Potent and Highly Selective Inhibitors of the Proteasome Trypsin-like Site by Incorporation of Basic Side Chain Containing Amino Acid Derived Sulfonyl Fluorides

Article abstract of DOI:10.1021/acs.jmedchem.8b00685

A unique category of basic side chain containing amino acid derived sulfonyl fluorides (SFs) has been synthesized for incorporation into new proteasome inhibitors targeting the trypsin-like site of the 20S proteasome. Masking the former α-amino functionality of the amino acid starting derivatives as an azido functionality allowed an elegant conversion to the corresponding amino acid derived sulfonyl fluorides. The inclusion of different SFs at the P₁ site of a proteasome inhibitor resulted in 14 different peptidosulfonyl fluorides (PSFs) having a high potency and an excellent selectivity for the proteolytic activity of the β2 subunit over that of the β5 subunit. The results of this study strongly indicate that a free N-terminus of PSFs inhibitors is crucial for high selectivity toward the trypsin-like site of the 20S proteasome. Nevertheless, all compounds are slightly more selective for inhibition of the constitutive over the immunoproteasome.

Full text of DOI:10.1021/acs.jmedchem.8b00685

Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES
Article
Potent and highly selective Inhibitors of the Proteasome
Trypsin-like site by Incorporation of Basic Side Chain
containing Amino Acid derived Sulfonyl Fluorides
Raik Artschwager, David Ward, Susan Gannon, Arwin J. Brouwer,
Helmus van de Langemheen, Hubert Kowalski, and Rob Liskamp
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b00685 • Publication Date (Web): 21 May 2018
Downloaded from http://pubs.acs.org on May 21, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Potent and highly selective Inhibitors of the
Proteasome Trypsinꢀlike site by Incorporation
of Basic Side Chain containing Amino Acid
derived Sulfonyl Fluorides
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
†
†
‡
Raik Artschwager, David J.Ward, Susan Gannon, Arwin J. Brouwer, Helmus van de
†
†
,†
Langemheen, Hubert Kowalski, Rob M. J. Liskamp*
†
School of Chemistry, Joseph Black Building, University Avenue, Glasgow, G12 8QQ,
United Kingdom
‡
Chemical Biology & Drug Discovery, Department of Pharmaceutical Sciences, Faculty of
Science, Utrecht University, P.O. Box 80082, NLꢀ3508 TB Utrecht, The Netherlands
ABSTRACT
A unique category of basic side chain containing amino acid derived sulfonyl fluorides (SFs)
has been synthesized for incorporation into new proteasome inhibitors targeting the trypsinꢀ
like site of the 20S proteasome. Masking the former αꢀamino functionality of the amino acid
starting derivatives as an azido functionality, allowed an elegant conversion to the
corresponding amino acid derived sulfonyl fluorides. The inclusion of different SFs at the P1
site of a proteasome inhibitor resulted in 14 different peptido sulfonyl fluorides (PSFs)
having a high potency and an excellent selectivity for the proteolytic activity of the β2
subunit over that of the β5 subunit. The results of this study strongly indicate that a free Nꢀ
terminus of PSFs inhibitors is crucial for high selectivity towards the trypsinꢀlike site of the
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 65
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
20S proteasome. Nevertheless, all compounds are slightly more selective for inhibition of the
constitutive over the immunoproteasome.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
The ubiquitinꢀproteasome pathway (UPS) comprises the main machinery for degrading
1
damaged, misfolded, pathogen derived and abnormal proteins in the cell. Therefore, the
proteasome plays a crucial role in the regulation of many cell cycle processes especially
2
involving antigen processing and apoptosis after protein quality control. Proteolysis of the
designated proteins is achieved by the 20S proteasome which consists of 4 stacked rings
comprising 28 subunits assembled in two outer αꢀrings and two inner βꢀrings. Within the
proteolytic βꢀrings of the 20S constitutive proteasome the β1c, β2c and β5c subunits are
found to show catalytic activity referred to as caspaseꢀlike activity (β1c), trypsinꢀlike
activity (β2c) and chymotrypsinꢀlike activity (β5c). Upon exposure to interferon gamma
(IFNγ) and/or tumor necrosis factorꢀα (TNFα) these subunits are substituted by β1i (LMP2),
β2i (MECLꢀ1) and β5i (LMP7), respectively, resulting in the soꢀcalled
3
immunoproteasome. Selective targeting of either constitutive or immunoproteasome
subunits is a particular challenge and opens up further possibilities for the development of
antiꢀcancer and antiꢀinflammatory therapeutic agents. This may be even extended to
4
development of parasiteꢀselective proteasome inhibitors. The development of proteasome
inhibitors has been an outstanding case showing that irreversible inhibitors may provide
5
unique advantages by forming longꢀlived ties with their target.
6
7
Since the approval of Bortezomib 1 in 2003 and Carfilzomib 2 in 2012 (Figure 1) for the
treatment of multiple myeloma many peptide based inhibitors containing different
ACS Paragon Plus Environment
2

Page 3 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
8
9,10
electrophilic traps have been reported. These next generations of inhibitors like ixazomib
1
1
12,13
oprozomib (currently in clinical trials) and LUꢀ102 (3)
(in preclinical testing) clearly
demonstrated that other "warheads" or electrophilic traps within a peptide based inhibitor can
provide attractive alternatives.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
We have initiated the exploration of the sulfonyl fluoride warhead for incorporation into
1
4
proteasome inhibitors and other proteases inhibitors leading to peptido sulfonyl fluorides
1
5,16
(PSFs).
Since then this electrophilic trap has undergone considerable development as it is
presently denoted as a “privileged warhead” in chemical biology, partly due to its
a,17
considerable aqueous stability and chemical reactivity in the respective target enzyme.
Apart from obtaining potent proteasome inhibitors, in particular CbzꢀLeu ꢀSF 4 with an IC ꢀ
4
50
1
5
value of 7 nM for the β5c subunit of the constitutive proteasome, it was found that the SFꢀ
warhead also endowed proteasome inhibitor 5 with a 25 fold higher selectivity for inhibition
of the β5i subunit of the immunoproteasome over the β5c subunit although with loss of
16
potency. This showed clearly that SFs were not merely acting as a powerful electrophilic
trap but also gave rise to higher selectivity by simply changing the electrophilic trap from an
epoxyketone to a sulfonyl fluoride.
a
Several transformations (e.g. oxidation, reduction, hydrolysis etc.) can be carried out in the
presence of an aromatic SF moiety, which is indicative of the difference in stability of
18
aliphatic and aromatic SFs.
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 65
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
Figure 1. Structures of Bortezomib (1), Carfilzomib (2), LUꢀ102 (3), CbzꢀLeu ꢀSF (4), β5i
3 3
immunoproteasome selective PSF (5) and N ꢀPheꢀLeu ꢀSF (6).
Although this was an important finding, we think that selectivity for a particular
proteasome subunit or in general a protease is largely determined by the character and
b
relative position of the P1 side chain with respect to the SF warhead. Therefore, we focused
our efforts on inhibitors with basic side chains at the P1 position to evaluate whether this
would be sufficient to confer selectivity of the resulting inhibitors for the proteasome trypsinꢀ
like site (β2) over the chymotrypsinꢀlike site (β5). Moreover, development of β2 selective
12,13
inhibitors will contribute to overcoming resistance against existing (β5) inhibitors.
Recently, we found that PSF 6 was a very potent proteasome inhibitor (IC50 110 nM for
15
19
β5c). The known inhibitor LUꢀ102 3 (IC50 3.8 nM for β2) has largely the same backbone
sequence but also contains a basic nonꢀnatural amino acid residue. This latter sequence was
used here for incorporation of an SF warhead leading to development of the synthesis and
investigation of the selectivity and potency of basic side chain containing PSF proteasome
inhibitors. The required development of amino acid derived sulfonyl fluorides containing a
basic side chain is described in this paper and their incorporation in PSF proteasome
b
Recently it was found in peptido vinyl sulfonyl fluorides (PVSFs) that although a PVSF
may be more reactive than a peptido sulfonyl fluoride (PSF), the sulfonyl fluoride warhead
part may occupy a less favorable P1' position because it is further positioned from the P1 side
20
chain, leading to a reduced inhibition.
ACS Paragon Plus Environment
4

