Design and synthesis of N-alkyldeoxynojirimycin derivatives with improved metabolic stability as inhibitors of BVDV and Tacaribe virus

Article abstract of DOI:10.1016/j.bmcl.2013.04.052

Novel N-alkyldeoxynojirimycins (NADNJs) based on our previous lead 3 were designed, synthesized and tested in metabolic assays and in virus cultures. NADNJs containing terminal tertiary benzamide, sulfonamide, urea, and oxazolidinone moieties were discove

Full text of DOI:10.1016/j.bmcl.2013.04.052

Accepted Manuscript
Design and synthesis of N-alkyldeoxynojirimycin derivatives with improved
metabolic stability as inhibitors of BVDV and Tacaribe virus
Yanming Du, Hong Ye, Fang Guo, Lijuan Wang, Tina Gill, Noshena Khan,
Andrea Cuconati, Ju-Tao Guo, Timothy M. Block, Jinhong Chang, Xiaodong
Xu
PII:
S0960-894X(13)00533-7
DOI:
http://dx.doi.org/10.1016/j.bmcl.2013.04.052
BMCL 20417
Reference:
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
6 February 2013
15 April 2013
22 April 2013
Please cite this article as: Du, Y., Ye, H., Guo, F., Wang, L., Gill, T., Khan, N., Cuconati, A., Guo, J-T., Block,
T.M., Chang, J., Xu, X., Design and synthesis of N-alkyldeoxynojirimycin derivatives with improved metabolic
stability as inhibitors of BVDV and Tacaribe virus, Bioorganic & Medicinal Chemistry Letters (2013), doi: http://
dx.doi.org/10.1016/j.bmcl.2013.04.052
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Design and synthesis of N-alkyldeoxynojirimycin derivatives with improved metabolic stability
as inhibitors of BVDV and Tacaribe virus
Yanming Du*,a, Hong Yea, Fang Guob, Lijuan Wangb, Tina Gillb, Noshena Khana, Andrea
Cuconatia, Ju-Tao Guob, Timothy M. Blocka,b, Jinhong Chang*,b, Xiaodong Xu*,a
a. Institute for Hepatitis and Virus Research, 3805 Old Easton Road, Doylestown, PA 18902; b.
Drexel Institute for Biotechnology and Virology Research, Drexel University College of
Medicine, Doylestown, Pennsylvania 18902
Keywords: N-alkyldeoxynojirimycin; sulfonamide; urea; oxazolidinone; metabolic stability;
antiviral; BVDV; Tacaribe
Abstract: Novel N-alkyldeoxynojirimycins (NADNJs) based on our previous lead 3 were
designed, synthesized and tested in metabolic assays and in virus cultures. NADNJs containing
terminal tertiary benzamide, sulfonamide, urea, and oxazolidinone moieties were discovered to
have improved metabolic stability compared to 3, while maintaining submicromolar EC50 against
BVDV and Tacaribe virus; and low cytotoxicity.

The metabolic stability of a drug has significant impact on both its safety and efficacy. As
a result, in vitro liver microsomal stability testing has been implemented in the early stages of
drug discovery to evaluate the metabolic liabilities of a drug candidate in the presence of
cytochrome P450 enzymes (CYPs).1 The results can be used as a criteria to select compounds for
advancement into further in-vitro bioactivity assays and in vivo animal models. For those
compounds that fail in the metabolic stability test, structural modification can be executed to
improve stability while maintaining attractive in vitro activities observed. Generally, a metabolic
site, either as a labile element or functional group, needs to first be identified, and this can be
achieved either by incubation with CYP isozymes followed by LC/MS2 analysis, or by
prediction3. Once this information is available, a metabolically more stable element2 or surrogate
group4 can be used to block the site. Several strategies have been employed for improving
metabolic stability, including reduction of lipophilicity, change of sterics and electronics,
introduction of conformational constraints, and alteration of the stereochemistry of the
compound.5 These methods can be also understood as to either change the reduction potential of
the labile site, or move the labile site away from the target binding position.
In our recent studies on deoxynojirimycin (1, DNJ) and its derivatives as glucosidase
inhibitors and broad spectrum antiviral agents against hemorrhagic fever viruses,6,7,8 compounds
2-4, with either aromatic ether or tertiary amide terminals, were identified as having
submicromolar EC50s against bovine viral diarrhea virus (BVDV), Dengue virus, and Tacaribe
virus, and low cellular toxicities. However, when subsequent in vitro ADMET studies were
performed,9 we found that compound 3 and 4 had lower microsomal stability (Table 1) as
compared to compound 2. Because these two unstable compounds share a common tertiary amide
scaffold, this feature became the focus of our structural optimization.

Figure 1. DNJ and active N-alkyl derivatives with aromatic ether and tertiary amide terminals
Liver Microsomal Stability Remaining
@ 60 min (%)
Compound
Human
91
Rat
79
Mouse
91
2
3
4
41.56
36.11
26.87
21.56
58.57
55.27
Table 1. Microsomal stability tests for N-alkyl DNJ derivatives with aromatic ether and tertiary amide terminals
Compared to compound 2, compound 4 has only one difference in its terminal structure,
which consists of a nitrogen heteroatom, a pivaloyl group, and an aromatic substitution. Because
the nitrogen atom is blocked by three bulky groups, we don’t expect that C-N bonds are likely to
be the source of liability. And arguably, the 2,4-difluorophenyl in compound 4 would have
similar microsomal stability to the 2,5-difluorophenyl in compound 2. Thus, we hypothesized
that the microsomal liability for both compound 3 and 4 was due to the acetyl of the pivaloyl
group. To improve the microsomal stability of this labile site, the lead compound, 3, was selected
as the starting point. We reasoned that we can change the reduction potential of the sp3
hybridized methyl group by either replacing it with a sp2 hybridized phenyl ring (5 in Figure 2) or
substituting the carbonyl in pivaloyl with a more electron withdrawing sulfonyl and phosphoryl
group (6, 7); We can also change the polarity environment around the methyl group by
introducing a hydrogen bonding donor NH to the carbonyl to form a urea (8), which is expected

to influence the interaction with the P-450 enzymes, therefore, to reduce the likelihood of leaving
the methyl group at the oxidation pocket. Another potential approach is to remove the alkyl
group connected to the carbonyl and introduce an oxygen atom in the form of oxazolidin-2-one,
as we know that the Boc group with the same carbamate moiety could bring potency to this series
of compounds.8 Thus, compounds 5-9 were prepared as described in Schemes 1.
Figure 2. Proposed analog structures of compound 3 with terminal group replaced.
The compounds 5-8 were synthesized in a general procedure, which allowed us to
diversify the acyl group replacement (Scheme 1). Synthesis began with the alkylation of
cyclohexylamine with ((6-bromohexyl)oxy)(tert-butyl)dimethylsilane, followed by derivation
with an acyl, sulfinyl, sulfonyl,10 phosphoryl chloride or isocyanate to give a protected tertiary
amine derivative 12. Subsequent deprotection with TBAF and oxidation of the resulting alcohol
13 generated the corresponding aldehyde 14. Reductive amination with DNJ provided the final
DNJ derivatives 5-8.

H
N
a
b
d
Br
TBDMSO
HO
TBDMSO
TBDMSO
10
12
11
13
R
N
R
N
c
OH
N
O
H
R
N
R
N
e
HO
HO
14
5-8
OH
Reagents and conditions: a, C6H11NH2, K2CO3, CH3CN, 80 oC; b, Et3N, R7COCl or R8SO2Cl, R9SOCl, (R10O)2POCl,
or R11NCO, CH2Cl2 or THF; c, TBAF, THF; d, PCC or Dess-Matin periodane; e, DNJ, AcOH, EtOH, Pd/C.
Scheme 1. General synthetic route for preparation of compound 5-8.
Compound 9 was synthesized by a similar procedure, in which compound 10 was reacted
with the appropriate oxazolidinones in the presence of LiHMDS in THF. Three compounds, 9a, 9b,
and 9c, were prepared following this procedure.
The compounds were evaluated for their antiviral activities against BVDV (a pestivirus)
in MDBK cells and tacaribe virus (an arenavirus) in Huh7.5 cells (Table 2).6,11 The compounds
that maintained the antiviral potency of lead compound 3 would be selected for metabolic
stability study. The antiviral activity was measured by yield reduction assay and expressed as
EC50 values; cytotoxicity was measured by an MTT assay and expressed as CC50 values.
Compared to compound 3, replacement of the tert-butyl group with either phenyl ring (5a) or
2,4,5-fluorophenyl ring (5b) retained the antiviral potency against both BVDV and Tacaribe virus
with low cellular toxicity. When sulfonamide 6a was introduced as a surrogate for amide in 3,
submicromolar EC50s were also obtained for BVDV and Tacaribe virus. Larger hydrophobic
groups contained as tert-butyl in 6c and phenyl in 6b were tolerated, but higher cellular toxicity
was seen with 6b. These results were different from the amide series in which larger hydrophobic
groups afforded better potency.8 Phosphoramide 7 reduced the activity against BVDV by more
than ten-fold (EC50 = 3.4 M), although it retained good potency against Tacaribe virus (EC50 =
0.4 M). Of the three urea analogs tested, only compound 8c resumed submicromolar activities
against both BVDV and Tacaribe virus (EC50 = 0.25, 0.74 M respectively) while exhibiting low
cellular toxicity, consistent with the results in the amide series. The unsubstituted urea 8a and n-

propyl containing compound 8b both showed weak activity against BVDV and Tacaribe virus,
suggesting that an increase in hydrophobic interaction benefits the antiviral activities of this
family of compounds.
Table 2. Inhibition and toxicity evaluation of amide derivatives
BVDV
Tacaribe
EC50
CC50
EC50
CC50
compound
R
(M)
(M)
(M)
(M)
3
0.2
> 500
> 500
0.2
0.6
500
5a
5b
0.32
> 500
0.27
> 500
0.6
> 500
6a
6b
0.2
0.5
> 500
200
0.7
1.6
> 500
280
6c
7
0.31
3.4
> 500
> 500
0.4
0.3
> 500
310

