Extending pummerer reaction chemistry: (±)-dibromoagelaspongin synthesis and related studies

Article abstract of DOI:10.1021/jo900283g

The sponge-derived alkaloid dibromoagelaspongin was prepared from a dihydrooroidin derivative by exploiting the Pummerer reaction twice in succession. Oxidative cyclization of the substrate's pyrrole-2-carboxamide function into the imidazole moiety was ac

Full text of DOI:10.1021/jo900283g

Extending Pummerer Reaction Chemistry:
(()-Dibromoagelaspongin Synthesis and Related Studies
Ken S. Feldman* and Matthew D. Fodor
Department of Chemistry, The PennsylVania State UniVersity, UniVersity Park, PennsylVania 16802
ksf@chem.psu.edu
ReceiVed February 10, 2009
The sponge-derived alkaloid dibromoagelaspongin was prepared from a dihydrooroidin derivative by
exploiting the Pummerer reaction twice in succession. Oxidative cyclization of the substrate’s pyrrole-
2-carboxamide function into the imidazole moiety was achieved in a regiospecific manner to establish
both C-N bonds to C(6) of the target.
Introduction
(2)3 is continually increasing as bioprospecting efforts expand
in kind. Monomeric, dimeric, and tetrameric cyclized derivatives
of 1/2 have been discovered to date,1,4 and there is no reason
to suspect that this family has been exhausted. The monomeric
cyclization/oxidation products of 1/2 that feature connections
between the imidazole and pyrrole cores represent one large
subclass of these species, as illustrated in Figure 1. Isomeriza-
tion/cyclization from 1 or oxidative cyclization from 2 leads
almost exclusively to attachment of N(14) to C(10) of the
2-aminoimidazole ring (cf. 3 and 4) with concomitant bonding
of the nucleophilic pyrrole to the nascent electrophilic C(6) of
a putative imidazoline intermediate (phakellin numbering).
Theme and variation are manifest through reaction of either the
pyrrole’s C(3) position (cf. 4, the isophakellin skeleton) or the
nitrogen (cf. 3, the phakellin skeleton), and by the incorporation
of heteroatoms at C(11) and/or C(12) of the precursor oroidin
skeleton. One sponge isolate, dibromoagelaspongin (6),5 stands
alone among these monomeric oroidin/dihydrooroidin deriva-
tives in that the initial connection between N(14) and the
2-aminoimidazole ring is forged to C(6) instead of C(10). In
addition, the three heteroatoms (nitrogens) attached to C(6) and
the two heteroatoms attached to C(10) require that cyclization
of oroidin must be accompanied by a formal oxidation as well,
a second point of distinction between 6 and the more numerous
The sponge-derived pyrrole-imidazole alkaloids represent yet
another example of Nature’s capacity for generating astonishing
structural diversity from a simple precursor.1 The roster of
cyclization/oxidation products formally derived from either of
the naturally occurring progenitors oroidin (1)2 or dihydrooroidin
(1) (a) Al Mourabit, A.; Potier, P. Eur. J. Org. Chem. 2001, 23, 7–243. (b)
Hoffmann, H.; Lindel, T. Synthesis 2003, 1753–1783. (c) Jacquot, D. E. N.;
Lindel, T. Curr. Org. Chem. 2005, 9, 1551–1565. (d) Weinreb, S. M. Nat. Prod.
Rep. 2007, 24, 931–948. (e) Pettit, G. R.; McNulty, J.; Herald, D. L.; Doubek,
D. L.; Chapuis, J.-C.; Schmidt, J. M.; Tackett, L. P.; Boyd, M. R. J. Nat. Prod.
1997, 60, 180–183. (f) Gautschi, J. T.; Whitman, S.; Holman, T. R.; Crews, P.
J. Nat. Prod. 2004, 67, 1256–1261.
(2) From Agelas oroides: (a) Forenza, S.; Minale, L.; Riccio, R.; Fattorusso,
E. J. Chem. Soc., Chem. Commun. 1971, 112, 9–1130. From Axinella Verrucosa
and Acanthella aurantiaca: (b) Cimino, G.; De Rosa, S.; De Stefano, S.;
Mazzarella, L.; Puliti, R.; Sodano, G. Tetrahedron Lett. 1982, 23, 767–768. From
Pseudaxinyssa cantharella: (c) De Nanteuil, G.; Ahond, A.; Guilhem, J.; Poupat,
C.; Tran Huu Dau, E.; Potier, P.; Pusset, M.; Pusset, J.; Laboute, P. Tetrahedron
1985, 41, 6019–6033. From Agelas mauritiana: (d) Tsukamoto, S.; Kato, H.;
Hirota, H.; Fusetani, N. J. Nat. Prod. 1996, 59, 501–503. From Acanthella carteri:
(e) Mancini, I.; Guela, G.; Amade, P.; Roussakis, C.; Francesco, P. Tetrahedron
Lett. 1997, 38, 6271–6274. From Agelas clathrodes, Agelas conifera, Agelas
dispar, and Agelas longissima: (f) Cafieri, F.; Carnuccio, R.; Fattorusso, E.;
Taglialatela-Scafati, O.; Vallefuoco, T. Bioorg. Med. Chem. Lett. 1997, 7, 2283–
2288. From Stylissa carteri: (g) Eder, C.; Proksch, P.; Wray, V.; Steube, K.;
Bringmann, G.; Van Soest, R. W. M.; Sudarsono; Ferdinandus, E.; Pattisina,
L. A.; Wiryowidagdo, S.; Moka, W. J. Nat. Prod. 1999, 62, 184–187. From
Agelas wiedenmeyeri: (h) Assmann, M.; Lichte, E.; Van Soest, R. W. M.; Ko¨ck,
M. Org. Lett. 1999, 1, 455–457. From Agelas sVentres: (i) Assmann, M.; Zea,
S.; Ko¨ck, M. J. Nat. Prod. 2001, 64, 1593–1595. From Stylissa massa: (j)
Tasdemir, D.; Mallon, R.; Greenstein, M.; Feldberg, L. R.; Kim, S. C.; Collins,
K.; Wojciechowicz, D.; Mangalindan, G. C.; Concepcion, G. P.; Harper, M. K.;
Ireland, C. M. J. Med. Chem. 2002, 45, 529–532. From Axinella damicornis:
(k) Hammami, S.; Ben Jannet, H.; Ciavatta, M. L.; Mollo, E.; Cimino, G.; Mighri,
Z. J. Soc. Algerienne Chim. 2006, 16, 79–89. From Stylissa caribica: (l)
Mohammed, R.; Peng, J.; Kelly, M.; Hamann, M. T. J. Nat. Prod. 2006, 69,
1739–1744.
(3) From Agelas nemoecinata: Pedpradab, S. Ph.D. thesis, Heinrich-Heine
Universita¨t, Du¨sseldorf, 2005.
(4) Grube, A.; Ko¨ck, M. Org. Lett. 2006, 8, 4675–4678.
(5) (a) Fedoreyev, S. A.; Ilyin, S. G.; Utkina, N. K.; Maximov, O. B.;
Reshetnyak, M. V.; Antipin, M. Y.; Struchkov, Y. T. Tetrahedron 1989, 45,
3487–3492. (b) Pin˜a, I. C.; White, K. N.; Cabrera, G.; Rivero, E.; Crews, P. J.
Nat. Prod. 2007, 70, 613–617.
10.1021/jo900283g CCC: $40.75
Published on Web 03/30/2009
2009 American Chemical Society
J. Org. Chem. 2009, 74, 3449–3461 3449

Feldman and Fodor
SCHEME 1. Speculation on the Subject of
Dibromoagelaspongin Biosynthesis
SCHEME 2. Alternative Biosynthesis Proposal for
Dibromoagelaspongin, Featuring Successive Formation of
Two Electrophilic Triazafulvene Units from
2-Aminoimidazole
FIGURE 1. Members of the tetracyclic family of monomeric oroidin/
suffers formal dehydration via initial N(14)/C(6) condensation
followed by N(1)/C(6) closure to complete the tetracyclic
skeleton.5a Oxidation at C(10) by undescribed means then
completes the biosynthesis. Maximov et al.’s proposal also
leaves open the possibility that dibromophakellin itself serves
dihydrooroidin derivatives.
phakellin/isophakellin species 3 and 4, respectively. In fact, the
triaminomethane unit within 6 has been described only rarely
in the natural products arena.6 Dibromoagelaspongin was
coisolated with a methyl ether derivative (OCH3 instead of OH),
although that species is likely just an artifact of the methanol-
mediated isolation procedure. Nevertheless, this observation
points to the facility of nucleophile exchange at C(10), which
in turn may inform synthesis planning for accessing the target
6. The hydrogenolysis product of 6 displays no optical activity
and is described as a racemate, although no optical activity
measurements were reported for dibromoagelaspongin itself.
This observation suggests, but does not demand, that 6 is
racemic in the sponge as well.
Many members of the tetracyclic oroidin-derived family of
sponge metabolites exhibit promising biological activity against
a range of diseases.1b For example, whereas dibromophakellin
(3a) generated only modest cytotoxicity data (ED50s of 15-20
µg/mL against several human cancer cell lines), 3a’s naturally
occurring formal hydrolysis product dibromophakellstatin (urea
instead of guanidine, not shown) displayed potent activity (ED50s
< 1 µg/mL against these same cell lines).1e The N(7) methylated
version 3c inhibits 12-human lipoxygenase (10 µM), an enzyme
implicated in cell proliferation.1f No biological activity studies
for dibromoagelaspongin (8) have been reported to date.
Biosynthesis postulates for this structurally unique member
6 of the oroidin family have followed two distinct lines of
thought, Scheme 1. The authors who first isolated and character-
ized 6 postulate that initial hydration/tautomerization of oroidin
proceeds to afford an imidazolinone intermediate 7, which
as
a precursor to 6 via a series of oxidations and
rearrangements.5a An alternative biosynthesis rationale is offered
by Potier and Al Mourabit,1a who consolidate their dibromo-
agelaspongin and dibromophakellin thinking by citing a common
nine-membered ring intermediate 8 for both targets. This species
is presumed to originate through nucleophilic addition of the
pyrrole’s N(1) to a transiently electrophilic C(6) derived from
the 2-amino-4-alkenyl-imidazole portion of oroidin. Subsequent
transannular cyclization of N(14) into (an sp2 version of) C(10)
would deliver the phakellin structure, whereas similar cyclization
of N(14) into C(6) as shown provides the agelaspongin skeleton.
Once again, a subsequent oxidation at C(10) by undescribed
means would formulate the complete natural product 6.
Neither of these biosynthesis proposals provides much
guidance for developing a rational chemical synthesis plan for
dibromoagelaspongin, given the mysterious C(10) oxidation that
is required in each case. However, an alternative biosynthesis
scheme, which utilizes two successive oxidative cyclizations
from dihydrooroidin (or an isomerization/cyclization followed
by an oxidative cyclization from oroidin), can be envisioned,
Scheme 2. In this hypothesis, the triazafulvene electrophile
illustrated in both 9 and 13 plays a prominent role in facilitating
bond formation to assemble the core tetracycle. The formation
of the distinctive N(14)/C(6) bond might occur directly as per
9 f 11, or indirectly via either N or C migration (N shown for
simplicity) within the intermediate phakellin-like spirocycle 10.
Both of these conceptualizations raise the specter of competition
with simple formation of the phakellin structure by addition of
the pyrrole’s N(1) to the C(6) electrophile in 10. Thus, an
(6) Sunazuka, T.; Shirahata, T.; Tsuchiya, S.; Hirose, T.; Mori, R.; Harigaya,
Y.; Kuwajima, I.; Omura, S. Org. Lett. 2005, 7, 941–943.
3450 J. Org. Chem. Vol. 74, No. 9, 2009