Page 5 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
inhibitors led to both potent and highly selective inhibitors of the proteasome's trypsinꢀlike (
β2) activity.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Basic side chain containing amino acid derived SFs 7 ꢀ 10.
RESULTS AND DISCUSSION
Synthesis of basic side chain containing amino acid derived sulfonyl fluorides
Considerations: Synthesis of an amino acid derived sulfonyl fluoride (SF) containing
a basic side chain, which upon incorporation in the remainder of the proteasome inhibitor
sequence should endow the resulting molecular construct with β2ꢀselectivity, was a
significant challenge. Because of the simultaneous presence of an electrophilic site ꢀ the
sulfonyl fluoride moiety ꢀ and two nucleophilic sites, that is, the former αꢀamino group and
the basic side chain, a suitable protecting group strategy was necessary. It was known from
our previous work that as long as the αꢀamino group is protected or protonated, it is possible
15
to leave the SFꢀmoiety intact and ultimately incorporate it in an inhibitor construct. The
desired amino acid derived SFs with a basic side chain for inhibitor construction are shown in
Figure 2. Based on the pKaꢀvalues of the side chain and the relative position of the amino
functionality with respect to the SFꢀelectrophile, it was expected that the guanidine
functionality will always remain protonated. Therefore, it will not react with an SFꢀmoiety
not even after the final deprotection step in the synthesis of the arginine building block 7
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 65
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(Scheme 1) and incorporation in PSFs 53ꢀ58 by peptide coupling reactions. The aromatic and
benzylic amino groups obtained by deprotection after incorporation of 9 and 10, respectively,
will not be available for an intraꢀmolecular reaction. This is probably also the case for the
lysine derived amino group obtained by deprotection after incorporation of 8, which apart
from being protonated at physiological pH can only give rise to the formation of an 8ꢀ
membered ring. Nevertheless, there are possibilities for intermolecular reactions and/or
(slow) hydrolysis (see stability experiments, Figure 5) and therefore amino groups of amino
SF derivatives 8ꢀ10 remained protected including the coupling steps leading to PSFs 59ꢀ67.
Synthesis: The syntheses of amino acid derived SFs 7ꢀ10 was carried out following
our earlier described general strategy with modifications involving the fluoronating agent and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
14
in light of the considerations above (Schemes 1 ꢀ 3). Starting from commercially available
arginine compound 11, which has protecting groups resistant to conditions used for the
introduction of the SF warhead, it was first converted to methyl ester 12. Reduction to
alcohol 13, preparation of mesylate 14 was followed by substitution to thioacetate 15. After
oxidation to the sulfonic acid derivative 16 the corresponding sulfonyl fluoride 17 was
21
obtained using XtalFluorꢀM™. Finally, simultaneous removal of the Cbz and Mtr
protective groups afforded the arginine derived SF 7 ready for coupling to the remainder of
the inhibitor sequence (Scheme 1)
Scheme 1. Synthesis of arginine derived sulfonyl fluoride (7).
ACS Paragon Plus Environment
6

Page 7 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
HN
N(H)Mtr
NH
HN
N(H)Mtr
NH
HN
N(H)Mtr
NH
HN
N(H)Mtr
NH
1
2
) EtOAc, KHSO
) SOCl , MeOH
4
NaBH , LiCl,
MsCl, NEt3,
Cs
2 3
CO , HSAc,
4
DMF
2
THF, EtOH
DCM
O
H N
3
O
OH
OMs
Cbz(H)N
Cbz(H)N
Cbz(H)N
Cbz(H)N
O
O
1
4 (94%)
11
12 (100%)
13 (95%)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Cl
2
HN
N(H)Mtr
NH
HN
N(H)Mtr
HN
N(H)Mtr
H
N
NH
2
NH
F
NH
NH
O
N
S
BF
4
1
) H
2
O
2
, AcOH
F
1) HBr/AcOH,
2) DowexꢀCl
2
) NaOAc
(XtalFluorꢀM )
®
Cl
H
SAc
5 (76%)
SO
3
Na
SO
2
F
SO
2
F
Cbz(H)N
Cbz(H)N
Et N 3HF,
Cbz(H)N
3
N
3
DCM, ꢀT
1
16
17 (37%, 2 steps)
7 (87%)
For the synthesis of the lysine derived SF 8 and the amino phenylalanine derived SFs 9 and
0 a different synthetic strategy was necessary. To avoid manipulation of two orthogonal
1
protecting groups present on both aminoꢀfunctional groups during or at the end of synthesis,
it was decided to mask the αꢀamino group as an azide functionality until the very end, that
is, after completion of the synthesis of the SFꢀwarhead. This strategy was successful and
allowed a relatively straightforward synthesis of the side chain protected azido amino acid
derived SFs 40, 41 and 42.
Scheme 2. Syntheses of azido precursor amino acids (19), (22) and (27).
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 65
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Briefly, Lysine derivative 18 and amino phenyl alanine derivative 20 were converted to the
corresponding αꢀazido derivatives 19 and 21 using the azido transfer reagent azidoꢀsulfonylꢀ
22
imidazole in the presence of cupric sulfate. The required aminoꢀmethyl phenylalanine
derivative 25 had to be prepared first in four steps from phenylalanine 23 similar to the
19
synthesis of Geurink et al. Compound 26 was then converted analogously to the required α
azido derivative 27 (Scheme 2). For these three αꢀazido derivatives 19, 22 and 27 an
identical series of synthetic steps was followed to obtain the desired substituted amino SFs 8 ꢀ
0 (Scheme 3). The steps for introduction of the sulfonic acid moiety comprised of reduction
ꢀ
1
of the methyl ester to amino alcohol 28 ꢀ 30, introduction of Msꢀleaving group to mesylates
ACS Paragon Plus Environment
8

Page 9 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
31 ꢀ 33, followed by substitution to thioacetates 34 ꢀ 36 and finally oxidation to afford the
sulfonates 37 ꢀ 39, which were immediately converted to the corresponding SFꢀderivatives 40
®
ꢀ
42 using XtalFluorꢀM . The combined oxidation ꢀ SF conversion is still a considerable
hurdle. Nevertheless, still decent yields of 33 ꢀ 42% of over two steps (average 58 ꢀ 65% per
step) were realized. Reduction of the azide group followed by protonation to the amino acid
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
SF derivatives 8 ꢀ 10 was carried out using Zinc powder in 15% TFA in AcOH. After
purification by semiꢀpreparative HPLC, SF derivatives were obtained as TFAꢀsalts in
moderate yields (54ꢀ59%). The overall yields of the SF warhead containing amino acid
derivatives were quite satisfactory, i.e. 11% (7ꢀsteps) for TFA·H
% (9ꢀsteps) for TFA·H NꢀPhe(4ꢀN(H)Cbz)ΨCH SO F (9) and 2% (10 steps) for
TFA·Phe(4ꢀCH N(H)CbzΨCH SO F (10).
2 2 2
NꢀLys(Cbz)ΨCH SO F (8),
7
2
2
2
2
2
2
Scheme 3. Completion of the synthesis of basic side chain containing amino acid derived
SFs 8 ꢀ 10.
Incorporation of amino SF derivatives toward syntheses of peptido sulfonyl fluorides
(PSFs)
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 65
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
For completion of the synthesis of the desired PSFs the amino acid derived SFs had to be
incorporated into suitable peptide sequences i.e. 47 and 52. These sequences were based on
15
the earlier developed powerful PSF proteasome inhibitors. Their syntheses are shown in
Scheme 4. Assembly of AcꢀPheꢀLeuꢀLeuꢀOH 47 was carried out on a 2ꢀchlorotrityl chloride
resin to afford 46 using a SPPS protocol and preparation of BocꢀPheꢀLeuꢀLeuꢀOH 51 was
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
c
carried out in solution. To prevent racemization at the Cꢀterminus, the methyl ester of 50
was saponified first before conversion of the amino terminus to an azide functionality in 52.
This azide containing precursor peptide was prepared, because the most active of our earlier
15
described proteasome inhibitors contained an azide functionality at the Nꢀterminus of PSF
6
. As it was found that PSFs require at least a capped Nꢀterminus for selectivity towards the
immunoproteasome over the constitutive proteasome, also PSFs with an acetylated and
unprotected Nꢀterminus were synthesized to investigate if this was also crucial for β2
16
specificity.
Scheme 4. Syntheses of the peptides necessary for incorporation of the amino acid
derived SFs.
c
There was no specific reason for synthesis of one tripeptide on the solid phase and the
other tripeptide in solution. Either method can be used for both tripeptides.
ACS Paragon Plus Environment
10

