Studies in chlorin chemistry. II. A versatile synthesis of dihydrodipyrrins

Article abstract of DOI:10.1021/jo050907l

Dihydrodipyrrins are key building blocks for the synthesis of hydroporphyrins, many of which have important biological activity. The title compounds were prepared in stereo- and regioselective fashion by a three-step sequence consisting of (1) Pd(0)-catal

Full text of DOI:10.1021/jo050907l

Studies in Chlorin Chemistry. II. A Versatile Synthesis of
Dihydrodipyrrins
William G. O’Neal, William P. Roberts, Indranath Ghosh, and Peter A. Jacobi*
Burke Chemical Laboratory, Dartmouth College, Hanover, New Hampshire 03755
peter.a.jacobi@dartmouth.edu
Received May 6, 2005
Dihydrodipyrrins are key building blocks for the synthesis of hydroporphyrins, many of which have
important biological activity. The title compounds were prepared in stereo- and regioselective fashion
by a three-step sequence consisting of (1) Pd(0)-catalyzed coupling-cyclization of 2-iodopyrroles with
γ-alkynoic acids to afford enelactones of the desired substitution pattern, (2) methylenation at the
lactone carbonyl group employing the Petasis reagent, and (3) in situ enol-ether hydrolysis and
amination of the resultant 1,4-diketone to close the pyrroline ring (nine examples). Yields for each
step were generally high, although in substrates not blocked by geminal substitution aromatization
to a dipyrromethane is a competing side reaction.
Introduction
Hydroporphyrins are partly reduced derivatives of the
porphyrin ring system, the most common of which are
members of the chlorin (1a, ∆7,8,12,13), isobacteriochlorin
(1b, ∆12,13), and bacteriochlorin (1c, ∆7,8) families (Figure
1; ∆ signifies the location of double bonds).1 Aromatic
tetrapyrroles of type 1a-c have diverse functions in
nature. Representative examples include chlorophyll a,
the ubiquitous chlorin that regulates photosynthesis in
green plants, algae, and cyanobacteria; bonellin (2), the
sex-differentiating chlorin of the marine worm Bonella
viridis;2 and siroheme (3), the isobacteriochlorin pros-
thetic group of numerous sulfite and nitrite reductases.3
In its reduced Fe-free form, 3 is also a key intermediate
in vitamin B12 biosynthesis.4 A number of hydroporphy-
rins are also of medical interest. These include members
of the chlorin and bacteriochlorin families, which because
of their favorable photophysical properties are attractive
(1) (a) Battersby, A. R. Nat. Prod. Rep. 2000, 17, 507. (b) Montforts,
F.-P.; Gerlach, B.; Ho¨per, F. Chem. Rev. 1994, 94, 327.
(2) Agius, L.; Ballantine, J. A.; Ferrito, V.; Jaccarini, V.; Murray-
Rust, P.; Pelter, A.; Psaila, A. F.; Schembri, P. J. Pure Appl. Chem.
1979, 51, 1847.
FIGURE 1. Some naturally occurring hydroporphyrins.
agents in tumor photodynamic therapy (PDT). This
technique employs photostimulated production of singlet
oxygen to selectively eradicate malignant tissue.5 In a
nonrelated area, bacteriochlorins of the tolyporphin (4)
(3) (a) Murphy, M. J.; Siegel, L. M. J. Biol. Chem. 1973, 248, 6911.
(b) Murphy, M. J.; Siegel, L. M.; Tove, S. R.; Kamin, H. Proc. Natl.
Acad. Sci. U.S.A. 1974, 71, 612.
(4) Battersby, A. R.; Frobel, K.; Hammerschmidt, F.; Jones, C. J.
Chem. Soc., Chem. Commun. 1982, 455.
10.1021/jo050907l CCC: $30.25 © 2005 American Chemical Society
Published on Web 07/29/2005
J. Org. Chem. 2005, 70, 7243-7251
7243

O’Neal et al.
SCHEME 1
SCHEME 2
class have been shown to reverse tumor multidrug
resistance (MDR) and may find use in cancer chemo-
therapy.6
In contrast to chlorophyll a, most naturally occurring
hydroporphyrins are available only in exceedingly small
quantities, which contributes to their continuing interest
as synthetic targets. Many de novo syntheses of 1a-c
employ dihydrodipyrrin intermediates of type 8, which
function as either “western” A,D- or “northern” A,B-ring
building blocks (Scheme 1). The earliest and still most
frequently applied syntheses of 8 involve reductive cy-
clization of nitroketones of general structure 7, them-
selves prepared by Michael addition of enones 5 with
pyrrole derivatives 6.7 Battersby in particular has made
elegant use of this strategy in a number of chlorin and
isobacteriochlorin syntheses, as well as in biosynthetic
studies on vitamin B12.7a-c More recently, Lindsey et al.
have adapted this approach to the synthesis of chlorins
bearing peripheral synthetic “handles”.7c,d Typically,
conjugate addition of 6a or 6b with mesityl oxide (5, C
) Me, D ) H), followed by reductive cyclization, affords
dihydrodipyrrins 8a,b (C ) Me, D ) H) in 15-40%
overall yields.7 The relatively modest yields for this
sequence are mainly due to difficulty controlling the key
reductive cyclization of 7 to 8, which invariably affords
products contaminated by over-reduction at C5-C6. This
problem is partly addressed by reoxidation of the derived
tetrahydrodipyrrins employing Pb(OAc)4, although
chemoselectivity is generally poor.7a,8 A more satisfactory
solution is to employ the C5-C6-reduced form of dihy-
drodipyrrins 8 to prepare tetrahydrobilenes, at which
point oxidation can be effected concomitant with ring
closure.7f In either case, the choice of groups C is very
limited and C8 must be monosubstituted. Finally, this
methodology fails for C5-meso-substituted dihydrodi-
pyrrins (i.e., R * H).8
Several strategies have been explored in attempts to
develop more general syntheses of dihydrodipyrrins 8.