Extending Pummerer Reaction Chemistry
SCHEME 3. Horne and Olofson’s Approach to
Dibromoagelaspongin
SCHEME 4. Biomimetic Retrosynthetic Analysis for
Dibromoagelaspongin (6) Based upon the Biosynthetic
Proposal Illustrated in Scheme 2
interesting aspect of this hypothesis is the basis for selective
agelaspongin construction rather than phakellin formation; a
priori it is not possible to invoke any explanation more
thoughtful than enzymatic intervention. As the chemistry
transpired (vide infra), a decisive role for protecting group(s)
on N(1) and N(9) in steering intermediates like 9/10 toward
the agelaspongin skeleton was revealed.
Horne and Olofson first considered this alternative biosyn-
thesis proposal for dibromoagelaspongin and it guided their
preliminary synthesis efforts in this area, Scheme 3.7 Treatment
of dihydrooroidin•HCl with base and the oxidant Br2 furnished
a product following acidification that they tentatively identified
as the tricyclic species 14. Unfortunately, they were not able to
advance this species to the agelaspongin skeleton via subsequent
transformations. Dibromoagelaspongin synthesis based on the
biosynthetic blueprint of Scheme 2 is not necessarily contrain-
dicated by the unpromising results of Scheme 3, provided that
alternative approaches to the second cyclization step (e.g., 14
f 6) are available. Thus, the formulation of an alternative
bicyclic core to that contained in 14, which features different
and more reactive electrophilic functionality, might be sufficient
to sustain a synthesis plan based upon the overall oxidative
cyclization strategy.
A comprehensive synthesis program for the oroidin-based
alkaloids based upon the premise that a Pummerer oxidation
sequence might serve as a useful surrogate for 2-aminoimidazole
f triazafulvene oxidation has been initiated.8 Application of
this chemistry to the dibromoagelaspongin problem is illustrated
in retrosynthetic format, Scheme 4.9 The hypothetical biosyn-
thesis described in Scheme 2 serves as the guiding theme for
this presumably biomimetic approach to the target. In this
chemistry, the sulfur moiety in 21 serves as the initial locus of
oxidation, thereby minimizing selectivity problems with alterna-
tive oxidation sites within this electron rich substrate. Then,
through either a vinylogous or an additive Pummerer reaction
mechanism, the sulfur’s oxidation charge is translocated to either
C(6) or C(10) of the imidazole nucleus (cf. 19), where
nucleophilic addition by the pendant N(14) nitrogen ensues. In
the context of dibromoagelaspongin synthesis, both the R1
substituent on N(9) and protection of the pyrrole’s nitrogen
provide the key to steering the transformation toward one or
the other regiochemical outcome (C(6) vs C(10) attachment,
vide infra). The desired bicycle 18, reminiscent of the related
species 14 reported by Horne and Olofson, is now poised for a
second Pummerer-mediated oxidative cyclization sequence to
close the fourth ring and deliver the intact agelaspongin
tetracycle of 15. In this plan, selective generation of the reactive
diazacyclopentadienethionium ion of 16 from the 2-thioimida-
zole unit in 18 sets up the final C-N closure, and hopefully
unreactive intermediates like the iminoguanidine of 14 can be
avoided. Finally, removal of the R1 protecting group, replace-
ment of the product’s SR unit with NH2 and OH-for-X exchange
(recall the facility of dibromoagelaspongin methyl ether forma-
tion upon exposure of 6 to methanol during isolation) within
15 will complete the installation of the functionalized 2-ami-
noimidazoline moiety of the target dibromoagelaspongin.
Results and Discussion
The entire dibromoagelaspongin synthesis project evolved
from a serendipitous observation that emerged from attempts
at securing the isophakellin skeleton from a dihydrooroidin
precursor. Specifically, the hypothesis that simply protecting
the pyrrole’s nitrogen would steer cyclization to C(3) and the
isophakellin structure was tested, Scheme 5. Initially, a series
of N(1) protected dihydrooroidin analogues 24a-24e were
prepared by previously described methodology.8g Treatment of
the exemplary species 24c with aqueous HCl discharged the
dimethylaminosulfonyl (DMAS) protecting group to afford the
unprotected imidazole-2-sulfinate. Exposure of this sulfoxide
to standard Pummerer initiation conditions (Tf2O, i-Pr2NEt) did
not afford any cyclized material; rather, simple triflation of the
free NH of the imidazole moiety led to the only characterized
product, 25. In an attempted workaround, reaction of the sulfide
corresponding to the desulfamoylation product of 24c with
PhI(CN)OTf8c led to consumption of starting material without
formation of any characterizable products. Since the free NH
of this substrate appeared to be problematic, recourse was made
to the simple expedient of leaving the DMAS group in place
during the Pummerer chemistry. This plan was not without its
concerns as well, as a lack of a proton to lose from the imidazole
nitrogen implies that the electrophilic Pummerer intermediate
(7) Olofson, A. S. Ph.D. thesis, Columbia University, 1998.
(8) (a) Feldman, K. S.; Vidulova, D. B. Org. Lett. 2004, 6, 1869–1871. (b)
Feldman, K. S.; Karatjas, A. G. Org. Lett. 2004, 6, 2849–2852. (c) Feldman,
K. S.; Vidulova, D. B. Tetrahedron Lett. 2004, 45, 5035–5037. (d) Feldman,
K. S.; Vidulova, D. B.; Karatjas, A. G. J. Org. Chem. 2005, 70, 6429–6440. (e)
Feldman, K. S.; Skoumbourdis, A. P. Org. Lett. 2005, 7, 929–931. (f) Feldman,
K. S.; Karatjas, A. G. Org. Lett. 2006, 8, 4137–4140. (g) Feldman, K. S.;
Skoumbourdis, A. P.; Fodor, M. D. J. Org. Chem. 2007, 72, 8076–8086.
(9) Feldman, K. S.; Fodor, M. D. J. Am. Chem. Soc. 2008, 130, 14964–
14965.
J. Org. Chem. Vol. 74, No. 9, 2009 3451

Feldman and Fodor
SCHEME 5. Approach to the Isophakellin Skeleton Leads
Unexpectedly to the Bicyclic Core of the Agelaspongin
Skeleton
SCHEME 6. Formation of the Fully Deprotected Bicycle of
Dibromoagelaspongin
SCHEME 7. Second Serendipitous Reaction: Formation of
the Dibromoagelaspongin Tetracycle
(cf. 19, Scheme 4) might be doubly charged. Nevertheless, some
precedent for this type of doubly cationic intermediate is
known,10 and so Pummerer-based oxidative cyclizations with
24a-24c were explored. These dihydrooroidin derivatives in
particular were selected with the expectation that the bromination
pattern on the pyrrole ring might influence the nucleophilicity
at C(3) in an exploitable manner. Oxidative cyclization did
proceed smoothly but in modest yield for the cases where at
least one pyrrole bromide was present, 24b and 24c. Remark-
ably, the cyclization proceeded with N(14) f C(6) bond
formation rather than the expected N(14) f C(10) closure
required for the isophakellin objective. Unfortunately, the
pyrrole’s C(3) position did not engage any intermediate elec-
trophile in productive cyclization, thus thwarting any further
progress toward the isophakellin series. Attempted PhI(CN)OTf-
mediated oxidative cyclization of the sulfide corresponding to
24c led to recovered starting material. Nevertheless, the end of
the isophakellin project marked the beginning of the dibromo-
agelaspongin work, since the cyclization product 26c maps
nicely onto that particular oroidin-derived skeleton.
deprotected imidazole-2-sulfide 27 ready for subsequent oxida-
tive cyclization chemistry. Preliminary scouting experiments
with Stang’s reagent (PhI(CN)OTf) and either 27 or an analogue
with the SO2NMe2 group in place were not fruitful, leaving
sulfide-to-sulfoxide oxidation followed by Pummerer chemistry
as the most promising alternative.
Continuation of this chemistry within the context of dibro-
moagelaspongin synthesis requires an unprotected pyrrole
nitrogen, and the prospects for identifying reaction conditions
for the deprotection of the N(1) methyl substituent within 26c
without compromising the structural integrity of the remainder
of the molecule were not good. Therefore, the more readily
deprotected MOM and SEM ethers 24d and 24e, respectively,
were prepared and examined in a Pummerer-mediated oxidative
cyclization, Scheme 6. The SEM-protected species 24d was the
clear winner among the substrates examined, as it provided the
desired bicyclic product 26e in the highest yield. The SEM
protecting group could be removed in a two-step procedure
followed by DMAS cleavage with mild acid to furnish the fully
Oxidation of the sulfide 27 to the sulfoxide 28 was achieved
with mCPBA; inclusion of Na2CO3 was critical for obtaining
the product in good yield (Scheme 7). Unfortunately, treatment
of this sulfoxide with the standard Pummerer initiator Tf2O/i-
Pr2NEt led to no more than nitrogen triflation to form 29a/29b,
a result reminiscent of the earlier Pummerer failure with the
dihydrooroidin derivative 24c (Scheme 5, 24c f 25). Attempts
to prepare the N-SO2NMe2 protected variant of 28 by either
direct sulfamoylation of 28 or by oxidation of a sulfide substrate
32 bearing the intact N-SO2NMe2 group, failed. However, the
synthesis plan took an unexpected turn when alternative
oxidation conditions with 27 were examined as part of a general
survey of reagents. Thus, exposure of sulfide 27 to TFA and
H2O2 (f CF3CO3H)11 at elevated temperature led to the
formation of trace amounts of a tetracyclic species 31 that only
(10) (a) Saito, S.; Sato, Y.; Ohwada, T.; Shudo, K. J. Am. Chem. Soc. 1994,
116, 2312–2317. (b) Yokoyama, A.; Ohwada, T.; Shudo, K. J. Org. Chem. 1999,
64, 611–617. (c) Saitoh, T.; Ichikawa, T.; Horiguchi, Y.; Toda, J.; Sano, T. Chem.
Pharm. Bull. 2001, 49, 979–984. (d) Saitoh, T.; Shikiya, K.; Horiguchi, Y.; Sano,
T. Chem. Pharm. Bull. 2003, 51, 667–672.
(11) Venier, C. G.; Squires, T. G.; Chen, Y. Y.; Smith, B. F. J. Org. Chem.
1982, 47, 3773–3774.
3452 J. Org. Chem. Vol. 74, No. 9, 2009

Extending Pummerer Reaction Chemistry
SCHEME 8. Formation of the Dibromoagelaspongin
Tetracycle by Oxidative Cyclization
followed from comparison of its spectral data to those of 33a.
A discussion of the possible mechanism(s) of this transformation
is deferred until later in this account. The anticipated lability
of the halides in 33a/33b (cf. formation of dibromoagelaspongin
methyl ether by treatment of dibromoagelaspongin with metha-
nol)5 was realized by the ease with which 33a and 33b were
converted to either the methyl ether 34 or the free alcohol 31.
All that remained to complete the synthesis of dibromoag-
elaspongin was the aforementioned SPh f NH2 transformation,
and now sufficient material was available to systematically
explore several options. In fact, this functional group interchange
completed the conversion of a thiophenyl derivative of dibro-
mophakellin into dibromophakellin itself, as reported earlier.8g
Toward this end, conversion of the thioamidine of 31 into a
urea moiety using CAN or HgCl2 was explored. Unfortunately,
the success realized in the phakellin series with these conditions
was not matched in the dibromoagelaspongin case, as starting
material was recovered in both cases, even under forcing
conditions. Several direct displacements of the thiophenyl unit
in 31 with NH3, EtCO2NH4, or NaN3 were attempted, but these
efforts either returned sulfide or complete decomposition
resulted. Indirect methods relying on initial sulfur oxidation prior
to substitution held promise,15 but conditions/reagents to effect
this transformation with 31 or 34b (SPh f SOPh or SO2Ph)
could not be identified. Again, starting material or decomposition
resulted from treatment of 34b with mCPBA, H2O2, t-BuOOH,
DMDO, Oxone, or NaIO4. At this point, it became apparent
that both 31 and 34b suffered from a insurmountable blend of
global sensitivity coupled with thioamidine refractivity compared
to the phakellin series, and the target molecule dibromoagela-
spongin could not be reached via this chemistry.
differed from the target dibromoagelaspongin by a (nontrivial)
SPh f NH2 exchange. The structure of 31 was secured by
analysis of the 13C NMR spectrum and HMQC/HMBC correla-
tions (see structure 31 for key data), and by comparison of the
spectral data to those reported for dibromoagelaspongin methyl
ether.5b Efforts to improve the yield of 31 went unrewarded,
and the small amount of material in hand precluded extensive
efforts to effect NH2 installation. Nevertheless, the acquisition
of 31 prompted a new direction in the attempts to force closure
of the final ring of dibromoagelaspongin. The mechanistic
course of this transformation remains a matter of speculation,
but plausible intermediates might include the epoxide 30 and
its derived imino alcohol 30a12 or the sulfoxide 28. Attempts
to promote Pummerer reaction with concomitant N(1)/C(6)
closure by treating 28 directly with various acids did not lead
to any cyclization products.
The use of halonium ion sources to promote oxidative
additions to imidazole derivatives within the oroidin series dates
back to Bu¨chi’s seminal work on the dihydrooroidin f
dibromophakellin transformation,13a and more recent efforts by
Horne have illustrated how broad the scope of these transforms
are.7,13b This background, along with the formation of 31 from
27, suggested that prospecting among “X+” sources might be
profitable, Scheme 8. The initial entry point utilized N-
chlorosuccinimide (NCS) with the free NH substrate 27 and
led to rapid consumption of substrate without concomitant
formation of any isolable/characterizable products. Recalling
the lessons learned in the dihydrooroidin series 24c-24e,
retention of the DMAS imidazole protecting group appeared to
be a logical modification of the system. In fact, this simple and
expedient alternative was quickly rewarded with a high-yielding
oxidative cyclization to afford the halogen-substituted tetracycles
33a and 33b from 32, itself already in hand from the earlier
studies. The structural assignment of 33a was suggested by the
position of characteristic 13C NMR signals relative to those of
31 and dibromoagelaspongin (see structure 33a for key data),
and was secured by single crystal X-ray analysis (see Supporting
Information).14 The structure of the bromo analogue 33b
The precedented procedures for effecting thioamidine f
guanidine conversion were pursued with S-methyl variants rather
than the S-phenyl unit of 31/34 (Scheme 9).15 Thus, a
workaround was implemented whereby the chemistry described
in Schemes 6 and 8 with the SPh-bearing imidazole substrate
was duplicated with an SCH3 analogue. This plan was not
without its risks, primarily in the initial Pummerer reaction
where now the relatively acidic methyl protons of the
-S+(X)-CH3 unit might deprotonate in competition with the
desired N-lone pair participation within the imidazole ring.
Nevertheless, this avenue was pursued commencing with the
known imidazole derivative 3516 and utilizing chemistry
developed in the SPh series. Fortunately, the Pummerer oxida-
tive cyclization of 38 proceeded cleanly and without any
evidence for untoward reaction at the S-CH3 group, so the
concerns expressed above appeared to be unfounded. The yield
of isolated bicycle 39 was rather modest, but that deficiency is
largely due to the compound’s instability to chromatographic
1
purification. The crude H NMR spectrum of 39 suggests that
the chemical yield is more in the 85% range, and carrying on
(13) (a) Bu¨chi, G.; Foley, L. H. J. Am. Chem. Soc. 1982, 104, 1776–1777.
(b) Olofson, A.; Yakushijin, K.; Horne, D. A. J. Org. Chem. 1998, 63, 1248–
1253.
(14) Cambridge Crystallographic Data Centre deposition number: CCDC
714079. The data can be obtained free from the Cambridge Crystallographic
Data Centre via http://www.ccdc.cam.ac.uk/data_request/cif.
(15) (a) Meketa, M. L.; Weinreb, S. M. Org. Lett. 2006, 8, 1443–1446. (b)
Chauviere, G.; Bouteille, B.; Enanga, B.; de Albuquerque, C.; Croft, S. L.;
Dumas, M.; Perie, J. J. Med. Chem. 2003, 46, 427–440. (c) Bierer, D. E.;
O’Connell, J. F.; Parquette, F. R.; Thompson, C. M.; Rapoport, H. J. Org. Chem.
1992, 57, 1390. (d) Ronne, E.; Grivas, S.; Olsson, K. Acta Chem. Scand. 1994,
48, 823–830.
(12) (a) Du, H.; He, Y.; Sivappa, R.; Lovely, C. J. Synlett 2006, 965–992.
(b) Sivappa, R.; Koswatta, P.; Lovely, C. J. Tetrahedron Lett. 2007, 48, 5771–
5775. (c) O’Malley, D. P.; Li, K.; Maue, M.; Zografos, A.; Baran, P. S. J. Am.
Chem. Soc. 2007, 129, 4762–4775.
(16) Lee, J. J.; Haung, L.; Zaw, K.; Bauer, L. J. Heterocyclic Chem. 1998,
35, 81–89.
J. Org. Chem. Vol. 74, No. 9, 2009 3453