Page 11 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1) 20% piperidine/DMF
2) FmocꢀPheꢀOH,
HCTU, DiPEA, DMF
1
) 20% piperidine/
DMF
O
DiPEA, DCM
Cl
OH
O
Fmoc(H)N
O
Fmoc(H)N
Fmoc(H)N
2) FmocꢀLeuꢀOH,
HCTU, DiPEA,
DMF
N
3) 20% piperidine/DMF
H
O
O
O
4) Ac O, DMF
2
4
3
44
45
O
O
O
O
H
N
TFA/TIS/H O
H
N
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
N
OH
Cl
= 2ꢀchloro trityl chloride resin
N
H
N
N
H
O
H
H
O
O
O
46
47
BocꢀLeuꢀOH,
HCTU, DiPEA,
DCM
LiCl, MeOH/
O
O
1) TFA, DCM
H
2
H O
Cl
OMe
Boc(H)N
OMe
N
OMe
H N
N
2) BocꢀPheꢀOH,
HBTU, Oxymapure®,
DiPEA, DCM
Boc(H)N
N
3
H
H
O
O
O
O
4
8
49 (94%)
50 (93%)
O
O
H
H
N
1) TFA, DCM
OH
N
N
N
H
OH
Boc(H)N
2) N SO Im HCl, K CO ,
N
3
3
2
2
3
H
O
4
7
(
x
x
%
)
CuSO
4
5H
2
O, MeOH
O
O
O
5
1 (80%)
52 (93%)
4
7
(
x
x
%
)
Introduction of SF warhead containing amino acid derivatives has always been viewed as
one of the most challenging steps in the total synthesis of PSFs. First, the coupling conditions
involving a nucleophilic amino group have to be selected in such a way that the SF
electrophile stays as much as possible intact. Second, there is the possibility of racemization
of the amino SF derivative in view of the electronꢀwithdrawing character of the SF moiety.
Therefore, several coupling reagents and conditions were attempted. These included HCTU,
®
®
DIC/Oxymapure , HBTU/Oxymapure , BOP and HATU. Using a different base, for
example NMM instead of DiPEA, during the coupling step did not affect the yields.
Nevertheless, DIC/Oxymapure and HATU in combination with DiPEA as a base gave the
24,25
best results in series in terms of yields and racemisation (see supporting information).
For the final successful preparation of the basic sideꢀchain containing proteasome containing
inhibitors 53 ꢀ 67, the Cbz and Boc protecting groups leading to PSFs 59, 60, 62, 63, 65 and
6
6 were removed by HBr in acetic acid. These deprotection conditions led to substitution of
the azide functionality in compounds like N ꢀPheꢀLeuꢀLeuꢀPhe(4ꢀNH )ꢀSF 64 by bromide.
3
2
Fortuitously, a 4 M HCl solution in dioxane led to the desired deprotected proteasome
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 65
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
inhibitor
Unfortunately, this method was not successful to afford the inhibitor N
NH )ꢀSF 64. All PSF inhibitors containing an arginine at the P1 position were coupled
N
3
ꢀPheꢀLeuꢀLeuꢀPhe(4ꢀCH
2
NH
2
)ꢀSF 67 and
N
3
ꢀPheꢀLeuꢀLeuꢀLysꢀSF 61.
3
ꢀPheꢀLeuꢀLeuꢀPhe(4ꢀ
2
successfully with unprotected arginine derived SF as it was anticipated that the guanidine
moiety stays protonated during the reaction and therefore would not react.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 5. Synthesis of the peptide sulfonyl fluorides (PSFs) and overview of the
structures of the inhibitors. Note: all amino and guanidine functionalities are
protonated.
®
1) HBTU, Oxymapure , DiPEA or
®
DIC, Oxymapure , DiPEA or
DIC, Oxymapure , NMM or
®
R
R
O
O
O
O
H
H
HATU, DiPEA
R4
R4
R5
A
N
2
SO F
N
2
SO F
N
N
N
OH
H N
2) HBr, HOAc or
3
H
H
H
O
4Μ HCl, dioxane
O
R5
7
: A= Cl, 8 - 10: A = TFA
53 - 67
R4
R5
R4
R5
A: Arginine derived series
HN
NH
NH
2
H
H
N
53
O
N
56
O
O
O
O
H
N
H
54
O
O
57
58
2
H N
R4
N
2
SO F
N
N
H
H
O
O
O
H
N
R5
55
N
3
NH
SO
2
NH
SO
2
NH
SO
2
B: Lysine derived series
O
O
O
O
O
O
H
H
H
N
H
N
H
2
N
N
2
F
N
2
F
N
3
2
F
N
N
H
N
H
N
H
N
H
N
H
H
O
O
O
O
5
9
60
61
C: Aminoꢀphenylalanine derived series
NH
2
NH
2
NH
2
O
O
O
O
O
O
H
H
N
H
N
H
N
H
2
N
N
SO
2
F
SO
2
F
N
3
2
SO F
N
N
N
H
N
H
N
H
N
H
H
H
O
O
O
O
6
2
63
64
D: Methyleneꢀaminoꢀphenylalanine derived series
NH
2
NH
2
NH
2
O
O
O
O
O
O
H
H
H
N
H
N
H
2
N
N
SO
2
F
N
SO
2
F
N
3
2
SO F
N
N
N
H
N
H
N
H
N
H
H
H
O
O
O
O
6
5
66
67
ACS Paragon Plus Environment
12

Page 13 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Biological evaluation of the PSFs
The IC50 values of the in vitro structure activity relationship studies (SAR) of the
synthesized peptido sulfonyl fluorides were determined from the inhibitory curves in Figure 3
using constitutive human proteasome and are summarized in Table 1. The residual activity of
the 20S proteasome activity was measured at time points using a fluorescent probe where a
decrease in fluorescence corresponds to lower residual proteasome activity and therefore
indicating a more potent inhibitor. In total, 14 basic amino acid derived peptido sulfonyl
fluorides were tested of which compounds 53, 54, 57, 62, 65, 66 and 67 showed IC50 values
below 250 nM for the trypsinꢀlike site. Particularly compounds 54, 65, 66 and 67 were the
most potent inhibitors of this series with IC50 values of 150 nM for 54, 140 nM for 65,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
119 nM for 66 and 130 nM for compound 67. Although compounds 60 and 63 had IC50
values higher than 1 µM, all other compounds had IC50 values lower than 1 µM for inhibiting
the trypsinꢀlike proteasome activity. Compounds 59, 65 and 67 could only be obtained as a
diasteromeric mixtures in which the amino sulfonyl fluoride residue had probably partially
racemised, nevertheless these PSFs still had relatively low IC50 values. In the arginine
derived PSFs the Cbzꢀprotecting group appeared to be beneficial for the potency when
compound 54 is compared to compound 55, 56, 57 or 58. For example, changing the Bocꢀ
group in PSF 55 to a Cbz protecting group in PSF 54 decreased the IC50 value by more than
500 nM. Due to the synthetic strategy of the PSFs 59 ꢀ 67, the syntheses of Cbz Nꢀtermini
protected analogs in this series of lysine, aminoꢀphenylalanine and 4ꢀaminomethylꢀ
d
phenylalanine derived PSFs was not feasible at this point. Comparing compounds 56 with 58
and 60 with 61, it becomes apparent that an azido functionality does enhance the potency of
d
The final deprotection step was removal of the sideꢀchain Cbzꢀprotecting group of the
amino protected sulfonyl fluoride, which also would remove any terminal Cbzꢀprotecting
group
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 65
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the inhibitor significantly (roughly a factor 2) compared to those of the acetyl capped
counterparts. The lysine and aminoꢀphenylalanine derived PSFs 59, 60, 61, 62 and 63 showed
in general the lowest potencies in this study indicating that the presence of stronger basic side
chains in the P1 position was more favorable.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The key goal of this study was the development of β2 selective (trypsinꢀlike) PSF
inhibitors. It was shown that, in general, except for the aniline derived PSFs 62 and 63, all
inhibitors possessed a moderate to very high β2 selectivity. Outstanding β2ꢀselectivity was
shown by PSF inhibitors having a free terminus as in the arginine derived PSF 57 (~600 fold
selectivity), the lysine derived PSF 59 (>1000 fold selectivity) and the methylene amino
phenyl alanine derived PSF 65 (~900 fold selectivity), emphasizing the essentiality of free
amino terminus with respect to this.
15
Originally, azide containing PSFꢀligands were developed to capture possibly formed
ligand/proteasome covalent adducts using the copper(I) catalyzed azideꢀalkyne cyclo addition
(click) reaction. However, later we found that formation of covalent adduct with the
16
proteasome active site is followed by elimination of the ligand . Nevertheless, PSF 6 having
15
an azido Nꢀterminus was uncovered as one of most active β5 proteasome inhibitors and
therefore azido containing PSF inhibitors were included in this study to evaluate whether this
also would be the case for β2 inhibition. Although the azido containing inhibitors (58, 61 and
65) were slightly more active or had a similar activity that Nꢀterminally protected PSFs, as
was mentioned above, a freeꢀamino terminus, generally led to both the most potent and
selective β2 compounds 57, 59, 62 and 65.
Although the β5ꢀinhibitory activity, that is inhibition of chymotrypsin activity, of all
compounds was understandably poor, perhaps with the exception of 62 and 63 (both IC50
ꢀ
values < 1 ꢁM), this activity improves slightly by having a protected or masked (azide) Nꢀ
terminus. This finding is in accordance with earlier findings that β5ꢀselectivity (towards the
ACS Paragon Plus Environment
14