Building upon earlier studies by Gossauer,9 Battersby et
al. showed that monothioimide derivatives of type 9
undergo condensation with cyano-stabilized Wittig re-
agents 10 to afford variable yields of dihydropyrro-
methenones 11 (R ) CN, X ) NH; Scheme 2).10
These last materials were then converted to dihy-
drodipyrrins 8 by a seven-step sequence initiated by
removal of the meso-cyano group.11 Among other ex-
amples, this strategy was employed in an enantioselec-
tive synthesis of the Fe-free form of siroheme (3), in which
the monothioimide starting material was prepared by
either of two routes: oxidative degradation of vitamin
12
(5) (a) Pandey, R. K.; Zheng, G. Porphyrins as Photosensitizers in
Photodynamic Therapy. In The Porphyrin Handbook; Kadish, K. M.,
Smith, K. M., Guilard, R., Eds.; Academic Press: San Diego, CA, 2000;
Vol. 6, pp 157-230. (b) Lavi, A.; Weitman, H.; Holmes, R. T.; Smith,
K. M.; Ehrenberg, B. Biophys. J. 2002, 82, 2101. (c) Bronshtein, I.;
Afri, M.; Weitman, H.; Frimer, A. A.; Smith, K. M.; Ehrenberg, B.
Biophys. J. 2004, 87, 1155.
B12 or in a 12-step sequence starting with R-malic
acid.13 In large part the success of this strategy depends
on the availability of monothioimides 9, which even in
simple cases are difficult to prepare in regioselective
(6) Prinsep, M. R.; Caplan, F. R.; Moore, R. E.; Patterson, G. M. L.;
Smith, C. D. J. Am. Chem. Soc. 1992, 114, 385.
(9) Gossauer, A.; Hinze, R.-P. J. Org. Chem. 1978, 43, 283.
(10) (a) Arnott, D. M.; Battersby, A. R.; Harrison, P. J.; Henderson,
G. B.; Sheng, Z.-C. J. Chem. Soc., Chem. Commun. 1984, 525. (b)
Battersby, A. R.; Turner, S. P. D.; Block, M. H.; Sheng, Z.-C.;
Zimmerman, S. C. J. Chem. Soc., Perkin Trans. 1 1988, 1577. (c)
Arnott, D. M.; Harrison, P. J.; Henderson, G. B.; Sheng, Z.-C.; Leeper,
F. J.; Battersby, A. R. J. Chem. Soc., Perkin Trans. 1 1989, 265.
(11) Battersby, A. R.; Block, M. H.; Leeper, F. J.; Zimmerman, S.
C. J. Chem. Soc., Perkin Trans. 1 1992, 2189.
(12) Block, M. H.; Zimmerman, S. C.; Henderson, G. B.; Turner, S.
P. D.; Westwood, S. W.; Leeper, F. J.; Battersby, A. R. J. Chem. Soc.,
Chem. Commun. 1985, 1061.
(13) Whittingham, W. G.; Ellis, M. K.; Guerry, P.; Henderson, G.
B.; Mu¨ller, B.; Taylor, D. A.; Leeper, F. J.; Battersby, A. R. J. Chem.
Soc., Chem. Commun. 1989, 1116.
(7) (a) Harrison, P. J.; Sheng, Z.-C.; Fookes, C. J. R.; Battersby, A.
R. J. Chem. Soc., Perkin Trans. 1 1987, 1667. (b) Battersby, A. R.;
Cutton, C. J.; Fookes, C. J. R.; Turner, S. P. D. J. Chem. Soc., Perkin
Trans. 1 1988, 1557. (c) Battersby, A. R.; Dutton, C. J.; Fookes, C. J.
R. J. Chem. Soc., Perkin Trans. 1 1988, 1569. (d) Strachan, J.-P.;
O’Shea, D. F.; Balasubramanian, T.; Lindsey, J. S. J. Org. Chem. 2000,
65, 3160. (e) Balasubramanian, T.; Strachan, J.-P.; Boyle, P. D.;
Lindsey, J. S. J. Org. Chem. 2000, 65, 7919. (f) Taniguchi, M.; Ra, D.;
Mo, G.; Balasubramanian, T.; Lindsey, J. S. J. Org. Chem. 2001, 66,
7342.
(8) Arnott, D. M.; Harrison, P. J.; Henderson, G. B.; Sheng, Z.-C.;
Leeper, F. J.; Battersby, A. R. J. Chem. Soc., Perkin Trans. 1 1989,
265.
7244 J. Org. Chem., Vol. 70, No. 18, 2005

A Versatile Synthesis of Dihydrodipyrrins
SCHEME 3a
fashion. In view of this difficulty, and also because of the
excessive number of manipulations following coupling,
the authors later characterized this approach as “experi-
mentally very demanding” and not suitable for larger-
scale preparations.14
In follow-up studies the Battersby group showed that
S-pyridylthioesters 12 undergo clean condensation with
pyrrole Grignard reagents of type 13c (Y ) Me), affording
good to excellent yields of acyl lactones 14c (Scheme 2;
R ) OH, X ) O, Y ) Me). Unfortunately, though, it
proved impossible to employ Grignard reagents 13a,b
having Y ) CO2t-Bu or H. This limitation necessitated
extensive modifications to both the C1- and C5-substitu-
ents in 14c before ultimate conversion to dihydrodipyr-
rins 8. In a noteworthy example of this methodology,
Battersby et al. developed a 14-step synthesis of the
chiral thioester derivative 12d beginning with L-glutamic
acid (Scheme 2).14,15 Thioester 12d was then converted
in 13 steps to the dihydrodipyrrin 8j, which was a key
intermediate in their synthesis of haem d1.
