Preparation and properties in vitro and in vivo of antitubercular pyrroles

Article abstract of DOI:10.1002/jhet.352

(Chemical Equation Presented) 1-Acylamino-2,5-dimethylpyrroles were prepared in the exploration of heterocyclic structures useful for their antitubercular activity. The pyrroles were conveniently formed from the reaction of aromatic acid hydrazides with hexane-2,5-dione in water or ethanol, without resorting to acid catalysis. In each case, the procedure provided a single pyrrole in pure form, and the product was identified without difficulty on the basis of highly characteristic spectrometric features. Some members of this class have significant activities against drug-resistant tuberculosis in vitro and offer substantial protection in a rigorous mouse model of the disease.

Full text of DOI:10.1002/jhet.352

May 2010
Preparation and Properties In Vitro and In Vivo of Antitubercular
Pyrroles
707
Michael J. Hearn,a* Michaeline F. Chen,a Marianne S. Terrot,a
Eleanor R. Webster,a and Michael H. Cynamonb
aDepartment of Chemistry, Wellesley College, Wellesley, Massachusetts 02481
bVeterans Affairs Medical Center, Syracuse, New York 13210
*E-mail: mhearn@wellesley.edu
Received August 18, 2009
DOI 10.1002/jhet.352
Published online 26 March 2010 in Wiley InterScience (www.interscience.wiley.com).
1-Acylamino-2,5-dimethylpyrroles were prepared in the exploration of heterocyclic structures useful
for their antitubercular activity. The pyrroles were conveniently formed from the reaction of aromatic
acid hydrazides with hexane-2,5-dione in water or ethanol, without resorting to acid catalysis. In each
case, the procedure provided a single pyrrole in pure form, and the product was identified without diffi-
culty on the basis of highly characteristic spectrometric features. Some members of this class have sig-
nificant activities against drug-resistant tuberculosis in vitro and offer substantial protection in a
rigorous mouse model of the disease.
J. Heterocyclic Chem., 47, 707 (2010).
INTRODUCTION
more necessary by the emergence of strains of M. tuber-
culosis that are drug resistant or exceptionally virulent
[6–9]. In the case of the most widely prescribed tubercu-
losis medication, isoniazid (isonicotinic acid hydrazide,
INH), resistant strains have appeared that require drug
concentrations 1000 times greater than those for suscep-
tible bacteria. This renders the drug ineffective as a ther-
apeutic mainstay for patients infected with those strains.
Against the urgency and seeming intransigence of this
situation, there is increasing evidence that heterocyclic
compounds will serve as significant leads for the discov-
ery of robust new antitubercular medications and as val-
uable probes for gaining a better understanding of the
life cycle of the pathogen [10–13]. Much, however,
remains to be learned about the scope of heterocyclic
structures that will possess this valuable antimicrobial
activity [14,15]. As early as 1953, it had been reported
that some pyrroles had antimycobacterial activity in
vitro [16–20], but this perceptive observation was not
given concerted follow-up at the time. More recently,
significant work has resumed on antitubercular drug
design using pyrroles as templates for synthesis [21–23],
including elegant molecular modeling studies in con-
junction with laboratory experiments [24,25]. Although
much ground is still to be covered, it seems clear that
the chemistry of pyrroles will provide both important
lead compounds in the search for new antitubercular
Worldwide, tuberculosis ranks first among the causes
of death from a communicable disease [1]. This killer’s
designation as an international public health threat by
the World Health Organization more than a decade ago
[2] was followed by the mobilization of considerable
resources from governments, pharmaceutical companies,
philanthropic foundations, and public–private partner-
ships [3]. Even so, advances thus far against the disease
have been hard-won, and fully one-third of the earth’s
population remains infected with the causative bacillus,
Mycobacterium tuberculosis [4]. In industrialized coun-
tries, the unexpected reemergence of tuberculosis has
come with substantial costs to the public health infra-
structure. Among the populations of developing nations,
the disease has persisted, and millions of deaths result
each year from such infection [5]. Although tuberculosis
can strike persons of any age, the sinister natural history
of the disease makes it particularly devastating to those
of middle years, during the most productive period of
their lives. For those who do not themselves succumb,
there is social and economic hardship as individuals,
families, and communities struggle to deal with the col-
lateral burdens of widespread infection. Progress in the
development of strong new treatment regimens for tu-
berculosis has now been made both more difficult and
C
V
2010 HeteroCorporation

708
M. J. Hearn, M. F. Chen, M. S. Terrot, E. R. Webster, and M. H. Cynamon
Vol 47
Scheme 1
medications and tools for probing the complex interac-
tions among pathogen, mammalian host, and drugs in
tuberculosis treatment modalities.
results on the preparation of these pyrroles III are sum-
marized in Table 1.
