Synthesis and antimycobacterial activity of 2,1′-dihydropyridomycins

Article abstract of DOI:10.1021/ml300385q

Dihydropyridomycins 2 and 3, which lack the characteristic enol ester moiety of the potent antimycobacterial natural product pyridomycin (1), have been prepared from L-Thr, R- and S-hydroxy isovaleric acid, and 3-pyridinecarboxaldehyde. The 2R isomer 2 shows only 4-fold lower anti-Mtb activity than 1, indicating that the enol ester moiety in the natural product is not critical for its biological activity. This finding establishes 2 as a potent and more practical lead for anti-TB drug discovery.

Full text of DOI:10.1021/ml300385q

Letter
pubs.acs.org/acsmedchemlett
Synthesis and Antimycobacterial Activity of 2,1′-
Dihydropyridomycins
Oliver P. Horlacher,† Ruben C. Hartkoorn,‡ Stewart T. Cole,‡ and Karl-Heinz Altmann*,†
†Institute of Pharmaceutical Sciences, Department of Chemistry and Applied Biosciences, Swiss Federal Institute of Technology
(ETH) Zurich, 8093 Zurich, Switzerland
̈
‡
́
́ ́
Ecole Polytechnique Federale de Lausanne (EPFL), Global Health Institute, 1015 Lausanne, Switzerland
S
* Supporting Information
ABSTRACT: Dihydropyridomycins 2 and 3, which lack the characteristic enol
ester moiety of the potent antimycobacterial natural product pyridomycin (1),
have been prepared from L-Thr, R- and S-hydroxy isovaleric acid, and 3-
pyridinecarboxaldehyde. The 2R isomer 2 shows only 4-fold lower anti-Mtb
activity than 1, indicating that the enol ester moiety in the natural product is not
critical for its biological activity. This finding establishes 2 as a potent and more
practical lead for anti-TB drug discovery.
KEYWORDS: natural products, InhA, pyridomycin, total synthesis, tuberculosis
decades to come, we have recently confirmed the in vitro
uberculosis (TB) is a frequently fatal infectious disease
Tthat causes more than 1.4 million deaths annually. TB was
considered to be well contained in the 1960s, but recent
decades have witnessed a resurgence of the disease, even in
industrialized countries, due to comorbidity with AIDS and the
emergence of multi- and extensively drug-resistant (MDR,
XDR) strains of the causative pathogen Mycobacterium
tuberculosis (Mtb).1 These developments were paralleled by a
decline in TB-directed drug discovery efforts; thus, as of today,
the last TB drug with a novel mode of action was launched
more than 40 years ago, and current combination therapy is
insufficient to eliminate XDR Mtb.2 As a consequence, there is
an urgent need for the development of new anti-TB agents that
can shorten the duration of treatment (current standard first
line therapy of TB extends over 6 months) and/or are active
against MDR and XDR bacteria.
antimycobacterial activity of 1, which we found to inhibit Mtb
growth with a minimal inhibitory concentration (MIC) of 0.3
μg/mL.6 Moreover, we have identified the molecular target of 1
as the mycobacterial NADH-dependent enoyl-[acyl-carrier-
protein] reductase (InhA),6 which is also the target of the
clinical TB drug isoniazid (INH) (after metabolic activation
and formation of a NADH adduct as the effective inhibitory
species).7,8 Pyridomycin (1) is a competitive inhibitor at the
NADH-binding site of InhA but has not shown cross-resistance
with INH.6 This suggests that the exact molecular interactions
of 1 with the NADH-binding site differ from those of the
NADH adduct of INH. Together with the fact that the
structure of 1 does not resemble any known TB drug, these
findings render pyridomycin an auspicious starting point for TB
drug discovery.
Only a single total synthesis of 1 has been reported in the
literature; significant difficulties were encountered in that work9
with regard to the establishment of the enol ester double bond
between C2 and C1′,10 a problem that could not be solved in a
fully satisfactory manner. In light of these difficulties, we felt
that the broader exploration of the pyridomycin (or a
pyridomycin-derived) scaffold for TB drug discovery would
greatly benefit from saturation of the double bond between C2
and C1′, provided that the enol ester moiety was not a critical
prerequisite for antimycobacterial activity per se. To explore
this question, 2′-desmethyl-2,1′-dihydropyridomycins 2 and 3
were targeted for synthesis and biological evaluation (Chart 1).