Feldman and Fodor
SCHEME 9. Preparation of an SCH3 Analogue of the
Dibromoagelaspongin Tetracycle
SCHEME 11. Mechanistic Dichotomy for the Halonium Ion
Induced Oxidative Cyclization of 32 and 40 into Tetracycles
33 and 41, Respectively
structural assignment; in particular a 0.6 ppm downfield shift
1
for the methyl group was observed, and the H NMR splitting
patterns become much more complex due to the introduction
of a new stereogenic center at sulfur. Finally, the mass spectrum
of this unpurified species indicated that one oxygen atom had
been incorporated in the starting material. The lability of 43
presaged successful replacement of the sulfur unit, and a
procedure adapted from work of Kita17 proved serviceable in
this transformation. Treatment of the crude sulfoxide with ZnI2
and TMSN3 led cleanly to the corresponding azide 44. Attempts
to use ammonia directly in this displacement did not provide
any isolable guanidine-containing product. Cleavage of the azide
moiety without competitive scission of the C-Br bonds was
required next. Although there are, in principle, procedures
involving R3P-type reagents that are documented to achieve this
goal, it turned out that chemistry no more complicated than
hydrogenolysis with Pd/H2 sufficed to deliver the free amine
without loss of Br. Finally, exposure of the TFA salt of the
crude hydrogenation product to hot water quantitatively effected
the OH-for-OCH3 exchange to deliver the desired target (()-
dibromoagelaspongin as its TFA salt. The spectral data of the
synthetic material matched in every way those data reported
for the natural isolate.
One final unresolved issue in this work remainssthe mech-
anism of the halonium ion-mediated oxidative cyclization of
32 and 40. In principle, two distinct and unrelated mechanisms
seem reasonable to consider, Scheme 11. The transform could
proceed through an orthodox Pummerer reaction where the
halonium ion serves as the initiator (47 f 48 f 49), or it could
proceed via direct imidazole halogenation followed by pyrrole
nitrogen trapping by the nascent C(6) electrophile (47 f 51 f
52). One approach to distinguishing between these two pathways
lies in the manner by which the halogen becomes attached to
C(10). In the Pummerer sequence, the halogen arrives as X- in
a second step following discharge from the activated sulfonium
ion intermediate 48. In the imidazole attack mechanism, the
halogen presents as X+ in the first step of the mechanism. Thus,
employing a different exogenous halide (Y-) in the presence
of electrophilic halide X+ may, in principle provide the means
SCHEME 10. Completion of the Dibromoagelaspongin
Synthesis
this material crude through the SEM deprotection steps led to
an overall yield for the three-step sequence (38 f 40) of 32%.
The second oxidative cyclization was mediated equally ef-
ficiently by NCS (92%) or by NBS (62%, Br replaces Cl in
41). Thus, through this 9-step sequence, the frontier of the work
was reached with the hopefully more tractable SCH3 group now
incorporated in the tetracycle 41.
The completion of the dibromoagelaspongin synthesis is
detailed in Scheme 10. Treatment of tetracycle 41 with acidic
methanol furnished the methyl ether 42 with concomitant loss
of the DMAS protecting group. Contrary to the SPh series,
substitution of water for methanol in this transformation did
not lead to a stable and isolable C(10) alcohol analogue of 42.
The basis for the difference in alcohol stability between the PhS-
and CH3S-containing species at present is unknown. Access to
42 sets the stage for the final test of the SCH3 hypothesisswill
this methyl analogue be susceptible to further sulfur oxidation
to furnish a demonstrably more labile sulfinyl moiety? It was
gratifying to observe that simple treatment of 42 with base and
mCPBA did deliver an unstable sulfoxide 43 in good yield.
Spectral data displayed by the crude material supported the
(17) Kita, Y.; Shibata, N.; Yoshida, N.; Tohjo, T. Tetrahedron Lett. 1991,
32, 2375–2378.
3454 J. Org. Chem. Vol. 74, No. 9, 2009

Extending Pummerer Reaction Chemistry
SCHEME 12. Mechanistic Probes of the Halonium Ion
Induced Oxidative Cyclization of 32/40
reagent ICl held promise in this regard, as this species is well
documented to react as an iodonium (I+) and not a chloronium
source.18 Initial experiments with NIS (4 equiv) demonstrated
that “I+” is sufficient to promote the oxidative cyclization of
the imidazole-2-methylsulfide 40 to provide the diiodotetracycle
56 in good yield, Scheme 12. Once again, initial electrophile
attack at the pyrrole unit trumped all other reactivity options;
reaction of 40 with 1 equiv of NIS led only to the pyrrole
iodination product 55. In the definitive experiment of this series,
treatment of imidazole-2-methylsulfide 40 with ICl (4 equiv)
afforded single isolable tetracyclic product 41, identical in all
respects to the product of 40 + NCS (Scheme 9), in excellent
yield. This species unequivocally bore a chloride at C(10), and
no pyrrole iodination took place. A control experiment (56 +
Bu4NCl) confirmed that 41 did not arise from initial C(10)
iodination followed by a subsequent halide exchange. This result
(Cl at C(10) of 41) can only be accommodated by the Pummerer
mechanism of Scheme 11; the alternative direct imidazole attack
mechanism would have placed an iodide at C(10).
The first total chemical synthesis of (()-dibromoagelaspongin
(6) has been accomplished via application of two successive
Pummerer reactions on a dihydrooroidin-based substrate. The
key recognition is that protection of the pyrrole nitrogen and
the imidazole nitrogen is sufficient to steer the initial Pummerer-
mediated oxidative cyclization to C(10) bond formation rather
than the precedented C(6) bond formation of the phakellin series
of oroidin derivatives. Argument-by-analogy from unambiguous
results in a model system support the hypothesis that the second
oxidative cyclization also proceeds through a Pummerer-type
process that features an underutilized halonium ion initiation
to trigger the multistep sequence and furnish the second N f
C(6) bond that characterizes this unique target structure. Overall,
this chemistry expands the repertoire of oroidin derivatives
accessible by synthesis.
to differentiate between the two mechanistic pathways. The
Pummerer pathway would yield product with halogen Y
incorporated at C(10), whereas the imidazole attack pathway
should only have X attached to C(10), provided that halogen
exchange pathways are not operational.
With these thoughts as background, exposure of 32 to NCS
(1 equiv) and LiBr (20 equiv) in a 1:1 mixture of CH2Cl2 and
THF led to the first surprise of this mechanistic inquiry; fairly
clean bromination of the pyrrole’s free position, 32 f 53,
Scheme 12. A small amount of the tetrabromo cyclized material
54 was isolated as well. Increasing the amount of NCS to 2
equiv led exclusively to this tetrabrominated cyclized material.
Whereas the formation of the C(10) brominated product 54
instead of the C(10) chloro alternative is consistent with the
prediction of the Pummerer mechanism and not the imidazole
attack alternative, the indisputable presence of some type of
electrophilic bromine species (NBS?, Br2?, BrCl?) as revealed
by the bromination of the pyrrole ring clearly calls that
conclusion into question.
Experimental Section
General Procedure 1: Coupling of Trichloroacetyl Pyrrole
Derivatives with Aminoimidazole Substrates. A stirring solution
of aminoimidazole in either acetonitrile or DMF (0.1 M) was treated
with Na2CO3 (1 equiv) followed by addition of the corresponding
acyl pyrrole (1 equiv). The resulting solution was held at room
temperature for 16 h. In cases where a solid precipitated, it was
collected and rinsed with water (20 mL) and then with ether (50
mL) to yield an analytically pure product. In all other cases where
no precipitate formed, the reaction mixture was concentrated in
vacuo to yield a dark-yellow oil. Purification of this oil by flash
column chromatography using the indicated eluent provided the
pure carboxamide.
General Procedure 2: Triflic Anhydride-Mediated Pum-
merer Cyclizations. A stirring solution of amidosulfoxide in
CH2Cl2 (0.05 M) was treated with i-Pr2NEt (2 equiv) and then
cooled to -78 °C. After 5 min, Tf2O (1.1 equiv) was slowly added
dropwise to the reaction mixture. After 10 min, the reaction mixture
was rapidly warmed to room temperature and poured into ice-cold
water. The resulting solution was partitioned between CH2Cl2 and
H2O, and the aqueous layer was extracted with CH2Cl2 (2×), dried
over Na2SO4, and concentrated to give a red oil. Purification of
this oil by flash column chromatography using the indicated eluent
provided the pure cyclization product.
Apparently, some sort of disproportionation (i.e., X+ + Y-
f X- + Y+) has compromised the results. Further insight into
this process was gained by the control experiments described
next. Mixing freshly recrystallized NBS with LiCl in d8-THF/
CD2Cl2 led to the second surprise of this series of experiments;
13C NMR monitoring of the mixture indicated that signals
corresponding to succinimide (δ 179.2 in this particular solvent
mixture) were present but neither NBS (δ 174.2) nor NCS (δ
171.8) were evident. In addition, the reaction solution im-
mediately turned orange, suggesting that Br2 or perhaps BrCl
had been formed. The complementary combination, NCS and
LiBr was explored next. In this instance, mixing freshly
recrystallized NCS with LiBr in d8-THF/CD2Cl2 again led to
instantaneous orange color formation. Furthermore, the 13C
NMR spectrum of this mixture did not display signals for NCS,
NBS or succinimide. Thus, these results were taken to indicate
that the mixtures NCS/LiBr or NBS/LiCl were not suitable for
probing the questions raised in Scheme 11.
Methyl-1H-pyrrole-2-carboxylic Acid [3-(2-Benzenesulfinyl-
3-dimethylsulfamoyl-3H-imidazol-4-yl)-propyl]-amide (24a). Fol-
(18) (a) Bell, A. S.; Campbell, S. F.; Morris, D. S.; Roberts, D. A.; Stefaniak,
M. H. J. Med. Chem. 1989, 32, 1552–1558. (b) Marsh, S. J.; Kartha, K. P. R.;
Field, R. A. Synlett 2003, 9, 1376–1378.
An alternative approach that would not be victimized by
halide exchange was required. The commercially available
J. Org. Chem. Vol. 74, No. 9, 2009 3455