In general, the preparation of a pyrrole by the reaction
of a 2,5-dione with an amine, the Paal-Knorr synthesis
[26], is sensitive to both the nucleophilicity of the amine
and the specific reaction conditions employed. These
issues have raised considerable interest within the syn-
thesis community [27], and some trends have become
apparent. Among aliphatic amines, for example, the
steric environment of the carbon bearing the nitrogen
appears to be crucial, with yields falling drastically as
the carbon becomes more highly substituted [28].
Among substituted anilines, low reactivity is observed
for those amines in which the aromatic ring is substi-
tuted with electron-withdrawing groups; thus 2,5-
dichloroaniline fails to react with hexane-2,5-dione,
whereas o-phenylenediamine is highly nucleophilic and
reacts rapidly with two equivalents of the hexane-2,5-
dione to produce a 1,2-dipyrrolyl benzene [28]. With
respect to reaction conditions, some researchers have
pointed out the accelerating role of acid catalysis [29],
while others have noted that the amount of acid used, if
any, should be adequate to allow for activation of the
carbonyl functions of the dione but not enough to fully
protonate the amine function and thereby render it non-
nucleophilic [30]. In the present work, we found that in
each case the terminal nitrogen of the hydrazide moiety
was sufficiently nucleophilic to permit reaction in water
or ethanol, without resorting to acid catalysis. This was
true even for examples in which a strongly electron-with-
drawing group was present, such as IIIf. The ability to
carry out the reaction under neutral conditions consider-
ably simplified the work-up and isolation of the products.
Within this context, we wished to explore, using up-
to-date methods, both the preparative chemistry and the
antitubercular behavior of a class of acylaminopyrroles
derived from acid hydrazides and diketones. We now
report on the convenient preparation and useful antimy-
cobacterial properties of a series of these compounds
(III). We have found that the compounds are readily
formed from aromatic carboxylic acid hydrazides (I)
and 2,5-hexanedione (II) in excellent purity as stable
highly crystalline solids (see Scheme 1). The reactions
take place to give a single product without complication,
and the resulting 1-acylamino-2,5-dimethylpyrroles may
be characterized in clear-cut ways by standard spectro-
metric techniques. In a representative example, com-
pound II was treated with INH (I, Ar ¼ 4-C5H4N, 1.00
equivalent) in refluxing water for 2 h. After cooling and
standing over night, the needles that formed were fil-
tered off to give III (Ar ¼ 4-C5H4N) in analytically
pure form, with the distinguishing spectrometric features
of the acylaminopyrrole. Notably, the features included:
(1) the NAH band appropriate for the amide link near
3225 wavenumbers in the infrared spectrum; (2) the am-
ide I and amide II peaks at 1673 and 1537 wavenum-
bers; (3) the NAH signal at d 11.5 ppm in the hydrogen
NMR spectrum, integrating for one hydrogen, quenched
by the addition of D2O; (4) the peak at d 5.7, a singlet
integrating for two hydrogens for the pyrrole ring pro-
tons; and (5) a signal at d 11.3 ppm in the carbon nmr
spectrum for the methyl groups at the 2- and 5-posi-
tions. The overall yield of the product was 67%. Our
Table 1
Pyrroles III.
Entry
Compound
Ar
% yield
mp (ꢀC)
mmax NAH (cmꢁ1
)
1H NMR, d
13C NMR, d
1
2
3
4
5
6
7
IIIa
IIIb
IIIc
IIId
IIIe
IIIf
IIIg
4-C5H4N
C6H5
4-ClC6H4
4-CH3C6H4
4-BrC6H4
67
89
69
91
77
75
30
151–153
175–177
185
187–189
207–210
218–221
173–174
3224
3261
3274
3281
3246
3278
3270
11.5
11.3
11.3
11.2
11.3
11.3
10.9
165
166
165
166
165
165
162
4-NO2C6H4
2-OH-4-NH2C6H3
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

May 2010
Preparation and Properties In Vitro and In Vivo of Antitubercular Pyrroles
709
Table 2
Table 4
In vitro activities: Minimum inhibitory concentrations in fully
susceptible Mtb Erdman.
In vivo activities of pyrroles.
Group
Log CFU/Lung
Entry
Compound
MIC (lg/mL)a
Untreated controls
INH controls
IIIa
6.53 6 0.16a
5.34 6 0.22a
5.01 6 0.73a
5.53 6 0.16b
1
2
3
4
5
6
7
IIIa
IIIb
IIIc
IIId
IIIe
IIIf
IIIg
3.2b
16
IIIg
32
>8
32
a Six mice/group.
b Four mice/group.
32
2.6c
tivity in vitro for both INH-susceptible and INH-resist-
ant strains.
a INH standard control MIC 0.06 lg/mL.
b Geometric mean of three determinations.
c Geometric mean of five determinations.