Page 15 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
5
chymotrypsinꢀlike site) requires at least a capped Nꢀterminus of the PFSs. Interestingly,
compound 63 is the only one in this series, which gave rise to stronger inhibition of β5ꢀ
activity as compared to inhibition of β2ꢀactivity. This might indicate that the hydrophobic
character of 'aniline' sideꢀchain plays a dominant role in determination of the selectivity
towards β5ꢀinhibition.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Inhibition of β2 (trypsinꢀlike) acitivity
Inhibition of β5 (chymotrypsinꢀlike) acitivity
1
00
100
75
5
3
4
5
3
5
75
55
54
55
56
57
58
5
6
50
57
50
5
8
25
25
0
0
1
10
100
1000
10000
100000
10
100
1000
10000 100000 1000000
c (inhibitor) [nM]
Inhibition of β2 (trypsinꢀlike) acitivity
c (inhibitor) [nM]
Inhibition of β5 (chymotrypsinꢀlike) acitivity
5
6
6
9
0
1
5
6
9
0
100
100
75
50
25
0
7
5
2
5
0
5
0
61
62
63
65
6
2
63
6
6
6
5
6
7
6
6
6
7
1
10
100
1000
10000
100000
1
10
100
1000
10000 100000 1000000
c (inhibitor) [nM]
c (inhibitor) [nM]
Figure 3. In vitro evaluation of PSFs using human constitutive 20S proteasome. Fluorogenic
substrates were selective for the respective trypsinꢀlike (BzꢀVGRꢀAMC) or chymotrypsinꢀ
like (SucꢀLLVYꢀAMC) subunit. Top: Arginine derived PSF´s. Bottom: Lysine, aminoꢀ
phenylalanine and 4ꢀaminomethylꢀphenylalanine derived PSF. Left: Trypsinꢀlike residual
enzyme activity. Right: Chymotrypsinꢀlike residual enzyme activity.
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 65
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1. Overview of the IC50 values of inhibition of the constitutive proteasome by
synthesized PSFs.
Compound
IC50 [nM]
IC50 (β5)/
IC50 (β2)
β5 (Chymotrypsinꢀ
like)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
5
R
R
R
β2 (Trypsinꢀlike)
5
3
Cbz(H)N
200 ± 100
10,000 ± 3,400
50
54
Cbz(H)N
Boc(H)N
Ac(H)N
150 ± 30
700 ± 200
980 ± 10
250 ± 45
350 ± 65
470 ± 110
1,500 ± 35
700 ± 45
220 ± 90
2,100 ± 35
n.d.
6,700 ± 1,600
27,800 ± 15,700
69,000 ± 29,000
140,000 ± 12,100
39,000 ± 8,000
>> 100 µM
45
40
5
5
5
5
6
7
70
H
2
N
560
110
> 1,000
300
96
58
N
3
5
6
6
6
9
0
1
2
H
2
N
Ac(H)N
>> 100 µM
N
3
67,000 ± 3,200
950 ± 65
H
2
N
4.3
63
Ac(H)N
210 ± 80
0.1
6
6
6
6
4
5
6
7
N
3
n.d.
n.d.
900
150
66
H
2
N
140 ± 7
125,000 ± 26,000
18,000 ± 380
8,900± 360
Ac(H)N
119 ± 7
N
3
130 ± 117
Compounds 59, 65 and 67 could only be obtained as a diasteromeric mixture in which the amino sulfonyl fluoride residue
1
had probably partially racemised as was apparent from HꢀNMR.
n.d.: not determined as compound 64 could not be synthesised.
A higher IC50 (β5)/ IC50 (β2) ratio indicates a higher β2 selectivity.
In this assay synthesized LUꢀ102 showed IC50 values of 10.8 ± 2.4 nM(β2) and 2,600 ± 500 nM (β5) with an IC50 (β5)/
1
9
IC50 (β2) of 241, which is agreement with the literature
1
6
Recently we found that PSFs showed selective inhibition of the immuno proteasome.
Therefore, a subset of the most powerful inhibitors were taken and evaluated in an immuno
ACS Paragon Plus Environment
16

Page 17 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
proteasome assay as in shown in table 2 and figure 4. The data showed that the selectivity of
1
9
β2 inhibition PSFs is maintained. However, in contrast to earlier developed PSFs selectivity
for inhibition of the immuno proteasome was virtually absent. This was also the case for Luꢀ
102, which although a highly potent inhibitor, also showed selectivity for constitutive
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
proteasome inhibition.
Inhibition of β2i (trypsinꢀlike) acitivity
Inhibition of β2i (trypsinꢀlike) acitivity
53
54
59
62
65
1
00
100
75
50
25
0
57
75
66
67
5
2
0
5
0
1
10
100
1000 10000 1000001000000
1
10
100
1000
10000 100000 1000000
c (inhibitor) [nM]
c (inhibitor) [nM]
Inhibition of β5i (chymotrypsinꢀlike) acitivity
Inhibition of β5i (chymotrypsinꢀlike) acitivity
53
59
1
00
54
100
75
50
25
0
6
2
57
65
7
5
2
5
0
5
0
66
67
1
10
100
1000
10000 100000 1000000
100
1000
10000
100000
1000000
c (inhibitor) [nM]
c (inhibitor) [nM]
Figure 4. In vitro evaluation of PSFs using human immuno 20S proteasome. Fluorogenic
substrates were selective for the respective trypsinꢀlike (BzꢀVGRꢀAMC) or chymotrypsinꢀ
like (SucꢀLLVYꢀAMC) subunit. Top: Arginine derived PSF´s. Bottom: Lysine, aminoꢀ
phenylalanine and 4ꢀaminomethylꢀphenylalanine derived PSF. Left: Trypsinꢀlike residual
enzyme activity. Right: Chymotrypsinꢀlike residual enzyme activity.
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 65
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 2. Overview of the IC50 values of inhibition of the immuno proteasome by selected
PSFs.
Compound
IC50 [nM]
IC50 (β5)/
IC50 (β2)
β5 (Chymotrypsinꢀ
like)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
5
R
R
R
β2 (Trypsinꢀlike)
5
3
Cbz(H)N
Cbz(H)N
1,110 ± 260
400± 130
14,000± 7,300
13
54
57
6,700 ± 12,800
>> 100 µM
17
2
10± 60
>1000
H
2
N
3
00 ± 60
>> 100 µM
>1000
2.3
5
6
9
2
H
2
N
2
00 ±60
450 ± 100
H
H
N
N
2
2
6
6
6
5
6
7
360± 130
395 ± 40
130 ± 10
136,500 ± 68,600
17,200 ± 6,700
3,100 ± 1,500
380
44
Ac(H)N
N
3
24
Compounds 59, 65 and 67 could only be obtained as a diasteromeric mixture in which the amino sulfonyl fluoride residue
1
had probably partially racemised as was apparent from HꢀNMR.
A higher IC50 (β5)/ IC50 (β2) ratio indicates a higher β2 selectivity.
In this assay synthesized LUꢀ102 showed IC50 values of 27.5 ± 6.9 nM(β2) and 1,800 ± 2,400 nM (β5) with an IC50 (β5)/
1
9
IC50 (β2) of 64, which is agreement with the literature
PSFs containing the SFꢀwarhead comprise relatively new inhibitor ligands of which
stability and reactivity properties have not been investigated as yet. In order to start
determination of the former properties we have evaluated the degree of hydrolysis of PSFs in
buffers of different pHs. Therefore, stability tests of several PSFs were carried out to study
the rate of hydrolysis of the SF moiety at a physiological pH (7.4) as well as at pH 6.5 and
pH 8.0. Acetyl capped PSFs were chosen to prevent a possible intermolecular reaction
ACS Paragon Plus Environment
18