As the target of 1 was unknown at the outset of this work, no
Pyridomycin (1) (Chart 1) is a bacterial natural product that
was first isolated from the Streptomyces strain 6706 in 1953.3,4
The compound was subsequently shown to exhibit significant
in vitro antimycobacterial activity and low systemic toxicity in
mice.5 While these findings were not further explored for
Chart 1
Received: November 10, 2012
Accepted: December 18, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ml300385q | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
a
predictions were possible with regard to the more favorable
configuration of the new chiral center at C2 of dihydropyr-
idomycins (if any), and both configurations needed to be
addressed; in addition, 2 and 3 are not formal reduction
products of 1 itself but incorporate a symmetrical iso-propyl
substituent at C2, thus avoiding uncertainties about the
preferred configuration of a chiral sec-butyl substituent.
Scheme 2
As illustrated in Scheme 1, the synthesis of analogues 2 and 3
was to be based on macrolactamization of amino acids 4/epi-4,
a
Scheme 1. Retrosynthesis of 2 and 3
a
Reagents and conditions: (a) (i) Ac-Gly-OH, NaOAc, Ac2O, 115 °C,
18 h; (ii) NaOAc, MeOH, rt, 72 h, 78%. (b) [Rh(COD)(R,R-
DIPAMP)]BF4 (0.1 mol %), H2 (5 bar), HBF4, MeOH, 50 °C, 18 h,
85% (87% ee). (c) SOCl2, MeOH, reflux, 18 h, 86% (crude). (d)
Benzaldehyde, NaCNBH3, molecular sieves 4 Å, MeOH/AcOH 10:1,
rt, 24 h, 88%. (e) LAH, Et2O, 0 °C, 30 min, 99%. (f) Dess−Martin
periodinane, CH2Cl2, 0 °C, 30 min. (g) c-Hex2BOTf, 15, Et3N,
CH2Cl2, −78 → 0 °C, 73% (2 steps), dr 5:1. (h) LiOH, THF/MeOH/
H2O 3:2:2, rt, 24 h, 96%. (i) (Trimethylsilyl)diazomethane, MeOH/
toluene 1:2, 0 °C, 81%. (k) H2 (1 bar), Pd/C, MeOH, rt, 67%. Ac,
acetyl; COD, 1,5-cyclooctadiene; R,R-DIPAMP, (1R,2R)-bis[(2-
methoxyphenyl)-phenylphosphino]ethane; and c-Hex2BOTf, dicyclo-
hexylboron trifluoromethanesulfonate.
a
Ac, acetyl; Bn, benzyl; TBS, t-butyldimethylsilyl; and PG, protecting
group or H.
which would be followed by deprotection of the exocyclic
amino group, coupling with 3-hydroxypicolinic acid, and final
deprotection of the hydroxyl group on C10. The cyclization
precursors 4/epi-4 can be further disconnected into acid 5 and
alcohols 6/epi-6 (Scheme 1), of which the latter would be the
result of the esterification of commercially available t-
butyloxycarbonyl (Boc)-L-Thr benzyl ester (9) with TBS-
protected hydroxy acids 8/ent-8, respectively. Acid 5 would be
obtained from aldehyde 7 either via auxiliary-based anti-
selective aldol reaction according to Masamune11 or by
stereoselective crotylboration,12 followed by appropriate func-
tional group manipulations. Lastly, aldehyde 7 was envisioned
to be accessed by enantioselective hydrogenation of a 3-pyridyl-
2-amino acrylic acid derivative.
respectively), neither of these conditions led to full substrate
turnover.20
Ester 12 was then transformed into the corresponding N,N-
bis-benzyl derivative 13 by acetamide cleavage with MeOH/
SOCl2 and reductive amination of the resulting free amino ester
with benzaldehyde in 76% overall yield. Reduction of the ester
moiety with lithium aluminum hydride (LAH) followed by
Dess−Martin oxidation21 of the ensuing alcohol then provided
aldehyde 7; to minimize the risk of epimerization, the latter was
not purified but used crude in subsequent reactions.22 Aldol
reaction of 7 with the Masamune reagent 1511 and
dicyclohexylboron trifluoromethanesulfonate (c-Hex2BOTf) as
the Lewis acid proceeded with good stereoselectivity (5:1 in
favor of the desired isomer over the sum of all other isomers)
and provided ester 16 in 52% yield (from 14) as a single isomer
after flash column chromatography. Cleavage of the auxiliary
under basic conditions then cleanly delivered acid 5 in excellent
yield (95%).