For testing of compounds IIIa and IIIg in vivo, we
used a short-course therapy model in mice, which has
previously been described in detail [32]. This method
provides an exacting test of the activities in vivo of
investigational drugs and has the benefit of cutting in
half the time required to produce data in animal studies,
compared to the more traditional murine models. Our
results are given in Table 4, in which the data are pre-
sented as the logarithms of bacterial colony-forming
units (log CFU) per mouse lung, with and without
administration of the candidate drugs. No overt toxicity
or ill effects were observed when the drugs were admin-
istered at the therapeutic doses (see Experimental). For
reference purposes, highly active drugs known to be
tuberculocidal, such as INH, generally display a drop of
one or more log CFU versus untreated controls when
examined in the model. On the other hand, less-active
drugs that have been characterized as tuberculostatic, as
opposed to tuberculocidal, such as p-aminosalicylic acid
(PAS), typically lead to a lowering of one-half log CFU
versus untreated controls (see Experimental). In compar-
ing the data for IIIa and IIIg with the untreated con-
trols, there has been a reduction of 1.00 log CFU or
more in each case, suggesting that both compounds IIIa
and IIIg are bactericidal in tuberculosis-infected mice.
Overall, the results in vitro and in vivo warrant further
development of the antitubercular properties of pyrroles
derived from aromatic acid hydrazides.
With respect to biological assessment, the pyrroles
displayed a range of activities against both laboratory
strains and drug-resistant clinical isolates of M. tubercu-
losis. Activities in vitro were initially determined for the
pyrroles as minimum inhibitory concentration (MIC)
values against M. tuberculosis strain Erdman (Mtb Erd-
man). The MIC represents the minimum concentration
of the compound necessary to inhibit growth. Mtb Erd-
man is a fully drug-susceptible laboratory strain, often
used in primary antimycobacterial assays. The determi-
nations were referred to the known antitubercular INH
as a standard control. Individual MIC values are
reported in Table 2. The geometric mean of the MICs
for the pyrroles tested was 12 lg/mL. For comparison
purposes, tuberculosis drugs currently used in the clinic
have MIC values typically ranging from 0.015 lg/mL
(rifabutin) to 50 lg/mL (pyrazinamide) [31]. Of particu-
lar interest were compounds IIIa and IIIg with the low-
est MICs of 3.2 and 2.6 lg/mL, respectively. These
highly effective pyrroles were selected for further evalu-
ation in a panel of three clinical strains isolated from
patients showing significant drug resistance to INH, and
the results are shown in Table 3. Compound IIIa was
slightly more effective than INH itself, and compound
IIIg clearly maintained its strong potency in the drug-re-
sistant bacteria, demonstrating essentially the same ac-
In conclusion, we have found that 1-acylamino-2,5-
dimethylpyrroles, of interest in the investigation of het-
erocyclic structures for their antitubercular activity, can
be conveniently prepared from the reaction of acid
hydrazides with hexane-2,5-dione in water or ethanol
without an acid catalyst. In each case, the process leads
to a single pyrrole in pure form, product isolation is
straightforward, and the resulting compound is readily
identified on the basis of highly characteristic spectro-
metric features. Some members of this class have useful
activities against drug-resistant tuberculosis in vitro and
offer excellent protection in an animal model of the
disease.
Table 3
In vitro activities: Minimum inhibitory concentrations in
drug-resistant clinical isolates.
Clinical isolates
Entry
1
Compound
303
1889
35829
IIIa
4
8
4
4
1
8
8
>8
2
INH control
IIIg
INH control
2
2
64
>64
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

710
M. J. Hearn, M. F. Chen, M. S. Terrot, E. R. Webster, and M. H. Cynamon
Vol 47
EXPERIMENTAL
PAS were dissolved in distilled water. Stock solutions were fil-
ter-sterilized by passage through a membrane filter 0.22-lm
pore size and stored at -20ꢀC until use. The drugs were pre-
pared each morning, before experimentation. With respect to
testing against these isolates, the MICs of all antimicrobial
agents were determined in modified 7H10 broth (7H10 agar
formulation with agar and malachite green omitted; pH 6.6)
supplemented with 10% Middlebrook oleic acid-albumin-dex-
trose-catalase (OADC) enrichment (Difco Laboratories,
Detroit, MI) and 0.05% Tween 80 [33]. The activities of the
antimicrobial agents were determined by a broth dilution
method [34]. The organism was grown in the modified 7H10
broth with 10% OADC enrichment and 0.05% Tween 80 on a
rotary shaker at 37ꢀC for 5 days. The culture suspension was
diluted in modified 7H10 broth to yield 100 Klett units/mL
(Photoelectric Colorimeter, Manostat Corporation, New York,
NY), or approximately 5 ꢂ 107 CFU/mL. The size of the inoc-
ulum was determined by titration and counting from triplicate
7H10 agar plates (BBL Microbiology Systems, Cockeysville,
MD) supplemented with 10% OADC enrichment. The plates
were incubated at 37ꢀC in ambient air for 4 weeks before
counting of the colonies. MIC values are reproducible to
within 1–2 dilutions. Results in vitro (strain H37Rv) were also
determined according to the fully-documented protocols of the
Tuberculosis Antimicrobial Acquisition and Coordinating Fa-
cility, of the National Institutes of Health [35].