Feldman and Fodor
lowing General Procedure 1, aminosulfoxide 22 (1.89 g, 5.30 mmol)
in 25 mL of DMF was treated with acyl pyrrole 23a (1.20 g, 5.30
mmol). Purification of the residue by flash column chromatography
(gradient, 1%f20% acetone/EtOAc) gave carboxamide 24a (1.85
g, 78%) as a tan solid. mp 70-72 °C; IR (thin film) 3347, 1637
pyrrole 23e (5.60 g, 11.2 mmol). Purification of the residue by flash
column chromatography (EtOAc) gave carboxamide 24e (4.62 g,
62%) as a yellow solid. mp 42-43 °C; IR (thin film) 3225, 1651
cm-1 1H NMR (360 MHz, CDCl3) δ 7.84-7.81 (m, 2H),
;
7.48-7.46 (m, 3H), 6.98 (s, 1H), 6.87 (t, J ) 5.4 Hz, 1H), 6.78 (s,
1H), 5.71 (s, 2H), 3.59 (t, J ) 7.9 Hz, 2H), 3.42 (t, J ) 6.5 Hz,
2H), 2.92 (s, 6H), 2.87-2.67 (m, 2H), 1.90 (quint, J ) 7.2 Hz,
2H), 0.88 (t, J ) 8.2 Hz, 2H), -0.05 (s, 9H); 13C NMR (100 MHz,
CDCl3) δ 160.1, 150.9, 142.9, 135.9, 131.7, 129.1, 129.0, 128.7,
126.2, 116.0, 110.7, 99.8, 75.0, 66.2, 38.8, 28.2, 23.0, 17.8, -1.5;
LRMS (ESI) m/z (relative intensity) 736.1 (100% M + H+); HRMS
(ESI) m/z calcd for [C25H36N5O5SiS2Br2]+, 736.094, found 736.0262.
4,5-Dibromo-1-methyl-N-(3-(2-(phenylsulfinyl)-1-(trifluoro-
methylsulfonyl)-1H-imidazol-5-yl)propyl)-1H-pyrrole-2-carbox-
amide (25). To a stirring solution of protected amide 24c (70 mg,
0.11 mmol) in 1 mL of THF was added aqueous HCl (1.5 M, 0.50
mL) and the reaction mixture was heated at reflux for 15 min. The
reaction solution was cooled to room temperature and poured into
ice-cold NaHCO3 (5 mL). The resulting aqueous solution was
extracted with EtOAc (2 × 20 mL), dried over Na2SO4, and
concentrated to give a yellow oil. This oil was immediately taken
up in 5 mL of CH2Cl2, cooled to -78 °C, and sequentially treated
with i-Pr2NEt (0.040 mL, 0.24 mmol) and Tf2O (0.020 mL, 0.13
mmol). After 10 min, the black solution was poured into 10 mL of
ice-cold H2O. The resulting solution was partitioned between
CH2Cl2 and H2O, and the aqueous layer was extracted with CH2Cl2
(2 × 10 mL), dried over Na2SO4, and concentrated to give a yellow
oil. Purification of this oil by flash column chromatography (70%
Et2O/hexanes) gave triflate 25 (21 mg, 29% over 2 steps) as a
colorless oil. IR (thin film) 1646 cm-1; 1H NMR (300 MHz, CDCl3)
δ 7.88-7.85 (m, 2H), 7.56-7.50 (m, 3H), 7.17 (s, 1H), 6.62 (s,
1H), 6.21 (t, J ) 5.8 Hz, 1H), 3.94 (s, 3H), 3.65 (q, J ) 6.7 Hz,
2H), 2.68 (t, J ) 7.4 Hz, 2H), 1.93 (quint, J ) 7.0 Hz, 2H); 13C
NMR (75 MHz, CDCl3) δ 160.5, 152.3, 145.8, 141.8, 132.7, 129.6,
127.4, 126.3, 118.4, 113.5, 111.6, 97.9, 38.2, 35.7, 28.1, 25.0;
LRMS (ESI) m/z (relative intensity) 644.9 (100% M + H+); HRMS
(ESI) m/z calcd for [C19H18N4O4Br2S2F3]+, 644.9088, found 644.9075.
R-SO2CF3 was not observed in the 13C spectrum due to insufficient
material.
cm-1 1H NMR (300 MHz, CDCl3) δ 7.86-7.81 (m, 2H),
;
7.49-7.46 (m, 3H), 7.00 (s, 1H), 6.69 (t, J ) 2.0 Hz, 1H), 6.53
(dd, J ) 4.0, 1.7 Hz, 1H), 6.11 (t, J ) 4.9 Hz, 1H), 6.05 (dt, J )
3.9, 2.6 Hz, 1H), 3.90 (s, 3H), 3.51-3.34 (m, 2H), 2.93 (s, 6H),
2.88-2.68 (m, 2H), 1.92 (quint, J ) 6.9 Hz, 2H); 13C (75 MHz,
CDCl3) δ 162.1, 151.0, 143.0, 136.0, 131.7, 129.13, 129.11, 127.9,
126.2, 125.4, 111.4, 107.2, 38.4, 38.0, 36.6, 28.6, 23.0; LRMS (ESI)
m/z (relative intensity) 464.1 (100% M + H+); HRMS (ESI) m/z
calcd for [C20H26N5O4S2]+, 464.1426, found 464.1440.
4-Bromo-1-methyl-1H-pyrrole-2-carboxylic Acid [3-(2-Ben-
zenesulfinyl-3-dimethylsulfamoyl-3H-imidazol-4-yl)-propyl]-
amide (24b). Following General Procedure 1, aminosulfoxide 22
(0.450 g, 1.26 mmol) in 5 mL of DMF was treated with Na2CO3
(0.147 g, 1.39 mmol) followed by acyl pyrrole 23b (0.424 g, 1.39
mmol). Purification of the residue by flash column chromatography
(gradient, 1f20% acetone/EtOAc) gave carboxamide 24b (0.372
g, 54%) as a white solid. mp 72-74 °C; IR (thin film) 3334, 1645
cm-1 1H NMR (300 MHz, CDCl3) δ 7.82-7.79 (m, 2H),
;
7.46-7.44 (m, 3H), 6.97 (s, 1H), 6.64 (d, J ) 1.8 Hz, 1H), 6.56
(d, J ) 1.8 Hz, 1H), 6.47 (t, J ) 5.8 Hz, 1H), 3.83 (s, 3H),
3.41-3.34 (m, 2H), 2.89 (s, 6H), 2.84-2.66 (m, 2H), 1.88 (quint,
J ) 7.0 Hz, 2H); 13C (75 MHz, CDCl3) δ 161.0, 150.8, 142.9,
136.0, 131.7, 129.1, 129.0, 127.0, 126.2, 126.1, 113.4, 94.2, 38.4,
38.0, 36.7, 28.2, 22.9; LRMS (ESI) m/z (relative intensity) 542.1
(100% M + H+); HRMS (ESI) m/z calcd for [C20H25N5O4S2Br]+,
542.0531, found 542.0560.
4,5-Dibromo-1-methyl-1H-pyrrole-2-carboxylic Acid [3-(2-
Benzenesulfinyl-3-dimethylsulfamoyl-3H-imidazol-4-yl)-propyl]-
amide (24c). Following General Procedure 1, aminosulfoxide 22
(0.450 g, 1.26 mmol) in 5 mL of DMF was treated with Na2CO3
(0.147 g, 1.39 mmol) followed by acyl pyrrole 24c (0.534 g, 1.39
mmol). Purification of the residue by flash column chromatography
(gradient, 1f20% acetone/EtOAc) gave carboxamide 24c (0.381
g, 49%) as a white solid. mp 61-62 °C; IR (thin film) 3337, 1648
cm-1 1H NMR (300 MHz, CDCl3) δ 7.84-7.79 (m, 2H),
;
4-(4-Bromo-1-methyl-1H-pyrrole-2-carbonyl)-2-phenylsulfan-
yl-4,5,6,7-tetrahydro-imidazo[4,5-b]pyridine-1-sulfonic Acid Dim-
ethylamide (26b). Following General Procedure 2, amidosulfoxide
24b (0.039 g, 0.072 mmol) in 1.4 mL of CH2Cl2 was treated with
i-Pr2NEt (24 µL, 0.14 mmol) and Tf2O (66 µL, 0.39 mmol).
Purification of the residue by flash column chromatography (2%
MeOH/CH2Cl2) gave cyclization product 26b (5 mg, 13%) as a
white gum. IR (thin film) 1637 cm-1; 1H NMR (360 MHz, CDCl3)
δ 7.34-7.29 (m, 5H), 6.57 (d, J ) 1.8 Hz, 1H), 6.16 (d, J ) 1.8
Hz, 1H), 3.77 (dd, J ) 7.2, 5.4 Hz, 2H), 3.62 (s, 3H), 3.02 (s, 6H),
2.90 (t, J ) 6.8 Hz, 2H), 2.07 (quint, J ) 6.5 Hz, 2H); 13C (75
MHz, CDCl3) δ 160.0, 140.3, 138.5, 132.8, 129.9, 129.0, 128.3,
127.6, 125.7, 117.2, 116.1, 94.1, 43.6, 38.5, 35.7, 23.2, 22.7; LRMS
(ESI) m/z (relative intensity) 524.0 100% (M + H+); HRMS (ESI)
m/z calcd for [C20H23N5O3S2Br]+, 524.0426, found 524.0442.
4-(4,5-Dibromo-1-methyl-1H-pyrrole-2-carbonyl)-2-phenyl-
sulfanyl-4,5,6,7-tetrahydro-imidazo[4,5-b]pyridine-1-sulfonic Acid
Dimethylamide (26c). Following General Procedure 2, amidosul-
foxide 24c (0.040 g, 0.064 mmol) in 1.3 mL of CH2Cl2 was treated
with i-Pr2NEt (21 µL, 0.13 mmol) and Tf2O (59 µL, 0.35 mmol).
Purification of the residue by flash column chromatography
(gradient, 0f5% MeOH/CH2Cl2) gave cyclization product 26c (17
7.52-7.43 (m, 3H), 6.98 (s, 1H), 6.66 (s, 1H), 6.34 (t, J ) 5.4 Hz,
1H), 3.91 (s, 3H), 3.49-3.34 (m, 2H), 2.93 (s, 6H), 2.89-2.67
(m, 2H), 1.90 (quint, J ) 7.0 Hz, 2H); 13C (75 MHz, CDCl3) δ
160.6, 150.9, 142.9, 136.0, 131.8, 129.2, 129.1, 127.4, 126.2, 113.6,
111.6, 97.8, 38.6, 38.1, 35.6, 28.4, 23.0; LRMS (ESI) m/z (relative
intensity) 620.0 (90% M + H+); HRMS (ESI) m/z calcd for
[C20H24N5O4S2Br2]+, 619.9636, found 619.9656.
4,5-Dibromo-1-methoxymethyl-1H-pyrrole-2-carboxylic Acid
[3-(2-Benzenesulfinyl-3-dimethylsulfamoyl-3H-imidazol-4-yl)-
propyl]-amide (24d). Following General Procedure 1, aminosul-
foxide 22 (0.450 g, 1.26 mmol) in 5 mL of DMF was treated with
Na2CO3 (0.147 g, 1.39 mmol) followed by acyl pyrrole 23d (0.626
g, 1.51 mmol). Purification of the residue by flash column
chromatography (gradient, 60f100% EtOAc/hexanes) gave car-
boxamide 24d (0.391 g, 48%) as a yellow solid. mp 62-64 °C; IR
(thin film) 3327, 1651 cm-1 1H NMR (360 MHz, CDCl3) δ
;
7.74-7.71 (m, 2H), 7.39-7.37 (m, 3H), 6.88 (s, 1H), 6.82 (t, J )
7.2 Hz, 1H), 6.72 (s, 1H), 5.63 (s, 2H), 3.32 (q, J ) 7.2 Hz, 2H),
3.25 (s, 3H), 2.81 (s, 6H), 2.75-2.68 (m, 2H), 1.81 (quint, J ) 7.2
Hz, 2H); 13C (100 MHz, CDCl3) δ 160.1, 150.9, 142.9, 135.9, 131.8,
129.2, 129.1, 128.5, 126.3, 115.8, 111.0, 100.0, 76.8, 56.2, 38.8,
38.0, 28.3, 23.0; LRMS (ESI) m/z (relative intensity) 672.0 (100%
M + Na+); HRMS (ESI) m/z calcd for [C21H26N5O5S2Br2]+,
649.9742, found 649.9749.
1
mg, 44%) as a white gum. IR (thin film) 1615 cm-1; H NMR
(300 MHz, CDCl3) δ 7.31 (s, 5H), 6.21 (s, 1H), 3.78 (dd, J ) 5.4,
3.5 Hz, 2H), 3.59 (s, 3H), 3.03 (s, 6H), 2.89 (t, J ) 6.6 Hz, 2H),
2.08 (quint, J ) 6.7 Hz, 2H); 13C (75 MHz, CDCl3) δ 159.2, 141.4,
138.2, 133.7, 129.6, 129.09, 129.05, 129.0, 128.9, 116.6, 116.1,
97.5, 43.3, 38.5, 35.1, 23.2, 22.6; LRMS (ESI) m/z (relative
intensity) 602.0 (100% M + H+); HRMS (ESI) m/z calcd for
[C20H22N5O3S2Br2]+, 601.9531, found 601.9524.
4,5-Dibromo-1-(2-trimethylsilanyl-ethoxymethyl)-1H-pyrrole-
2-carboxylic Acid [3-(2-Benzenesulfinyl-3-dimethylsulfamoyl-
3H-imidazol-4-yl)-propyl]-amide (24e). Following General Pro-
cedure 1, aminosulfoxide 22 (3.60 g, 10.1 mmol) in 50 mL of
acetonitrile was treated with Na2CO3 (1.20 g, 11.3 mmol) and acyl
3456 J. Org. Chem. Vol. 74, No. 9, 2009