In this article, we describe a versatile synthesis of
dihydrodipyrrins that is regio- and stereoselective, ac-
commodates a wide variety of pyrrole- and meso-substit-
uents, and can be adapted to prepare homochiral dihy-
drodipyrrins.
a Reagents and conditions: i. Pd(PPh3)4, BnNEt3Cl, CH3CN,
NEt3, reflux. ii. NH3, THF,-33°C to room temperature; iii. Mont-
morillonite K-10, THF, room temperature. iv. Lawesson’s reagent,
toluene, 100°C. v. MeI, THF, reflux. vi. ZnI2, THF, MeMgBr, room
temperature, then 19a-Z, PdCl2(PPh3)2, 70°C.
Results and Discussion
Method A. In a recent article, we described efficient
syntheses of E-enelactones 17-E, which were prepared
in 85-96% yield by Pd(0)-catalyzed coupling/cyclization
of iodopyrrole 15 with alkyne acids 16 (Scheme 3).16
Lactones 17-E were subsequently converted to E,Z-
mixtures of thioimidates 19 by a four-step sequence
involving aminolysis/cyclodehydration to afford lactams
18, followed by thiolactam formation and S-methylation.16b
We intended to convert both isomers of thioimidates 19
to the corresponding dihydrodipyrrins 8, potential inter-
mediates in a new chlorin synthesis. Following this
strategy, methylation of 19a-Z with the reagent system
Pd(0)/MeZnI gave an 88% yield of 8a as a single isomer.
Surprisingly, however, 19a-E failed to react under these
conditions and many others.
Eventually this difference was traced to a selective
activating effect of Zn, which serves to polarize the
thioimidate C-S bond in 19a-Z by chelation (Figure 2).17
Control experiments showed that such activation is
necessary for oxidative insertion of Pd(0). Chelation of
this type is not possible for 19a-E, which is also sterically
prohibited from simple complexation (cf. Zn-19-E; for a
detailed discussion of this mechanism see ref 17). Con-
sequently, the desired cross-coupling reaction fails. Com-
pounding this problem, the ratio of 19-Z:19-E decreases
FIGURE 2. Selective Zn(2) activation of 19a-Z (R ) Me).
with increasing size of the meso-substituent (i.e., H >
Me . Ph), making it impractical to incorporate larger
alkyl groups. This is because with small R-groups the
planar Z-configuration is stabilized by hydrogen bond-
ing,17 but as R increases in size this effect becomes less
important. In contrast, the twisted E-isomer 19-E can
accommodate large meso-substituents with minimal non-
bonded interactions (in a closely related example the
rings are skewed by 52°; cf. ref 17). For R ) H the
Z-thioimidate 19c-Z is obtained nearly exclusively by the
route outlined in Scheme 3, while for R ) Ph the E:Z
ratio approaches 50:50. Unfortunately, under the cou-
(14) Mackman, R. L.; Micklefield, J.; Block, M. H.; Leeper, F. J.;
Battersby, A. R. J. Chem. Soc., Perkin Trans. 1 1997, 2111.
(15) Mickelfield, J.; Beckmann, M.; Mackman, R. L.; Block, M. H.;
Leeper, F. J.; Battersby, A. R. J. Chem. Soc., Perkin Trans. 1 1997,
2123.
(16) (a) Jacobi, P. A.; Lanz, S.; Ghosh, I.; Leung, S. H.; Lower, F.;
Pippin, D. Org. Lett. 2001, 3, 831. (b) Enelactones 17 are obtained
exclusively as the E-isomers under kinetic control, as dictated by
mechanistic considerations.17 The stereochemistry of lactams 18 and
thioimidates 19 depends mainly upon steric factors (cf. Figure 2).
Dihydrodipyrrins 8 generally strongly favor the Z-stereochemistry due
to intramolecular H-bonding.
(17) Ghosh, I.; Jacobi, P. A. J. Org. Chem. 2002, 67, 9304.
J. Org. Chem, Vol. 70, No. 18, 2005 7245

O’Neal et al.
SCHEME 4
SCHEME 5
pling conditions 19-Z and 19-E do not equilibrate. With
an unavoidable loss of material at such a late stage we
set out to develop a more efficient synthesis of dihy-
drodipyrrins 8.
tuted lactams 18 (Y ) H) were either unreactive or
suffered slow decomposition employing either the Petasis
reagent (TiCp2Me2)21 or the generally more reactive
Tebbe reagent [Cp2TiCH2ClAl(CH3)].22 The most effective
lactam activating groups are reported to be methyl- and
tert-butylcarbamates,20a and we therefore prepared the
corresponding enelactams 18 with Y ) CO2Me and Y )
CO2t-Bu. In both cases, these materials were difficult to
access because of competing reaction at the pyrrole
nitrogen. In any event modifications of this type were
not productive since we still could detect no formation of
enamines 23. Finally, we made many efforts to prepare
N-trimethylsilylated derivatives 18 where Y ) TMS,
since lactams of this type are known to undergo Petter-
son-like olefination with MeLi (i.e., 18 f 24 f 8).23
Unfortunately, however, these attempts were uniformly
unsuccessful, perhaps due to steric hindrance.
Method D. In contrast to the low reactivity of enelac-
tams 18, enelactones 17 were excellent substrates for
methylenation (Scheme 6). This transformation was first
accomplished employing the Tebbe reagent, in which case
it was best to directly hydrolyze the crude enol ethers
25 using a biphasic system of dichloromethane and 50%
aqueous TFA. Partly this was because of the cumbersome
workup involved in isolating 25. Nevertheless, on rela-
tively small scales this procedure routinely afforded
overall yields of 80-85% in the conversion of lactone 17a
to diketone 20a. Similar results were obtained with the
Petasis reagent, which in our experience is more easily
prepared, less sensitive to air and moisture, and more
economical. Therefore, we employed this reagent for
optimization studies and scale-up with other lactones.