8a
8b
14.2
15.8
> 500
> 500
100
25
> 500
> 500
8c
0.25
> 500
0.74
> 500
The effect of the tertiary amide to oxazolidinone conversion is illustrated in Table 3.11
Compound 9a exhibited reasonable activity against BVDV (EC50 = 0.8 M), but not against
Tacaribe virus. It is interesting that this compound is about four-fold more potent than its
diastereomer 9b, which had weaker activity against BVDV (EC50 = 3.5 M). Notably, the
benzofused oxazolidinone 9c showed improved potencies against both BVDV and Tacaribe virus
(EC50 = 0.4, 0.4 M respectively) and low cellular toxicity (CC50 > 500 M), indicating a
compact scaffold is favored in this family of the compounds.
Table 3. Evaluation of DNJs with oxazolidinoe terminal
BVDV
Tacaribe
EC50
CC50
EC50
CC50
compound
R12
(M)
(M)
(M)
(M)
9a
0.8
3.5
> 500
> 500
ND
ND
ND
ND
9b

9c
0.4
> 500
0.4
> 500
ND, not determined.
With some new leads with comparable activities to the early lead 3 in hand, we next
selected six compounds representing benzamide, sulfonamide, urea, and oxazolidinone families
and carried out metabolic stability studies.9 We found that the benzamide 5a exhibited more than
twice the stability of 3 in human, rat, and mouse microsome assays (Table 4). The methyl
sulfonamide 6a also possessed increased metabolic stability in human liver microsome assay.
However, the sulfonamide 6b and 6c, with bulkier phenyl or tert-butyl group displayed no
improvement relative to compound 3. This probably is due to the different oxidation potential or
trajectory of the sulfonamide substituent from that of the amide,12 which might deliver the phenyl
or tert-butyl group into a new and better oxidation position in the pocket of P450. On the other
hand, the urea 8c exhibited excellent metabolic stability (93%, 92%, and 87% in human, rat and
mouse microsome assay, respectively), even although the tert-butyl group was unaltered and the
length of this side chain was increased by one NH group. We propose that the presence of the
additional NH group changed binding conformation and physicochemical and electronic
environment around the tert-butyl and carbonyl groups, possibly keeping the tert-butyl group
away from the optimal oxidation position. The oxazolidinone 9c, on the contrary, eliminated the
corresponding tert-butyl group as in lead 3, resulting in similar microsomal stability (100%, 87%,
and 87% in human, rat and mouse microsome assay, respectively). This observation suggests that
the target interacting conformation of 9c is similar to that of the 3.
In summary, different approaches have been used in improving the metabolic stability of
the early lead 3, with some proving to be successful. The minimalist approach was to entirely
remove or trim away the labile saturated group (such as in 9c), although structural modification at
other parts of the molecule may be needed to compensate for loss of potency. Introduction of an
additional polar atom also worked well (such as in 8c), because this new polar group can change
the binding conformation and electronics of the molecule itself through formation of hydrogen
bonds, and this change may protect the labile group escape from the oxidative process.
Table 4. Evaluation of microsomal stability of new DNJ derivatives.
compound
Liver Microsomal Stability Remaining