Extending Pummerer Reaction Chemistry
4-(4,5-Dibromo-1-methoxymethyl-1H-pyrrole-2-carbonyl)-2-
phenylsulfanyl-4,5,6,7-tetrahydro-imidazo[4,5-b]pyridine-1-sul-
fonic Acid Dimethylamide (26d). Following General Procedure
2, amidosulfoxide 24d (0.097 g, 0.15 mmol) in 1.4 mL of CH2Cl2
was treated with i-Pr2NEt (49 µL, 0.30 mmol) and Tf2O (137 µL,
0.810 mmol). Purification of the residue by flash column chroma-
tography (1:1 EtOAc/hexanes) gave cyclization product 26d (16
SEM deprotection of bicycle 26e (1.00 g, 1.39 mmol), the crude
DMAS protected bicycle 32 (714 mg) was taken up in 50 mL of
THF and aqueous HCl (1.5 M, 27 mL) was added to the reaction
solution. After 16 h at room temperature, the reaction mixture was
carefully poured into 50 mL of ice-cold NaHCO3. The resulting
mixture was partitioned between EtOAc and H2O and the aqueous
layer was extracted with EtOAc (2 × 50 mL). The combined
organic fractions were dried with Na2SO4 and concentrated to give
a yellow oil. Purification of this oil by flash column chromatography
(Et2O) gave the deprotection product 27 (537 mg, 80% over the 3
step sequence) as a yellow solid. mp 122-124 °C; IR (thin film)
1592 cm-1; 1H NMR (300 MHz, d4-MeOH) δ 7.33-7.21 (m, 5H),
6.65 (br s, 1H), 3.77 (br s, 2H), 2.71 (t, J ) 6.5 Hz, 2H), 2.06
(quint, J ) 6.3 Hz, 2H); 13C (75 MHz, d6-DMSO) δ 157.8, 137.8,
135.6, 129.6, 129.5, 127.7, 127.5, 126.6, 118.8, 116.6, 104.3, 97.8,
44.4, 22.9 19.4; LRMS (ESI) m/z (relative intensity) 480.9 (100%
M + H+); HRMS (ESI) m/z calcd for [C17H15N4OSBr2]+, 480.9333,
found 480.9348.
1
mg, 17%) as a white gum. IR (thin film) 1615 cm-1; H NMR
(360 MHz, CDCl3) δ 7.33-7.30 (m, 2H), 7.29-7.26 (m, 3H), 6.25
(s, 1H), 5.49 (s, 2H), 3.77 (dd, J ) 7.2, 3.6 Hz, 2H), 3.08 (s, 6H),
2.90 (t, J ) 7.2 Hz, 2H), 2.70 (s, 3H), 2.08 (quint, J ) 7.2 Hz,
2H); 13C (75 MHz, CDCl3) δ 158.8, 142.3, 138.0, 134.4, 129.2,
129.14, 129.07, 128.3, 117.3, 116.4, 109.5, 99.3, 76.9, 55.6, 43.1,
38.5, 23.1, 22.5; LRMS (ESI) m/z (relative intensity) 632.0 (100%
M + H+); HRMS (ESI) m/z calcd for [C21H24N5O4S2Br2]+,
631.9636, found 631.9597.
4-[4,5-Dibromo-1-(2-trimethylsilanyl-ethoxymethyl)-1H-pyr-
role-2-carbonyl]-2-phenylsulfanyl-4,5,6,7-tetrahydro-imidazo[4,5-
b]pyridine-1-sulfonic Acid Dimethylamide (26e). Following
General Procedure 2, amidosulfoxide 24e (0.990 g, 1.34 mmol) in
100 mL of CH2Cl2 was treated with i-Pr2NEt (465 µL, 2.81 mmol)
and Tf2O (453 µL, 2.67 mmol). Purification of the residue by flash
column chromatography (3:2 Et2O/hexanes) gave cyclization
product 26e (0.505 g, 52%) as a yellow solid. mp 86-88 °C; IR
(2-Benzenesulfinyl-1,5,6,7-tetrahydro-imidazo[4,5-b]pyridin-
4-yl)-(4,5-dibromo-1H-pyrrol-2-yl)-methanone (28). A stirring
solution of bicycle 27 (200 mg, 0.416 mmol) in 20 mL of CH2Cl2
at room temperature was sequentially treated with Na2CO3 (93 mg,
0.87 mmol) and then mCPBA (103 mg, 0.416 mmol). After 30
min, the reaction mixture was poured into ice-cold NaHCO3 (20
mL). The resulting solution was partitioned between CH2Cl2 and
H2O and the aqueous layer was extracted with CH2Cl2 (2 × 25
mL), dried over Na2SO4 and concentrated to give a yellow oil.
Purification of this oil by flash column chromatography (gradient,
80f100% EtOAc/hexanes) gave sulfoxide 28 (147 mg, 71%) as a
1
(thin film) 1615 cm-1; H NMR (360 MHz, CDCl3) δ 7.34-7.27
(m, 5H), 6.23 (s, 1H), 5.53 (s, 2H), 3.78 (dd, J ) 5.4, 3.2 Hz, 2H),
3.18 (t, J ) 8.3 Hz, 2H), 3.05 (s, 6H), 2.91 (t, J ) 6.8 H, 2H),
2.08 (quint, J ) 6.1 Hz, 2H), 0.68 (t, J ) 8.3 Hz, 2H), -0.15 (s,
9H); 13C NMR (90 MHz, CDCl3) δ 160.5, 142.8, 139.5, 135.1,
130.6, 130.5, 130.1, 129.8, 118.7, 118.0, 111.1, 100.7, 75.9, 66.7,
44.8, 40.0, 24.6, 24.1, 18.7, 0.00; LRMS (ESI) m/z (relative
intensity) 718.0 (100% M + H+); HRMS (ESI) m/z calcd for
[C25H34N5O4SiS2Br2]+, 718.0188, found 718.0145.
1
tan solid. mp 110-112 °C; IR (thin film) 3112, 1581 cm-1; H
NMR (400 MHz, d4-MeOH) δ 7.71-7.69 (m, 2H), 7.56-7.51 (m,
3H), 6.56 (s, 1H), 4.00 (m, 1H), 3.81 (m, 1H), 2.67 (t, J ) 6.7 Hz,
2H), 1.99 (quint, J ) 7.2 Hz, 2H); 13C (90 MHz, d6-DMSO) δ
158.7, 142.7, 141.8, 135.3, 132.4, 130.4, 127.0, 125.1, 121.7, 117.2,
106.2, 98.8, 46.1, 22.8, 20.2; LRMS (ESI) m/z (relative intensity)
4-(4,5-Dibromo-1H-pyrrole-2-carbonyl)-2-phenylsulfanyl-
4,5,6,7-tetrahydro-imidazo[4,5-b]pyridine-1-sulfonic Acid Dim-
ethylamide (32). A stirring solution of SEM-protected bicyclopy-
rrole 26e (1.94 g, 2.70 mmol) in 30 mL of CH2Cl2 was cooled to
0 °C and BF3Et2O (3.39 mL, 27.0 mmol) was added dropwise to
the reaction mixture. After an additional 2 h at 0 °C, the reaction
mixture was allowed to warm to room temperature and held there
for 3 h before being poured into 50 mL of ice-cold H2O. The
resulting solution was partitioned between CH2Cl2 and H2O, and
the aqueous layer was extracted with CH2Cl2 (2 × 30 mL). The
combined organic fractions were dried over Na2SO4 and then
concentrated to give a yellow oil. This oil was immediately taken
up into 5 mL of THF and sequentially treated with ethylenediamine
(362 µL, 5.40 mmol) followed by TBAF (1.0 M in THF, 27 mL,
27 mmol), and then the solution was heated at reflux for 30 min.
The reaction mixture was allowed to cool to room temperature and
then was poured into an ice-cold saturated NH4Cl solution. The
resulting mixture was partitioned between EtOAc and H2O and the
aqueous layer was extracted with EtOAc (2 × 20 mL). The
combined organic fractions were dried over Na2SO4 and concen-
trated to give a red oil. Purification of this oil by flash column
chromatography (Et2O) gave deprotection product 32 (994 mg, 63%
over 2 steps) as a yellow solid. When the crude pyrrole deprotection
product 32 was carried on through the next step (removal of the
DMAS group), the overall yield of the three-step sequence was
496.9 (100%
M +
H+); HRMS (ESI) m/z calcd for
[C17H15N4O2SBr2]+, 496.9282, found 496.9325.
(2-Benzenesulfinyl-1-trifluoromethanesulfonyl-1,5,6,7-tetrahy-
dro-imidazo[4,5-b]pyridin-4-yl)-(4,5-dibromo-1H-pyrrol-2-yl)-meth-
anone (29a) and (2-Benzenesulfinyl-1-trifluoromethanesulfonyl-
1,5,6,7-tetrahydro-imidazo[4,5-b]pyridin-4-yl)-(4,5-dibromo-
1-trifluoromethanesulfonyl-1H-pyrrol-2-yl)-methanone (29b).
Following General Procedure 2, bicycle 28 (0.020 g, 0.040 mmol)
in 8 mL of CH2Cl2 was treated with i-Pr2NEt (14 µL, 0.084 mmol)
and Tf2O (13 µL, 0.042 mmol). Purification of the residue by flash
column chromatography (gradient, 30f60% Et2O/hexanes) gave
monotriflation product 29a (5 mg, 20%) as a white tacky solid and
bistriflation product 29b (9 mg, 30%) as a colorless oil. Spectral
data for 29a: IR (thin film) 3172, 1607 cm-1; 1H NMR (400 MHz,
d6-DMSO) δ 12.8 (br s, 1H), 7.71-7.55 (m, 5H), 6.55 (d, J ) 2.7
Hz, 1H), 3.98-3.90 (br m, 1H), 3.62-3.57 (br m, 1H), 2.81 (br
m, 2H), 2.11-1.93 (br m, 2H); 13C (75 MHz, d6-DMSO) δ 158.4,
149.1, 141.1, 139.5, 132.4, 129.4, 126.8, 126.4, 120.1, 117.0, 106.0,
98.3, 44.4, 22.0, 20.6; LRMS (ESI) m/z (relative intensity) 628.7
(100% M + H+); HRMS (ESI) m/z calcd for [C18H14N4O4F3S2Br2]+,
628.8775, found 628.8779. R-SO2CF3 was not observed in 13C
spectrum due to insufficient material. Spectral data for 29b: IR (thin
1
film) 1654 cm-1; H NMR (400 MHz, CDCl3) δ 7.66-7.55 (m,
1
80%. mp 160-162 °C; IR (thin film) 3164, 1594 cm-1; H NMR
5H), 6.66 (br s, 1H), 4.03 (br m, 1H), 3.74 (br m, 1H), 2.90 (s,
2H), 2.20-2.03 (br m, 2H); 13C NMR (75 MHz, CDCl3) δ 157.1,
151.5, 139.0, 133.3, 132.5, 129.6, 125.7, 121.05, 121.03, 118.2,
116.76, 116.72, 42.4, 21.1, 20.9; LRMS (ESI) m/z (relative
intensity) 760.8 (100% M + H+); HRMS (ESI) m/z calcd for
[C19H13N4O6S3Br2F6]+, 760.8268, found 760.8254. R-SO2CF3 was
not observed in 13C spectrum due to insufficient material.
(300 MHz, CDCl3) δ 11.4 (br s, 1H), 7.52-7.50 (m, 2H), 7.40-7.33
(m, 3H), 6.76 (app d, J ) 2.8 Hz, 1H), 3.89 (dd, J ) 7.5, 5.6 Hz,
2H), 3.07 (s, 6H), 2.90 (t, J ) 6.6 Hz, 2H), 2.04 (quint, J ) 6.2
Hz, 2H); 13C NMR (75 MHz, d6-DMSO) δ 158.2, 138.1, 137.6,
132.0, 130.2, 129.0, 128.3, 127.5, 118.1, 116.3, 104.7, 97.8, 43.8,
38.1, 22.8, 22.2; LRMS (ESI) m/z (relative intensity) 587.9 (100%
M + H+); HRMS (ESI) m/z calcd for [C19H20N5O3S2Br2]+,
587.9374, found 587.9399.
1H,8H-Imidazo[4,5-b]pyrrolo[1′,2′:3,4]imidazo[1,2-a]pyridin-
8-one, 2-(phenylthio)-10,11-dibromo-3a,4,5,6-tetrahydro-3a-hy-
droxy- 31. Method A. A stirring solution of bicycle 27 (97 mg,
0.20 mmol) in 5 mL of CH2Cl2 was treated with trifluoroperoxy-
(4,5-Dibromo-1H-pyrrol-2-yl)-(2-phenylsulfanyl-1,5,6,7-tet-
rahydro-imidazo[4,5-b]pyridin-4-yl)-methanone (27). Following
J. Org. Chem. Vol. 74, No. 9, 2009 3457