Method B. In principle, dihydodipyrrins 8a-c might
be prepared in a more direct fashion by condensation of
ammonia with diketones 20a-c, themselves obtained by
ring opening of enelactones 17a-c with methyl anion or
an equivalent species (Scheme 4). This approach was
explored extensively with MeLi and MeMgBr but with
little success. Invariably these reactions gave a complex
mixture of multiple addition products containing only
small amounts of the desired diketones 20. Somewhat
better results were obtained with sterically hindered
nucleophiles, including lithiotrimethylsilylmethane (LiCH2-
SiMe3). This reagent is known to convert secondary and
tertiary esters to methyl ketones,18 and it gave essentially
quantitative yields of trimethylsilyl ketones 21b,c upon
reaction with enelactones 17b,c. Unfortunately, though,
these key intermediates were relatively unstable, espe-
cially with respect to intramolecular aldol condensation.
Compound 21c (R ) H), for example, produced significant
quantities of cyclopentenone 22c upon attempted ring
closure with a variety of ammonia sources. The aldol
reaction pathway appears to be facilitated by the TMS
group, since the parent methyl ketones 20, once formed,
are relatively easy to handle (vide infra). Most likely this
is due to Si-assisted dehydration of the initial aldol
product, for which there is ample precedent (see also
Scheme 5).19 Finally, since the use of such highly nucleo-
philic reagents might be incompatible with substrates
containing other ester groups, this route was put aside.
Method C. We also explored the possibility that
enelactams 18 might be directly converted to dihy-
drodipyrrins 8 by methylenation followed by tautomer-
ization (Scheme 5, 18 f 23 f 8). A number of closely
related transformations have been reported with both
amides and lactams, although activation by N-substitu-
tion is usually required.20 In the present case unsubsti-
We took advantage of a number of modifications
reported by Payack et al., including that in which TiCp2-
Me2 is prepared in situ as a toluene solution rather than
isolating the solid reagent.24 Although the concentration
of this solution can be assayed, in practice this was
(21) Petasis, N. A.; Bzowej, E. I. J. Am. Chem. Soc. 1990, 112, 6392.
(22) (a) Pine, S. H.; Pettit, R. J.; Geib, G. D.; Cruz, S. G.; Gallego,
C. H.; Tijerina, T.; Pine, R. D. J. Org. Chem. 1985, 50, 1212. (b) Tebbe,
R. N.; Parshall, G. W.; Reddy, G. S. J. Am. Chem. Soc. 1978, 100, 3611.
(23) (a) Hua, D. H.; Miao, S. W.; Bharathi, S. N.; Katsuhira, T.;
Bravo, A. A. J. Org. Chem. 1990, 55, 3682. (b) Ahn, Y.; Cardenas, G.
I.; Yang, J.; Romo, D. Org. Lett. 2001, 3, 751.
(18) Demuth, M. Helv. Chim. Acta 1978, 61, 3136.
(19) (a) Hudrlik, P. F.; Peterson, D. J. Am. Chem. Soc. 1974, 97,
1464. (b) Hudrlik, P. F.; Peterson, D. Tetrahedron Lett. 1974, 1133.
(c) Utimoto, K.; Obayashi, M.; Nozaki, H. J. Org. Chem. 1976, 41, 2940.
(20) (a) Petasis, N. A.; Lu, S.-P. Tetrahedron Lett. 1995, 36, 2393.
(b) Herdeis, C.; Heller, E. Tetrahedron: Asymmetry 1993, 4, 2085.
7246 J. Org. Chem., Vol. 70, No. 18, 2005

A Versatile Synthesis of Dihydrodipyrrins
SCHEME 6a
procedure routinely afforded 80-85% yields of Z-dihy-
drodipyrrin 8b following chromatography.16b Moreover,
we were able to streamline this synthesis to a very
efficient “one-pot” transformation leading directly from
enol ether 25b to 8b. In this modification 25b was
dissolved in DMF and treated with 1 M HCl, which
effected rapid in situ hydrolysis to diketone 20b with
little or no decomposition (stronger acidic conditions,
including 50% aqueous TFA, caused significant darken-
ing). After hydrolysis was complete (TLC), the reaction
mixture was treated with excess NH4OH and allowed to
continue stirring at room temperature. Finally, upon
consumption of 20b (TLC), the reaction was acidified to
pH 1 with 6 M HCl to afford ∼85% of dihydrodipyrrin
8b on gram scales (conditions A). Not surprisingly, a
mixture of E/Z-isomers of 8b is initially formed under
these conditions. However, upon equilibration the Z-
isomer is formed almost exclusively.16b
At this point we were disappointed to find that the
conditions optimized for 25b were not uniformly ap-
plicable. Despite repeated efforts, enol ether 17a (R )
Me) produced only trace amounts of dihydrodipyrrin 8a,
while enol ether 17c (R ) H) suffered extensive decom-
position. These results are illustrative of a common
phenomenon encountered in our study of these ring
systems, in that closely related substrates can have
markedly different reactivity profiles. In many cases,
including 17a and 17c, these differences depend mainly
upon the nature of the meso-substituent. For 17a,c, our
failure to effect amination probably reflects the sterically
less-hindered (and more reactive) nature of intermediate
diketones 20a,c, which in NH4OH undergo competing
aldol reactions (cf. Schemes 4 and 6). Working from this
premise, we were able to minimize side reactions with
minor adjustments, substituting NH4OAc/NEt3 for the
NH4OH employed with 25b. Under these conditions,
decomposition was greatly reduced and Z-dihydrodi-
pyrrins 8a-c were obtained in 70-80% yield after
warming at 55 °C (conditions B). In similar fashion, we
synthesized dihydrodipyrrins 8d,e utilizing the appropri-
ate iodopyrroles 15 and alkyne acids 16 (Table 1).