In brief, for the short-course therapy studies in vivo, 4-
week-old female C57BL/6 mice (Charles River, Wilmington,
MA) were infected intranasally with approximately one million
viable M. tuberculosis organisms. As a negative control,
groups of infected but untreated mice were sacrificed at the
initiation of therapy. The known antitubercular drug INH was
used as a positive control. Treatment began 1 day post-infec-
tion and was administered for 2 days. The agents were intro-
duced by gavage: the INH control and a given pyrrole were
dosed daily at 25 and 100 mg/kg of body weight, respectively.
No overt toxicity or ill effects were noted at these doses. Mice
were sacrificed by carbon dioxide inhalation 3 days post-infec-
tion. Their right lungs were removed aseptically and were
ground in a tissue homogenizer (IdeaWorks! Laboratory Devi-
ces, Syracuse, NY). The number of viable organisms was
determined by titration on 7H10 agar plates. The plates were
incubated at 37ꢀC in ambient air for 4 weeks before counting
of the bacterial colonies. Data are presented as the logarithms
of colony-forming units. The complete procedure for the short-
course therapy model has been reported [32] in detail. As
standards of reference, such tuberculocidal drugs as INH gen-
erally display a drop of one or more log CFU versus untreated
controls in the model. By way of contrast, PAS is tuberculo-
static, as distinct from tuberculocidal, typically leading to a
falling-off of one-half log CFU versus untreated controls. A
typical data set for PAS is as follows: dose 500 mg PAS/kg/
day, six mice per group, untreated controls 5.78 6 0.38 log
CFU, PAS 5.32 6 0.22 log CFU, positive control (INH) 4.44
6 0.62 log CFU. The use of experimental animals complied
with institutional policies and federal guidelines.
General methods. Elemental analyses were carried out by
Galbraith Laboratories, Knoxville, TN. Melting points (mp;
ꢀC) were taken in open capillary tubes using a Mel-Temp ap-
paratus (Laboratory Devices, Cambridge, MA) and are cor-
rected. Infrared (IR) spectra were recorded on a Perkin-Elmer
1600 Fourier transform spectrometer as KBr pellets or Nujol
mulls or on a Perkin-Elmer Spectrum One Fourier transform
spectrophotometer fitted with a universal attenuated total re-
flectance sampling accessory, reported in wavenumbers (m,
cm–1). Most reactants, reagents, and solvents were obtained
from Aldrich Chemical Company (Milwaukee, WI) and Lan-
caster Synthesis (Windham, NH) and were used as received.
Methyl 4-aminosalicylate was purchased from Nantong Chang
Chemicals, Peking, China. Nuclear magnetic resonance (NMR)
spectra were taken on a Bruker 300 Fourier transform instru-
ment as dilute solutions in dimethyl sulfoxide-d6 (DMSO-d6)
or chloroform-d, recorded at 300 MHz (1H NMR) or 75 MHz
(13C NMR) and are reported in parts per million delta (d)
downfield from internal tetramethylsilane as reference, with
coupling constants given in cycles per second (cps). In some
proton spectra, only signals in the region 0–10 ppm are reported.
Appropriate solvent blanks were recorded to account for water
and DMSO. Gas chromatography (GC) and low-resolution mass
spectrometry (LR-MS) analyses were recorded with an HP 5890
Series II Plus gas chromatograph, using a crosslinked 5% di-
phenyl- 95% dimethylsiloxane column from Hewlett Packard,
and an HP 5972 electrical ionization mass detector. High-resolu-
tion mass spectroscopy (HR-MS) data were obtained using the
JEOL HX-110 double focusing mass spectrometer of the Michi-
gan State University Mass Spectrometry Facility, courtesy of Pro-
fessor D. Gage and Ms. B. Chamberlin. This magnetic sector
instrument is equipped with a fast atom bombardment (FAB) ion-
ization source. It is capable of performing high-resolution (peak
matching) and tandem mass spectrometry. Safety notes: gloves
were worn during the chemical syntheses, and the reactions were
carried out in the hood. In general, any scale up of preparations
of compounds with relatively high proportions of nitrogen and
oxygen was done with due caution. No specific safety problems
were encountered with the methods given below. No attempt was
made to optimize yields. To monitor the progress of the pyrrole-
forming reactions, a thin-layer chromatography (TLC) system
was devised. In general, we used plastic sheets coated with silica
gel (Merck-Darmstadt), eluted with absolute ethanol, with starting
materials the subjects of parallel reference runs. Details are given
in individual procedures, where applicable, but typical Rf values
were 0.5 for the acid hydrazide and 0.8 for the acylaminopyr-
roles. The dione was not observed in these runs.