Feldman and Fodor
acetic acid (4 M, 50 µL, 0.20 mmol) and the reaction solution was
warmed to 40 °C and held at this temperature for 16 h. The reaction
mixture was then allowed to cool to room temperature and poured
into saturated aqueous NaHCO3 (5 mL). The resulting solution was
partitioned between CH2Cl2 and H2O and the aqueous layer was
extracted with CH2Cl2 (2 × 5 mL). The organic fractions were
combined, dried over Na2SO4 and concentrated to give a yellow
oil. Purification of this oil by flash column chromatography gave
hydroxytetracycle 31 (6 mg, 5%) and recovered starting material
27 (41 mg).
Method B. To a stirring solution of bromotetracycle 33b (100
mg, 0.150 mmol) in 10 mL of THF was added 10 mL of a 1.5 M
HCl solution. The reaction mixture was heated to 65 °C for 16 h
after which time it was cooled to room temperature and then poured
into 10 mL of ice-cold NH4OH. The resulting solution was
partitioned between EtOAc and H2O and the aqueous layer was
extracted with EtOAc (2 × 20 mL). The combined organic fractions
were dried over Na2SO4 and concentrated to give a yellow oil.
Purification of this oil by flash column chromatography (Et2O then
EtOAc) gave hydroxytetracycle 31 (52 mg, 70%) as a white oil.
IR (thin film) 3216, 1693 cm-1; 1H NMR (300 MHz, d4-MeOH) δ
7.69-7.66 (m, 2H), 7.51-7.42 (m, 3H), 6.65 (s, 1H), 3.93 (m,
1H), 3.16 (m, 1H), 2.05-1.93 (m, 2H), 1.73-1.60 (m, 2H); 13C
NMR (125 MHz, d6-DMSO) δ 169.8, 156.9, 134.8, 129.5, 129.3,
126.9, 126.5, 107.0, 102.6, 102.8, 100.9, 91.4, 35.9, 33.3, 17.2;
LRMS (ESI) m/z (relative intensity) 496.9 (100% M + H+); HRMS
(ESI) m/z calcd for [C17H15N4O2SBr2]+, 496.9282, found 496.9284.
m/z (relative intensity) 665.8 (100% M + H+); HRMS (ESI) m/z
calcd for [C19H19N5O3S2Br3]+, 665.8479, found 665.8492.
Tetracyclic Methyl Ether (34a). A stirring solution of chlo-
rotetracycle 33a (3.0 mg, 0.0048 mmol) in 750 µL of MeOH was
treated with AgNO3 (15 mg, 0.088 mmol) and the reaction mixture
was held at room temperature for 16 h after which time it was
heated to 45 °C for an additional 1 h. The resulting blue solution
was allowed to cool to room temperature and then concentrated.
The residue was dissolved in 5 mL of CHCl3 and poured into H2O
(5 mL). The resulting solution was partitioned between CHCl3 and
H2O, and the organic layer was washed with H2O (1 × 5 mL),
dried over Na2SO4, and then concentrated to give a colorless oil.
Purification of this oil by flash column chromatography (hexanes
then Et2O) gave methyl ether 34a (1.6 mg, 54%) as a white
semisolid. IR (thin film) 1712 cm-1; 1H NMR (300 MHz, CDCl3)
δ 7.59-7.56 (m, 2H), 7.42-7.34 (m, 3H), 6.66 (s, 1H), 4.08 (dt,
J ) 13.1, 9.2 Hz, 1H), 3.10 (s, 6H), 3.01 (m, 1H), 2.99 (s, 3H),
2.76 (m, 1H), 2.06 (m, 1H), 1.87-1.75 (m, 2H); 13C (90 MHz,
CDCl3) δ 167.9, 157.2, 135.4, 130.3, 129.4, 127.3, 126.6, 109.2,
104.8, 103.1, 100.3, 98.8, 52.3, 38.4, 34.3, 25.1, 16.6; LRMS (ESI)
m/z (relative intensity) 617.9 (100% M + H+); HRMS (ESI) m/z
calcd for [C20H22N5O4S2Br2]+, 617.9480, found 617.9460.
3H,8H-Imidazo[4,5-b]pyrrolo[1′,2′:3,4]imidazo[1,2-a]pyridin-
8-one, 2-(Phenylthio)-10,11-dibromo-3a,4,5,6-tetrahydro-3a-
methoxy-TFA Salt (34b). To a stirring solution of chlorotetracycle
33a (0.150 g, 0.240 mmol) in 4.8 mL of MeOH was added HCl
(conc, 240 µL), and the reaction solution was held at room
temperature for 16 h after which time it was concentrated to ∼1
mL and then diluted with 30 mL of CH2Cl2. This organic solution
was washed with NaHCO3 (1 × 20 mL), dried over Na2SO4, and
concentrated to give a yellow oil. Purification of this oil by flash
column chromatography (gradient, 80f100% Et2O/hexanes) gave
methyl ether 34b (86 mg, 59%) as a white solid after conversion
to its TFA salt. [Note: Fractions from column chromatography
containing the desired product were combined and TFA (100 µL)
3H,8H-Imidazo[4,5-b]pyrrolo[1′,2′:3,4]imidazo[1,2-a]pyridine-
3-sulfonamide, 10,11-dibromo-3a-chloro-3a,4,5,6-tetrahydro-
N,N-dimethyl-2-(phenylthio)-8-oxo- (33a). A stirring solution of
bicycle 32 (25 mg, 0.042 mmol) in 1 mL of CH2Cl2 was treated
with NCS (11 mg, 0.084 mmol). After 5 min, the reaction mixture
was diluted with CH2Cl2 to a volume of 20 mL and poured into 10
mL of ice-cold H2O. The organic layer was sequentially washed
with H2O (1 × 10 mL), saturated aqueous NaHCO3 (1 × 10 mL),
1 M H3PO4 (1 × 10 mL), dried over Na2SO4 and concentrated to
give a yellow oil. Purification of this oil by flash column
chromatography (3:2 Et2O/hexanes) gave tetracycle 33a (17 mg,
65%) as a white solid. A small portion of this solid was dissolved
in acetone and allowed to evaporate slowly over 72 h, which
resulted in the growth of an X-ray quality crystal. mp 146-148
was added before concentrating in vacuo.] IR (thin film) 1694 cm-1
;
1H NMR (360 MHz, CDCl3) δ 7.79-7.76 (m, 2H), 7.64-7.54 (m,
3H), 6.72 (s, 1H), 3.98 (dt, J ) 13.0, 8.6 Hz, 1H), 3.56 (s, 3H),
3.25 (m, 1H), 2.37 (m, 1H), 2.14 (m, 1H), 1.83-1.65 (m, 2H); 13
C
(90 MHz, CDCl3) δ 174.7, 158.8, 137.1, 133.4, 131.9, 127.9, 123.0,
110.5, 106.4, 105.7, 97.1, 95.3, 52.3, 37.6, 29.4, 17.3; LRMS (ESI)
m/z (relative intensity) 510.9 (100% M + H+); HRMS (ESI) m/z
calcd for [C18H17N4O2SBr2]+, 510.9439, found 510.9445.
°C; IR (thin film) 1717 cm-1 1H NMR (400 MHz, CDCl3) δ
;
7.59-7.57 (m, 2H), 7.46-7.38 (m, 3H), 6.66 (s, 1H), 4.08 (dt, J
) 13.2, 8.7 Hz, 1H), 3.16 (s, 6H), 3.11-2.96 (m, 2H), 2.16 (ddd,
J ) 12.3, 8.0, 3.2 Hz, 1H), 1.90 (m, 1H), 1.72 (m, 1H); 13C (100
MHz, CDCl3) δ 168.7, 157.1, 135.4, 130.6, 129.5, 127.2, 125.9,
109.4, 105.2, 102.7, 100.1, 90.6, 38.2, 37.1, 35.6, 17.7; LRMS (ESI)
m/z (relative intensity) 621.9 (100% M + H+); HRMS (ESI) m/z
calcd for [C19H19N5O3S2ClBr2]+, 621.8985, found 621.9014.
5-(3-Amino-propyl)-2-methanesulfinyl-imidazole-1-sulfonic Acid
Dimethylamide (36). A stirring solution of sulfide 3515 (20.8 g,
94.0 mmol) in 500 mL of THF was cooled to -78 °C and n-BuLi
(45.0 mL, 113 mmol) was added dropwise to the reaction solution
over 15 min. After an additional 45 min, the resulting brown
solution was treated with 1-chloro-3-iodopropane (11.9 mL, 113
mmol), and the reaction mixture was allowed to slowly warm to
room temperature over 16 h after which time 200 mL of saturated
NH4Cl and 100 mL of H2O were added sequentially. The resulting
solution was partitioned between Et2O and H2O and the aqueous
layer was extracted with Et2O (1 × 150 mL), dried over Na2SO4,
and concentrated to give a dark-yellow oil. Purification of this oil
by flash column chromatography (gradient, 33%f40% EtOAc/
hexanes) gave the chloropropyl imidazole product (25.6 g, 91%)
as a pale-yellow oil. IR (thin film) 2930 cm-1; 1H NMR (360 MHz,
CDCl3) δ 6.61 (s, 1H), 3.46 (t, J ) 6.5 Hz, 2H), 2.81 (s, 6H), 2.76
3H,8H-Imidazo[4,5-b]pyrrolo[1′,2′:3,4]imidazo[1,2-a]pyridine-
3-sulfonamide, 3a,10,11-tribromo-3a,4,5,6-tetrahydro-N,N-di-
methyl-2-(phenylthio)-8-oxo- (33b). A stirring solution of bicycle
32 (10 mg, 0.017 mmol) in 1 mL of CH2Cl2 was treated with NBS
(6.0 mg, 0.034 mmol). After 5 min, the reaction mixture was diluted
with CH2Cl2 to a volume of 20 mL and poured into 10 mL of ice-
cold H2O. The organic layer was sequentially washed with H2O (1
× 10 mL), saturated aqueous NaHCO3 (1 × 10 mL), 1 M H3PO4
(1 × 10 mL), dried over Na2SO4, and concentrated to give a yellow
oil. Purification of this oil by flash column chromatography (14:
5:1 Et2O/hexanes/CH2Cl2) gave tetracycle 33b (6.6 mg, 62%) as a
white solid which was unavoidably contaminated by a small amount
of tetrabromo compound 54. Spectral Data for 33b: mp 143-144
(t, J ) 7.2 Hz, 2H), 2.43 (s, 3H), 1.96 (quint, J ) 6.5 Hz, 2H); 13
C
(90 MHz, CDCl3) δ 146.9, 133.0, 127.4, 43.7, 37.9, 30.6, 23.1,
14.8; LRMS (ESI) m/z (relative intensity) 298.1 (100% M + H+);
HRMS (ESI) m/z calcd for [C9H17N3O2S2]+, 298.0451, found
298.0465.
°C; IR (thin film) 1716 cm-1 1H NMR (360 MHz, CDCl3) δ
;
7.60-7.56 (m, 2H), 7.45-7.36 (m, 3H), 6.67 (s, 1H), 4.11 (dt, J
) 13.0 7.9 Hz, 1H), 3.17 (s, 6H), 3.10 (m, 1H), 3.01 (m, 1H), 2.46
To a stirring solution of this chloride (25.0 g, 83.9 mmol) in
420 mL of DMF was added potassium phthalate (18.6 g, 101
mmol). The resulting suspension was heated to 80 °C and held at
this temperature for 16 h after which time the reaction mixture was
cooled to room temperature and filtered to remove any remaining
(ddd, J ) 14.4, 11.6, 3.2 Hz, 1H), 1.82 (m, 1H), 1.70 (m, 1H); 13
C
(75 MHz, CDCl3) δ 168.7, 157.2, 135.5, 130.6, 129.6, 127.1, 125.9,
109.3, 105.2, 102.4, 100.2, 83.6, 39.6, 38.2, 36.2, 18.6; LRMS (ESI)
3458 J. Org. Chem. Vol. 74, No. 9, 2009