A diverse group of meso-groups R is accommodated by
this methodology, ranging from H to Ar to n-alkyl. The
most common side reaction observed in the Pd(0)-
catalyzed reaction of iodopyrroles 15 and alkyne acids
16 was reduction of 15 to the corresponding R-unsubsti-
tuted pyrroles (i.e., I ) H in 15), which with Pd[P(Ph)3]4/
MeCN/NEt3 ranged up to 20%. However, C-I bond
reduction was minimized by employing Pd2(dba)3 with
two molecular equivalents of P(Ph)3 or, alternatively,
employing a large excess of K2CO3 in DMF with Pd-
[P(Ph)3]4. Mechanistic studies of this unusual Pd-
promoted reductive deiodination, which if optimized could
prove useful, are underway.25
a Reagents and conditions: i. TiCp2Cl2, MeLi, toluene, 0°C, 1
h, then 17, cat. TiCp2Cl2, 80 °C, 6 h, then MeOH, H2O, NaHCO3,
40 °C, 14 h. ii. THF, 1 M HCl, room temperature, 1 h. iii. NH3,
THF, -33 °C to room temperature. iv. Montmorillonite K-10, THF,
room temperature. v. Tebbe reagent, THF, -40 °C, then 1:1 TFA:
H2O, CH2Cl2, 80%. vi. TiCp2Me2, THF, 65 °C, then 1:1 TFA/H2O,
CH2Cl2, 61%.
generally not necessary. Even in excess the Petasis
reagent is sufficiently selective that reaction can be
prevented at other potentially labile sites in the molecule
(vide infra). We also confirmed Payack’s observation that
small amounts of titanocene dichloride significantly
reduced the reaction period, affording complete conver-
sion after 6 h at 80 °C. Finally, the large volumes of
petroleum ether or pentane required to precipitate
titanium residues during workup were avoided by using
Payack’s improved isolation procedure, in which decom-
position of residual titanium species was accomplished
employing aqueous NaHCO3/methanol. Following this
protocol, we were able to obtain reproducible yields of
enol ethers 25a-c in the range of 87-92%. Alternatively,
hydrolysis of the crude products in biphasic dichlo-
romethane/aqueous TFA gave diketones 20a-c in 75-
82% overall yield.
In early experiments, we carried out the conversion of
diketones 20a-c to dihydrodipyrrins 8a-c employing
liquid ammonia followed by dehydration with Mont-
morillonite clay (Scheme 6, 80 - 86% yield). However,
this method was less suitable for larger scale prepara-
tions and was often accompanied by decomposition. We
also hoped to avoid the use of liquid ammonia. Utilizing
diketone 20b, we screened a large number of ammonia
sources and solvent combinations to effect ring closure
and subsequent dehydration. Of these, the combination
of 14.8 N NH4OH in DMF produced the best results and
was also experimentally simple. Typically a solution of
20b in DMF was stirred at room temperature with an
excess of NH4OH until initial hemiaminal formation was
complete (6-8 h, TLC). The reaction was then acidified
with 6 M HCl, which effected clean dehydration. This
Lipophilic substituents such as n-C5H11 and n-C10H21
are of interest in designing more effective chromophores
for PDT (vide supra), since they often have a favorable
impact on partitioning of tetrapyrroles between normal
and cancerous tissue,5a as well as on cell membrane
penetration.5b,c The alkyne acids 16d and 16e required
to introduce these substituents were readily prepared
(24) (a) Payack, J. F.; Hughes, D. L.; Cai, D.; Cottrell, I. F.;
Verhoeven, T. R. Organic Syntheses; Wiley & Sons: New York, 2002;
Collect. Vol. 79, p 19. (b) Payack, J. F.; Huffman, M. A.; Cai, D.;
Hughes, D. L.; Collins, P. C.; Johnson, B. K.; Cottrell, I. F.; Tuma, L.
D. Org. Process Res. Dev. 2004, 8, 256.
(25) For a similar observation, see: Leung, S. H.; Edington, D. G.;
Griffith, T. E.; James, J. J. Tetrahedron Lett. 1999, 40, 7189.
J. Org. Chem, Vol. 70, No. 18, 2005 7247

O’Neal et al.
TABLE 1. Isolated Yields for 17, 25, and 8a-i
entry
substituents
yield 17 (%)
yield 25 (%)
yielda 8 (%)
group I
a
b
c
d
e
f
g
h
i
A,B,D ) Me; C ) H; R ) Me
A,B,D ) Me; C ) H; R ) Ph
A,B,D ) Me; C,R ) H
A,B,D ) Me; C ) H; R ) C5H11
A,B,D ) Me; C ) H; R ) C10H21
A ) CH2CH2CO2Me () PMe); B,C ) Me; D,R ) H
A,B,C ) Me; D ) H; R ) H
A,B ) Me; C,D ) H; R ) H
A,B ) Me; C,D ) H; R ) Ph
96
85
98
77
69
73e
74
68
71
92
87
87
76
75
80
86
72
55
75c
85b, 79c
71c
60c
66c
group II
83d
65c, 82d
53c, 42d
45c, 0d
group III
a Yields refer to isolated and purified Z-8 only and in most cases are not optimized. An additional 0-10% of 8 was obtained as the
E-isomer. b Prepared according to conditions A. c Prepared according to conditions B. d Prepared according to conditions C. e Battersby et
al. prepared the Z-isomer corresponding to lactone 17f in 12 steps from 3,3-dimethylglutaric acid by a modification of the route described
in Scheme 2.14
SCHEME 7
SCHEME 8
Enelactones 17a-e differ only in the meso-substituent
R, having A, B, D ) Me and C ) H (cf. Table 1). However,
17f contains a propionate ester (PMe) that might undergo
competitive Petasis reaction (Scheme 8). In practice, this
group presented few complications so long as reaction
was stopped immediately upon consumption of starting
lactone 17f (TLC). Prolonged reaction times led to the
formation of the corresponding bis-enol ether derivative
30. However, even this side reaction might be eliminated
employing a “sacrificial” ester as very recently described
by Payack et al.24b
In all cases where D ) methyl the transformation of
diketones 20 to 8 was reliable, employing our standard
reaction conditions (NH4OAc/NEt3/DMF/55 °C, conditions
B), and afforded good overall yields of dihydrodipyrrins
8 (Table 1). With other regioisomers, however, it was
necessary to devise modified reaction conditions to avoid
side reactions. For example, the diketones 20f and 20g
(D ) H) were prone to competitive aldol condensation
and afforded significant quantities of byproducts 22f and
22g (Scheme 9). We attribute this change in reactivity
pattern to the fact that the initial aldol products 31f,g
are relatively strain-free, containing no adjacent quater-
nary centers. In contrast, the regioisomeric diketone 20c
is much less reactive toward aldol condensation to give
32c, which introduces significant steric strain. The only
exception thus far observed to this pattern is with the
silicon-substituted diketone 21c (cf. Scheme 4), which we
believe represents a special case. Also of interest, the
alternative aldol pathway involving nucleophilic partici-
pation of the methyl ketone is operative only under
from the commercially available propargyl alcohol de-
rivatives 26d and 26e, following the sequence outlined
in Scheme 7. Other alkyne acids were either com-
mercially available (16h), known from the literature (16a,
16c, 16g, and 16i)26,29b or produced in straightforward
fashion. For example, 16b (R ) Ph) was prepared in high
yield by Sonogashira coupling of iodobenzene with the
known alkyne ester 29b,26a followed by base hydrolysis.