The predicted absolute configuration of the C10 stereocenter
in 5 (pyridomycin numbering) was confirmed by Mosher ester
analysis23 of its methyl ester (prepared with TMS-diazo-
methane in 81% yield), while the relative configuration of the
C9, C10, and C11 stereotriad was validated by X-ray
crystallography of lactam 17 (which formed spontaneously
upon debenzylation of the methyl ester of 5 by catalytic
hydrogenation in MeOH; Scheme 2).
The synthesis of acid 5 was initiated by Erlenmeyer azlactone
synthesis13 from 3-pyridinecarboxaldehyde (10) and N-acetyl-
glycine, which produced, after azlactone opening with NaOAc/
MeOH, the Z-configured ester 11 as the sole isomer in 78%
yield (Scheme 2). The subsequent stereoselective hydro-
genation was best performed with [Rh(COD)(R,R-DIPAMP)]-
14,15
BF4
as a catalyst in the presence of HBF4 as a
noncomplexing acid.16 These conditions provided the desired
S-configured amino acid with high selectivity (87% ee) in 85%
isolated yield at a catalyst load of only 0.1%.17 While reactions
18,19
performed with [Rh(COD)(R-propranolol)]BF4
in combi-
nation with HBF4 (3% catalyst, 55 °C) or with [Rh(COD)-
(R,R-DIPAMP)]BF4 in the absence of HBF4 (2% catalyst, 55
°C) likewise gave good selectivities (ee values of 85 and 87%,
As an alternative to the aldol-based strategy depicted in
Scheme 2, we have also investigated the installation of the
B
dx.doi.org/10.1021/ml300385q | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
a
stereocenters at C10 and C11 by means of substrate-directed
crotylation of aldehyde 7, employing the crotyl chromium
reagent derived from CrCl2 and crotyl bromide12 (Scheme 3).
Scheme 4
a
Scheme 3
a
Reagents and conditions: (a) Crotyl bromide, CrCl2, THF, rt, 3 h,
67% from 14, dr 15:1. (b) (i) AD-mix α, t-BuOH/H2O 1:1, 3 days;
(ii) NaIO4, CH2Cl2/H2O 3:2, rt, 2 h; (iii) NaClO2, NaH2PO4, t-
BuOH/H2O 4:1, rt, 2 h, 45%. rt, room temperature.
The reaction provided the desired homoallylic alcohol 18
with excellent diastereoselectivity (dr 15:1) in 67% yield over
two steps from alcohol 14. Dihydroxylation of the double bond
with AD-mix-α,24 periodate cleavage of the resulting diol, and
Pinnick−Kraus oxidation of the ensuing aldehyde then gave the
desired acid 5 in 45% overall yield as a single isomer (from 18).
Overall, this approach gave acid 5 in 30% yield from alcohol 14;
it is thus somewhat less efficient than the aldol-based route via
ester 16, which gave 5 in 50% overall yield.
The elaboration of acid 5 into dihydropyridomycin 2
commenced with its esterification with alcohol 6; the latter
was obtained by Yamaguchi type esterification25 of TBS-
protected (R)-hydroxy isovaleric acid (8) with Boc-Thr-OBn
(9) followed by TBS removal with HF·pyridine in 63% overall
yield (Scheme 4).
Unexpectedly, ester formation between 5 and 6 was found to
be highly challenging, with virtually no reaction being observed
for a number of common condensing agents [1-tert-butoxy-2-
tert-butoxycarbonyl-1,2-dihydroisoquinoline (BBDI), N-cyclo-
hexyl-N′-(2-morpholinoethyl)carbodiimide methyl-p-toluene-
sulfonate (CMC), N,N′-dicyclohexyl-carbodiimide (DCC), 1-
ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDCI), and
O-(benzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium tetra-
fluoroborate (TBTU)]. Efficient ester formation occurred
only under highly optimized reaction conditions, which entailed
reaction of 5 with 2,4,6-trichlorobenzoyl chloride25 at −78 °C
(to form the mixed anhydride), subsequent simultaneous
addition of alcohol 6 and 4-dimethylaminopyridine (DMAP) in
toluene, and slow warming of the reaction mixture to −35 °C.
Using this procedure, the desired ester 19 could be obtained in
yields of up to 50% (Scheme 4) (64% for epi-19).26
Gratifyingly, subsequent global debenzylation to amino acid 4
proceeded smoothly, which set the stage for the crucial ring-
closing reaction. Efficient cyclization of 4 was achieved by
treatment of the compound with O-(7-azabenzotriazol-1-yl)-
N,N,N′,N′-tetramethyluronium hexafluorophosphate (HATU)
and N,N-diisopropylethylamine (DIEA) in 1% DMF/dichloro-
methane at high dilution, which delivered depsipeptide 20 in
63% yield (79% for epi-20). In light of the efficiency of this
process, no other coupling reagents were investigated for this
transformation. Cleavage of the Boc-group and coupling of 3-
hydroxypicolinic acid to the free amine 21 (HATU/DIEA)
finally furnished (2R)-2′-desmethyl-2,1′-dihydropyridomycin
(2) in 52% yield over two steps. The 2S diastereomer 3 was
obtained from (S)-hydroxy isovaleric acid (ent-8) in a
completely analogous way.
a
Reagents and conditions: (a) Boc-L-Thr-OBn (9), 2,4,6-trichlor-
obenzoyl chloride, Et3N, toluene, DMAP, rt, 18 h, 73%. (b)
HF·pyridine, THF, 0 °C → rt, 16 h, 86%. (c) Acid 5, 2,4,6-
trichlorobenzoyl chloride, Et3N, toluene, −78 °C, 5 min, then alcohol
6, toluene, DMAP, −78 → −35 °C, 43 h, 50%. (d) H2 (1 bar), Pd/C,
MeOH, rt, 5 h, quantitative. (e) HATU, DIEA, CH2Cl2, DMF (1%),
10−3 M, rt, 18 h, 63%. (f) TFA, CH2Cl2, 0 °C → rt, 3 h, quantitative
(crude). (g) 3-Hydroxypicolinic acid, HATU, DIEA, acetonitrile, rt, 18
h, 52% (2 steps). DIEA, N,N-diisopropylethylamine; DMAP, 4-
dimethylaminopyridine; HATU, O-(7-azabenzotriazol-1-yl)-
N,N,N′,N′-tetramethyluronium hexafluorophosphate; and rt, room
temperature.
retains most of the activity of natural 1 (4-fold loss in potency;
Table 1). On the basis of this finding, the presence of an enol
Table 1. Antimycobacterial Activity of 1−3 and InhA
Inhibition
a
b
compd
MIC (μg/mL)
Ki (μM)
pyridomycin (1)
0.39
1.56
12.5
4.8 1.1
71.0 7.9
ND
2
3
a
Minimal inhibitory concentrations (MIC) were measured against Mtb
strain H37Rv using the resazurin method. Data are average values from
two independent experiments. Inhibition constants (Ki) were
b
measured against InhA (S94A) with 2-trans-octenoyl-CoA or 2-trans-
dodecenoyl-CoA as the substrate. Data are average values from three
independent experiments SDs (with each experiment carried out in
triplicate). ND, no activity detectable up to a concentration of 75.7
μM. For details, see ref 6 and the Supporting Information.
ester moiety is not a critical requirement for the antimyco-
bacterial activity of pyridomycin-derived structures. At the same
time, a clear activity difference is apparent between 2 and 3,
with the latter being 32-fold less potent against Mtb H37Rv
than 1 (8-fold potency difference to 2), thus pointing to the
importance of the configuration of the newly created chiral
center at C2. The difference in antibacterial activity between 2
and 3 is also reflected in their differential ability to inhibit the
pyridomycin target InhA in vitro (Table 1). Unfortunately, no
Ki value against InhA could be determined for 3, due to its
Intriguingly, the assessment of the antimycobacterial activity
of dihydropyridomycins 2 and 3 revealed that 2R isomer 2
C
dx.doi.org/10.1021/ml300385q | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
limited solubility; no inhibition of InhA was detectable at the
solubility limit for the compound (ca. 75 μM). The difference
in Ki values between 2 and pyridomcyin (ca. 15-fold)
significantly exceeds the difference in antibacterial activity (4-
fold difference in MIC values). The reasons for this apparent
discrepancy are unknown at this point but may be related to
target-independent parameters such as differences in bacterial
cell wall penetration or intracellular metabolism.
To rationalize the differences in InhA inhibition between 2
and 3 at the structural level, we have performed preliminary
docking studies.27 However, in the absence of structural
information on the InhA−pyridomycin complex, attempts to
dock the compounds into the extensive NADH-binding pocket
of InhA did not produce any conclusive results.
In summary, we have achieved the total synthesis of two
close structural analogues of the bacterial metabolite 1 that lack
the characteristic enol ester moiety of the natural product.
While both analogues 2 and 3 retain the ability of 1 to inhibit
Mtb growth, 2R isomer 2 is more potent, with its MIC for Mtb
H37Rv being increased only 4-fold relative to 1. Because of the
improved synthetic accessibility of 2 over natural 1, this
discovery should facilitate future SAR work and, potentially, the
development of pyridomycin-derived drug candidates for TB
treatment. The synthesis and biological evaluation of new
pyridomycin variants that are based on the macrocyclic scaffold
of analogue 2 are currently ongoing in our laboratories.
phate; c-Hex2BOTf, dicyclohexylboron trifluoromethanesulfo-
nate; LAH, lithium aluminum hydride; TBS, t-butyldimethyl-
silyl; TBTU, O-(benzotriazol-1-yl)-N,N,N′,N′-tetra-
methyluronium tetrafluoroborate
REFERENCES
■
(1) Glaziou, P.; Floyd, K.; Raviglione, M. Global Burden and
Epidemiology of Tuberculosis. Clin. Chest Med. 2009, 30, 621−636.
(2) Cegielski, J. P. Extensively Drug-Resistant Tuberculosis: “There
Must Be Some Kind of Way Out of Here”. Clin. Infect. Dis. 2010, 50,
195−200.
(3) Maeda, K.; Kosaka, H.; Okami, Y.; Umezawa, H. A New
Antibiotic, Pyridomycin. J. Antibiot., Ser. A 1953, 6, 140.
(4) The structure and absolute configuration of 1 were first reported
in 1967: Koyama, G.; Iitaka, Y.; Maeda, K.; Umezawa, H. Structure of
Pyridomycin. Tetrahedron Lett. 1967, 37, 3587−3590. Note that the
structural drawing of 1 in this paper depicts a R configuration for both
C9 and C10. In contrast, the drawing of the X-ray structure of the
molecule in Figure 1 of the same paper shows these chiral centers to
be present in a S,S configuration. As in the new analogues 2 and 3 that
are reported here for the first time, C9 and C10 were both S-
configured in the synthetic pyridomycin reported by Kinoshita et al.
(ref 9); the synthetic material was found to be identical spectroscopi-
cally with natural pyridomycin.
(5) Kuroya, M.; Takahashi, B.; Hinuma, Y.; Yashima, T.; Watanabe,
K.; Hamada, S. An Antituberculous Antibiotic from Streptomyces sp.
Studies on the Streptomyces Antibiotics. XXXII. J. Antibiot., Ser. A
1954, 7, 58−59.
(6) Hartkoorn, R. C.; Sala, C.; Neres, J.; Pojer, F.; Magnet, S.;
Mukherjee, R.; Uplekar, S.; Boy-Rottger, S.; Altmann, K.-H.; Cole, S.
T. Towards a New Tuberculosis Drug: PyridomycinNature’s
Isoniazid. EMBO Mol. Med. 2012, 4, 1032−1042.
(7) Banerjee, A.; Dubnau, E.; Quemard, A.; Balasubramanian, V.;
Um, K. S.; Wilson, T.; Collins, D.; de Lisle, G.; Jacobs, W. R., Jr. InhA,
a Gene Encoding a Target for Isoniazid and Ethionamide in
Mycobacterium Tuberculosis. Science 1994, 263, 227−230.
(8) Vilcheze, C.; Wang, F.; Arai, M.; Hazbon, M. H.; Colangeli, R.;
Kremer, L.; Weisbrod, T. R.; Alland, D.; Sacchettini, J. C.; Jacobs, W.
R., Jr. Transfer of a Point Mutation in Mycobacterium Tuberculosis
InhA Resolves the Target of Isoniazid. Nat. Med. 2006, 12, 1027−
1029.
(9) Kinoshita, M.; Nakata, M.; Takarada, K.; Tatsuta, K. Total
Synthesis of Pyridomycin. Tetrahedron Lett. 1989, 30, 7419−7422.
(10) Apart from the problem of controlling double bond geometry,
even the introduction of the C2-C1′ double bond as such proved to be
difficult and could not be accomplished with high efficiency. See also
Kinoshita, M.; Mori, Y. Bull. Soc. Chem. Jpn. 1985, 58, 3298−3308.
(11) Abiko, A.; Liu, J.-F; Masamune, S. The Anti-Selective Boron-
Mediated Asymmetric Aldol Reaction of Carboxylic Esters. J. Am.
Chem. Soc. 1997, 119, 2586−2587.
ASSOCIATED CONTENT
* Supporting Information
■
S
Synthetic procedures and analytical data for all new
compounds. Crystallographic information on 17. This material
is available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Tel: +41-44-6337390. Fax: +41-44-6331369. E-mail: karl-
heinz.altmann@pharma.ethz.ch.
■
Funding
This work was supported by the European Community
Framework Programme 7, More Medicines For Tuberculosis
(MM4TB), grant agreement no. 260872.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We are indebted to Dr. Bernhard Pfeiffer (ETH Zurich) for
■
̈
NMR support, to Louis Bertschi from the ETHZ LOC MS
Service for HRMS spectra, and to Dr. B. Schweizer for X-ray
crystallographic data. We also thank Dr. Joao Neres (EPFL) for
providing the substrates for the InhA assay and Dr. Sean Ekins
(CDD) for preliminary modeling studies.
(12) Ciapetti, P.; Falorni, M.; Taddei, M. CrCl2 Mediated Addition of
Allylic Halides or Phosphates to N-Protected Alpha-Amino Aldehydes.
Stereocontrolled Synthesis of a New Core for C-2 Symmetric HIV-
protease Inhibitors. Tetrahedron 1996, 52, 7379−7390.
(13) Humphrey, C. E.; Furegati, M.; Laumen, K.; La Vecchia, L.;
Leutert, T.; Muller-Hartwieg, J. C. D.; Vogtle, M. Optimized Synthesis
̈
̈
ABBREVIATIONS
of L-m-Tyrosine Suitable for Chemical Scale-up. Org. Process Res. Dev.
2007, 11, 1069−1075.
■
BBDI, 1-tert-butoxy-2-tert-butoxycarbonyl-1,2-dihydroisoquino-
line; Bn, benzyl; Boc, t-butyloxycarbonyl; CMC, N-cyclohexyl-
N′-(2-morpholinoethyl)carbodiimide methyl-p-toluenesulfo-
nate; COD, 1,5-cyclooctadiene; DCC, N,N′-dicyclohexyl-
carbodiimide; DIEA, N,N-diisopropylethylamine; R,R-DI-
PAMP, (1R,2R)-bis[(2-methoxyphenyl)-phenyl-phosphino]-
ethane; DMAP, 4-dimethylaminopyridine; EDCI, 1-ethyl-3-
(3-dimethylaminopropyl)-carbodiimide; HATU, O-(7-azaben-
zotriazol-1-yl)-N,N,N′,N′-tetramethyluronium hexafluorophos-
(14) Vineyard, B. D.; Knowles, W. S.; Sabacky, M. J. Alpha-Amino-
Acids by Catalytic Asymmetric Hydrogenation. J. Mol. Catal. 1983, 19,
159−69.
(15) Vineyard, B. D.; Knowles, W. S.; Sabacky, M. J.; Bachman, G. L.;
Weinkauff, D. J. Asymmetric HydrogenationRhodium Chiral
Bisphosphine Catalyst. J. Am. Chem. Soc. 1977, 99, 5946−5952.