Biological assessments. M. tuberculosis ATCC 35801
(strain Erdman) was obtained from the American Type Culture
Collection (ATCC, Manassas, VA). Drug-resistant clinical iso-
lates of M. tuberculosis (isolates 303, 1889, and 35829) were
kindly provided by Dr. Sheldon Morris, United States Food
and Drug Administration. The mutational origins of the drug
resistance of these organisms have been characterized and data
are available from the authors upon request. PAS was a gift of
Jacobus Pharmaceutical Company. INH was purchased from
Sigma Chemical Company (St. Louis, MO). For testing, a
given investigational compound was dissolved in dimethyl
sulfoxide and subsequently diluted in distilled water. INH and
1-(4-Pyridoyl)amino-2,5-dimethylpyrrole (IIIa). Isonicotinic
acid hydrazide (0.82 g, 6 mmol) was dissolved in deionized
distilled water (25 mL). 2,5-Hexanedione (0.68 g, 6 mmol)
was added and the mixture was heated at a gentle boil with
reflux for 2 h. The mixture was then allowed to cool
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

May 2010
Preparation and Properties In Vitro and In Vivo of Antitubercular Pyrroles
711
overnight, after which time crystals of IIIa formed in solution
and were collected by vacuum filtration. The mother liquor
was allowed to evaporate on a watch glass and the resulting
crystals were collected separately. The total yield was 67%,
mp 151–153ꢀC; IR 3224, 3041, 2923, 1673, 1595, 1537, 1490,
allowed to cool and placed on ice. To the reaction mixture on
ice was added 50 mL of cold water, provoking the gradual
precipitation of IIId. An additional 100 mL of cold water was
necessary to keep the solution thin enough to pour into the fil-
ter. Vacuum filtration of the precipitate yielded tiny, fluffy,
off-white crystals, 1.25 g (91%); mp 187–189ꢀC; IR 3281,
2980, 2935, 1665, 1610, 1531, 1491, 1324, 1301, 1276; 1H
NMR (DMSO-d6) d 11.2 (s, 1 H, quenched with D2O), 7.9 (d,
J ¼ 6 cps, 2 H), 7.4 (d, J ¼ 6 cps, 2 H), 5.7 (s, 2 H), 2.4 (s, 3
H), 2.0 (s, 6 H); 13C NMR (DMSO- d6) d 166, 143, 130, 129,
128, 127, 103, 22, 11; major fragments in LR-MS m/z 119, 94;
HR-MS (fast atom bombardment MHþ) calculated for
C14H17N2O 229.1341, found 229.1337.
1
1410, 1322, 1289; H NMR (DMSO-d6) d 11.5 (s, 1 H, disap-
peared after D2O shake), 8.8 (d, J ¼ 6 cps, 2 H), 7.9 (d, J ¼ 6
cps, 2 H), 5.7 (s, 2 H), 2.0 (s, 6 H); 13C NMR (DMSO-d6) d
165, 151, 139, 127, 122, 104, 11.
Anal. Calcd. for C12H13N3O: C, 66.96; H, 6.09. Found: C,
66.84; H, 6.18.
1-Benzoylamino-2,5-dimethylpyrrole (IIIb). Benzoic acid
hydrazide (0.82 g, 6 mmol) was heated to boiling in 35 mL
deionized, distilled water with reflux. 2,5-Hexanedione (0.82 g,
7.2 mmol) was dissolved in 10 mL deionized, distilled water and
added drop wise to the reflux flask without ever halting boiling.
Reflux continued for 23 h. After cooling, crystals of IIIb
appeared in solution and were collected by vacuum filtration
(89%). A second crop of crystals was harvested by evaporation
of the mother liquor (combined yield 97%). After recrystalliza-
tion in a mixed system of ethanol/water, crystals were clean,
very fine, and just slightly off-white in color, mp 175–177ꢀC; IR
3261, 3061, 2919, 1684, 1602, 1580, 1536, 1488, 1440, 1282;
1H NMR (DMSO-d6) d 11.3 (s, 1 H, quenched by D2O), 8.0 (d,
second peak split, J ¼ 6, 2 cps, 2 H), 7.7 (tt, J ¼ 6, 2 cps, 1 H),
7.5 (t, J ¼ 6 cps, 2 H), 5.7 (s, 2 H), 2.1 (s, 6 H); 13C NMR
(DMSO-d6) d 166, 133, 132, 129, 128, 127, 103, 11; major frag-
ments in LR-MS m/z 105, 94; HR-MS (fast atom bombardment,
MHþ) calculated for C13H15N2O 215.1184, found 215.1174.
Anal. Calcd. for C13H14N2O: C, 72.87; H, 6.59. Found: C,
72.69; H, 6.61.
Anal. Calcd for C14H16N2O: C, 73.65; H, 7.06. Found: C,
73.76; H, 7.13.
1-(4-Bromobenzoyl)amino-2,5-dimethylpyrrole (IIIe). 4-Bro-
mobenzoic hydrazide (1.29 g, 6 mmol) was reacted with 2,5-
hexanedione (0.82 g, 7.2 mmol) in the manner described for
the synthesis of IIIc. After 1 h of reflux, a fine suspended solid
clouded the mixture, which the addition of 100 mL ethanol did
not clear. The solid was not filterable from the reaction mix-
ture. TLC analysis (silica gel, absolute ethanol) revealed a
mixture of three species: 4-bromobenzoic hydrazide (Rf 0.5),
IIIe (Rf 0.8), and a third compound (Rf 0.7) postulated to be a
reactive intermediate, as suggested in the work of Amarnath
et al. [36] for a different family of pyrroles. After 27 h of
reflux, the solid had disappeared, and TLC monitoring began
to show a preponderance of IIIe, with small amounts of the
other two compounds still present. Reflux was stopped after 47
h, no solid having reappeared in the intervening day. Crystals
were precipitated from solution by the addition of cold water
to the reaction mixture on ice, with a total yield of 1.34 g
(77%); mp 207–210ꢀC; IR 3246, 1660, 1588, 1534, 1521,
1-(4-Chlorobenzoyl)amino-2,5-dimethylpyrrole (IIIc). 4-
Chlorobenzoic hydrazide (1.02 g, 6 mmol) was dissolved in
absolute ethanol (100 mL) and heated to a gentle boil with reflux.
2,5-Hexanedione (0.82 g, 7.2 mmol) was dissolved in absolute
ethanol (10 mL) and added dropwise to the hydrazide solution.
Reflux was continued for 50 h, during which time no solid pre-
cipitated from solution. TLC was used to monitor reaction pro-
gress and showed gradual progression from 4-chlorobenzoic hy-
drazide (Rf 0.5) to IIIc (Rf 0.8). The mixture was cooled to room
temperature and kept over night. Excess ethanol was boiled off
and cold water added, causing the precipitation of white solid
IIIc. After vacuum filtration and drying, this solid dissolved eas-
ily and cleanly in DMSO, yield 69%, consistently isolated as the
hemihydrate and analyzed as such, mp 185ꢀC with decomposi-
tion; IR 3449, 3274, 2998, 2961, 1656, 1623, 1594, 1521, 1483,
1
1465, 1377, 1322, 1300, 1268; H NMR (DMSO-d6) d 11.3 (s,
1 H, quenched with D2O), 7.9 (dd, J ¼ 6, 2 cps, 2 H), 7.8 (dd,
J ¼ 6, 2 cps, 2 H), 5.7 (s, 2 H), 2.0 (s, 6 H); 13C NMR
(DMSO-d6) 165, 132, 131, 130, 127, 126, 104, 11; major frag-
ment in LR-MS m/z 94; HR-MS (fast atom bombardment
MHþ) calculated for C13H14N2OBr 293.0290, found 293.0281.
Anal. Calculated for C13H13N2OBr: C, 53.26; H, 4.47.
Found: C, 53.36; H, 4.57.
1-(4-Nitrobenzoyl)amino-2,5-dimethylpyrrole (IIIf). 4-Nitro-
benzoic hydrazide (1.09 g, 6 mmol) was reacted with 2,5-hex-
anedione (0.82 g, 7.2 mmol) following the procedure for the
synthesis of (IIIc). Solid clouded the reflux solution after 1 h;
addition of 100 mL ethanol did not improve homogeneity. Af-
ter 66 h, cloudiness persisted but TLC analysis revealed the
material present to be entirely IIIf, with no traces of starting
material or an intermediate. Reflux was stopped, and the solu-
tion was filtered while hot. After cooling, cold water was used
to provoke the precipitation of IIIf from the mother liquor,
with a yield of 1.16 g (75%); mp 218–221ꢀC (recrystallization
from ethanol and water); IR 3278, 1674, 1604, 1523, 1468,
1343, 1270; 1H NMR (DMSO-d6) d 11.3 (s, 1 H, quenched
with D2O), 8.4 (dd, J ¼ 6, 2 cps, 2 H), 8.2 (dd, J ¼ 6, 2 cps,
2 H), 5.7 (s, 2 H), 2.1 (s, 6 H); 13C NMR (DMSO-d6) d 165,
150, 138, 129, 127, 124, 104, 11; major fragment in LR-MS
m/z 94; HR-MS (fast atom bombardment MHþ) calculated for
C13H14N3O3 258.0879, found 258.0876.
1
1448, 1299, 1272; H NMR (DMSO-d6) d 11.3 (s, 1 H, quenched
with D2O), 8.0 (dd, J ¼ 6, 2 cps, 2 H), 7.6 (dd, J ¼ 6, 2 cps, 2
H), 5.7 (s, 2 H), 2.0 (s, 6 H); 13C NMR (DMSO-d6) d 165, 137,
131, 130, 129, 128, 127, 104, 11; major fragments in LR-MS m/z
139, 94; HR-MS (fast atom bombardment MHþ) calculated for
C12H14N2OCl: 249.0795, found 249.0801.
Anal. Calcd. for C12H13N2OCl x 0.5H2O: C, 60.58; H, 5.48.
Found: C, 60.29; H, 5.19.