(26) (a) Magnus, P.; Slater, M. J.; Principe, L. M. J. Org. Chem.
1989, 54, 5148. (b) Jacobi, P. A.; Liu, H. J. Org. Chem. 1999, 64, 1778.
(c) Skuballa, W.; Raduechel, B.; Vorbrueggen, H.; Elger, W.; Loge, O.;
Schillinger, E. (Schering.-G) Ger. Offen. 2,729,960, Jan 18, 1979; Chem.
Abstr. 1979, 90, 151671. (d) Doering, W. E.; Yamashita, Y. J. Am.
Chem. Soc. 1983, 105, 5368. (e) Rosenblum, S. B.; Huynh, T.; Alfonso,
A.; Davis, H. R. Tetrahedron 2000, 56, 5735.
(27) O’Neal, W. G.; Roberts, W. P.; Ghosh, I.; Jacobi, P. A. Studies
in Chlorin Chemistry. III. A Scalable Synthesis of Chlorins. Manuscript
in preparation.
(28) Jacobi, P. A.; Martinelli, M. J.; Polanc, S. J. Am. Chem. Soc.
1984, 106, 5594.
(29) (a) Schreiber, S. L.; Klimas, M. T.; Sammakia, T. J. Am. Chem.
Soc. 1987, 109, 5749. (b) Jacobi, P. A.; Pippin, D. Org. Lett. 2001, 3,
827 and references therein.
7248 J. Org. Chem., Vol. 70, No. 18, 2005

A Versatile Synthesis of Dihydrodipyrrins
SCHEME 9
SCHEME 10
down to 0 °C. Moreover, starting material was still
present after 72 h. Similar results were realized with
NH4+EtCO2- in DMF (conditions C), which while experi-
mentally more convenient afforded no improvement in
rate or selectivity relative to NH4OAc. After 24 h at 0
°C, 20h gave a 42% yield of dihydrodipyrrin 8h together
with trace amounts of 33h. In contrast, diketone 20i
proved to be essentially unreactive at 0 °C, with little 8i
evident after one week. On the basis of these experi-
ments, the most convenient route to dihydropyrrins 8h,i
remains the simple modification to conditions B described
above, where aminolysis is carried out at room temper-
ature (cf. Scheme 10). In any event, complications of this
type will be rare since in most cases aromatization is
blocked by geminal alkyl substituents.
Thus far we have identified three structural categories
of enol ethers 25 (groups I-III, Table 1), each of which
has distinct reactivity patterns that must be taken into
account for efficient conversion to 8. For clarity, we
summarize in Figure 3 our optimized conditions for the
one-pot conversion of each of these variants of 25 to
dihydrodipyrrins 8. Enol ethers of group I are perhaps
the most straightforward, since the geminal Me-substit-
uents at C-8 effectively block competing aldol reactions
(cf. bottom of Scheme 9). Of these, enol ether 25b (R )
Ph) is a particularly favorable substrate and is converted
to dihydrodipyrrin 8b in excellent overall yield employing
conditions A. The remaining members of this group are
somewhat more prone to side reactions and require a
buffered medium for aminolysis (conditions B; NH4OAc/
NEt3 replaces NH4OH; 25-55 °C). Enol ethers of group
II have geminal Me-substituents at C-7, and their hy-
drolysis products 20 undergo competitive aldol condensa-
tion linking C-5 and C-9 (cf. Scheme 9). This side reaction
is effectively eliminated by carrying out aminolysis under
conditions C, which in particular avoids basic conditions
(NH4OAc or NH4+EtCO2-/DMF/0 °C; no NEt3). Given that
most naturally occurring hydroporphyrins have geminal
substituents at C-7, conditions C may prove to be most
useful for the preparation of dihydrodipyrrin precursors
special circumstances.27 In the event, we eventually found
that both categories of aldol reaction are largely circum-
vented by effecting the aminolysis of diketones 20f,g at
0 °C with NH4OAc in DMF, omitting added NEt3. Even
better results were achieved employing the more soluble
(and anhydrous) ammonium propionate (NH4+EtCO2-),28
which afforded 80-85% yields of 8f,g after 18 h in DMF
at 0 °C (conditions C).
A complication of a different nature was encountered
with enol ethers 25h and 25i, where the initially formed
dihydrodipyrrins 8h,i underwent facile aromatization to
the dipyrromethanes 33h,i (Scheme 10). In our first
experiments enol ethers 25h,i were dissolved in DMF and
hydrolyzed in situ with 1 M HCl as described previously.
After diketone formation was complete (TLC) the reaction
mixture was treated with excess NH4OAc/NEt3 and
heated to 55 °C (conditions B). Under these conditions,
the sole identifiable products were dipyrromethanes 33h
(59%) and 33i (71%), respectively. In the case of enol
ether 25h aromatization was effectively eliminated by
carrying out aminolysis at ambient temperature and
limiting reaction time to 4 h. Longer reaction times led
to increasing formation of 33h even at room temperature.
Using the same reagents, we found that enol ether 25i
afforded 45% of dihydrodipyrrin 8i together with 10% 33i,
but the final aminolysis was much slower (∼48 h). To
better understand these transformations, diketones 20h,i
were isolated and subjected to a variety of cyclization
conditions. Diketone 20h proved to be significantly more
reactive toward aminolysis with NH4OAc in DMF, even
in the absence of NEt3. However, no particular benefit
was realized with this modification, which still afforded
small amounts of dipyrromethane 33h at temperatures
J. Org. Chem, Vol. 70, No. 18, 2005 7249

O’Neal et al.
syntheses of 2 and other naturally occurring hydropor-
phyrins are currently in progress.
Experimental Section
Representative Procedure for Syntheses of Dihy-
drodipyrrins 8. Dihydrodipyrrin 8b: 5-[(4,4-Dimethyl-
5-oxo-dihydro-furan-2-ylidene)-phenyl-methyl]-3,4-di-
methyl-1H-pyrrole-2-carboxylic Acid tert-Butyl Ester
(17b).16 A solution of 4.00 g (12.5 mmol) of iodopyrrole 15a,
3.70 g (18.3 mmol) of alkyne acid 16b, and 4.46 g (14.0 mmol)
of BnNEt3Cl in 146 mL of CH3CN and 29 mL NEt3 was purged
with argon for 10 min and then treated with 1.42 g (1.1 mmol)
of Pd(PPh3)4. The resulting solution was flushed with argon
and heated at reflux for 17 h. At the end of this period, the
reaction was concentrated to dryness under reduced pressure
and the residue was partitioned between CH2Cl2 and water.
The organic layer was washed with water, dried over MgSO4,
filtered, and concentrated under reduced pressure. The residue
was purified by flash chromatography (silica gel, EtOAc/
hexanes ) 1:20) to give 4.2 g (85%) of lactone 17b as a colorless
crystalline solid, mp 129 °C; Rf (1:5 EtOAc/hexanes) 0.60; IR
1
(thin film) 3289, 3056, 1800, 1672 cm-1; H NMR (500 MHz,
CDCl3) δ 1.38 (s, 6H), 1.60 (s, 9H), 1.91 (s, 3H), 2.33 (s, 3H),
2.72 (s, 2H), 7.31-7.39 (m, 5H), 8.59 (bs, 1H); 13C NMR (300
MHz, CDCl3) δ 9.7, 11.0, 25.0, 28.7, 39.7, 41.8, 80.9, 109.8,
119.9, 120.5, 126.1, 127.3, 128.4 (2C), 128.8 (2C), 128.9, 135.9,
147.2, 161.6, 179.9; Anal. Calcd. for C24H29NO4: C, 72.89; H,
7.39; N, 3.54. Found: C, 72.71; H, 7.43; N, 3.55.
5-[(4,4-Dimethyl-5-methylene-dihydro-furan-2-ylidene)-
phenyl-methyl]-3,4-dimethyl-1H-pyrrole-2-carboxylic Acid
tert-Butyl Ester (25b). A suspension of 1.47 g (5.97 mmol)
of titanocene dichloride in 16 mL of anhydrous toluene was
treated with 8.2 mL (13.1 mmol) of 1.6 M methyllithium in
diethyl ether over 5 min at 0 °C. After stirring the solution
for 1 h at 0 °C, we quenched the reaction with 14 mL of 6%
NH4Cl. The organic layer was separated and washed sequen-
tially with water and brine. It was then dried over Na2SO4
and filtered to give an orange solution of dimethyl titanocene.
This solution was treated with 0.502 g (1.26 mmol) of lactone
17b and 19 mg (0.076 mmol) of titanocene dichloride and
heated in the dark at 80 °C for 6 h. At the end of this period,
the flask was cooled to room temperature and 1.5 mL of
methanol, 63 mg of NaHCO3, and 15 µL of water were added.
The solution was stirred for 12 h at 40 °C. The resulting green
solution was filtered through a pad of Celite, which was further
washed with hexanes. If necessary, the filtrate was filtered a
second time, and the solvents were then removed under
reduced pressure. The resulting oil was purified by flash
chromatography (silica gel, EtOAc/hexanes ) 1:10 with 1%
NEt3) to give 0.43 g (87%) of enol ether 25b as a colorless
crystalline solid, mp 146-147 °C. Rf (1:4 EtOAc/hexanes) 0.51;
IR (thin film) 3303, 2970, 1660, 1240, 1129 cm-1; 1H NMR (500
MHz, CDCl3) δ 1.25 (s, 6H), 1.57 (s, 9H), 1.85 (s, 3H), 2.29 (s,
3H), 2.45 (s, 2H), 4.08 (d, J ) 2.44, 1H), 4.51 (d, J ) 2.44,
1H), 7.30 (m, 5H), 8.35 (br s, 1H); 13C NMR (500 MHz, CDCl3)
δ 9.8, 11.1, 27.5, 28.8, 39.5, 44.3, 80.6, 82.0, 104.1, 119.7, 126.2,
126.3, 128.3, 128.5, 131.1, 137.4, 154.1, 161.6, 170.5 (one
overlapping aromatic signal). Anal. Calcd for C25H31NO3: C,
76.30; H, 7.94; N, 3.56; O, 12.20. Found: C, 76.25; H, 7.98; N,
3.59.