Extending Pummerer Reaction Chemistry
solids. The filtrate was collected and partitioned between CH2Cl2
(500 mL) and H2O (100 mL). The aqueous layer was extracted
with CH2Cl2 (1 × 500 mL) and the combined organic fractions
were sequentially washed with H2O (2 × 500 mL) and brine
(saturated, 500 mL), dried over Na2SO4 and concentrated to give a
brown oil. Purification of this oil by flash column chromatography
(gradient, 30f90% EtOAc/hexanes) gave the phthalimide product
(18.6 g, 54%) as a white solid. mp 108-109 °C; IR (thin film)
4-[4,5-Dibromo-1-(2-trimethylsilanyl-ethoxymethyl)-1H-pyr-
role-2-carbonyl]-2-methylsulfanyl-4,5,6,7-tetrahydro-imidazo[4,5-
b]pyridine-1-sulfonic Acid Dimethylamide (39). To a stirring
solution of amidosulfoxide 38 (0.200 g, 0.296 mmol) in 5.9 mL of
CH2Cl2 was added 2,6-lutidine (72 µL, 0.62 mmol) and the reaction
mixture was cooled to -78 °C. After 5 min, Tf2O (55 µL, 0.33
mmol) was added and the reaction solution turned from colorless
to light yellow. After 5 additional min, the reaction mixture was
diluted with CH2Cl2 to a total volume of 15 mL and then poured
into ice-cold H2O (5 mL). The resulting solution was partitioned
between CH2Cl2 and H2O and the aqueous layer was extracted with
CH2Cl2 (1 × 10 mL). The organic fractions were combined, dried
with Na2SO4, and concentrated to give a yellow foam. Purification
of this foam by flash column chromatography (1:1:1 Et2O/hexanes/
CH2Cl2) gave bicycle 39 (88 mg, 45%) as a tacky white gum. IR
1
1711 cm-1; H NMR (360 MHz, CDCl3) δ 7.86-7.83 (m, 2H),
7.73-7.70 (m, 2H), 6.78 (app t, J ) 1.1 Hz, 1H), 3.78 (t, J ) 6.8
Hz, 2H), 2.93 (s, 6H), 2.76 (app dt, J ) 8.3, 1.1 Hz, 2H), 2.57 (s,
3H), 2.06-1.98 (quint, J ) 7.6 Hz, 2H); 13C (90 MHz, CDCl3) δ
168.4, 147.4, 134.0, 133.9, 132.1, 127.5, 123.2, 38.2, 37.4, 27.4,
23.8, 15.3; LRMS (ESI) m/z (relative intensity) 409.2 (100% M +
H+); HRMS (ESI) m/z calcd for [C17H21N4O4S2]+, 409.1004, found
409.1011.
1
(thin film): 1616 cm-1; H NMR (360 MHz, CDCl3) δ 6.40 (s,
1H), 5.55 (s, 2H), 3.85-3.82 (m, 2H), 3.53 (t, J ) 8.3 Hz, 2H),
2.95 (s, 6H), 2.84 (t, J ) 7.2 Hz, 2H), 2.27 (s, 3H), 2.05 (quint, J
) 6.1 Hz, 2H), 0.84 (t, J ) 8.3 Hz, 2H), -0.09 (s, 9H); 13C NMR
(90 MHz, CDCl3) δ 159.4, 143.8, 137.6, 129.1, 116.7, 115.8, 108.5,
99.0, 75.2, 65.9, 43.8, 38.3, 22.8, 22.2, 17.6, 15.0, -1.52; LRMS
(ESI) m/z (relative intensity) 656.1 (100%, M + H+); HRMS (ESI)
m/z calcd for [C20H32N5O4S2Br2Si]+, 656.0005, found 656.0032.
4-(4,5-Dibromo-1H-pyrrole-2-carbonyl)-2-methylsulfanyl-
4,5,6,7-tetrahydro-imidazo[4,5-b]pyridine-1-sulfonic Acid Dim-
ethylamide (40). Following the Pummerer-mediated cyclization of
amidosulfoxide 38 (3.20 g, 4.73 mmol) the crude material 39 (3.71
g) was immediately taken up in 100 mL of CH2Cl2 and TFA (20
mL) was added to the reaction mixture. After 24 h, the reaction
solution was diluted with CH2Cl2 to a volume of 250 mL and poured
into 100 mL of an ice-cold dilute NaHCO3 solution. Caution: This
acid-base reaction is very exothermic and care should be taken
during the entire workup process.
The resulting solution was partitioned between CH2Cl2 and H2O,
and the aqueous layer was extracted with CH2Cl2 (2 × 150 mL),
dried with Na2SO4, and concentrated to give a red foam. This foam
was immediately taken up in 20 mL of THF, and ethylenediamine
as a formaldehyde trap19 (1.58 mL, 23.7 mmol) was added to the
reaction mixture followed by the addition of TBAF (1.0 M, 46 mL,
46 mmol). The reaction mixture was heated to reflux and held at
that temperature for 20 min after which it was cooled to room
temperature and poured into 50 mL of a dilute ice-cold NH4Cl
solution. The resulting solution was partitioned between EtOAc (50
mL) and H2O, and the aqueous layer was extracted with EtOAc (3
× 50 mL). The organic fractions were combined, dried with
Na2SO4, and concentrated to give dark-brown foam. Purification
of this foam by flash column chromatography (1:1 Et2O/CH2Cl2)
gave the deprotection product 40 (811 mg, 32% over 3 steps) as a
white solid. mp. 184-186 °C; IR (thin film): 3166, 1599 cm-1; 1H
NMR (360 MHz, CDCl3) δ 12.6 (br s, 1H), 6.87 (s, 1H), 3.97-3.94
(m, 2H), 3.01 (s, 6H), 2.84 (t, J ) 6.5 Hz, 2H), 2.68 (s, 3H), 2.04
(quint, J ) 6.8 Hz, 2H); 13C NMR (75 MHz, CDCl3) δ 158.3, 144.1,
136.8, 126.7, 120.1, 118.1, 104.5, 99.9, 44.1, 38.4, 22.7, 22.5, 15.7;
LRMS (ESI) m/z (relative intensity) 525.9 (100%, M + H+); HRMS
(ESI) m/z calcd for [C14H18N5O3S2Br2]+, 525.9218, found 525.9233.
3H,8H-Imidazo[4,5-b]pyrrolo[1′,2′:3,4]imidazo[1,2-a]pyridine-
3-sulfonamide, 10,11-Dibromo-3a-chloro-3a,4,5,6-tetrahydro-
N,N-dimethyl-2-(methylthio)-8-oxo- (41). Method A: To a stirring
solution of fused bicycle 40 (0.460 g, 0.872 mmol) in 18 mL of
CH2Cl2 was added N-chlorosuccinimide (NCS) (128 mg, 0.959
mmol). After 1 min, the reaction mixture was poured into 20 mL
of H2O. The resulting solution was partitioned between CH2Cl2 and
H2O, and the aqueous layer was extracted with CH2Cl2 (1 × 20
mL). The organic fractions were combined, dried with Na2SO4, and
concentrated to give chlorotetracycle 41 (452 mg, 92%) as a cream
colored solid. A small amount of this material was purified by flash
column chromatography (Florasil, 60% Et2O/hexanes) for charac-
A stirring solution of this phthalimidosulfide (0.630 g, 1.54
mmol) in 30 mL of CH2Cl2 was cooled to 0 °C and treated with
mCPBA (75% purity, 0.390 g, 1.70 mmol). After 15 min, the
reaction solution was warmed to room temperature and held at this
temperature for an additional 1 h after which time the reaction
mixture was poured into aqueous NaHCO3 (30 mL). The resulting
solution was partitioned between CH2Cl2 and H2O and the aqueous
layer was extracted with CH2Cl2 (1 × 30 mL). The organic fractions
were combined, dried with Na2SO4 and concentrated to give a
yellow oil. Purification of this oil by flash column chromatography
(80% EtOAc/hexanes then 25% acetone/EtOAc) gave the phthal-
imidosulfoxide product (0.646 g, 99%) as a tacky white gum; IR
1
(thin film) 1714 cm-1; H NMR (360 MHz, CDCl3) δ 7.81-7.77
(m, 2H), 7.71-7.68 (m, 2H), 7.01 (s, 1H), 3.76 (t, J ) 6.8 Hz,
2H), 2.98 (s, 3H), 2.95 (s, 6H), 2.89 (m, 1H), 2.72 (m, 1H),
2.09-1.99 (m, 2H); 13C NMR (90 MHz, CDCl3) δ 168.2, 150.7,
135.7, 134.0, 131.8, 128.5, 123.1, 40.9, 38.0, 37.1, 27.0, 23.1;
LRMS (ESI) m/z (relative intensity) 425.1 (100% M+H+); HRMS
(ESI) m/z calcd for [C17H21N4O5S2]+: 425.0953, found 425.0951.
To a stirring solution of this phthalimide (11.0 g, 25.9 mmol) in
500 mL of EtOH was added anhydrous hydrazine (16.3 mL, 518
mmol). The reaction solution was heated at reflux for 30 min after
which time a white precipitate had formed. The reaction mixture
was then cooled to room temperature and partitioned between H2O
(300 mL) and CHCl3 (400 mL). The aqueous layer was extracted
with CHCl3 (1 × 100 mL) and the combined organic fractions were
washed with brine (1 × 150 mL), dried over Na2SO4 and
concentrated to give amine 36 (4.20 g, 55%) as a colorless oil which
required no further purification. IR (thin film) 3342 cm-1; 1H NMR
(360 MHz, CDCl3) δ 6.91 (s, 1H), 2.93 (s, 3H), 2.91 (s, 6H),
2.86-2.61 (m, 4H), 1.79-1.70 (m, 2H), 1.33 (br s, 2H); 13C (90
MHz, CDCl3) δ 150.2, 136.7, 128.2, 41.1, 40.7, 37.9, 31.4, 22.8;
LRMS (ESI) m/z (relative intensity) 295.1 (100% M + H+); HRMS
(ESI) m/z calcd for [C9H19N4O3S2]+, 295.0899, found 295.0917.
4,5-Dibromo-1-(2-trimethylsilanyl-ethoxymethyl)-1H-pyrrole-
2-carboxylic Acid [3-(3-dimethylsulfamoyl-2-methanesulfinyl-
3H-imidazol-4-yl)-propyl]-amide (38). A solution of SEM pro-
tected pyrrole 23e (0.306 g, 0.612 mmol) in 5 mL of DMF was
added to a neat mixture of amine 36 (0.150 g, 0.510 mmol) and
Na2CO3 (0.60 g, 0.56 mmol) via cannula. After 15 min, the reaction
solution was concentrated to give a brown oil. Purification of this
oil by flash column chromatography (EtOAc then 25% acetone/
EtOAc) gave amidosulfoxide 38 (0.269 g, 78%) as a white gum.
IR (thin film): 1651 cm-1; 1H NMR (360 MHz, CDCl3) δ 7.01 (s,
1H), 6.80 (t, J ) 5.4 Hz, 1H), 6.79 (s, 1H), 5.73 (s, 2H), 3.61 (t,
J ) 7.9 Hz, 2H), 3.46 (ddd, J ) 12.6, 6.8, 2.9 Hz, 2H), 3.03 (s,
3H), 2.99 (s, 6H), 2.91 (m, 1H), 2.76 (m, 1H), 1.96 (quint, J ) 7.2
Hz, 2H), 0.89 (t, J ) 8.2 Hz, 2H), -0.03 (s, 9H); 13C NMR (90
MHz, CDCl3) δ 160.1, 150.5, 136.3, 128.8, 128.7, 116.0, 110.8,
99.9, 75.1, 66.3, 40.8, 38.9, 38.2, 28.5, 23.2, 17.8, -1.46; LRMS
(ESI) m/z (relative intensity) 674.0 (100%, M + H+); HRMS (ESI)
m/z calcd for [C20H34N5O5S2Br2Si]+, 674.0137, found 674.0142.
(19) Rawal, V. H.; Cava, M. P. J. Am. Chem. Soc. 1986, 108, 2110–2112.
J. Org. Chem. Vol. 74, No. 9, 2009 3459