1-(4-Toluoyl)amino-2,5-dimethylpyrrole (IIId). 4-Toluic
acid hydrazide (0.90 g, 6 mmol) was reacted with 2,5-hexane-
dione (0.82 g, 7.2 mmol), in a manner similar to that for IIIc.
Reflux continued for 70 h, during which TLC was used to
track reaction progress and no solid appeared in solution. After
70 h, TLC (silica gel, absolute ethanol) showed no remaining
traces of the 4-toluic hydrazide. The reaction mixture was
Anal. Calculated for C13H13N3O3: C, 60.23; H, 5.05. Found:
C, 60.30; H, 5.17.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

712
M. J. Hearn, M. F. Chen, M. S. Terrot, E. R. Webster, and M. H. Cynamon
Vol 47
[9] Miesel, L.; Rozwarski, D.; Sacchettini, J.; Jacobs, W. In
Genetics and Tuberculosis, Novartis Foundation Symposia; Chadwick,
D. J., Ed.; Wiley: Chichester, England, 1998; pp 209–227.
[10] Billington, D.; Coleman, M.; Ibiabuo, J.; Lambert, P.; Rath-
bone, D.; Tims, K. Drug Des Discov 1998, 15, 269.
1-(4-Aminosalicyloyl)amino-2,5-dimethylpyrrole (IIIg). 4-
Aminosalicylic acid hydrazide (ASAH) was obtained (58%) in
pure form from the hydrazinolysis of methyl 4-aminosalicylate
by a method that we have previously described in detail [37].
ASAH thus obtained (1.67 g, 10.0 mmoles) was weighed into a
100 mL round-bottom flask fitted for reflux with a temperature-
controlled heating mantle, reflux condenser, and magnetic stir-
rer. Deionized distilled water (25 mL) was added, and the mix-
ture was brought to the boil. Just below the boiling point, the
mixture was heterogeneous, but a clear solution was obtained at
the boiling point. 2,5-Hexanedione (1.14 g, 10.0 mmoles) was
added and the mixture was refluxed for 2 h. Heating was
stopped, and stirring was continued for an hour as the mixture
slowly cooled. The resulting white solid was filtered off to pro-
vide the title compound IIIg (0.722 g, 30%), mp 173–174ꢀC;
IR 3371, 3270, 1612, 1494, 1385, 1334, 1312, 1270, 1200,
1161, 1012, 967, 833, 786, 757, 692; 1H NMR (DMSO-d6) d
12.1 (s, 1H), 10.9 (s, 1H), 7.6 (d, J ¼ 7 cps, 1H), 6.2 (d, J ¼ 7
cps, 1H), 6.1–6.0 (overlapping s, 3H), 5.7 (s, 2H), 2.0 (s, 6H);
13C NMR (DMSO-d6) d 169, 162, 155, 129, 127, 106, 103,
101, 99, 11; HR-MS (fast atom bombardment MHþ) calculated
for C13H16N3O2 246.1243, found 246.1244.
[11] Pathak, A.; Pathak, V.; Seitz, L.; Suling, W.; Reynolds, R.
J Med Chem 2004, 47, 273.
[12] Janin, Y. Bioorg Med Chem 2007, 15, 2479.
[13] de Souza, M.; Pais, K.; Kaiser, C.; Peralta, M.; Ferreira, M.
L.; Lourenc¸o, M. C. S. Bioorg Med Chem 2009, 17, 1474.
[14] Hearn, M. J.; Webster, E. R.; Cynamon, M. H. J Heterocycl
Chem 2005, 42, 1225.
[15] Hearn, M. J.; Chanyaputhipong, P. J Heterocycl Chem
1995, 32, 1647.
[16] Youmans, G.; Doub, L.; Youmans, A. The Bacteriostatic
Activity of 3500 Organic Compounds for Mycobacterium tuberculosis
var. hominis; Chemical-Biological Coordination Center, National
Research Council: Washington, DC, 1953; compound 2603.
[17] Cymerman-Craig, J.; Willis, D. J Chem Soc 1955, 4315.
[18] Gazave, J.; Buu-Hoi, N.; Xuong, N.; Mallet, J.; Pillot, J.;
Savel, J.; Dufraisse, G. Therapie 1957, 12, 486.
[19] Yale, H.; Losee, K.; Martins, J.; Holsing, M.; Perry, F.;
Bernstein, J. J Am Chem Soc 1953, 75, 1933.
Anal. Calcd. for C13H15N3O2: C, 63.66; H, 6.16. Found: C,
63.48; H, 6.28.
[20] Zsolnai, T. Zentralblatt fuer Bakteriologie, Parasitenkunde,
Infektionskrankheit und Hygiene, Abteilung 1: Medizinisch-Hygienische
Bakteriologie, Virusforschung und Parasitologie, Orginale 1959, 175, 269.
[21] Hearn, M. J.; Terrot, M. S. Abstracts of Papers, American
Chemical Society 221st National Meeting, San Diego, California,
2001, CHED-156.