Page 19 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
involving the Nꢀterminus and the SFꢀmoiety. Of course intermolecular reactions involving
the amino groups in 60, 63 and 66 cannot be ruled out but are highly unlikely, because they
are largely protonated at the studied pH's or poorly nucleophilic (63). An intermolecular
reaction involving the guanidinium functionality in 56 is even less likely because it is
virtually completely protonated at the studied pH range. So it was assumed that a possible
decrease in stability of the SFꢀmoiety is due to an increased inclination towards hydrolysis. In
order to study only the hydrolytic stability and to exclude nucleophilic reactions of buffer
components (for example tris(hydroxylmethyl)aminomethane in "Tris" buffer), phosphate
buffered saline (PBS) was chosen as a buffer system.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
pH 6.5
pH 7.4
pH 8.0
1
00
100
75
50
25
0
100
75
50
25
0
7
5
0
5
5
6
6
0
56
60
63
56
60
63
25
0
63
0
2
4
6
8
10
12
0
2
4
6
8
10
12
0
2
4
6
8
10
12
time [h]
time [h]
time [h]
Figure 5. PBS buffer stability test at pH 6.5, 7.4 and 8.0. Hydrolysis of selected PSFs was
measured via analytical HPLC over 12 h.
As anticipated, hydrolysis of PSFs was slower at slightly acidic pH (6.5) compared to
physiological pH (7.4) and slightly basic pH (8.0). Even after 12 h about 75% of the initial
used PSF remained unaffected at pH 6.5. After 1 h at pH 7.4 90% of the PSF was still intact.
Thus, during an incubation time of one hour for the biological evaluation of a PSF, a large
majority is still available for interaction with the proteasome. After 6 h at pH 7.4, > 50% of
the PSFs were still intact. However, rapid hydrolysis was observed at pH 8.0 at which 50% of
the PSFs were hydrolyzed after 2 h (Figure 5).
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 65
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
CONCLUSIONS
We have shown that our general route for the preparation of amino acid derived SFs could
be extended to SF derivatives containing nucleophilic groups in their side chains in the
presence of the SF electrophilic trap. The route to the SF derived from arginine 7 was largely
based on our earlier developed synthesis. The preparation of SF derivatives containing an
amino group in their side chains was achieved using a judicious strategy masking the αꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
amino group as an azide functionality, thereby circumventing complicated (de)protection
strategies. This resulted in the successful and convenient synthesis of lysine derived SF 8,
aminoꢀphenylalanine 9 and methylene aminoꢀphenylalanine derived SF 10 ready for
incorporation in proteasome inhibitor sequences.
Although incorporation of the SF warhead into a peptide inhibitor sequence remains a
challenging step in the total synthesis of PSFs, we have successfully synthesised several
potent and highly selective PSFs capable of inhibiting the β2 (trypsinꢀlike) activity over the
β5 (chymotrypsinꢀlike) activity for the first time.
In addition, we have established the crucial role of the Nꢀterminus. PSFs with a free amino
terminus gave rise to highly selective β2 proteasome inhibitors. Having an azide functionality
as Nꢀterminus did not lead to a tremendous enhanced potency as was found earlier with the
β5ꢀproteasome inhibitors. Although the described PSFs were less potent than e.g. LUꢀ102
their selectivity for β2 was similar. Both PSFs and LUꢀ102 did not show selective inhibition
of the immunoproteasome. Since PSF gave rise to permanent inhibition of the proteasome by
ligandꢀinduced crosslinking of the active site, it would be interesting to evaluate the
significance of proteasome inhibition by irreversible inhibitors versus reversible inhibitors.
In view of the promising electrophilic trap properties leading to highly active and selective
compounds, we embarked on initial stability studies in buffer of the SFꢀwarhead. It was
found that the stability at physiological pH was quite satisfactory. Clearly, this is just the start
ACS Paragon Plus Environment
20

Page 21 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
toward gaining insights in the behaviour of these relatively novel aliphatic ꢀamino acid
derivedꢀ SFs and PSFs. Thus, there will be significant chemical challenges ahead with respect
to modulating both reactivity and stability. This research will be guided by future studies
directed towards investigations of cellular permeability and stability of the most promising
PSFs described above.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
EXPERIMENTAL PROCEDURES
General procedures. All starting materials, reagents and solvents were obtained from
commercial sources and used as received. Dry solvents were obtained from a
TM
PureSolv 500 MD solvent purification system. Reactions requiring dry conditions were
performed in heatꢀgun dried glassware. All reactions were performed at ambient temperature
unless stated otherwise. Reactions in solution were monitored by TLC analysis on Merck preꢀ
coated silica gel 60 F254 (0.25 mm) glass backed plates. Spots were visualised by UV light
(254 and 366 nm) and by heating plates after dipping in a ninhydrine or cerium/molybdenum
solution. Column chromatography was performed on Siliaflash® P60 (40 ꢀ 63 µm) from
Silicycle (Canada). Petroleum ether (40 ꢀ 60 °C fraction) and nꢀhexane were used for flash
1
13
19
column chromatography. HꢀNMR, CꢀNMR and FꢀNMR spectra were recorded on a
Bruker DPX 400 spectrometer or Bruker 500 spectrometer with chemical shift values
reported in parts per million (ppm) relative to TMS (δ
H
= 0.00 and δ
C
= 0.0) or residual
= 39.52) as
CHCl
3
(δ
H
= 7.28 and δ
C
1
= 77.16) or residual d
6
ꢀDMSO (δ = 2.50 and δ
H
C
13
standard. Assignments of H and CꢀNMR signals are based on twoꢀdimensional COSY,
HSQC, HMBC, DEPT and DEPTQ experiments, respectively. Highꢀresolution electrospray
ionization (ESI) mass spectra were measured on a Bruker micrOTOFꢀQ II in positive or
negative mode and calibrated with an ESI tuning mix from Agilent Technologies. Infrared
spectra were recorded using a Shimadzu FTIR 8400S apparatus. Optical rotations were
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 65
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
determined as solutions irradiating with the sodium D line (λ = 589 nm) using an Auto pol
−1
2
−1
V polarimeter. [α] values are given in units 10 deg cm g . Semiꢀpreparative high
D
performance liquid chromatography (HPLC) was performed on an Agilent Technologies
ꢀ1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1260 Infinity system (12.5 mLmin ). Analytical HPLC chromatograms were recorded on a
ꢀ1
Shimadzu Prominence system (1 mLmin ). Buffers used for HPLC: buffer A (0.1% TFA in
MeCN:H O v/v 5:95) and buffer B (0.1% TFA in MeCN:H O v/v 95:5). Semiꢀpreparative
2
2
runs started with an isocratic flow of buffer A (100% for 5 min), followed by a linear
gradient of buffer B (in 60 min to X%). Subsequently, an isocratic flow of buffer B (100%
for 5 min) was performed followed by a linear gradient to buffer A (in 5 min to 100%). Runs
ended with an isocratic flow of buffer A (100% for 5 min). Used columns are mentioned in
the supporting information. In addition, all HPLC chromatograms and retention times of all
purified compounds are supplied in the supporting information. All compounds have purities
>
95%.
Proteasome inhibitory assays. Inhibition of the proteasome enzymatic activity was
TM
determined using the VIVAdetect 20S Assay Kit PLUS (Viva bioscience, UK) utilizing a
Clariostar microplate reader (BMG LABTECH, Germany). All working solutions were
freshly prepared for each measurement. Kinetic enzyme assays were performed using 96ꢀ
wells Corning half area plates using 50 µL of total amount of liquid. Incubation of all
measured inhibitors was at ambient temperature on a shaker for 60 min. Fluorescence
measurements were carried out at λ = 360 nm and λ = 460 nm at 25 °C for 2 h. All
ex
ex
assays were carried out in duplicate with three repetitions. Each well contained 35 µL
TM
VIVAdetect buffer. The final enzyme concentration in a well was 2.5 nM (5 µL of a
TM
25 nM enzyme working solution in VIVAdetect
buffer, prepared from 1 mg/mL 20S
proteasome). Final substrate concentration was 100 µM (5 µL of a 1 mM substrate working
TM
solution in VIVAdetect buffer; SucꢀLLVYꢀAMC for the chymotrypsinꢀlike activity; Bzꢀ
ACS Paragon Plus Environment
22

Page 23 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
VGRꢀAMC for the trypsinꢀlike activity). In order to use a final minimal concentration of
DMSO stock solutions of each inhibitor were prepared: Arginine derived PSF´s were
dissolved in 10% DMSO/H
aminoꢀphenylalanine and 4ꢀmethyleneꢀaminoꢀphenylalanine derived PSF´s were dissolved in
0% DMSO/H O. Dilution series of inhibitors were prepared using the appropriate stock
2 2
O, lysine derived PSF´s were dissolved in 30% DMSO/H O, 4ꢀ
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
2
solution. For the positive controls 5 µL of the stock DMSOꢀpercentage solution was added
instead of the respective inhibitor solution MG132 (CbzꢀLeuꢀLeuꢀLeucinal) was used as a
negative control with a final concentration of 5 µM (5 µL of a 50 µM working solution,
TM
supplied in the VIVAdetect 20S Assay Kit PLUS). Final inhibitor concentrations were:
4
1
00 µM, 200 µM, 100 µM, 50 µM, 10 µM, 5 µM, 2.5 µM, 1 µM, 750 nM, 500 nM, 250 nM,
00 nM, 50 nM, 10 nM, 1 nM, 0.1 nM. The inhibitory activities of the compounds were
expressed as IC50 values. IC50 values were obtained by plotting the residual percentage of
enzymatic activity against the logarithm of the inhibitor concentrations. Experimental data
were fitted to the equation %Residual Activity = 100/(1 + 10^((Log IC Logc(inhibitor)) *
5
0
Hill Slope)) using GraphPad Prism software version 5.
Stability evaluation in buffer. Compounds 56, 60 and 61 were dissolved in 90 µL DMSO
and added to an aqueous 1 x PBS buffer solution (prepared from 10 x PBS buffer solution
(Gibco) adjusted with aq. 2
M HCl or 2 M NaOH solution, respectively; pH 6.5 or 7.4 or 8.0,
910 µL) resulting in a final concentration of ~1 mM. The hydrolysis was monitored via
analytical HPLC (C ) over 12 h. 0 h was measured directly after the addition of the PBS
A3
buffer solution and was taken as the reference peak.
Synthesis. The synthetic procedures and characterization data of crucial synthetic
intermediates as well as those of the final PSFs are described below. All other synthetic
procedures and characterization data are included in the supporting information, which also
contains all NMR spectra, MSꢀdata and HLPC traces of the final PSFs.
ACS Paragon Plus Environment
23

2HCl·H
2
NꢀArgꢀψ
[CH
2
SO
2
]ꢀF (7). CbzꢀArg(Mtr)ꢀψ[CH
2 2
SO ]ꢀF (17) (310 mg, 541 µmol,
Journal of Medicinal Chemistry
Page 24 of 65
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
.00 eq.) was dissolved in CH
2
Cl (8 mL) and 33% HBr/AcOH solution (8 mL) was added.
2
The reaction mixture was stirred for 1 h and the solvents removed in vacuo. H O (20 mL)
2
®
was added and the aqueous layer extracted with EtOAc (20 mL) and treated with Dowex
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
×8 chloride form (777 mg) for 10 min. The mixture was filtered and freeze dried. The crude
1
product was obtained as a yellowish solid (166 mg, 555 µmol, 100%). HꢀNMR (500 MHz,
d ꢀDMSO, 298 K): δ (ppm) = 8.63 (s, 3H, NH αC), 7.79 (t, J = 5.9 Hz, 1H, CNHCH ),
6
H
3
2
3
4
.48 (dt, J = 15.5, 6.0 Hz, 1H, CH
a
SO
2
F), 4.40 (dt, JH,H,F = 15.5, 5.5 Hz, 1H, CH
b
SO
2
F),
3
.86 ꢀ 3.72 (m, 1H, H
3
NαCH), 3.22 ꢀ 3.05 (m, 2H, αCNHCH
2
), 1.87 ꢀ 1.72 (m, 2H, α
). CꢀNMR (126 MHz, d ꢀDMSO, 298 K): δ
F), 46.3 (αCH), 40.6 (α
13
CHCH
2
), 1.72 ꢀ 1.55 (m, 2H, αCHCH
2
CH
2
6
C
(
ppm) = 157.4 (NHCNH
2
), 52.3 (d, JC,S,F = 14.8 Hz, CH
2
SO
2
19
CNHCH ), 29.3 (NHCH CH ), 24.3 (αCNHCH CH ). FꢀNMR (471 Hz, d ꢀDMSO,
2
2
2
2
2
6
2
98 K): δ
F
(ppm) = 60.3 (s, 1F, SO
2
F). HRMS (ESI positive) calc. for C
6
H
16
N
4
O
2
SF
+
[M+H] 227.0973, found 227.0971.
2 2 3 2 2
TFA·HꢀLys(Cbz)ꢀψ[CH SO ]ꢀF (8). N ꢀLys(Cbz)ꢀψ[CH SO ]ꢀF (40) (300 mg, 837 µmol,
1
1
.00 eq.) was dissolved in AcOH (12.5 mL). Next, zinc powder (547 mg, 8.37 mmol,
0.0 eq.) and TFA (1.9 mL) were added and the reaction mixture was stirred overnight at rt.
The solvents were removed in vacuo and AcOH (10 mL) was added. The crude product was
purified via semiꢀpreparative HPLC (0 to 100% B, CP1) and fractions containing the pure
product were pooled and lyophilized. The pure product was obtained as a white solid
1
(
212 mg, 474 µmol, 57%). HꢀNMR (400 MHz, CDCl
3
, 298 K):
δ
H
(ppm) = 8.39 (s, 3H,
), 5.01 (s, 2H, CH
F), 4.25 (dt, J = 15.6, 5.7 Hz, 1H, CH SO F),
CH), 3.04 ꢀ 2.92 (m, 2H, NHCH ), 1.79 ꢀ 1.68 (m, 2H, αCHCH ),
+
NH
3
), 7.41 ꢀ 7.28 (m, 5H, ArꢀH (Cbz)), 7.24 (t, J = 5.6 Hz, 1H, NHCH
2
2
(
Cbz)), 4.41 (dt, J = 15.6, 5.6 Hz, 1H, CH
a
SO
2
b
2
3
.79 ꢀ 3.65 (m, 1H, NH
3
2
2
ACS Paragon Plus Environment
24