FIGURE 3. Summary of conditions A-C.
SCHEME 11
of such compounds. The last category of enol ethers 25
consists of the unsubstituted members of group III, whose
amination products 8h,i are susceptible to aromatization.
Although conditions C partly mitigate this problem, the
most convenient means of avoiding aromatization is by
employing conditions B at 25 °C.
Finally, this methodology should be readily adaptable
to the synthesis of homochiral dihydrodipyrrins 8, limited
only by the availability of the requisite chiral alkyne acids
16. For this purpose, we are developing efficient protocols
for synthesizing these substrates, utilizing either a
Nicholas-Schreiber reaction29 or a variant of Ireland’s
enantioselective ester-enolate Claisen rearrangement.30a
For example, in companion studies we recently described
the enantioselective synthesis of alkyne acid 16j (ee >
95%),30b which we are employing as a ring-C synthon in
our work on cobyric acid (Scheme 11). Alkyne acid 16j
also has the proper substitution pattern for preparing
the A,B-ring portion of bonellin (2). Enantioselective
3,4-Dimethyl-5-[phenyl-(4,4,5-trimethyl-3,4-dihydro-
pyrrol-2-ylidene)-methyl]-1H-pyrrole-2-carboxylic Acid
tert-Butyl Ester (8b). Conditions A. A solution of 0.95 g
(2.41 mmol) of 25b in 23 mL of DMF was treated with 1.2 mL
(1.21 mmol) of 1 M HCl and stirred at room temperature until
TLC indicated complete formation of the intermediate diketone
[∼1 h; Rf (1:4 EtOAc/hexanes) 0.22]. The solution was then
treated with 3.3 mL (48.2 mmol) of 14.8 M NH4OH and stirred
at room temperature until TLC indicated complete consump-
tion of the diketone (∼6 h). 6 M HCl (7.2 mL; 43.4 mmol) was
then added until a pH of 1 was achieved, and the solution was
(30) (a) Ireland, R. E.; Mueller, R. H. J. Am. Chem. Soc. 1972, 94,
5897. (b) Ireland, R. E.; Varney, M. D. J. Am. Chem. Soc. 1984, 106,
3668. (c) Ireland, R. E.; Wipf, P.; Armstrong, J. D., III. J. Org. Chem.
1991, 56, 650. (d) Koch, G.; Janser, P.; Kottirsch, G.; Romero-Giron,
E. Tetrahedron Lett. 2002, 43, 4837 and references therein. (e) Jacobi,
P. A.; Tassa, C. Org. Lett. 2003, 5, 4879.
7250 J. Org. Chem., Vol. 70, No. 18, 2005

A Versatile Synthesis of Dihydrodipyrrins
stirred for a further 14 h at room temperature. The resulting
solution was poured into cold, saturated KHCO3 and extracted
with CH2Cl2 (3 × 10 mL). The combined organic extracts were
washed sequentially with water and brine, dried over Na2SO4,
filtered, and concentrated under reduced pressure. The result-
ing oil was purified by flash chromatography (silica gel, EtOAc/
hexanes ) 1:4 to 1:1) to give 0.81 g (86%) of imine 8b as a
crystalline solid, mp 184-185 °C. Rf (1:4 EtOAc/hexanes) 0.47;
IR (thin film) 3286 (br), 1692, 1429, 1173 cm-1; 1H NMR (500
MHz, CDCl3) δ 1.14 (s, 6H), 1.21 (s, 3H), 1.61 (s, 9H), 2.18 (s,
3H), 2.20 (s, 3H), 2.36 (s, 2H), 7.26 (m, 5H), 11.59 (br s, 1H);
13C NMR (500 MHz, CDCl3) δ 10.2, 10.4, 16.0, 25.9, 28.75,
28.78, 44.6, 48.2, 79.8, 119.5, 119.8, 120.9, 127.2, 128.7, 129.9,
130.9, 140.2, 150.5, 161.3, 187.7. Anal. Calcd for C25H32N2O2:
C, 76.49; H, 8.22; N, 7.14; O, 8.15. Found: C, 76.56; H, 8.59;
N, 6.78.
[∼6 h; Rf (1:4 EtOAc/hexanes) 0.47]. The reaction was then
diluted with 10% KH2PO4 (2.4 mL) and extracted with CH2-
Cl2 (3 × 5 mL). The combined organic extracts were washed
sequentially with water (2 × 10 mL) and brine, dried over Na2-
SO4, filtered, and concentrated under reduced pressure. The
resulting oil was purified by flash chromatography (silica gel,
EtOAc/hexanes ) 1:9) to give 0.079 g (79%) of imine 8b as a
crystalline solid that was identical in all respects to the
compound prepared using conditions A described in the
previous paragraph.
Acknowledgment. Financial support of this work
by the National Institutes of Health, NIGMS Grant No.
GM38913, is gratefully acknowledged. William G. O’Neal
is the recipient of an NSF predoctoral fellowship.
Conditions B. A solution of 0.102 g (0.25 mmol) of enol
ether 25b in 2.4 mL of DMF was treated with 21 µL (0.13
mmol) of 6 M HCl and stirred until formation of the diketone
was complete by TLC [∼1 h; Rf (1:4 EtOAc/hexanes) 0.22]. The
reaction was then treated with 0.40 g (5.1 mmol) of NH4OAc
and 0.70 mL (5.1 mmol) of NEt3. The resulting solution was
heated at 55 °C until product formation was complete by TLC
Supporting Information Available: Experimental de-
tails, characterization data, and NMR spectra for all new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
JO050907L
J. Org. Chem, Vol. 70, No. 18, 2005 7251