Acknowledgments. The authors thank Jacobus Pharmaceutical
Company, Princeton, New Jersey, for their generous support of
this work and Dr. David Jacobus for valuable discussions. This
work was also supported by the Global Alliance for Tuberculosis
Drug Development and by grant 1 R15 AI48397-01 from the Divi-
sion of Acquired Immunodeficiency Syndrome, National Institute
of Allergy and Infectious Diseases (NIAID), National Institutes of
Health (NIH). High resolution mass spectra were determined at the
NIH Mass Spectrometry Facility at Michigan State University,
East Lansing, Michigan, USA. The authors acknowledge Dr. Shel-
don Morris, United States Food and Drug Administration, for the
clinical isolates of drug-resistant M. tuberculosis. The authors
thank the staff of the Tuberculosis Antimicrobial Acquisition and
Co-ordinating Facility, administered by the Southern Research
Institute, Birmingham, Alabama, USA, under a research and devel-
opment contract with NIAID, NIH. MST is grateful to the Sequella
Global Tuberculosis Foundation for a student internship under the
direction of Dr. Clifton E. Barry III, NIH.
[22] Bijev, A. Arzneimittelforschung 2009, 59, 34.
[23] Arora, S.; Sinha, N.; Sinha, R.; Uppadhayaya, R.; Modak,
V.; Tilekar, A. 44th Interscience Conference on Antimicrobial Agents
and Chemotherapy, Washington, DC, 2004; Abstract Number F-1115.
[24] Biava, M.; Porretta, G. C.; Poce, G.; Logu, A. D.; Saddi, M.;
Meleddu, R.; Manetti, F.; Rossi, E. D.; Botta, M. J Med Chem 2008, 51, 3644.
[25] Biava, M.; Porretta, G. C.; Poce, G.; Supino, S.; Deidda,
D.; Pompei, R.; Molicotti, P.; Manetti, F.; Botta, M. J Med Chem
2006, 49, 4946.
[26] Paal-Knorr Synthesis. In The Organic Chemistry Portal,
2009. Available at: www.organic-chemistry.org/namedreactions/paal-
knorr-pyrrole-synthesis.shtm. Accessed July 31, 2009.
[27] Bellina, F.; Rossi, R. Tetrahedron 2006, 62, 7213.
[28] Buu-Hoi, N.; Xuong, N. J Org Chem 1955, 20, 850.
[29] Banik, B.; Samajdar, S.; Banik, I. J Org Chem 2004, 69, 213.
[30] Just, P. E.; Chane-Ching, K. I.; Lacaze, P. C. Tetrahedron
2002, 58, 3467.
[31] Global Alliance for Tuberculosis Drug Development.Tuber-
culosis 2008, 88, 85.
REFERENCES AND NOTES
[32] Shoen, C.; DeStefano, M.; Sklaney, M.; Monica, B.; Slee,
A.; Cynamon, M. J Antimicrob Chemother 2004, 53, 641.
[33] Vestal, A. L. Procedures for the Isolation and Identification
of Mycobacteria; Laboratory Division, National Communicable Dis-
ease Center, Public Health Service: Atlanta, Georgia, 1969; publication
number 1995, pp 113–115.
[1] Di Perri, G.; Bonora, S. J Antimicrob Chemother 2004, 54,
593.
[2] World Health Organization. TB: A global emergency; Ge-
neva, Switzerland, 1994; Report No. 14977.
[3] Hampton, T. JAMA 2004, 291, 2529.
[4] Global Alliance for Tuberculosis Drug Development.
Scientific blueprint for TB drug development, 2001. Available at:
www.tballiance.org/downloads/publications/TBA_Scientific_Blueprint.pdf.
Accessed July 6, 2009.
[34] Wong, C. S.; Palmer, G. S.; Cynamon, M. H. J Antimicrob
Chemother 1988, 22, 863.
[35] Secrist, J.; Anathan, S.; Kwong, C.; Maddry, J.; Reynolds,
R.; Poffenberger, A.; Michael, M.; Miller, L.; Krahenbuhl, J.; Adams,
L.; Biswas, A.; Franzblau, S.; Rouse, D.; Winfield, D.; Brooks, J.;
Orme, I. Antimicrob Agents Chemother 2001, 45, 1943.
[36] Amarnath, V.; Amarnath, K.; Valentine, W.; Eng, M.; Gra-
ham, D. Chem Res Toxicol 1995, 8, 234.
[5] Cahn, P.; Perez, H.; Ben, G.; Ochoa, C. J Int Assoc Physi-
cians AIDS Care 2003, 2, 106.
[6] Laughon, B. Curr Topics Med Chem 2007, 7, 463.
[7] Barry, C. E. Issues Infect Dis 2003, 2, 137.
[8] Davies, P.; Yew, W. Expert Opin Ther Targets 2003, 12,
1297.
[37] Hearn, M. J.; Chen, M. F.; Cynamon, M. H.; Wang’ondu,
R.; Webster, E. R. J Sulfur Chem 2006, 27, 149.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet