Design, Synthesis, and Bioactivation of O-Glycosylated Prodrugs of the Natural Nitric Oxide Precursor Nω-Hydroxy- l -arginine

Article abstract of DOI:10.1021/acs.jmedchem.6b00810

Naturally occurring N^ω-hydroxy-l-arginine (NOHA, 1) is the best substrate of NO synthases (NOS). The development of stable and bioavailable prodrugs would provide a pharmacologically valuable strategy for the treatment of cardiovascular diseases that are associated with endothelial dysfunction. To improve NOHAs druglike properties, we demonstrate that O-substitution by (glycosylic) acetal formation greatly increased the chemical stability of the hydroxyguanidine moiety and provided a nontoxic group that could be easily bioactivated by glycosidases. A straightforward synthetic concept was devised and afforded a series of diversely substituted prodrugs by O-conjugation of the hydroxyguanidine moiety with different monosaccharides. Systematic exploration of their bioactivation profile revealed that glucose-based prodrugs were more efficiently bioactivated than their galactose counterparts. NOS-dependent cytosolic NO release was quantified by automated fluorescence microscopy in a cell-based assay with murine macrophages. Glucose-based prodrugs performed particularly well and delivered cellular NO levels comparable to 1, demonstrating proof-of-concept.

Full text of DOI:10.1021/acs.jmedchem.6b00810

Subscriber access provided by Northern Illinois University
Article
Design, Synthesis, and Bioactivation of O-Glycosylated Prodrugs
#
of the Natural Nitric Oxide-Precursor N-Hydroxy-L-Arginine
Felix-A. Litty, Julia Gudd, Ulrich Girreser, Bernd Clement, and Dennis Schade
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00810 • Publication Date (Web): 22 Aug 2016
Downloaded from http://pubs.acs.org on August 23, 2016
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 free 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 accessible to all
readers and 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.
Journal of Medicinal Chemistry 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 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design, Synthesis, and Bioactivation of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
OꢀGlycosylated Prodrugs of the Natural Nitric
ω
OxideꢀPrecursor N ꢀHydroxyꢀ
LꢀArginine
1
1
1
1
*,2
FelixꢀA. Litty‡ , Julia Gudd‡ , Ulrich Girreser , Bernd Clement , Dennis Schade
‡
both authors contributed equally
corresponding author
*
1
Department of Pharmaceutical and Medicinal Chemistry, Pharmaceutical Institute, Christianꢀ
AlbrechtsꢀUniversity of Kiel, Gutenbergstraße 76, 24118 Kiel, Germany
2
TU Dortmund, Department of Chemistry & Chemical Biology, OttoꢀHahnꢀStraße 6, 44227
Dortmund
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 45
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
ABSTRACT
ω
Naturally occurring N ꢀhydroxyꢀLꢀarginine (NOHA, 1) is the best substrate of NO synthases
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
(NOS). The development of stable and bioavailable prodrugs would provide a pharmacologically
valuable strategy for the treatment of cardiovascular diseases that are associated with endothelial
dysfunction. To improve NOHAs drugꢀlike properties, we demonstrate that Oꢀsubstitution by
(glycosylic) acetal formation greatly increased the chemical stability of the hydroxyguanidine
moiety and provided a nonꢀtoxic group which could be easily bioactivated by glycosidases. A
straightforward synthetic concept was devised and afforded a series of diversely substituted
prodrugs by Oꢀconjugation of the hydroxyguanidine moiety with different monosaccharides.
Systematic exploration of their bioactivation profile revealed that glucoseꢀbased prodrugs were
more efficiently bioactivated than their galactose counterparts. NOSꢀdependent cytosolic NO
release was quantified by automated fluorescence microscopy in a cellꢀbased assay with murine
macrophages. Glucoseꢀbased prodrugs performed particularly well and delivered cellular NO
levels comparable to 1 demonstrating proofꢀofꢀconcept.
ACS Paragon Plus Environment

Page 3 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Cardiovascular diseases (CVDs) account for almost 40% mortality in the Western world,
1
representing the major cause of death. It is wellꢀrecognized that low endogenous nitric oxide
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
NO) levels are closely linked to the pathology of a variety of CVDs, with endothelial
2
,3
dysfunction and atherosclerosis representing hallmarks of such disease states. However, NO is
4
,5
involved in a plethora of physiological and pathophysiological processes. Therefore, it is
highly desirable to rather selectively affect pathophysiologically altered processes of NO
generation and NO availability, thereby providing a safe and efficient therapy. As a consequence,
a great deal of research has focused in the past decades on pharmacological modulation of the
6
predominant enzymes involved in the regulation of NO levels. The nitric oxide synthases
(NOSs) represent the key enzymes of NO formation as they catalyze the fiveꢀelectron oxidation
ω
of
Lꢀarginine to
L
ꢀcitrulline and NO via N ꢀhydroxyꢀ
Lꢀarginine (NOHA, 1) as a catalytic
7
intermediate (Figure 1). Although NOSs are attractive targets, their active sites have
demonstrated very little promiscuity, and to date only few alternative substrates have been
8
identified. In addition to representing the best substrate for NOSs, 1 mainly releases NO only
where it is ultimately needed, i.e. at the NOSꢀsiteꢀofꢀaction, as opposed to other NO donors (e.g.,
9
nitrates). Furthermore, 1 is a potent inhibitor of arginases (Figure 1). These enzymes catalyze
the conversion of Lꢀarginine to Lꢀornithine, and thus, can potentially deplete Lꢀarginine as a
1
0
substrate for NOSꢀmediated NO generation. An increased expression and activity of arginase is
11
indeed closely linked to atherosclerotic changes and endothelial dysfunction. Hence, 1 exhibits
a unique dual and selective mode of action for potential therapeutic applications in several
CVDs.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 45
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
ω
Figure 1. N ꢀHydroxyꢀ
L
ꢀarginine (1, NOHA) acts as a key intermediate of NOS catalysis and is
a potent inhibitor of arginases; CYP, cytochrome P450;
Lꢀarg,
Lꢀarginine; Lꢀcit, Lꢀcitrulline;
ω
Lꢀorn,
L
ꢀornithine; NO, nitric oxide; NOHA, N ꢀhydroxyꢀ
Lꢀarginine; NOS, nitric oxide
synthases.
However, direct application of 1 is limited by its poor drugꢀlike properties. In our previous
6,12
work we pursued different strategies to overcome its chemical and metabolic liabilities.
We
demonstrated that Oꢀ and Nꢀsubstitution of the Nꢀhydroxyguanidine moiety greatly increased
1
2
hydrolytic and oxidative stability. Unfortunately, in vitro experiments later revealed that all Nꢀ
substituted prodrugs (i.e., carbamates) were not bioactivated which in turn would prevent 1 of
being accepted as a NOS substrate for the desired release of NO (data not shown). Building on
our observation that sole Oꢀsubstitution efficiently protects from oxidation and hydrolytic
degradation, we next aimed at designing carbamateꢀfree derivatives. Moreover, simple Oꢀalkyl
prodrugs were supposed to be avoided as they required involving CYP enzymes in the
13
bioactivation cascade, thereby increasing risks for drugꢀdrug interactions.
ACS Paragon Plus Environment

Page 5 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Here, we introduce a concept that afforded chemically stable prodrugs of 1 which are taken up
by cells and bioactivated in a monooxygenaseꢀindependent fashion while releasing degradation
products that are uncritical from a toxicological point of view. Specifically, Oꢀacetalic
incorporation of the labile Nꢀhydroxyguanidine group into monosaccharides (i.e., galactose and
glucose) furnished stable prodrug candidates that are bioactivated by glycosidases (Figure 2).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Illustration of a novel prodrug concept for 1 by Oꢀglycosylation.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 45
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
RESULTS AND DISCUSSION
Synthesis of O-glycosylated prodrugs of 1
Within our longꢀstanding efforts in designing various N ꢀ and N ꢀsubstituted
δ
ω
L
ꢀarginines we
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
2,14–16
have utilized several synthetic strategies.
A number of Oꢀalkylated and Nꢀacylated
derivatives of 1 were already accessible by reaction of carbamoylated thioureas with the desired
hydroxylamine in the presence of a desulfurizing agent (i.e., EDCI). This strategy only furnished
prodrugs that were Nꢀcarbamoylated (e.g., Eoc or Cbz group) which required harsh conditions
for their removal. Thus, for the herein described novel carbamateꢀfree prodrug candidates, an
orthogonal protecting group was needed that would allow efficient synthesis of the
Nꢀhydroxyguanidine group but also ensure mild cleavage conditions that were compatible with
the carbohydrate conjugate and
αꢀamino acid protecting groups. We reasoned that the Alloc
protecting group could fulfill these requirements by a nucleophilic attack of a palladium(0)
catalyst to afford πꢀallyl palladium complexes which can then be trapped by mild nucleophiles
such as 5,5ꢀdimethylcyclohexaneꢀ1,3ꢀdione (dimedone).
For a series of ethyl ester prodrugs (i.e., 3, 5, 6, 8), protected Lꢀornithine 12 was prepared in
two steps from commercially available 10. First, 10 was treated with phthalic anhydride in
tetrachloroethylene and then esterified under Steglich conditions with 4ꢀDMAP, EtOH and
1
7,18
EDCI.
The Alloc group could then be introduced by removing the phthalimide with
hydrazine and directly reacting the intermediate ornithine ethyl ester with freshly prepared
1
9,20
allyloxycarbonyl isothiocyanate (AllocNCS, 13)
to furnish thiourea 14b in moderate yield
(
40%). Attempts to isolate the intermediate failed largely due to rapid cyclization to a δꢀlactam.
α
For the desired free
α
ꢀamino acid prodrug candidates (i.e., 2, 4, 7, 9), N ꢀBocꢀ
L
ꢀornithine tertꢀ
butyl ester 15 was reacted with 13 to obtain thiourea 14a in excellent yield (94%).
ACS Paragon Plus Environment