@ 60 min (%)
Rat
Human
41.56
82.56
95.55
12.96
50.19
93.56
100.82
Mouse
58.57
97.37
86.88
42.79
65.38
87.58
86.66
26.87
3
77.42
5a
6a
6b
6c
8c
9c
65.66
3.83
12.14
92.76
86.91
Thus, through modifying the terminal tertiary amide structure of the early lead 3, we have
identified several scaffolds, including benzamide, sulfonamide, urea, and oxazolidinone with
improved metabolic stability. The newly prepared DNJ analogs, 5a, 6a, 8c, and 9c, also
maintained good antiviral activity against both BVDV and Tacaribe virus.
Acknowledgements
We would like to thank the Defense Threat Reduction Agency (DTRA) for financial support
[HDTRA1-09-CHEM-BIO-BAA]. We are also grateful to Dr. William A. Kinney for his
constructive comments on the manuscript, and to Ms. Nancy Young for her coordination work.
References
1.
a) Tsaioun, K.; Kates, S. A. ADMET For Medicinal Chemists, A Practical Guide;
John Wiley & Sons: Hoboken, NJ 2011. b) Ackley, D. C.; Rockich, K. T.; Baker, T. R.
Optimization in Drug Discovery, Ed by Yan. Z; Caldwell, G. W; 2004, 151.
2.
3.
Chen, T.; Benmohamed, R.; Kim, J.; Smith, K.; Amante, D.; Morimoto, R. I.; Kirsch,
D. R.; Ferrante, R. J. and Silverman, R. B. J. Med. Chem., 2012, 55, 515.
a) Trunzer, M.; Faller, B.; and Zimmerlin, A. J. Med. Chem., 2009, 52, 329; b) Castro-
Perez, J. M. Drug Discovery Today, 2007, 12, 249.
4.
a) Radi, M.; Dreassi, E.; Brullo, C.; Crespan, E.; Tintori, C.; Bernardo, V.; Valoti, M.;
Zamperini, C.; Daigl, H.; Musumeci, F.; Carraro, F.; Naldini, A.; Filippi, I.; Maga, G.;
Schenone, S.; and Botta, M. J. Med. Chem., 2011, 54, 2610; b) La Regina, G.; Bai, R.;

Rensen, W.; Coluccia, A.; Piscitelli, F.; Gatti, V.; Bolognesi, A.; Lavecchia, A.;
Granata, I.; Porta, A.; Maresca, B.; Soriani, A.; Iannitto, M. L.; Mariani, M.; Santoni,
A.; Brancale, A.; Ferlini, C.; Dondio, G.; Varasi, M.; Mercurio, C.; Hamel, E.; Lavia,
P.; Novellino, E.; and Silvestri, R. J. Med. Chem., 2011, 54, 8394.
5.
a) St. Jean, Jr., D. J.; Fotsch, C. J. Med. Chem., 2012, 55, 6002, and references therein.
b) Nassar, A. F. Drug Metabolism Handbook, Ed. By Nassar, A. F.; Hollenberg, P. F.;
Scatina, J. 2009, 253. c) Meanwell, N. A. Chem. Res. Toxicol., 2011, 24, 1420.
6.
a)Block, T. M.; Jordan, R. Antiviral. Chem. Chemother. 2001, 12, 317. b) Mehta, A.;
Ouzounov, S.; Jordan, R.; Simsek, E.; Lu, X.; Moriarty, R. M.; Jacob, G.; Dwek, R.
A.; Block, T. M. Antiviral. Chem. Chemother. 2002, 13, 299; c) Gu, B., Mason, P.;
Wang, L.; Norton, P.; Bourne, N.; Moriarty, R.; Mehta, A.; Despande, M.; Shah, R.;
Block, T. Antiviral. Chem. Chemother. 2007, 18, 49; d) Chang, J., Wang, L.; Ma, D.;
Qu, X.; Guo, H.; Xu, X.; Mason, P. W; Bourne, N.; Moriarty, R.; Gu, B.; Guo, J.-T.;
Block, T. M. Antimicrob. Agents Chemother. 2009, 1501.
7.
8.
9.
Yu, W.; Gill, T.; Wang, L.; Du, Y.; Ye, H.; Qu, X.; Guo, J.-T.; Cuconati, A.; Zhao, K.;
Block, T. M.; Xu, X.; Chang, J. J. Med. Chem., 2012, 55, 6061.
Du, Y.; Ye, H.; Gill, T.; Wang, L.; Guo, F.; Cuconati, A.; Block, T. M.; Chang, J.; Xu,
X. Bioorg. Med. Chem. Lett., 2013, 23, 2172.
Microsomal stability test was conducted by Pharmaron. 0.5mg/ml of microsomes were
incubated with test compounds (2μM concentration) at 37°C in presence of 1mM
NADPH. Reactions without NADPH serve as negative controls.
10. Intermediate 12d was prepared through a two step reaction: Reaction of secondary
amine 11 with 2-Methyl-propane-2-sulfinyl chloride gave a sulfinamide, which was
oxidized to sulfonamide 12d by H2O2.
11. MDBK cells were infected with BVDV at MOI of 1. Huh7 cells were infected with
Tacaribe virus at MOI of 0.01. Cells were treated with test compounds at
concentrations ranging from 0.01 to 100 μM, immediately after incubation with virus.
BVDV and Tacaribe titers were determined by standard virus yield reduction assay in
MDBK and Vero cells, respectively. EC50 values were calculated from dose-dependent
virus yield reduction curves. CC50 values were determined using MTT (Sigma) assay in
MDBK and Huh7 cells, respectively.
12. Du, Y.; Creighton, C. J.; Yan, Z.; Gauthier, D. A.; Dahl, J. P.; Zhao, B.; Belkowski, S.
M.; Reitz, A. B. Bioorg. Med. Chem., 2005, 13, 5936.