Page 25 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
3
1.48 ꢀ 1.32 (m, 4H, NHCH
2
CH
56.6 (C=O), 137.7 (CꢀAr), 128.8, 128.2, 128.2 (CHꢀAr), 65.6 (CH
5.0 Hz, CH SO F), 46.6 (NH CH), 40.3 (NHCH ), 31.8 (NH CHCH ), 29.3 (CH ), 21.5
2
CH
2
). CꢀNMR (126 MHz, CDCl
3
, 298 K):
c
δ (ppm) =
1
1
2
(Cbz)), 52.3 (d, JC,F =
2
2
3
2
3
2
2
1
9
(CH
2
3 F 2
). FꢀNMR (471 MHz, CDCl , 298 K): δ (ppm) = 62.7 (SO F). HRMS (ESI positive)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
+
calc. for C14
H
23
N
2
O
4
SF [M+H] 333.1279, found 333.1260.
TFA·HꢀPhe(4ꢀNHCbz)ꢀψ[CH SO ]ꢀF (9). N ꢀPhe(4ꢀNHCbz)ꢀψ[CH SO ]ꢀF (41)
2
2
3
2
2
(
314 mg, 799 µmol, 1.00 eq.) was dissolved in AcOH (12 mL). Next, zinc powder (522 mg,
7.99 mmol, 10.0 eq.) and TFA (1.73 mL) were added and the reaction mixture was stirred
overnight at rt. The solvents were removed in vacuo and AcOH (10 mL) was added. The
crude product was purified via semiꢀpreparative HPLC (0 to 100% B, CP1) and fractions
containing the pure product were pooled and lyophilized. The pure product was obtained as a
1
white solid (209 mg, 468 µmol, 59%). HꢀNMR (400 MHz, d
6
ꢀDMSO, 298 K):
δ
H
(ppm) =
+
3
9.84 (s, 1H, NH), 8.20 (br s, 3H, NH
), 7.46 (d, J = 8.4 Hz, 2H, CH (Phe)), 7.44 ꢀ 7.32 (m,
(Cbz)), 4.24 (app dt, J
15.5, 5.3 Hz, 1H, CH SO F), 4.16 (app dt, J = 15.5, 6.0 Hz, 1H, CH SO F), 4.02 ꢀ 3.93 (m,
5H, ArꢀH (Cbz)), 7.23 (d, J = 8.4 Hz, 2H, CH (Phe)), 5.15 (s, 2H, CH
2
=
a
2
b
2
+
13
1H, NH
3 3 2 6
αCH), 3.03 ꢀ 2.92 (m, 2H, N αCHCH ). CꢀNMR (101 MHz, d ꢀDMSO,
298 K): δc (ppm) = 153.8 (C=O), 138.9 (CꢀAr), 137.1 (CꢀAr), 130.5 (CHꢀAr (Phe)), 128.9
(
CHꢀAr (Cbz)), 128.6 (CHꢀAr (Cbz)), 128.5 (CHꢀAr (Cbz)), 128.0 (CꢀAr), 119.0 (CHꢀAr
1
9
(
Phe)), 66.2 (CH
2
(Cbz)), 52.2 (d, J = 15.7 Hz, CSO
2
F), 47.8 (αCH), 37.3 (αCH CH
2
). Fꢀ
NMR (377 MHz, CDCl
positive) calc. for C17
TFA·H NꢀPhe(4ꢀCH
3
, 298 K):
δ
F
(ppm) = 60.6 (t, J = 5.8 Hz, 1F, SO F). HRMS (ESI
2
+
H
19
N
2
O
4
SFNa [M+Na] 389.0942, found 389.0926.
2
2
NHCbz)ꢀψ[CH
2
SO
2
]ꢀF (10). N
3
ꢀPhe(4ꢀCH
2
NHCbz)ꢀψ[CH
2
SO
2
]ꢀF
(42) (62 mg, 153 µmol, 1.00 eq.) was dissolved in AcOH (2.3 mL). Next, zinc powder
(99.8 mg, 1.53 mmol, 10.0 eq.) and TFA (342 µL) were added and the reaction mixture was
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 65
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
stirred overnight at rt. Then, the solvents were removed in vacuo and AcOH (5 mL) was
added. The crude product was purified via semiꢀpreparative HPLC (0 to 100% B, CP1) and
fractions containing the pure product were pooled and lyophilized. The pure product was
1
obtained as a white solid (40.9 mg, 82.7 µmol, 54%). HꢀNMR (400 MHz, d ꢀDMSO,
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
+
2
98 K):
δ
H
(ppm) = 8.53 (br s, 3H, NH
3
), 7.86 (t, J = 6.3 Hz, 1H, NHCH
2
), 7.41 ꢀ 7.20 (m,
, CH SO F),
.12 ꢀ 3.97 (m, 1H, NH CH), 3.17 ꢀ 2.97 (m, 2H, αCHCH Phe). CꢀNMR (101 MHz, d ꢀ
9H, ArꢀH (Cbz), CH (Phe)), 5.05 (s, 2H, CH
2
(Cbz)), 4.31 ꢀ 4.16 (m, 4H, NHCH
2
2
2
1
3
4
3
2
6
DMSO, 298 K):
δ
c
(ppm) = 156.9 (C=O), 139.5 (CꢀAr), 137.6 (CꢀAr), 133.3 (CꢀAr), 130.0
NHCbz])), 128.8 (CHꢀAr (Cbz)), 128.3 (CHꢀAr (Phe[4ꢀCH NHCbz])),
(CHꢀAr (Phe[4ꢀCH
28.2 (CHꢀAr (Cbz)), 127.9 (CHꢀAr (Cbz)), 65.9 (CH
CH SO F), 47.7 (αCHCH Phe), 43.9 (CH
(471 MHz, d ꢀDMSO, 298 K):
calc. for C18
CbzꢀArg(Mtr)ꢀψ[CH
added to a mixture of N ꢀZꢀN ꢀ(4ꢀmethoxyꢀ2,3,6ꢀtrimethylbenzenesulfonyl)ꢀLꢀarginine
2
2
1
2
(Cbz)), 52.1 (d, J = 15.4 Hz,
19
2
2
2
2
NHCbz), 37.5 (αCHCH
2
Phe). FꢀNMR
6
δ
F
2
(ppm) = 60.6 (s, J = 5.6 Hz, SO F). HRMS (ESI positive)
+
H
21
N
2
O
4
SFNa [M+Na] 403.1098, found 403.1091.
2
O]ꢀH (13). Sodium borohydride (1.81 g, 47.9 mmol, 2.50 eq.) was
α
ω
methyl ester 12 (10.2 g, 19.2 mmol, 1.00 eq.) and LiCl (2.03 g, 47.9 mmol, 2.50 eq.) in dry
THF (40 mL) at rt and was stirred for 15 min. EtOH (55 mL) was added carefully and the
resulting cloudy mixture stirred for 5 h (TLC controlled). The reaction mixture was cooled to
0
°C and quenched with saturated NH
4
Cl (45 mL) and H
2
O (13 mL). The mixture was
. The
extracted with EtOAc (3 × 100 mL) and the combined organic phases dried over MgSO
4
solvent was removed in vacuo and the crude product purified via column chromatography
1
(
EA). The pure product was obtained as a white solid (9.19 g, 18.1 mmol, 95%). HꢀNMR
3 H
(400 MHz, CDCl , 298 K): δ (ppm) = 7.29 ꢀ 7.14 (m, 5H, ArꢀH (Cbz)), 6.42 (s, 1H, ArꢀH
(Mtr)), 6.26 ꢀ 6.03 (m, 3H, 3 × NH (guanidine)), 5.49 (d, J = 8.5 Hz, 1H, HNαCH), 4.96 (s,
ACS Paragon Plus Environment
26