Feldman and Fodor
terization purposes. Method B: To a stirring solution of fused bicycle
40 (12.1 mg, 0.0229 mmol) in 850 µL of CH2Cl2 was added ICl
(0.0916 mmol, 146 µL of a freshly prepared 0.628 M solution in
CH2Cl2). The reaction solution immediately turned from colorless
to dark red. After 1 min, the reaction mixture was diluted with
CH2Cl2 to a volume of 20 mL and poured into 15 mL of NaHCO3
(sat.). The resulting solution was partitioned between CH2Cl2 and
H2O, and the aqueous layer was extracted with CH2Cl2 (1 × 20
mL). The organic fractions were combined, dried with Na2SO4, and
concentrated to give a black oil. Purification of this oil by flash
column chromatography (hexanes then Et2O) gave chlorotetracycle
41 (10.9 mg, 84%) as a light yellow solid whose spectral data were
identical with that of the product from Method A. mp. 177-178
°C; IR (thin film): 1714 cm-1; 1H NMR (360 MHz, CDCl3) δ 6.72
(s, 1H), 4.15 (dt, J ) 13.0, 9.0 Hz, 1H), 3.16-3.01 (m, 2H), 3.07
(s, 6H), 2.50 (s, 3H), 2.16 (ddd, J ) 15.0, 12.0, 3.2 Hz, 1H), 1.89
(m, 1H), 1.69 (m, 1H); 13C NMR (90 MHz, CDCl3) δ 169.6, 157.2,
127.0, 109.5, 105.3, 102.8, 100.0, 90.5, 38.1, 37.0, 35.8, 17.7, 15.7;
LRMS (ESI) m/z (relative intensity) 559.9 (35%, M + H+); HRMS
(ESI) m/z calcd for [C14H17N5O3S2Br2Cl]+, 559.8828, found
559.8854.
Dibromoagelaspongin•TFA Salt (6•CF3CO2H). To a stirring
solution of azidotetracycle 44 (89 mg, 0.20 mmol) in MeOH:THF
(8:2, 10 mL total volume) was added 10% Pd/C (9 mg) and the
reaction solution was placed under an atmosphere of H2 (balloon).
After 16 h, the reaction mixture was filtered through a thin pad of
Celite that was then washed with an additional 10 mL of MeOH.
TFA (100 µL) was added to the filtrate which, upon concentration,
gave dibromoagelaspongin methyl ether (103 mg, 100%) as its TFA
salt. A portion of this material (28 mg, 0.054 mmol) in 3 mL of
H2O was heated in a sealed tube at 90 °C and held at this
temperature for 16 h. The reaction solution was then cooled to room
temperature and concentrated to give a colorless oil. This oil was
redissolved in 10 mL of MeOH and 50 µL of TFA was added. The
solution was concentrated to give dibromoagelaspongin (6) (26 mg,
1
99%) as its TFA salt. IR (thin film): 3183, 1682 cm-1; H NMR
(300 MHz, d4-MeOH) δ 6.78 (s, 1H), 4.05 (m, 1H), 3.25 (m, 1H),
2.15 (m, 1H), 2.04 (m, 1H), 1.81-1.66 (m, 2H); 13C NMR (90
MHz, d4-MeOH) δ 159.1, 158.7, 128.1, 110.5, 106.3, 105.4, 93.4,
91.2, 37.3, 34.5, 18.0; LRMS (ESI) m/z (relative intensity) 404.2
(100%, M + H+); HRMS (ESI) m/z calcd for [C11H12N5O2Br2]+,
403.9358, found 403.9354.
2-Phenylsulfanyl-4-(3,4,5-tribromo-1H-pyrrole-2-carbonyl)-
4,5,6,7-tetrahydro-imidazo[4,5-b]pyridine-1-sulfonic Acid Dim-
ethylamide (53). A stirring solution of bicycle 32 (15 mg, 0.025
mmol) in a 1:1 mixture of CH2Cl2:THF (1 mL total volume) was
treated with LiBr (43 mg, 0.50 mmol) and then with NCS (3.3 mg,
0.025 mmol). After 5 min, the reaction mixture was diluted with
CH2Cl2 to a volume of 20 mL and poured into 10 mL of ice-cold
H2O. The resulting solution was partitioned between CH2Cl2 and
H2O and the organic layer was sequentially washed with H2O (1
× 10 mL), saturated aqueous NaHCO3 (1 × 10 mL), 1 M H3PO4
(1 × 10 mL), dried over Na2SO4 and concentrated to give a yellow
oil. Purification of this oil by flash column chromatography (3:2
Et2O/hexanes) gave tribromobicycle 53 (6 mg, 36%) as a yellow
oil along with bromotetracycle 54 (1 mg, 5%). Spectral data for
53: IR (thin film) 1600 cm-1; 1H NMR (360 MHz, CDCl3) δ 10.2
(br s, 1H), 7.30 (s, 5H), 3.89 (dd, J ) 5.4, 4.0 Hz, 2H), 3.04 (s,
6H), 2.90 (t, J ) 6.5 Hz, 2H), 2.04 (quint, J ) 6.1 Hz, 2H); 13C
NMR (90 MHz, CDCl3) δ 157.9, 141.4, 137.6, 133.4, 129.2, 129.0,
128.9, 126.8, 117.5, 103.9, 103.3, 102.7, 43.6, 38.7, 23.1, 22.7;
LRMS (ESI) m/z (relative intensity) 665.9 (100% M + H+); HRMS
(ESI) m/z calcd for [C19H19N5O3S2Br3]+, 665.8479, found 665.8476.
3H, 8H-Imidazo[4,5-b]pyrrolo[1′,2′:3,4]imidazo[1,2-a]pyri-
dine-3-sulfonamide, 3a,10,11,12-tetrabromo-3a,4,5,6-tetrahydro-
N,N-dimethyl-2-(phenylthio)-8-oxo- (54). A stirring solution of
bicycle 32 (10 mg, 0.017 mmol) in a 1:1 mixture of CH2Cl2:THF
(1 mL total volume) was treated with LiBr (30 mg, 0.34 mmol)
and then with NCS (4.5 mg, 0.034 mmol). After 5 min, the black
reaction mixture was diluted with CH2Cl2 to a volume of 20 mL
and poured into 10 mL of ice-cold H2O. The resulting solution
was partitioned between CH2Cl2 and H2O, and the organic layer
was sequentially washed with H2O (1 × 10 mL), saturated aqueous
NaHCO3 (1 × 10 mL), 1 M H3PO4 (1 × 10 mL), dried over Na2SO4
and concentrated to give a black oil. Purification of this oil by flash
column chromatography (14:5:1 Et2O/hexanes/CH2Cl2) gave tet-
rabromotetracycle 54 (8.4 mg, 66%) as a yellow oil. IR (thin film)
3H,8H-Imidazo[4,5-b]pyrrolo[1′,2′:3,4]imidazo[1,2-a]pyridin-
8-one, 2-(Methylthio)-10,11-dibromo-3a,4,5,6-tetrahydro-3a-
methoxy- (42). To a stirring solution of chlorotetracycle 41 (0.130
g, 0.231 mmol) in MeOH/THF (2.5:1, 7 mL total) was added HCl
(conc, 100 µL) and reaction solution was heated to reflux. After
3 h, the reaction solution was cooled to room temperature and
concentrated. The residue was dissolved in 50 mL of CH2Cl2 and
washed with NaHCO3 (1 × 20 mL). The organic fraction was dried
over Na2SO4 and concentrated to give a yellow oil. Purification of
this oil by flash column chromatography (Et2O) gave methyl ether
42 (70 mg, 67%) as a white solid. mp 203-204 °C; IR (thin film)
1
3218, 1684 cm-1; H NMR (300 MHz, d8-THF+ DCl) δ 6.71 (s,
1H), 3.93 (m, 1H), 3.44 (s, 3H), 3.36 (m, 1H), 2.86 (s, 3H), 2.64
(m, 1H), 2.09 (m, 1H), 1.72-1.61 (m, 2H); 13C NMR (75 MHz,
d8-THF) δ 175.6, 157.9, 127.6, 109.9, 105.5, 105.2, 96.3, 93.5,
52.8, 37.5, 29.0, 16.8, 14.7; LRMS (ESI) m/z (relative intensity)
449.0 (100%
M +
H+); HRMS (ESI) m/z calcd for
[C13H15N4O2Br2S]+, 448.9282, found 448.9250.
3H,8H-Imidazo[4,5-b]pyrrolo[1′,2′:3,4]imidazo[1,2-a]pyridin-
8-one, 2-Azido-10,11-dibromo-3a,4,5,6-tetrahydro-3a-methoxy-
(44). A stirring solution of sulfide 42 (400 mg, 0.889 mmol) in 20
mL of CH2Cl2 was treated with mCPBA (219 mg, 0.889 mmol).
After 30 min, the reaction mixture was poured into 20 mL of a
saturated NaHCO3 solution. The resulting solution was partitioned
between CH2Cl2 and H2O, and the aqueous layer was extracted
with CH2Cl2 (2 × 20 mL). The organic fractions were combined,
dried with Na2SO4, and concentrated to give a yellow oil (330 mg)
that was used without further purification. A portion of the crude
sulfoxide (200 mg) was immediately up in 20 mL of MeCN and
treated with ZnI2 (10 mg). After 2 min, TMSN3 (306 µL, 2.33
mmol) was added to the reaction solution followed by the addition
of H2O (50 µL). After 5 min, the resulting yellow solution was
diluted with EtOAc to a total volume of 40 mL and poured into 20
mL of a saturated NaHCO3 solution. The resulting solution was
partitioned between EtOAc and H2O, and the aqueous layer was
extracted with EtOAc (1 × 20 mL). The organic fractions were
combined, dried with Na2SO4, and concentrated to give a yellow
oil. Purification of this oil by flash column chromatography
(gradient, 70% Et2O/hexanes to 100% Et2O) gave azidotetracycle
44 (89 mg, 37% over 2 steps) as a tacky white solid. IR (thin film):
2146, 1713 cm-1; 1H NMR (360 MHz, d4-MeOH) δ 6.77 (s, 1H),
4.04 (m, 1H), 3.35 (s, 3H), 2.31 (m, 1H), 2.10 (m, 1H), 1.84-1.78
(m, 2H); 13C NMR (75 MHz, d6-DMSO) δ 160.3, 155.8, 126.1,
109.2, 104.8, 104.6, 100.0, 93.3, 53.0, 36.3, 28.4, 16.2; LRMS (ESI)
m/z (relative intensity) 444.1 (30%, M + H+); HRMS (ESI) m/z
for [C12H12N7O2Br2]+, 443.9419, found 443.9448. One proton signal
at ∼3.30 ppm is not reported due to overlap with solvent.
1
1716 cm-1; H NMR (360 MHz, CDCl3) δ 7.58-7.56 (m, 2H),
7.45-7.38 (m, 3H), 4.12 (dt, J ) 13.3, 7.9 Hz, 1H), 3.17 (s, 6H),
3.10 (ddd, J ) 14.5, 6.5, 2.5 Hz, 1H), 3.00 (m, 1H), 2.44 (ddd, J
) 14.4, 7.9, 3.2 Hz, 1H), 1.89-1.78 (m, 1H), 1.74-1.63 (m, 1H);
13C NMR (75 MHz, CDCl3) δ 169.1, 156.0, 135.5, 130.7, 129.6,
125.8, 124.9, 108.7, 102.8, 100.5, 97.6, 83.2, 39.5, 38.2, 36.2, 18.5;
LRMS (ESI) m/z (relative intensity) 743.8 (100% M + H+); HRMS
(ESI) m/z calcd for [C19H18N5O3S2Br4]+, 743.7585, found 743.7524.
4-(4,5-Dibromo-3-iodo-1H-pyrrole-2-carbonyl)-2-methylsul-
fanyl-4,5,6,7-tetrahydro-imidazo[4,5-b]pyridine-1-sulfonic Acid
Dimethylamide (55). To a stirring solution of bicycle 40 (9.0 mg,
0.017 mmol) in 1 mL of CH2Cl2 was added NIS (3.8 mg, 0.017
mmol). After 5 min, the reaction mixture was diluted with CH2Cl2
3460 J. Org. Chem. Vol. 74, No. 9, 2009

Extending Pummerer Reaction Chemistry
to a volume of 20 mL and the yellow solution was poured into
saturated NaHCO3 (10 mL). The reaction solution was partitioned
between CH2Cl2 and H2O, and the aqueous layer was extracted
with CH2Cl2 (1 × 20 mL), dried over Na2SO4 and concentrated to
give a yellow oil. Purification of this oil by flash column
chromatography (80% Et2O/hexanes) gave iodopyrrole 55 (9.0 mg,
sequentially washed with H3PO4 (1 M, 1 × 20 mL) followed by
NaHCO3 (sat., 20 mL), dried with Na2SO4, and concentrated to
give diodotetracycle 56 (52 mg, 64%) as a yellow oil which required
1
no further purification. IR (thin film): 1638 cm-1; H NMR (300
MHz, CDCl3) δ 4.27 (quint, J ) 6.5 Hz, 1H), 3.07 (m, 1H), 3.10
(s, 6H), 2.85 (dt, J ) 6.2, 4.4 Hz, 2H), 2.51 (s, 3H), 1.69-1.50
(m, 2H, overlaps with H2O peak); 13C NMR (75 MHz, CDCl3) δ
169.6, 156.7, 128.3, 113.5, 101.8, 101.1, 65.2, 63.0, 44.1, 38.1,
37.3, 19.9, 16.0; LRMS (ESI) m/z (relative intensity) 777.8 (35%,
M + H+); HRMS (ESI) m/z calcd for [C14H16N5O3S2Br2I2]+,
777.7151, found 777.7192.
81%) as a white solid. mp 142-144 °C; IR (thin film) 1618 cm-1
;
1H NMR (400 MHz, CDCl3) δ 12.1 (br s, 1H), 3.90 (br m, 2H),
3.00 (s, 6H), 2.87 (t, J ) 6.4 Hz, 2H), 2.52 (s, 3H), 2.14-2.08 (br
m, 2H); 13C NMR (75 MHz, CDCl3) δ 158.5, 144.1, 136.9, 128.7,
117.9, 109.1, 103.1, 75.9, 43.9, 38.6, 22.62, 22.60, 15.7; LRMS
(ESI) m/z (relative intensity) 651.9 (100% M + H+); HRMS (ESI)
m/z calcd for [C14H17N5O3S2Br2I]+, 651.8184, found 651.8187.
3H,8H-Imidazo[4,5-b]pyrrolo[1′,2′:3,4]imidazo[1,2-a]pyridine-
3-sulfonamide, 10,11-Dibromo-3a,4,5,6-tetrahydro-3a,9-diiodo-
N,N-dimethyl-2-(methylthio)-8-oxo- (56). To a stirring solution
of fused bicycle 40 (59 mg, 0.11 mmol) in 6 mL of CH2Cl2 was
added NIS (99 mg, 0.44 mmol). The reaction solution instantly
changed from colorless to red. After 1 min, the reaction mixture
was poured into 20 mL of NaHCO3 (sat.). The resulting solution
was partitioned between CH2Cl2 and H2O, and the aqueous layer
was extracted with CH2Cl2 (1 × 25 mL). The organic fractions
were combined and dried with Na2SO4. This dried solution was
poured through a thin pad of silica and additional CH2Cl2 (100
mL) was used to rinse the silica until no colored impurities
continued to elute. The filtrate was discarded. Next, the silica was
rinsed with Et2O (50 mL), and the filtrate was concentrated to give
a yellow oil. This oil was redissolved in CH2Cl2 (20 mL) and
Acknowledgment. We thank the National Science Founda-
tion for support of this work through CHE 0808983 and for
support of the Huck Institutes of the Life Sciences X-ray
crystallography facility through CHE 0131112.
Supporting Information Available: General Experimental
procedures; copies of 1H and 13C NMR spectra for 24a-e, 25,
26b-e, 27-29a/b, 31-33a/b, 34a/b, 36, 42 and 53-55; X-ray
1
crystallographic data for 33a. Note that copies of H and 13C
NMR spectra for 38-41, 44, 56 and 6 were submitted as
Supporting Information along with the publication listed in ref
9. This material is available free of charge via the Internet at
http://pubs.acs.org.
JO900283G
J. Org. Chem. Vol. 74, No. 9, 2009 3461