Page 7 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
O
O
O
O
i
ii
H N
OH
N
OH
NHBoc
N
OEt
NHBoc
2
NHBoc
O
O
1
0
11
12
1
t
R = Bu or Et
iii-iv
AllocNCS (13)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
R =
OAc
AcO
AcO
O
S
O
iv
O
Gal
Glc
Alloc
t
OR1
H N
O Bu
NHBoc
N
N
*
*
2
AllocNCS (13)
OAc
H
H
NHBoc
OAc
1
t
1
4a: R = Bu
1
5
1
O
14b: R = Et
AcO
AcO
OAc
O
R2
v
NH2
O
3
R =
1
6
OH
OH
HO
HO
R2
R2
O
Gal
O
O
N
*
*
N
O
OH
O
vi
Alloc
OR1
OR1
H N
N
N
H
N
H
2
H
NHBoc
NHBoc
Glc
HO
HO
1
t
1
t
19a-b: R = Bu
17a-b: R = Bu
OH
1
1
2
0a-b: R = Et
18a-b: R = Et
4
R = H or Et
viii
vi-vii
R2
O
R3
O
R3
x 2 HCl
N
x 2 HCl
O
O
N
O
N
O
ix (2 and 4)
OR4
O- Na+
H N
N
H
H2N
N
OEt
H N
N
2
2
H
H
NH2
NH2
R2
NH2
R3
R2
R4
R4
R3
7
9
Gal
Glc
2
3
Gal
H
4
5
Glc
Glc Et
H
6
8
Gal
Glc
Gal Et
Scheme 1. Synthetic routes to diverse Oꢀglycosylated prodrugs of 1. (i) phthalic anhydride,
C Cl , 100 °C, 6 h (61%); (ii) EtOH (5 equiv.), 4ꢀDMAP (40 molꢀ%), EDCI, DCM, r.t., 24 h
2
4
(
54%); (iii) N H ×H O (15 equiv.), DCM/EtOH 1:1, r.t., 2.5 h; (iv) allyloxycarbonyl
2 4 2
isothiocyanate (13), DCM, 0 °C–r.t., 2 h (94%, 14a and 40% for 14b in one pot, 2ꢀsteps); (v)
DIPEA, EDCI, DCM, 48–72 h, r.t. (73–93%); (vi) Pd(0)[Ph P] , dimedone (8 equiv.), THF, r.t.,
3
4
3
0 min (73–90%); (vii) acetyl chloride, EtOH abs., argon, 0 °C–r.t., 48 h (82–93%); (viii) 1.
HCl for 20 min, Et O abs., argon, ꢀ10 °C, 2. 4–8 °C overnight (75–97%); (ix) NaOMe, MeOH,
(
g)
2
0
°C–r.t., 2.5 h (90%–quant.).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 45
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
Next, the fully protected Nꢀhydroxyguanidine prodrug precursors 17a-b and 18a-b were
obtained in good yields (73–93%) by reacting thioureas 14a-b with the desired peracetyl
2
21,22
protected 1ꢀaminooxysugars 16 (R = Gal, Glc),
DIPEA and EDCI in dichloromethane.
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
Importantly, chemoselective cleavage of Alloc was achieved in good yields (73–90%) with
Pd(0)[P(Ph) ] and dimedone under mild conditions in THF without compromising the other
3
4
2
3
protecting groups or the Oꢀacetalic function.
In view of the lability of the ꢀglycosylic bond towards acidic conditions, final deprotection of
the ꢀamino acid moiety represented a crucial step but was achieved under waterꢀfree conditions
with gaseous HCl in Et O. The desired prodrugs 2-5 precipitated at 4–8 °C overnight as
β
α
2
3
5
dihydrochloride salts (determined by quantitative Cl NMR spectroscopy, see Supporting
Information). Prodrugs 6 and 8 were obtained from 18a-b by in situ generation of HCl in
absolute EtOH which ensured retaining the
αꢀcarboxylic ester while all acetyl groups could be
removed via reꢀesterification. The preparation of fully deprotected prodrug candidates 7 and 9
2
4
was achieved from 2 and 4 using sodium methoxide in MeOH (Scheme 1).
In previous work we found that 1 and other Nꢀhydroxyguanidines only existed in the
25,26
oximeꢀtype structure and not in the hydroxylamineꢀtype tautomer.
Here, we additionally
examined the cisꢀtrans (E/Z) isomerism of several prodrugs and synthetic precursors (i.e., 18a)
1
1
15
by H, HꢀNOESY and N NMR. The data revealed that in all analyzed compounds the sugar
moiety was located in transꢀ (E) position to the α−amino acid (for detailed spectroscopic data
see Supporting Information).
ACS Paragon Plus Environment

Page 9 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Chemical stability
We performed a systematic analysis on the stability profile of the new prodrugs using several
Oꢀgalactosyl derivatives as representative examples. The primary aim was to investigate whether
the novel O,Oꢀacetal consisting of an Nꢀhydroxyguanidine and a monosaccharide represents a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
stabilityꢀdetermining structural moiety. Hence, prodrug 7, with a free
αꢀamino acid and the
deprotected galactosyl residue, was tested under various conditions. As shown in Figure 3, this
prodrug was highly stable at pH 2.0, 7.4 and 9.0 over 24 hours at 37 °C. This result suggested a
surprisingly high chemical stability of these novel carbohydrate conjugates even under acidic
conditions. Moreover, our previous work revealed that 1 is very labile at basic pH but can be
1
2
stabilized by Oꢀalkylation. Since prodrug 7 is also very stable at pH 9 it seems that the same
effect can also be achieved by Oꢀacetalic substitution with carbohydrates.
Furthermore, we addressed which carboxylic esters are more susceptible to chemical
hydrolysis by focusing on the simple ethyl ester 6 as well as the perꢀacetylated galactosyl
prodrug 2. Interestingly, the ethyl ester in 6 was much faster hydrolyzed at pH 9 and pH 7.4
(t1/2 = ca. 10–12 hours) compared to the acetyl group(s) in 2 (t1/2 > 24 h) likely due to
neighboring effects of the ꢀamino group and/or for steric reasons. It should be noted that a slow
α
chemical hydrolysis of the different esters is welcome in view of a slow and sustained release of
the active drug 1 (= bioactivation).
Overall, all three prodrug candidates exhibited a sufficient stability profile allowing the
conduction of various in vitro experiments. These compounds would also be wellꢀsuited for any
in vivo application, even for an ultimately desired peroral application.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 45
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
Figure 3. Chemical stability of the Oꢀgalactosyl prodrugs 2, 6 and 7. Prodrugs (250 ꢀM) were
incubated at 37 °C in potassium phosphate buffer (10 mM) and analyzed by HPLC. Each data
point represents mean ± SD (n = 3).
In vitro bioactivation
Since our initially developed Nꢀcarbamateꢀcontaining prodrugs of 1 were not bioactivated, it
was essential to investigate the release of 1 from the herein disclosed carbamateꢀfree prodrug
series. Depending on the respective prodrug candidate, the required bioactivation consists of
ester as well as acetal hydrolysis as depicted in Figure 4a for several galactose conjugates (i.e., 2,
3, 6, 7). Glucoseꢀconjugated prodrugs of 1 were expected to be similarly bioactivated by the
action of βꢀglucosidase instead of βꢀgalactosidase. To systematically evaluate the proposed steps
of bioactivation along with substrate specificity, a selection of distinct galactoseꢀ (2, 6 and 7) and
glucoseꢀbased (4 and 9) prodrugs were tested with various enzyme sources (Figure 4bꢀd).
First, it was addressed whether a perꢀacetylated galactose would still be accepted by
β
ꢀgalactosidase. In fact, prodrug 2 did not show any turnover by βꢀgalactosidase in contrast to
the unsubstituted galactose conjugate 7 (Figure 4b, left). This result turned out to be in sharp
ACS Paragon Plus Environment

Page 11 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
contrast to the substrate specificity of
of both the perꢀacetylated glucose prodrug 4 and the unsubstituted prodrug 9 (Figure 4d).
Moreover, ꢀglucosidase accepted not only glucoseꢀbased prodrugs as substrates but also cleaved
βꢀglucosidase which efficiently catalyzed the conversion
β
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
off the galactosyl residue from prodrug 7 at comparable turnover rates (Figure 4d). This broad
substrate specificity is wellꢀdocumented in the literature and seems to also account for the herein
27,28
introduced compounds.
Next, a stepwise bioactivation was simulated by incubating ethyl ester prodrug 6 together with
carboxylesterase and
bioactivation, intermediately formed prodrug 7 was detected in addition to the actual drug 1.
However, 7 was formed at much lower rates than 1 suggesting that ꢀgalactosidase already
converted the ethyl ester 6 to a high extent (see Supporting Information). Furthermore, the
simultaneous release of ꢀgalactose as a cleavage product upon bioactivation of 7 was verified
βꢀgalactosidase (Figure 4b, right). In accordance with the proposed steps of
β
D
and quantified using a colorimetric enzymeꢀbased assay (Figure 4c).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 45
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
Figure 4. In vitro bioactivation of Oꢀglycosyl prodrugs of 1. a) Proposed steps of bioactivation
and metabolic intermediates as exemplified for galactoseꢀbased prodrugs 2, 3, 6 and 7, b) HPLCꢀ
based quantification of 1 after incubation of prodrugs 2, 6 or 7 (500 ꢀM) with: 25 U/mL
β
ꢀgalactosidase (left), 33 U/mL
β
ꢀgalactosidase and 133 U/mL carboxylesterase (right), c)
ꢀgalactose after incubation of prodrug 7 (10 mM) with
ꢀgalactosidase (59 U/mL) d) HPLCꢀbased quantification of 1 after incubation of prodrugs 4 or 9
Colorimetric quantification of
D
β
(500 ꢀM) with 33 U/mL
β
ꢀglucosidase. Each data point represents mean ± SD (n = 2ꢀ3).
ACS Paragon Plus Environment

Page 13 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In vitro bioactivation studies were complemented by also quantifying the generation of NO
from a oneꢀpot biochemical activation cascade employing carboxylesterase, βꢀgalactosidase and
NOS catalysis (see Supporting Information for details). The galactose prodrugs series was
examined in this setꢀup (i.e., 3, 4, 6, 7). Interestingly, only the deacetylated prodrugs 6 and 7
were able to release NO to a similar extent as 1 (Supporting Information Figure S3). The data are
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
in accordance with the above described findings that
βꢀgalactosidase does not accept
perꢀacetylated carbohydrate prodrugs. Therefore, a rateꢀlimiting bioactivation step for these
prodrugs is the chemical and enzymatic hydrolytic cleavage of the acetyl esters.
Taken together, all tested prodrugs were sufficiently bioactivated by carboxylesterase and
glycosidases with sustained release of 1. The data suggested that glucoseꢀbased prodrug
candidates might be more rapidly and more efficiently bioactivated under in vivo conditions
since βꢀglucosidase appeared to exhibit broader substrate specificity.
Cellular uptake and NO release
After demonstrating sufficient chemical stability of the new Oꢀglycosyl prodrugs of 1 as well
as their bioactivation in vitro, the next step was to evaluate their uptake and bioactivation in
living cells. To address this question, an imageꢀbased cellular assay was devised that quantifies
the inꢀcell generation of NO by automated fluorescence microscopy. We used 4ꢀaminoꢀ5ꢀ
methylaminoꢀ2‘,7‘difluorofluorescein diacetate (DAFꢀFMꢀDA) as a nonꢀfluorescent, membraneꢀ
permeable compound that is activated by cellular esterases to form DAFꢀFM. In the presence of
oxygen, DAFꢀFM reacts with NO to form a fluorescent benzotriazole (DAFꢀFMꢀT)
2
9,30
(
Figure 5a).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 45
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
Several imaging assays have been described in the literature using DAFꢀFMꢀDA as a probe to
detect NO but most of them investigate signal transduction pathways that affect NOS expression.
As a consequence, relatively large changes in NO production are typically detected 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
31,32
quantified.
However, Melisi et al. were able to quantify the intracellular release of NO from
3
1
carboxyl esterꢀgalactose conjugates of Dꢀ and Lꢀarginine. In a similar fashion, we established a
reliable method for quantifying intracellular NO by fluorescence microscopy in mediumꢀ
throughput (i.e., 96ꢀwell plateꢀformat) for best possible comparability between different groups
of treatment. Briefly, murine macrophages (J744A.1 cells) were stimulated with
lipopolysaccharide for 24 hours to induce iNOS expression which ensured sufficient levels of
NOS for later catalysis of the intracellularly released substrate 1 (Figure 5a). Next, cells were
incubated with prodrug candidates for 2.5 hours and then treated with DAFꢀFMꢀDA and
Hoechst 33342 before automated image acquisition (Figure 5b, representative images). No signs
of toxicity were observed at concentrations up to 1 mM as judged by cell morphology (i.e., cell
shape, blebbing, apoptosis) and their proliferative capacity (i.e., cell numbers) compared to
vehicle treated controls. About 1000 cells per well/condition (six images per well) were captured
providing a high validity of the addressed biological question. NOꢀdependent inꢀcell
fluorescence was normalized to total cell number and vehicle control (= DPBS buffer). Another
important control was to verify that the quantified NO signals were due to NOSꢀdependent
bioactivation. Thus, all prodrugs and 1 were also incubated together with the NOS inhibitor
nitroarginine methyl ester (LꢀNAME). As summarized in Figure 5c, coꢀincubation with LꢀNAME
reduced all quantified NO signals which were significant for the galactosyl prodrug 3 and all
prodrugs from the glucosyl series (4, 5, 8, 9).
ACS Paragon Plus Environment

Page 15 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
On a technical note, we were not able to establish a robust assay with a decent dynamic range
in cellular systems that would more authentically recapitulate the pathology to be targeted, i.e.
endothelial dysfunction in endothelial cell lines (HUVEC cells, EA.hy926). 1 (and also arginine)
did not yield sufficient intracellular NO for imageꢀbased quantification under various conditions,
such as increasing NOS expression/activity by transient overexpression, treatment with
acetylcholine, bradykinine, calcium ionophores and/or through tetrahydrobiopterin
supplementation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Imageꢀbased quantification of cellular NO release by Oꢀglycosyl prodrugs of 1; a)
Schematic illustration of prodrug uptake, bioactivation, inꢀcell release of NO and its detection in
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 45
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 form of DAFꢀFMꢀT; b) Representative fluorescence microscopy images of 1ꢀtreated J744A.1
macrophages after Hoechst staining at 377/447 nm (upper panel, DAPI channel) and 472/520 nm
(lower panel, GFP channel), yellow bar = 100 µm; c) Average intracellular fluorescence intensity
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
of prodrugꢀtreated cells, data normalized to cell number and NOꢀrelease by 1 (= 100%), shown
as differences (∆) to buffer control, mean ± SEM. Left: galactosyl conjugates (n = 9), right:
glucosyl conjugates (n = 6). * = significant difference to buffer control with p < 0.05.
Compounds 1-9 and LꢀNAME were applied at a final concentration of 1 mM.
Interestingly, all glucose conjugates showed a superior behavior regarding intracellular NO
release compared to the galactose conjugates (Figure 5c). Only prodrug 3 from the galactosyl
series achieved approx. 75% of the effect of 1 as the “endogenous” physiological source of NO.
In contrast, all prodrugs from the glucosyl series (i.e., 4, 5, 8, 9) achieved a similar level of NO
release compared to 1. For both series there seems to be a tendency of increased NO generation
with higher substituted prodrugs.
Concluding from these results, Oꢀglycosyl prodrugs of choice for in vitro and in vivo
pharmacology applications are derivatives from the glucosyl series. None of these prodrug
candidates are able to generate higher cellular NO levels compared to 1. In general, it might be
difficult to achieve a higher cellular uptake than that driven by amino acid transporters.
Therefore, an excellent outcome of these studies is that all Oꢀglucosyl prodrugs are at least
similarly well absorbed as 1, even with a protected (4, 5) or unprotected sugar moiety (8, 9) and
independent of a free
αꢀamino acid (9) or when the αꢀcarboxy group is esterified (8). For
instance, in the case of 8 and 9 it is possible that glucose transporters served as alternate uptake
routes and eventually compensated for the lack of amino acid transporterꢀmediated cellular
uptake.
ACS Paragon Plus Environment

Page 17 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CONCLUSIONS
Since 1 represents the best known substrate for NOSs it ensures an NO release as close to
physiological conditions as possible, thereby minimizing unwanted biological effects of the
pleiotropic signaling molecule NO. Moreover, 1 exhibits potent arginase inhibiting properties,
thus representing a unique pharmacological profile for the treatment of cardiovascular diseases
that are associated with impaired NO availability. In an attempt to overcome 1’s problematic
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
physicochemical properties,
a
prodrug strategy was devised that stabilized the
Nꢀhydroxyguanidine group via Oꢀglycosylation.
As the preparation of Nꢀcarbamateꢀfree prodrugs was a key requirement, Alloc was employed
as an orthogonal protecting group that allowed efficient preparation of the substituted Nꢀ
hydroxyguanidines and warranted cleavage without compromising the Oꢀglycosylic bond or the
other protecting groups. The newly introduced Oꢀglycosyl prodrugs were exceptionally stable,
even at pH 2. In vitro experiments revealed that glucoseꢀbased prodrugs are more efficiently
bioactivated than their galactose counterparts, mainly taking advantage of the broader substrate
specificity of βꢀglucosidases. Moreover, proofꢀofꢀconcept could be shown in a cellꢀbased system
with murine macrophages employing a mediumꢀthroughput imaging assay. Especially the
glucoseꢀbased prodrugs performed very well and led to cellular NO levels comparable to 1
indicating a likewise efficient cellular uptake and sufficient bioactivation.
In summary, we introduced a novel prodrug concept for Nꢀhydroxyguanidines as exemplified
with the physiological NO precursor 1. The presented approach successfully tackled several
shortcomings that limit widespread therapeutic application of 1 for CVDs that are associated
with endothelial dysfunction. In particular, glucose conjugates (i.e., 4, 5, 8, 9) represent attractive
agents as they are not only chemically stable but also enable a slow, prolonged and sustained
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 45
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
release of 1 while being efficiently taken up by cells. As a function of the degree of substitution,
a more or less pronounced chemical retardation can eventually be achieved depending on the
desired onset of action and/or required routes of administration.
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
Finally, it should be stressed that the described prodrug approach holds promise as a general
strategy for stabilization and chemical retardation of prodrugs for guanidino groupꢀcontaining
drug candidates.
ACS Paragon Plus Environment

Page 19 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
EXPERIMENTAL
General
1
13
15
35
Melting points were uncorrected. H (300 MHz), C (75 MHz), N (30 MHz) and Cl (29
MHz) NMR spectra were recorded on a Bruker Avance III, 300, spectrometer at 298 K equipped
with Bruker Topspin 2.1 Software. Chemical shifts (δ values) were quoted in ppm relative to
TMS as an internal standard, 3ꢀ(trimethylsilyl)ꢀ1ꢀpropanesulfonic acid sodium salt (TPS) as an
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
external standard for experiments in D O or alternatively, relative to the solvent residual signal.
2
Spectra interpretation was performed by first order analysis. Signal assignments were obtained
1
1
13
ꢀ
35
by the help of H, HꢀCOSY and CꢀHSQC and ꢀHMBC spectra. Quantification of Cl by Cl
NMR was verified by a threeꢀpoint calibration with NaCl in D O. Low resolution mass spectra
2
were recorded using a Bruker amazon SL system with LC coupling, electrospray ionization, in
the positive mode. Recordings of exact mass spectra were performed by the Department of
Physical Chemistry, ChristianꢀAlbrechtsꢀUniversity of Kiel, on a 7.05 Tesla Bruker APEX III
FTꢀICR mass spectrometer in ESIꢀpositive mode; substances were dissolved in EtOH and diluted
with a solvent mixture containing MeOH, H O and formic acid (49.9/49.9/0.2) to a concentration
2
of about 100 pmol/µL. Elemental analyses were performed on a CHNS analysator (HEKAtech
GmbH) by the Department of Inorganic Chemistry, ChristianꢀAlbrechtsꢀUniversity of Kiel. IR
spectra were recorded on a Shimadzu IRAffinityꢀ1S FTIR spectrometer equipped with MIRacle
10 Single Reflection ATR Accessory. Reactions were monitored by TLC on precoated silica gel
plates (SiO , 60, F ). All compounds could be detected by either UV detection or by ninhydrine
2
254
spray or Ce(IV)ꢀsulfate staining and heating to 120 °C. Purification of synthesized compounds
was carried out on a CombiFlash Rf, version 1.8.2, flashꢀchromatography apparatus using
Interchim PFꢀ30SIHPꢀJP/12G or 40G or PFꢀ15SIHP/12G silica gel columns. Reverse phase
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 45
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
chromatography was performed by column chromatography using silica gel 60 silanized (0.063ꢀ
0.200 mm, Merck). All starting materials were commercially available and used without further
α
purification. N ꢀ(tertꢀButyloxycarbonyl)ꢀ
Lꢀornithine (CAS Registry Number 21887ꢀ64ꢀ9) 10 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
α
N ꢀ(tertꢀButyloxycarbonyl)ꢀ
Lꢀornithineꢀtertꢀbutylester (CAS Registry Number 214629ꢀ97ꢀ7) 15
was purchased from Bachem. All solvents were distilled and dried according to standard
procedures.
Chemical synthesis
α
δ
17
α
N -(tert-Butyloxycarbonyl)-N -phthalimido-
L
-ornithine
(11).
N ꢀ(tertꢀ
Butyloxycarbonyl)ꢀ ꢀornithine 10 (1.9 g, 8.4 mmol) and phthalic anhydride (2.2 g, 15 mmol)
L
were suspended in 20 mL of chloroform and 150 mL of tetrachloroethylene. This suspension was
heated for 2 h at 60 °C and additional 6 h at 100 °C. The clear reaction mixture was concentrated
in vacuo to afford the crude product. Purification was carried out by flash column
chromatography (silica gel, CH Cl ꢀMeOH, 0ꢀ5%). Yield: 1.9 g (61%) of a white crystalline
2
2
solid. mp. 126 °C. TLC: R = 0.54 (silica gel, CH Cl /MeOH, 9:1). The spectroscopic data were
f
2
2
1
in agreement with those reported. H NMR (300 MHz, CDCl ):
δ/ppm = 1.43 (s, 9 H, C(CH3)3),
3
3
1.65ꢀ2.00 (m, 4H,
β
,
γ
ꢀCH ), 3.72 (t, J = 6.8 Hz, 2H,
δ
ꢀCH ), 4.14ꢀ4.37 (m, 1H,
αꢀCH ), 5.16 (d,
2
2
3
J = 8.2 Hz, 1H, NH), 7.68ꢀ7.75 (m, 2H, ArH), 7.80ꢀ7.86 (m, 2H, ArH), 9.36 (br s, 1H, OH).
1
3
C NMR (75 MHz, CDCl3):
δ/ppm = 37.4 (δꢀCH ), 29.8 (βꢀCH ), 28.3 (C(CH ) ), 24.8 (γꢀ
2 2 3 3
CH ), 53.0 (
α
ꢀCH), 80.3 (C(CH ) ), 123.3 (2 × ArCH), 132.0 (2 × ArC), 134.0 (2 × ArCH),
2
3 3
ꢀ1
155.6 (COꢀBoc), 168.4 (2 × COꢀPht), 176.5 (COOH). IR (ATR):ꢀ ꢁꢂ /cm = 2976, 1771, 1697,
+
1
514, 1396, 1366, 1159, 1047, 718. MS (ESI): m/z = 725 [2 × M + H] , 625 [2 × M – C H –
4
8
ACS Paragon Plus Environment

Page 21 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
+
+
+
CO + H] , 363 [M + H] , 263 [M – C H – CO + H] . Anal. calcd. for C H N O (362.38): C
2
4
8
2
18 22
2
6
5
9.66, H 6.12, N 7.73; found: C 59.21, H 6.15, N 7.58.
α
δ
N -(tert-Butyloxycarbonyl)-N -phthalimido-L-ornithine ethyl ester (12). 2.2 g (6.0 mmol)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
of the protected ornithine 11 was dissolved in 30 mL of dry DCM. To this solution 0.3 g DMAP
(2.4 mmol, 40 molꢀ%), 1.8 mL of dry EtOH (30 mmol) and 1.2 g (6.3 mmol) EDCI were added.
The reaction mixture was stirred magnetically for 24 h at r.t. and subsequently diluted with 25
mL of DCM and washed with 25 mL of 1% HCl, water and brine, each. The organic phase was
dried with Na SO and concentrated in vacuo. Further purification was performed by flash
2
4
column chromatography (silica gel, cyclohexaneꢀEtAc, 0ꢀ30%). Yield: 1.3 g (54%) of a white
1
solid. mp. 82 °C. TLC: R = 0.40 (cyclohexane/EtAc, 2:1). H NMR (300 MHz, CDCl ):
δ /ppm
f
3
3
=
1.26 (t, J = 7.2 Hz, 3H, CH ꢀCH ), 1.43 (s, 9H, C(CH ) ), 1.62ꢀ1.94 (m, 4H, β, γꢀCH ), 3.71
2
3
3 3
2
3
3
(t, J = 6.8 Hz, 2H,
δ
ꢀCH ), 4.18 (q, J = 7.2 Hz, 2H, CH ꢀCH ), 4.24ꢀ4.35 (m, 1H,
αꢀCH ), 5.06 (
2
2
3
3
13
br d, J = 8.1 Hz, 1H, NH), 7.68ꢀ7.75 (m, 2H, ArH), 7.81ꢀ7.87 (m, 2H, ArH). C NMR (75
MHz, CDCl3):
δ/ppm = 14.2 (CH ꢀCH ), 24.7 (γꢀCH ), 28.3 (C(CH ) ), 30.2 (βꢀCH ), 37.5 (δꢀ
2 3 2 3 3 2
CH ), 53.2 ( ꢀCH), 61.4 (CH ꢀCH ), 79.9 (C(CH ) ), 123.3 (2 × ArCH), 132.1 (2 × ArC), 134.0
α
2
2
3
3 3
(2 × ArCH), 155.4 (COꢀBoc), 168.4 (2 × COꢀPht), 172.5 (COOEt). MS (ESI): m/z = 413 [M +
+
+
Na] , 291 [M – C H – CO + H] . Anal. Calcd for C H N O (390.44): C 61.53, H 6.71, N
4
8
2
20 26
2
6
7.17. Found: C 61.73, H 7.32, N 6.94.
α
ω
N -(tert-butyloxycarbonyl)-N -allyloxycarbonyl-L-thiocitrulline tert-butyl ester (14a).
α
2
.3 g (7.0 mmol) of N ꢀBocꢀ
Lꢀornithine tertꢀbutylester hydrochloride 15 were dissolved in 250
mL of dry DCM. The solution was cooled to 0 °C, 1.1 mL (8.0 mmol) TEA were added and 40
mL of a 0.5 M solution (in dry DCM) of 13 (20 mmol) dropwise over 30 min. The reaction
mixture was stirred for 2 h while the solution was being warmed up to r.t. The mixture was
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 45
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
concentrated in a vacuum to approx. one third of the original volume and washed with 30 mL of
1
% HCl, water and brine, each. The organic phase was dried with Na SO and concentrated in a
2
4
1
vacuum. The thiourea 14a was at this point already pure by > 96% ( H NMR) and further
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
purified by flash column chromatography (silica gel, cyclohexane/EtAc, 0ꢀ30%). Yield: 2.9 g of
1
a colourless oil (94%). TLC: R = 0.44 (cyclohexane/EtAc, 3:1). H NMR (300 MHz, CDCl ):
f
3
δ
/ppm = 1.47, 1.50 (2 × s, 2 × 9H, 2 × C(CH ) ), 1.65ꢀ1.76 (m, 4H,
β
,
γ
ꢀCH ), 3.67 (pseudo q,
3
3
2
3
3
2H,
δ
ꢀCH ), 4.14ꢀ4.27 (m, 1H,
α
ꢀCH), 4.65 (dt, J = 5.7, 1.4 Hz, 2H, CH =CHꢀCH ), 5.10 (d, J
2
2
2
3
=
7.9 Hz, 1H, NHꢀBoc), 5.29ꢀ5.40 (m, 2H, CH =CHꢀCH ), 5.90 (ddt, J = 17.2, 10.4, 5.7 Hz, 1H,
2 2
1
3
CH=CH ), 8.11 (br s, 1H, NH), 9.66 (br s, 1H, NHꢀCH ). C NMR (75 MHz, CDCl ):
δ/ppm =
2
2
3
24.9 (
γ
ꢀCH ), 28.0, 28.3 (2 × C(CH ) ), 30.3 (
β
ꢀCH ), 45.1 (
δ
ꢀCH ), 53.5 (
αꢀCH), 67.0 (OꢀCH2),
2
3 3
2
2
7
9.8, 82.2 (2 × C(CH ) ), 119.5 (CH=CH ), 130.9 (CH=CH ), 152.4 (COꢀAlloc), 155.3 (COꢀ
3
3
2
2
t
ꢀ1
Boc), 171.5 (COO Bu), 179.2 (C=S). IR (ATR):ꢀ ꢁꢂ /cm = 3292, 2976, 1711, 1520, 1366, 1219,
+
+
+
1
152, 1028, 773. MS (ESI): m/z = 454 [M + Na] , 432 [M + H] , 376 [M – 2 × C H + H] , 276
4 8
+
[
M – 2 × C H −CO + H] . Anal. Calcd for C H N O S · 2 H O (467.59): C 48.81, H 7.98, N
4
8
2
19 33
3
6
2
8.99. Found: C 49.03, H 7.90, N 8.59.
α
ω
N -(tert-butyloxycarbonyl)-N -allyloxycarbonyl-
L
-thiocitrulline ethyl ester (14b). To a
stirred solution of 541 mg (1.4 mmol) of the protected ornithine 12 in 20 mL of a mixture of
DCM/EtOH 1:1 was added 1 mL (20 mmol, 15 eq.) hydrazine hydrate (98%) . After 2.5 h
precipitated phthalhydrazide was filtered off and the residue was washed three times with 5 mL
of DCM. The filtrate was mixed with 15 mL of saturated NaHCO solution and shaken. The
3
phases were separated, and the aqueous phase was extracted twice with 10 mL of DCM. Finally,
the pooled organic phases were washed with 15 mL of brine, dried with Na SO and chilled on
2
4
ice to 0 °C. To this preꢀcooled solution was added dropwise during about 30 min 5 mL of a 0.5
ACS Paragon Plus Environment

Page 23 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
M stock solution of 13 (in DCM). The reaction mixture was stirred for additional 2 h at r.t. The
solvent was removed in vacuo and the greasy product purified by flash column chromatography
(silica gel, cyclohexane/EtAc, 0ꢀ25%). Yield: 224 mg of a colourless oil (40%). TLC: R = 0.33
f
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
3
(
cyclohexane/EtAc, 3:1). H NMR (300 MHz, CDCl ):
δ
/ppm = 1.29 (t, J = 7.2 Hz, 3H,
3
CH CH ), 1.45 (s, 9H, C(CH ) ), 1.63ꢀ1.96 (m, 4H, β, γꢀCH ), 3.68 (pseudo q, 2H, δꢀCH ), 4.21
2 2
2
3
3 3
3
3
4
(
q, J = 7.2 Hz, 2H, CH CH ), 4.26ꢀ4.38 (m, 1H, αꢀCH), 4.65 (ddd, J = 5.8, J = 1.7, 1.4 Hz, 2H,
2 3
3
2
3
4
CH =CHꢀCH ), 5.10 (d, J = 8.1 Hz, 1H, NHꢀBoc), 5.31 (ddd, J = 2.7, J = 10.5, J = 1.2 Hz,
2
2
cis
2
3
4
1
H, CH =CHꢀCH ), 5.37 (ddd, J = 3.4, J
= 17.2, J = 1.4 Hz, 1H, CH =CHꢀCH ) 5.90 (ddt,
trans 2 2
2
2
3
13
J = 17.2, 10.5, 5.7 Hz, 1H, CH=CH ), 8.25 (br s, 1H, NH), 9.67 (br s, 1H, NHꢀCH ). C NMR
2
2
(75 MHz, CDCl3):
δ
/ppm = 14.2 (CH CH ), 24.3 (
γ
ꢀCH ), 28.3 (C(CH ) ), 30. (
β
ꢀCH ), 45.0 (
δꢀ
2
3
2
3 3
2
CH ), 53.1 (
α
ꢀCH), 61.5 (CH CH ), 67.0 (OꢀCH ), 80.0 (2 × C(CH ) ), 119.5 (CH=CH ), 130.9
2
2 3 2 3 3 2
ꢀ1
(
CH=CH ), 152.5 (COꢀAlloc), 155.4 (COꢀBoc), 172.4 (COOEt), 179.2 (C=S). IR (ATR):ꢀ ꢁꢂ /cm
2
=
3364, 3173, 2980, 1746, 1715, 1545, 1520, 1242, 1201, 1155, 1030, 671. MS (ESI): m/z = 426
+
+
+
+
[M + Na] , 404 [M + H] , 348 [M – C H + H] , 304 [M – C H − CO + H] . Anal. Calcd for
4
8
4
8
2
C H N O S · 1.2 H O (425.12): C 48.03, H 7.45, N 9.88 S 7.54. Found: C 47.94, H 6.60, N
17
29
3
6
2
9
.70, S 7.56.
N-Allyloxycarbonylisothiocyanate (13) Compound 13 was prepared as described previously
19,20
with allyl chloroformate (50 mmol) as reagent.
Yield: 5.9 g of a bright yellow oil (83%).
1
3
TLC: R = 0.83 (cyclohexane/EtAc, 4:1). H NMR (300 MHz, CDCl ):
δ
/ppm = 4.67 (ddd, J =
f
3
4
2
3
4
5
.9, J =1.5, 1.1 Hz, 2H, CH =CHꢀCH ), 5.34 (ddd, J = 2.5 Hz, J = 10.4 Hz, J = 1.1 Hz, 1H,
2
2
cis
2
3
4
CH =CHꢀCH ), 5.40 (ddd, J = 3.0 Hz, J
= 17.1 Hz, J = 1.4 Hz, 1H, CH =CHꢀCH ), 5.93
2 2
2
2
trans
3
13
(
ddt, J = 17.2, 10.4, 5.7 Hz, 1H, CH=CH ). C NMR (75 MHz, CDCl ):
δ
/ppm = 69.4 (OꢀCH2),
2
3
1
20.2 (CH=CH ), 130.3 (CH=CH ), 147.1 (N=C=S), 149.9 (C=O).
2
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 45
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
-Aminooxy-2,3,4,6-tetra-O-acetyl-
β
-
D
-galactopyranose (16, Gal) Galactose derivative 16
1
2,22
was prepared as described previously.
Yield (6 mmol batch): 2.1 g of a white foamy solid
1
(
96%; lit.: 67%). mp. 50 °C. TLC: R = 0.39 (cyclohexane/EtAc, 3:1). H NMR (300 MHz,
f
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
CDCl3):
δ/ppm = 1.99, 2.06, 2.08, 2.16 (4 × s, 4 × 3H, 4 × COCH ), 3.98 (td, J = 6.7, 1.2 Hz,
3
3
3
1
H, 5'ꢀCH), 4.12ꢀ4.23 (m, 2H, 6'ꢀCH ), 4.69 (d, J = 8.3 Hz, 1H, 1'ꢀCH), 5.05 (dd, J = 10.4, 3.5
2
3
3
Hz, 1H, 3'ꢀCH ), 5.26 (dd, J = 10.4, 8.3 Hz, 1H, 2'ꢀCH), 5.40 (dd, J = 3.5, 1.1 Hz, 1H, 4'ꢀCH),
2
13
5.83 (br s, 2H, NH ). C NMR (75 MHz, CDCl ):
δ/ppm = 20.5, 20.6, 20.7, 20.8 (4 × COCH3),
2
3
6
1
1.0 (6'ꢀCH ), 67.0 (4'ꢀCH), 67.3 (2'ꢀCH), 70.7 (5'ꢀCH), 71.0 (3'ꢀCH), 103.9 (1'ꢀCH), 169.7,
2
+
+
70.1, 170.2, 170.4 (4 × COCH ). MS (ESI): m/z = 364 [M + H) , 331 [C H O ] .
3
14 19
9
1-Aminooxy-2,3,4,6-tetra-O-acetyl-β-D-glucopyranose (16, Glc) Glucose derivative 16 was
2
1,22
prepared as described previously.
Yield (3.8 mmol batch): 1.2 g of a white solid (88%). mp.
1
1
23 °C. TLC: R = 0.83 (cyclohexane/EtAc, 4:1). H NMR (300 MHz, CDCl ):
δ/ppm = 2.01,
f
3
3
2
.03, 2.07, 2.10 (4 × s, 4 × 3H, 4 × COCH ), 3.75 (ddd, J = 10.0, 4.5, 2.3 Hz, 1H, 5'ꢀCH), 4.17
3
2
3
2
3
3
(dd, J = 12.2, J = 2.4 Hz, 1H, 6'aꢀCH ), 4.30 (dd, J= 12.2, J = 4.5 Hz, 1H, 6'bꢀCH ),4.72 (d, J
2
2
3
3
=
8.3 Hz, 1H, 1'ꢀCH), 5.07 (dd, J = 9.5, 2.1 Hz, 1H, 4'ꢀCH), 5.10 (dd, J = 9.5, 3.4 Hz, 1H, 2'ꢀ
3
13
CH), 5.23 (pseudo t, J = 9.4 Hz, 1H, 3'ꢀCH), 5.84 (br s, 2H, NH ). C NMR (75 MHz, CDCl ):
2
3
δ
/ppm = 20.6, 20.7, 20.8 (4 × COCH ), 61.8 (6'ꢀCH ), 68.2 (4'ꢀCH), 69.6 (2'ꢀCH), 71.8 (5'ꢀCH),
3 2
7
2.9 (3'ꢀCH), 103.4 (1'ꢀCH), 169.4, 169.5, 170.2, 170.7 (4 × COCH ). MS (ESI): m/z = 364 [M +
3
+
+
H) , 331 [C H O ] .
1
4
19
9
General procedure for the preparation of fully protected NOHA prodrug precursors
(
17a-b, 18a-b). 3 mmol of thiourea 14a-b, 3.4 mmol of 16 (Gal or Glc) and 3.4 mmol DIPEA
were dissolved in 30 mL of dry DCM. The batch was cooled to 0 °C, 3.4 mmol EDCI were
added and the mixture magnetically stirred for 72 h at r.t. The solution was diluted with approx.
ACS Paragon Plus Environment

Page 25 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
0 mL of DCM and washed with 30 mL of 1% HCl, water and brine, each. The organic phase
was dried with Na SO and evaporated to dryness. Purification of the crude products was carried
2
4
out by flash column chromatography (silica gel, cyclohexane/EtAc, 0ꢀ30%).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
α
ω
ω'
N -(tert-Butyloxycarbonyl)-N -allyloxycarbonyl-N -(2,3,4,6-tetra-O-acetyl-
β
-D
-galactopyranos-1-yloxy)-
L
-arginine tert-butyl ester (17a). Yield (3 mmol batch): 1.74 g of
1
a white foamy solid (76%). mp. 78 °C. TLC: R = 0.49 (cyclohexane/EtAc, 1:1). H NMR (300
f
MHz, CDCl3): δ/ppm =1.44, 1.47 (2 × s, 2 × 9H, 2 × C(CH ) ), 1.54ꢀ1.90 (m, 4H, β, γꢀCH2),
3 3
3
2
.00, 2.04, 2.07, 2.16 (4 × s, 4 × 3H, 4 × COCH ), 3.10 (pseudo q, 2H,
δ
ꢀCH ), 3.98 (td, J = 6.9,
3
2
3
1.0 Hz, 1H, 5'ꢀCH), 4.12ꢀ4.23 (m, 3H, 6'ꢀCH2,
α
ꢀCH ), 4.63 (dt, J = 5.6, 1.4 Hz, 2H, CH =CHꢀ
2
3
CH ), 4.87 (d, J = 8.4 Hz, 1H, 1'ꢀCH), 5.05ꢀ5.12 (m, 2H, 3'ꢀCH , NHꢀBoc), 5.23ꢀ5.43 (m, 4H,
2
2
3
3
2', 4'ꢀCH, CH =CHꢀCH ), 5.93 (ddt, J = 17.2, 10.4, 5.7 Hz, 1H, CH=CH ), 6.45 (br t, J = 5.2,
2
2
2
1
3
1
H, NHꢀCH ), 7.76 (br s, 1H, NH). C NMR (75 MHz, CDCl ):
δ/ppm = 20.66, 20.69, 20.7,
2
3
2
0.9 (4 × COCH ), 24.8 (δꢀCH ), 28.0, 28.3 (2 × C(CH ) ), 30.3 (βꢀCH ), 40.5 (δꢀCH ), 53.7 (αꢀ
2 3 3 2 2
3
CH), 61.0 (6'ꢀCH ), 66.5 (OꢀCH ), 66.9, 68.0, 70.5, 70.9 (4', 2', 5', 3'ꢀCH), 79.7, 81.9 (2 ×
2
2
C(CH ) ), 102.5 (1'ꢀCH), 118.9 (CH=CH ), 131.5 (CH=CH ), 150.3 (C=N), 152.8 (COꢀAlloc),
3
3
2
2
t
155.4 (COꢀBoc), 170.07, 170.13, 170.3, 170.25, 170.34 (4 × COCH ), 171.7 (COO Bu). MS
3
+
+
(
ESI): m/z = 783 [M + Na] , 761 [M + H] . IR (ATR): 3450, 2980, 1732, 1645, 1366, 1215,
ꢀ1
1
151, 1043, 772 cm . Anal. Calcd for C H N O (760.79): C 52.80, H 6.86, N 7.24. Found: C
33 52 4 16
52.34, H 6.67, N 7.28.
α
ω
ω'
N -(tert-Butyloxycarbonyl)-N -allyloxycarbonyl-N -(2,3,4,6-tetra-O-acetyl-
β
-D
-glucopyranos-1-yloxy)-
L
-arginine tert-butyl ester (17b). Yield (0.9 mmol batch): 548 mg
1
of a white foamy solid (81%). TLC: R = 0.42 (cyclohexane/EtAc, 1:1). H NMR (300 MHz,
f
3
CDCl3):
δ
/ppm = 1.30 (t, J = 7.2 Hz, 3H, CH CH ), 1.44, 1.46 (2 × s, 2 × 9H, 2 × C(CH ) ),
2
3
3 3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 45
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
.55ꢀ1.90 (m, 4H,
2H,
CH CH ), 4.33 (dd, J = 12.5, J = 4.3 Hz, 1H, 6'bꢀCH ), 4.89 (d, J = 8.2 Hz, 1H, 1'ꢀCH), 5.02ꢀ
β
,
γ
ꢀCH ), 2.02, 2.03, 2.08, 2.09 (4 × s, 4 × 3H, 4 × COCH ), 3.02ꢀ3.16 (m,
2
3
3
δ
ꢀCH ), 3.77 (ddd, J = 10.3, 4.2, 2.3 Hz, 1H, 5'ꢀCH), 4.04ꢀ4.23 (m, 4H, 6'aꢀCH , αꢀCH,
2
2
2
3
3
2
3
2
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
3
5
1
2
.18 (m, 3H, 2', 4'ꢀCH, NHꢀBoc), 5.25 (dd, J = 9.4, 7.9 Hz, 1H, 3'ꢀCH), 6.50 (br t, J = 5.1 Hz,
1
3
H, NHꢀCH ), 7.63 (br s, 1H, NH). C NMR (75 MHz, CDCl ): δ/ppm = 14.3 (CH CH ), 20.61,
2 3
2
3
0.63, 20.7, 20.8, (4 × COCH ), 24.9 (γꢀCH ), 28.0, 28.3 (2 × C(CH ) ), 30.3 (βꢀCH ), 40.5 (δ ꢀ
2 3 3 2
3
CH ), 53.7 (αꢀCH), 61.8 (6'ꢀCH ), 62.2 (CH CH ), 68.3, 70.2, 71.7, 72.9 (4', 2', 5', 3'ꢀCH), 79.3,
2 2 3
2
81.9 (2 × C(CH ) ), 102.0 (1'ꢀCH), 150.7 (C=N), 153.1 (COꢀEoc), 155.4 (COꢀBoc), 169.5, 169.8,
3 3
t
+
+
170.2, 170.7 (4 × COCH ), 171.7 (COO Bu). MS (ESI): m/z = 771 [M + Na] , 749 [M + H] . IR
3
ꢀ1
(ATR):ꢀ ꢁꢂ /cm = 1740, 1717, 1645, 1506, 1464, 1366, 1213, 1151, 1034, 905, 773. Anal. Calcd
for C H N O (748.78): C 51.33, H 7.00, N 7.48. Found: C 51.35, H 7.26, N 7.22.
3
2
52
4
16
α
ω
ω'
N -(tert-Butyloxycarbonyl)-N -allyloxycarbonyl-N -(2,3,4,6-tetra-O-acetyl-
β
-D
-galactopyranos-1-yloxy)-
L
-arginine ethyl ester (18a). Yield (1.0 mmol batch): 653 mg of
1
a white foamy solid (93%). mp. 65 °C. TLC: R = 0.31 (cyclohexane/EtAc, 1:1). H NMR (300
f
3
MHz, CDCl3):
δ
/ppm = 1.28 (t, J = 7.2 Hz, 3H, CH CH ), 1.45 (s, 9H, C(CH ) ), 1.59ꢀ1.94 (m,
2
3
3 3
4
3
5
H,
β
,
γ
ꢀCH ), 2.00, 2.04, 2.11, 2.16 (4 × s, 4 × 3H, 4 × COCH ), 3.10 (pseudo q, 2H,
δꢀCH2),
2
3
3
3
.99 (td, J = 6.9, 1.0 Hz, 1H, 5'ꢀCH), 4.07ꢀ4.34 (m, 5H, 6'ꢀCH ,
α
ꢀCH, CH CH ), 4.66 (dt, J =
2
2
3
3
.6, 1.4 Hz, 2H, CH =CHꢀCH ), 4.87 (d, J = 8.2 Hz, 1H, 1'ꢀCH), 5.03ꢀ5.13 (m, 2H, 3'ꢀCH, NHꢀ
2
2
3
Boc), 5.22ꢀ5.43 (m, 4H, 2', 4'ꢀCH, CH =CHꢀCH ), 5.92 (ddt, J = 17.2, 10.4, 5.7 Hz, 1H,
2
2
3
13
CH=CH ), 6.44 (br t, J = 5.2, 1H, NHꢀCH ), 7.76 (br s, 1H, NH). C NMR (75 MHz, CDCl ):
2
2
3
δ/ppm =14.2 (CH CH ), 20.57, 20.64, 20.67, 20.94 (4 × COCH ), 25.0 (
γ
ꢀCH ), 28.3 (C(CH ) ),
2
3
3
2
3 3
30.0 (
β
ꢀCH ), 41.1 (
δ
ꢀCH ), 53.2 (
α
ꢀCH), 60.9 (6'ꢀCH ), 61.4 (CH CH ), 66.8 (4'ꢀCH), 67.0 (Oꢀ
2
2
2
2
3
CH ), 67.6, 68.4, 70.7 (2', 5', 3'ꢀCH), 79.7 (C(CH ) ), 103.1 (1'ꢀCH), 119.4 (CH=CH ), 131.5
2
3 3
2
ACS Paragon Plus Environment

Page 27 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
CH=CH ), 150.1 (C=N), 152.8 (COꢀAlloc), 155.4 (COꢀBoc), 169.96, 170.14, 170.26, 170.31 (4
2
+
+
ꢀ1
×
COCH ), 171.5 (COOEt). MS (ESI): m/z = 755 [M + Na] , 733 [M + H] . IR (ATR):ꢀ ꢁꢂ /cm =
3
3
400, 1744, 1651, 1506, 1367, 1215, 1161, 1047, 910, 772. Anal. Calcd for C H N O · 1.6
31 48 4 16
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
H O (761.57): C 48.89, H 6.78, N 7.36. Found: C 48.64, H 6.98, N 6.88.
2
α
ω
ω'
N -(tert-Butyloxycarbonyl)-N -allyloxycarbonyl-N -(2,3,4,6-tetra-O-acetyl-
β
-D
-glucopyranos-1-yloxy)-
L
-arginine ethyl ester (18b). Yield (0.56 mmol batch): 300 mg of a
1
white foamy solid (73%). TLC: R = 0.31 (cyclohexane/EtAc, 1:1). H NMR (300 MHz, CDCl ):
f
3
3
δ
/ppm = 1.28 (t, J = 7.2 Hz, 3H, CH CH ), 1.45 (s, 9H, C(CH ) ), 1.55ꢀ1.91 (m, 4H,
β
,
γ
ꢀCH2),
2
3
3 3
3
2.02, 2.03, 2.06, 2.07 (4 × s, 4 × 3H, 4 × COCH ), 3.05ꢀ3.15 (m, 2H,
δ
ꢀCH ), 3.77 (ddd, J =
3
2
3
10.2, 4.5, 2.3 Hz, 1H, 5'ꢀCH), 4.10ꢀ4.36 (m, 5H, 6'ꢀCH2,
α
ꢀCH, CH CH ), 4.62 (dt, J = 5.7, 1.7
2
3
3
Hz, 2H, CH =CHꢀCH ), 4.89 (d, J = 8.2 Hz, 1H, 1'ꢀCH), 5.06ꢀ5.16 (m, 3H, 2', 4'ꢀCH, NHꢀBoc),
2
2
3
5
.22ꢀ5.41 (m, 3H, 3'ꢀCH, CH =CHꢀCH ), 5.92 (ddt, J = 17.2, 10.4, 5.7 Hz, 1H, CH=CH ), 6.45
2
2
2
3
13
(
br t, J = 5.3, 1H, NHꢀCH ), 7.72 (br s, 1H, NH). C NMR (75 MHz, CDCl ):
δ
/ppm = 14.2
2
3
(
CH CH ), 20.61, 20.63, 20.7, 20.9, (4 × COCH ), 25.0 (
γ
ꢀCH ), 28.3 (C(CH ) ), 30.1 (
β
ꢀCH2),
2
3
3
2
3 3
4
0.4 (δꢀCH ), 53.3 (αꢀCH), 61.8 (CH CH ), 61.9 (6'ꢀCH ), 66.6 (OꢀCH ), 68.3, 70.2, 71.7, 72.8
2 2 3 2 2
(4', 2', 5', 3'ꢀCH), 79.9 (C(CH ) ), 102.0 (1'ꢀCH), 119.0 (CH=CH ), 131.4 (CH=CH ), 150.5
3 3 2 2
(C=N), 152.8 (COꢀAlloc), 155.4 (COꢀBoc), 169.5, 169.9, 170.2, 170.7 (4 × COCH ), 172.6
3
+
+
ꢀ1
(COOEt). MS (ESI): m/z = 755 [M + Na] , 733 [M + H] . IR (ATR):ꢀ ꢁꢂ /cm = 3450, 1738, 1717,
1
645, 1456, 1366, 1213, 1161, 1034, 908, 772. Anal. Calcd for C H N O (732.74): C 50.82,
31 48 4 16
H 6.60, N 7.65. Found: C 50.78, H 6.74, N 8.08.
General procedure for the preparation of carbamate-free NOHA prodrug precursors
(19a-b, 20a-b). 1.0 mmol of the completely protected compound 17a-b or 18a-b, 8 mmol (8
equiv.) dimedone and 0.1 mmol (10 molꢀ%) Pd(0)[Ph P] were dissolved in 20 mL of THF and
3
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 45
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 reaction mixture was stirred for 30 min at r.t. The solvent was removed under reduced
pressure and the solid residue dissolved in 30 mL of EtAc. Solid dimedone was filtered off and
the organic phase was washed three times with 10 mL of diluted aqueous NH ꢀsolution (3%),
3
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
once with 10 mL of water and brine. The organic phase was dried with Na SO and concentrated
2
4
in a vacuum. The crude product was further purified by flash column chromatography (silica gel,
DCM/MeOH, 0ꢀ5% for about 25 min).
α
ω
N -(tert-Butyloxycarbonyl)-N -(2,3,4,6-tetra-O-acetyl-β-D-galactopyranos-1-yloxy)-L-
arginine tert-butyl ester (19a). Yield: 588 mg of a light yellow solid (90%). mp. 95 °C. TLC:
1
R = 0.48 (DCM/MeOH, 95:5). H NMR (300 MHz, CDCl ):
δ/ppm =1.44, 1.46 (2 × s, 2 × 9H, 2
f
3
×
C(CH ) ), 1.53ꢀ1.83 (m, 4H,
β
,
γ
ꢀCH ), 1.99, 2.03, 2.07, 2.18 (4 × s, 4 × 3H, 4 × COCH ), 2.98ꢀ
3
3
2
3
3
3
.06 (m, 2H,
m, 3H, 6'ꢀCH2, αꢀCH), 4.33 (br s, 2H, NH ), 4.87 (d, J = 8.3 Hz, 1H, 1'ꢀCH), 5.03ꢀ5.30 (m, 3H,
2
δꢀCH ), 3.59 (br s,1H, NHꢀCH ), 3.98 (td, J = 6.9, 1.0 Hz, 1H, 5'ꢀCH), 4.10ꢀ4.24
2 2
3
(
3
13
2
', 3'ꢀCH, NHꢀBoc), 5.40 (dd, J = 3.5, 1.0 Hz, 1H, 4'ꢀCH). C NMR (75 MHz, CDCl ):
δ/ppm =
3
2
0.6, 20.7, 21.0, (4 × COCH ), 25.4 (
γ
ꢀCH ), 28.0, 28.4 (2 × C(CH ) ), 30.8 (
β
ꢀCH ), 41.5 (δ ꢀ
3
2
3 3
2
CH ), 53.4 (αꢀCH), 61.0 (6'ꢀCH ), 66.9, 68.1, 70.4, 71.0 (4', 2', 5', 3'ꢀCH), 79.7, 82.1 (2 ×
2
2
C(CH ) ), 102.5 (1'ꢀCH), 155.6 (C=N), 156.6 (COꢀBoc), 170.1, 170.3, 170.4 (4 × COCH ), 171.7
3
3
3
t
15
(
COO Bu). N NMR (30 MHz, 25 °C, 0.2% MeNO in CDCl as external standard):
2
3
δ
/ppm = ꢀ328.4 (NH ), ꢀ321.2 (NHꢀCH ), ꢀ294.8 (NHꢀBoc), the nitrogen of the oxime moiety
2 2
ꢀ1
was not observed. IR (ATR):ꢀ ꢁꢂ /cm = 3450, 2950, 1744, 1714, 1645, 1516, 1366, 1217, 1152,
+
+
1
043, 908. MS (ESI): m/z = 677 [M + H] . HRMS (ESI) m/z for C H N O [M + H] :
29 48 4 14
677.3239, found: 677.3219.
α
ω
N -(tert-Butyloxycarbonyl)-N -(2,3,4,6-tetra-O-acetyl-
β
-D
-glucopyranos-1-yloxy)-
L
-
arginine tert-butyl ester (19b). Yield: 582 mg of a light yellow solid (89%). mp. 94 °C. TLC:
ACS Paragon Plus Environment

Page 29 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
R = 0.31 (DCM/MeOH, 95:5). H NMR (300 MHz, CDCl ):
δ
/ppm = 1.44, 1.47 (2 × s, 2 × 9H,
f
3
2 × C(CH ) ), 1.51ꢀ1.87 (m, 4H, β, γꢀCH ), 2.01, 2.02, 2.06, 2.07 (4 × s, 4 × 3H, 4 × COCH ),
3
3
2
3
3
2.96ꢀ3.23 (m, 2H,
δꢀCH ), 3.70 (br s,1H, NHꢀCH ), 3.77 (ddd, J = 10.1, 4.2, 2.4 Hz, 1H, 5'ꢀCH),
2 2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
4.07ꢀ4.22 (m, 2H, 6'aꢀCH2,
α
ꢀCH), 4.27ꢀ4.45 (m, 3H, 6'bꢀCH , NH ), 4.88 (d, J = 8.6 Hz, 1H,
2 2
1
3
1
'ꢀCH), 5.03ꢀ5.29 (m, 4H, 2', 3', 4'ꢀCH, NHꢀBoc). C NMR (75 MHz, CDCl ):
δ/ppm = 20.62,
3
2
0.65, 20.8, 20.9 (4 × COCH ), 25.6 (γꢀCH ), 28.0, 28.4 (2 × C(CH ) ), 30.7 (
β
ꢀCH ), 41.2 (δꢀ
3
2
3 3
2
CH ), 53.0 (
α
ꢀCH), 61.8 (6'ꢀCH ), 68.3, 70.3, 71.6, 72.9 (4', 2', 5', 3'ꢀCH), 80.0, 82.2 (2 ×
2
2
C(CH ) ), 102.1 (1'ꢀCH), 155.7 (C=N), 156.8 (COꢀBoc), 169.5, 169.9, 170.2, 170.8 (4 ×
3
3
t
ꢀ1
COCH ), 171.7 (COO Bu). IR (ATR):ꢀ ꢁꢂ /cm = 1744, 1714, 1643, 1639, 1520, 1366, 1215, 1161,
3
+
1112, 1096, 1034, 908. MS (ESI): m/z = 677 [M + H] . HRMS (ESI) m/z for C H N O [M +
2
9
48
4
14
+
H] : 677.3239, found: 677.3223.
α
ω
N -(tert-Butyloxycarbonyl)-N -(2,3,4,6-tetra-O-acetyl-
β
-D
-galactopyranos-1-yloxy)-
L
-
arginine ethyl ester (20a). Yield (0.40 mmol batch): 220 mg of a light yellow solid (73%).
1
3
TLC: R = 0.55 (DCM/MeOH, 95:5). H NMR (300 MHz, CDCl ):
δ
/ppm = 1.29 (t, J = 7.2 Hz,
f
3
3H, CH CH ), 1.44 (s, 9H, C(CH ) ), 1.56ꢀ1.91 (m, 4H, β, γꢀCH ), 1.99, 2.04, 2.09, 2.15 (4 × s, 4
2
2
3
3 3
3
×
3H, 4 × COCH ), 3.04ꢀ3.29 (m, 2H,
δ
ꢀCH ), 4.00 (td, J = 6.9, 1.0 Hz, 1H, 5'ꢀCH), 4.12ꢀ4.35
3
2
3
(m, 5H, 6'ꢀCH2,
α
ꢀCH, CH CH ), 4.73 (br s, 2H, NH ), 4.87 (d, J = 8.3 Hz, 1H, 1'ꢀCH), 5.07
2
3
2
3
3
(
dd, J = 10.5, 3.4 Hz, 1H, 3'ꢀCH), 5.20ꢀ5.31 (m, 2H, 2'ꢀCH, NHꢀBoc), 5.41 (dd, J = 3.4, 1.0 Hz,
1
3
1H, 4'ꢀCH). C NMR (75 MHz, CDCl ):
δ
/ppm = 14.2 (CH CH ), 20.60, 20.66, 20.7, 21.0, (4 ×
3
2
3
COCH ), 25.6 (
γ
ꢀCH ), 28.3 (C(CH ) ), 30.5 (
β
ꢀCH ), 41.2 (
δ
ꢀCH ), 52,7 (
α
ꢀCH), 60.9 (6'ꢀCH2),
3
2
3 3
2
2
61.6 (CH CH ), 66.9 (4'ꢀCH), 67.9, 70.7, 70.8 (2', 5', 3'ꢀCH), 80.3 (C(CH ) ), 103.0 (1'ꢀCH),
2
3
3 3
1
55.9 (C=N),157.1 (COꢀBoc), 170.1, 170.2, 170.3 (4 × COCH ), 172.5 (COOEt). IR
3
ꢀ1
+
(
ATR):ꢀ ꢁꢂ /cm = 1748, 1713, 1368, 1215, 1161, 1055, 955, 743. MS (ESI): m/z = 649 [M + H] .
ACS Paragon Plus Environment