Page 27 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2H, OCH
H, CH OH), 3.15 ꢀ 3.04 (m, 2H, NHCH
Mtr)), 2.03 (s, 3H, ArꢀCH (Mtr)), 1.52 ꢀ 1.32 (m, 4H, αCHCH
(101 MHz, CDCl , 298 K): δ (ppm) = 158.5 (C=O), 157.0 (Cꢀguanidine), 156.4 (CꢀAr),
38.5 (CꢀAr), 136.4 (CꢀAr (Cbz)), 133.3 (CꢀAr), 128.5, 128.1, 128.0 (CHꢀAr (Cbz)), 124.9
2
Ph), 3.73 (s, 3H, OCH
3
(Mtr)), 3.61 ꢀ 3.41 (m, 3H, HNαCH, CH
2
OH), 3.21 (br s,
1
2
2
), 2.57 (s, 3H, ArꢀCH (Mtr)), 2.50 (s, 3H, ArꢀCH
3
3
13
(
3
2 2
CH ). CꢀNMR
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
C
1
(
CꢀAr), 111.8 (CHꢀAr (Mtr)), 66.8 (CH (Cbz)), 64.7 (CH OH), 55.4 (OCH ), 41.0 (NHCH ),
2
2
3
2
2
8.5 (αCHCH
2
CH
2
), 25.6 (αCHCH
2
CH
2
), 24.1 CH
3
3 3
(Mtr)), 18.3 CH (Mtr)), 11.9 (CH
+
(Mtr)). HRMS (ESI positive) calc. for C24
H
34
N
4
O
3
S [M+Na] 529.2091, found 529.2073.
CbzꢀArg(Mtr)ꢀψ[CH
2
O]ꢀMs (14). Methanesulfonyl chloride (1.72 mL, 22.2 mmol,
O]ꢀH (13) (8.64 g,
(3.1 mL, 22.2 mmol, 1.30 eq.)
1
.30 eq.) was added dropwise to a solution of CbzꢀArg(Mtr)ꢀψ[CH
7.1 mmol, 1.00 eq.) in dry CH2CL2 (200 mL) at 0 °C. NEt
2
1
3
was added dropwise and the reaction mixture was stirred for 2 h at rt (TLC controlled). After
the reaction was finished the organic phase was washed with an aqueous 1 M KHSO
solution (250 mL), H O (250 mL) and brine (250 mL). The organic layer was dried over
4
2
MgSO
4
, the solvent was removed in vacuo and the crude product obtained as a white solid
1
(9.36 g, 16.0 mmol, 94%). HꢀNMR (400 MHz, CDCl , 298 K): δ (ppm) = 7.31 ꢀ 7.20 (m,
3
H
5
H, ArꢀH (Cbz)), 6.44 (s, 1H, ArꢀH (Mtr)), 6.11 (br s, 2H, 2 × NH (guanidine)), 5.99 (s, 1H,
NH (guanidine)), 5.46 (d, J = 7.7 Hz, 1H, HNαCH), 5.01 (d, J = 12.2 Hz, 1H, OCH Ph),
4.95 (d, J = 12.2 Hz, 1H, OCH Ph), 4.11 (dd, J = 10.3, 4.3 Hz, 1H, CH OSO CH ), 4.05 (dd,
J = 9.9, 4.3 Hz, 1H, CH OSO ), 3.85 ꢀ 3.76 (m, 1H, HNαCH), 3.74 (s, 3H, OCH
Mtr)), 3.17 ꢀ 3.05 (m, 2H, HNCH
a
b
a
2
3
b
2
CH
3
3
(
2
), 2.86 (s, 3H, SO
2
CH
3
), 2.57 (s, 3H, ArꢀCH
3
(Mtr)), 2.50
1
3
(s, 3H, ArꢀCH (Mtr)), 2.04 (s, 3H, ArꢀCH (Mtr)), 1.57 ꢀ 1.37 (m, 4H, αCHCH2CH2). Cꢀ
3
3
NMR (101 MHz, CDCl
3
, 298 K): δ
C
(ppm) = 158.6 (C=O), 156.4 (Cꢀguanidino), 156.3 (Cꢀ
Ar), 138.5 (CꢀAr), 136.6 (CꢀAr), 136.3 (CꢀAr), 128.5, 128.2 , 128.0 (CHꢀAr (Cbz)), 124.9
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 65
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(CꢀAr), 111.8 (CHꢀAr (Mtr)), 71.1 (CH
2
OSO
2
CH
3
), 66.9 (OCH
2
(Cbz)), 55.5 (OCH
3
), 50.1 (
), 24.1
αCH), 40.7 (NHCH ), 37.2 (SO CH
2
2
3
), 28.1 (αCHCH
2
CH
2
), 25.5 (αCHCH CH
2
2
(
CH (Mtr)), 18.3 (CH (Mtr)), 12.0 (CH (Mtr)). HRMS (ESI positive) calc. for
3
3
3
+
C
25
H
36
N
4
O
8
S
2
Na [M+Na] 607.1867, found 607.1853
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CbzꢀArg(Mtr)ꢀψ[CH
added to a suspension of Cs
atmosphere. Most of the Cs
[CH O]ꢀMs (14) (9.28 g, 15.9 mmol, 1.00 eq.) in DMF (38 mL) under argon atmosphere. The
reaction mixture was stirred overnight at rt with the flask covered in aluminium foil. EtOAc
160 mL) and H O (160 mL) were added to the reaction mixture and the organic layer
washed with aq. solution of 1 M KHSO (160 mL), aq. solution of 1 M NaHCO (160 mL),
2
S]ꢀAc (15). Thioacetic acid (2.27 mL, 31.7 mmol, 2.00 eq.) was
CO (5.17 g, 15.9 mmol, 1.00 eq.) in DMF (10 mL) under argon
CO was dissolved when added to a solution of CbzꢀArg(Mtr)ꢀψ
2
3
2
3
2
(
2
4
3
4
brine (160 mL) and dried over MgSO . The solvent was removed in vacuo and the crude
product purified via column chromatography (EA:Petrolium ether v/v 8:2 → EA). The pure
1
product was obtained as a yellowish solid (6.85 g, 12.1 mmol, 76%). HꢀNMR (400 MHz,
CDCl
5.97 (m, 3H, 3 × NH (guanidine)), 5.15 (d, J = 9.2 Hz, 1H, HNαCH), 5.00 (d, J = 12.4 Hz,
1H, OCH Ph), 4.92 (d, J = 12.4 Hz, 1H, OCH Ph), 3.72 (s, 3H, OCH (Mtr)), 3.68 ꢀ 3.57 (m,
H, HNαCH), 3.13 ꢀ 2.99 (m, 2H, HNCH ), 2.89 (dd, J = 14.0, 4.7 Hz, 1H, CH SC(O)CH ),
.79 (dd, J = 14.0, 7.9 Hz, 1H, CH SC(O)CH ), 2.58 (s, 3H, ArꢀCH (Mtr)), 2.51 (s, 3H, Arꢀ
(Mtr)), 2.19 (s, 3H, SC(O)CH ), 2.03 (s, 3H, ArꢀCH (Mtr)), 1.51 ꢀ 1.28 (m, 4H, α
). CꢀNMR (101 MHz, CDCl , 298 K): δ (ppm) = 196.1 (SC(O)CH ), 158.4
C=O), 156.6 (Cꢀguanidino), 156.3 (CꢀAr), 138.5 (CꢀAr), 136.5 (CꢀAr), 136.4 (CꢀAr), 133.6
CꢀAr), 128.5, 128.1, 127.8 (CHꢀAr (Cbz)), 124.8 (CꢀAr), 111.7 (CHꢀAr (Mtr)), 66.7 (OCH
), 51.1 (CH S), 40.8 (NHCH ), 33.8 (NHαCH), 31.8 (αCHCH CH ),
3 H
, 298 K): δ (ppm) = 7.30 ꢀ 7.15 (m, 5H, ArꢀH (Cbz)), 6.43 (s, 1H, ArꢀH (Mtr)), 6.27
ꢀ
a
b
3
1
2
2
a
3
b
3
3
CH
3
3
3
13
CHCH
2
CH
2
3
C
3
(
(
2
(Cbz)), 55.4 (OCH
3
2
2
2
2
ACS Paragon Plus Environment
28

Page 29 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
30.5 (SC(O)CH
Mtr)). HRMS (ESI negative) calc. for C26
CbzꢀArg(Mtr)ꢀψ[CH SO ]ꢀF (17). CbzꢀArg(Mtr)ꢀψ[CH
1.00 eq.) was dissolved in AcOH (40 mL) and 30% H
The reaction mixture was stirred for 48 h and additional H
3
), 25.7 (αCHCH
2
CH
2
), 24.1 (CH
3
(Mtr)), 18.3 (CH
3
(Mtr)), 12.0 (CH
3
–
(
H
36
N
4
O
6
S
2
[M−H] 563.2003, found 563.1980.
2
2
2
S]ꢀAc (15) (6.63 g, 11.7 mmol,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
O
2
aq. solution (13.5 mL) was added.
aq. solution (3.5 mL) was
2
O
2
added. NaOAc (963 mg, 11.7 mmol, 1.00 eq.) was added and the mixture stirred for 1 h.
DMF (20 mL) was added and the solution was concentrated in vacuo until a quarter of the
volume. This procedure was repeated 3 more times and finally DMF removed completely.
2
After coꢀevaporation with H O (2 × 100 mL) the crude product was lyophilized and 16
obtained as a white solid (6.82 g). The crude sodium salt 16 (600 mg, 1.01 mmol, 1.00 eq.)
®
was dissolved in dry CH
2
Cl
2
(25 mL) under argon atmosphere and XtalFluorꢀM (442 mg,
1
.82 mmol, 1.80 eq.) and 3HF·NEt (7.1 µL, 43.6 µmol, 0.04 eq.) were added. The reaction
3
mixture was heated to 40 °C and stirred overnight. Silica was added to quench the reaction
and the solvent removed in vacuo. Column chromatography (EA) yielded the pure product as
1
a white solid (320 mg, 559 µmol, 55% over two steps). HꢀNMR (400 MHz, CDCl
3
, 298 K):
δ
H
(ppm) = 7.37 ꢀ 7.24 (m, 5H, ArꢀH (Cbz)), 6.51 (s, 1H, ArꢀH (Mtr)), 6.36 ꢀ 6.04 (m, 3H, 3
NH (guanidine)), 5.89 (br s, 1H, HNαCH), 5.13 ꢀ 4.99 (m, 2H, CH (Cbz)), 4.20 ꢀ 4.00
m, 1H, HNαCH), 3.80 (s, 3H, OCH (Mtr)), 3.65 (dd, J = 15.2, 6.9 Hz, 1H, CH SO F), 3.52
×
2
(
3
a
2
ꢀ
3.39 (m, 1H, CH SO F), 3.15 (br s, 2H, HNCH ), 2.63 (s, 3H, ArꢀCH (Mtr)), 2.56 (s, 3H,
b 2 2 3
13
3 3 2 2
ArꢀCH (Mtr)), 2.10 (s, 3H, ArꢀCH (Mtr)), 1.79 ꢀ 1.38 (m, 4H, αCHCH CH ). CꢀNMR
(126 MHz, CDCl , 298 K): δ (ppm) = 158.7 (C=O), 156.3 (Cꢀguanidino), 156.0 (CꢀAr),
3
C
1
38.4 (CꢀAr), 136.6 (CꢀAr), 136.1 (CꢀAr), 132.9 (CꢀAr), 128.5 (CHꢀAr (Cbz)), 128.2 (CHꢀAr
Cbz)), 127.9 (CHꢀAr (Cbz)), 125.0 (CꢀAr), 111.8 (CH (Mtr)), 67.1 (CH (Cbz)), 55.4
OCH ), 54.5 (d, J = 13.2 Hz, CH SO F), 47.2 (αCH), 40.4 (αCHCH CH CH ), 30.5 (α
(
(
2
3
C,F
2
2
2
2
2
ACS Paragon Plus Environment
29