8 assay.
20. Allimuthu D, Adams DJ. ACS Chem Biol. 2017;12:2124.
Conclusions
21. Perri E, Parakh S, Atkin J. Expert Opin Ther Tar. 2017;21:37.
22. Garbi N, Tanaka S, Momburg F, Hämmerling GJ. Nat Immunol. 2006;7:93.
23. Kaplan A, Gaschler MM, Dunn DE, et al. Proc Natl Acad Sci U S A. 2015;112:E2245.
24. Serafim RAM, Pernichelle FG, Ferreira EI. Expert Opin Drug Dis. 2017;12:941.
25. Paulsen CE, Carroll KS. Chem Rev. 2013;113:4633.
We have developed a reversible modulator, 16F16A-NO, which could
enter into the active cavity of PDI. We further showed that 16F16A-NO
could release NO rapidly in vitro, inhibit PDI enzymatic activity effectively
and cause S-nitrosylation of PDI. Moreover, we also found that 16F16A-
NO exhibited antiproliferation activity to cancer cell lines. However,
further studied are needed to substantiate our speculation. This hy-
bridization may open new complementary or alternative approaches to
discover and develop more effective and safer PDI reversible inhibitors for
therapeutic applications to understand and treat cancer diseases.36
26. Jaffrey SR, Erdjument-Bromage H, Ferris CD, Tempst P, Snyder SH. Nat Cell Biol.
2001;3:193.
27. Ohwada T, Miura M, Tanaka H, et al. J Am Chem Soc. 2001;123:10164.
28. Wang C, Li W, Ren J, et al. Antioxid Redox Signal. 2013;19:36.
29. Weng M, Yang X, Ni Y, et al. Sensor Actuat B-Chem. 2019;283:769.
30. Wang L, Du W, Hu Z, Uvdal K, Li L, Huang W. Angew Chem Int Ed. 2019;58:14026.
31. Li L, Sun H. Biochemistry. 2019. https://doi.org/10.1021/acs.biochem.9b00831.
32. Guo H, Li Z. MedChemComm. 2017;8:1585.
33. Lang W, Yuan C, Zhu B, et al. Org Biomol Chem. 2019;17:3010.
34. Dong J, Hong D, Lang W, et al. ChemBioChem. 2019;20:1155.
35. Ying S, Du S, Dong J, et al. Chin Chem Lett. 2019;30:67.
Declaration of Competing Interest
36. Qin X, Yu C, Wei J, et al. Adv Mater. 2019. https://doi.org/10.1002/adma.
201902791.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
4