(16) HBF4 is assumed to protonate the pyridine nitrogen, which
avoids complexation and concomitant inactivation of the rhodium
catalyst.
D
dx.doi.org/10.1021/ml300385q | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
(17) For a previous synthesis of 12 using [Rh(COD)(R,R-
DIPAMP)]BF4, see Bozell, J. J.; Vogt, C. E.; Gozum, J. Transition-
Metal-Assisted Asymmetric-Synthesis of Amino-Acid-AnalogsA
New Synthesis of Optically Pure D-Pyridylalanine and L-Pyridylalanine.
J. Org. Chem. 1991, 56, 2584−2587.
(18) Krause, H. W.; Foken, H.; Pracejus, H. Aminophosphine
Phosphinites of Propranolol as Ligands for Rhodium Catalyzed
Asymmetric Hydrogenation of Dehydroamino Acids. New J. Chem.
1989, 13, 615−620.
(19) Dobler, C.; Kreuzfeld, H. J.; Michalik, M.; Krause, H. W.
Unusual Amino Acids. 7. Asymmetric Synthesis of 3- and 4-
Pyridylalanines. Tetrahedron: Asymmetry 1996, 7, 117−125.
(20) Variation of temperature, catalyst load, or H2 pressure did not
lead to full conversion in the case of the [Rh(COD)(R-propranolol)]
BF4 catalyst. Full conversion, however, was crucial due to the
inseparability of 11 and 12.
(21) Dess, D. B.; Martin, J. C. Readily Accessible 12-I-5 Oxidant for
the Conversion of Primary and Secondary Alcohols to Aldehydes and
Ketones. J. Org. Chem. 1983, 48, 4155−4156.
(22) In contrast to 7, the corresponding N-acetyl or N-Alloc
aldehydes were not feasible intermediates for the efficient elaboration
of acid 5, due to a pronounced tendency for rapid isomerization in
subsequent reactions and even upon workup after oxidation.
(23) Dale, J. A.; Dull, D. L.; Mosher, H. S. α-Methoxy-α-
Trifluoromethylphenylacetic Acid, a Versatile Reagent for the
Determination of Enantiomeric Composition of Alcohols and Amines.
J. Org. Chem. 1969, 34, 2543−2549.
(24) Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. A.;
Hartung, J.; Jeong, K. S.; Kwong, H. L.; Morikawa, K.; Wang, Z. M.;
Xu, D.; Zhang, X.-L. The Osmium-catalyzed Asymmetric Dihydrox-
ylation: A New Ligand Class and a Process Improvement. J. Org.
Chem. 1992, 57, 2768−2771.
(25) Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M.
Rapid Esterification by Means of Mixed Anhydride and Its Application
to Large-Ring Lactonization. Bull. Chem. Soc. Jpn. 1979, 52, 1989−
1993.
(26) The difficulties encountered for the ester formation between 5
and 6 are unlikely to be related to the presence of the free hydroxyl
group on C10 of 5 (pyridomycin numbering). In general, 5 could be
reisolated unchanged from unsuccessful esterification experiments,
except for an attempted DCC-mediated coupling, which gave the
corresponding N-acyl urea. The inherently low reactivity of the C10
hydroxyl group is also reflected in the fact that the conversion of 5 into
the corresponding t-butyldimethylsilyl (TBS) or TMS ether was not
possible in preparatively viable yields.
(27) Docking was performed using LibDock in Discovery Studio 3.5
(Accelyris, San Diego). The wild-type InhA (InhADWTdec11_ref-
mac19_coot) was obtained as a PDB file and then run through the
“Prepare proteins” protocol in Discovery Studio 2.5.5 (Accelrys). A
binding site sphere of 13.99 Å was generated around NADH. The
docking tolerance was 0.25, the docking preference was High Quality,
the conformation method was Fast, and the minimization algorithm
was steepest descent (RMSD cutoff 1, max steps 1000, RMS gradient
0.001, force field CHARMm, and nonbond list radius = 14).
Pyridomycin (1), 2, and 3 were docked into the protein structure.
E
dx.doi.org/10.1021/ml300385q | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX