Synthesis and hepatic metabolism of xanthobilirubinic acid regioisomers

Article abstract of DOI:10.1007/s00706-006-0543-8

A set of four regioisomeric dipyrrinone propionic acids has been synthesized and their hepatic metabolism examined in rats: xanthobilirubinic acid and pseudo-xanthobilirubinic acid each with a propionic acid on a pyrrole ring; exo-ψ-xanthobilirubinic acid

Full text of DOI:10.1007/s00706-006-0543-8

Monatshefte f€ur Chemie 137, 1463–1476 (2006)
DOI 10.1007/s00706-006-0543-8
Synthesis and Hepatic Metabolism
of Xanthobilirubinic Acid Regioisomers
Stefan E. Boiadjiev1, Brian A. Conley1, Justin O. Brower1,
Antony F. McDonagh2, and David A. Lightner1;ꢀ
1
Department of Chemistry, University of Nevada, Reno, Nevada, USA
2
Division of Gastroenterology and the Liver Center, University of California, San Francisco,
California, USA
Received April 27, 2006; accepted May 11, 2006
Published online October 16, 2006 # Springer-Verlag 2006
Summary. A set of four regioisomeric dipyrrinone propionic acids has been synthesized and their
hepatic metabolism examined in rats: xanthobilirubinic acid and pseudo-xanthobilirubinic acid each
with a propionic acid on a pyrrole ring; exo- -xanthobilirubinic acid and endo- -xanthobilirubinic
acid, each with a propionic acid transposed to a lactam ring. After intravenous injection all four iso-
mers were excreted to some degree in unchanged form in bile in normal rats. Xanthobilirubinic acid,
the structurally closest dipyrrinone to bilirubin, and exo- -xanthobilirubinic acid were excreted almost
entirely in unchanged form. However, a small fraction of xanthobilirubinic acid acyl glucuronide was
also detected. More extensive acyl glucuronidation was observed for pseudo-xanthobilirubinic acid,
and endo- -xanthobilirubinic acid underwent slow metabolism to unidentified more polar products
that did not seem to be glucuronides.
Keywords. Pyrrole; Dipyrrinone; NMR.
Introduction
Dipyrrinones are yellow, non-fluorescent dipyrrolic pigments with an extensive
system of conjugated double bonds, and they comprise the core chromophores of
the yellow-orange tetrapyrrole pigment of jaundice and mammalian bile: bilirubin
(Fig. 1A) [1, 2]. They have captured interest as simple analogs of bilirubin for use in
model studies of chemical reactions [3], in the preparation of highly-fluorescent
analogs [4], in photochemistry [5], and in hepatic metabolism [4, 6]. One dipyrri-
none, xanthobilirubinic acid (XBR, Fig. 1B) has for the past three decades stood
out as a bilirubin model compound and also as a standard for HPLC analysis of
bilirubin glucuronides [7]. With respect to the location of a propionic acid at a
ꢀ
Corresponding author. E-mail: lightner@unr.nevada.edu

1464
S. E. Boiadjiev et al.
€
Fig. 1. (A) Bilirubin and (B) its dipyrrole cleavage products isolated in 1911: colorless ‘‘Bilirubinsaure’’
€
€
(H. Fischer) or ‘‘Bilinsaure’’ (O. Piloty), and yeꢁllow ‘‘Xanthobilirubinsaure’’ (Fischer) or ‘‘Dehydro-
€
Bilinsaure’’ (Piloty): a: HI in CH3CO2H at 100 C followed by HI-PH3; b: CH3ONa=CH3OH, 220–
230ꢁC or 0.1 M KMnO4 at 7ꢁC; (C) the target XBR regioisomers of the current work
‘‘pyrrole’’ ꢀ-position, there are four regioisomers, including XBR (with its propionic
acid at C(8)). One regioisomeric analog of XBR (1), with the C(7) methyl and C(8)
propionic acid switched, we have named pseudo-XBR (2, -XBR, Fig. 1C). Switch-
ing the C(3) ethyl and C(8) propionic acid of XBR gives a third regioisomer (4),
which we call endo- -XBR, and switching the C(2) methyl and C(3) propionic acid
of the latter provides a fourth regioisomer, 3 (exo- -XBR).
Dipyrrinones first appeared in the chemical literature in the form of bilinic
acid [8a] or bilirubinic acid [9] and dehydrobilinic acid [8b] or XBR [9] nearly
100 years ago in O. Piloty’s and H. Fischer’s attempts to determine the structure
of bilirubin. Bilirubinic acid and XBR are interconvertible by redox reactions. As
C17 compounds, they represent one-half of the bilirubin structure, and their con-
stitutional structures were characterized long ago. After several missteps originat-
ing from the degradative cleavage of mesobilirubin [9c], Fischer, Bartholom€aus,
and R€ose postulated the hydroxypyrrole tautomer as the structure of bilirubinic
acid and the correct (modern) lactam structure for XBR (Fig. 1B) in 1913 [10],
albeit without designating its stereochemistry at C(4). Curiously, subsequent to

Xanthobilirubinic Acid Regioisomers
1465
1914, Fischer drew only the lactim form of XBR through the end of his enor-
mously prolific career (which ended with his death on March 31, 1945 in Munich).
Although Fischer’s early interest (1911–1914) was in the structure of bile pig-
ments, especially bilirubin, work on the structure of the latter appeared to wane in
the Fischer labs after 1915 (while research on the structure of the heme porphyrin
accelerated) with a few publications on the reduction of bilirubin in the mid-
1920s. After having solved the structure of hemin, for which the 1930 Nobel
Prize in Chemistry was awarded to Fischer, he resumed the structure deter-
mination of bilirubin and pursued the structure determination and synthesis of
chlorophyll.
Because bilirubinic acid and XBR represented only one-half of the bilirubin struc-
ture, Fischer sought to determine the other half, again by degradation [11]. These
studies of the early 1930s led Fischer to corroborate the structure of XBR by syn-
thesis [12] using a methodology that also allowed for the synthesis of an important
regioisomer of XBR that differed only in switching the locations of the methyl and
ethyl groups on the lactam ring (iso-XBR, Fig. 2A) [13]. Put into practice, the
methodology also led to the equivalent regioisomer of -XBR: iso- -XBR (Fig. 2A),
as well as 3. The advances in synthesis of the late 1920s and early 1930s that made
dipyrrinone synthesis possible involved condensation of an appropriate ꢁ-H pyr-
role with a designed ꢁ-pyrrolealdehyde to afford a dipyrrylmethene as illustrated
for the synthesis of XBR in Fig. 2B [12, 13].
Our interest in dipyrrinones 1–4 stems from their role as models for probing
and elucidating the molecular requirements of substrate in the hepatic elimination
of bilirubin. In particular, these bilirubin analogs serve as probes of the sensitivity
of uptake transporters and glucuronosyl transferase isoforms to the location of the
propionic acid group of the pigment. XBR (1) correlates structurally with natural
bilirubin-IXꢁ (Fig. 1A), while 3 correlates structurally with bilirubin-IXꢂ, and 2
and 4 correlate with bilirubin-IXꢀ and ꢃ.
Fig. 2. (A) Fischer dipyrrinones synthesized, along with XBR, in the structure proof of mesobili-
rubin; (B) outline of the synthetic methodology developed by the Fischer group for the syntheses of
XBR and the dipyrrinones of (A)

1466
S. E. Boiadjiev et al.
Results and Discussion
Synthesis Aspects
The synthesis of 1 has been reported previously and makes use of improved and
modified preparations based on Fischer’s later work [14]. Thus, as outlined in
Scheme 1, the bromomethylenepyrrolinone 5 was condensed with acid 6a in re-
fluxing, acidic CH3OH to afford high yields of the methyl ester 7a of 1 [15, 16].
Similarly, the methyl ester of 2 was prepared by condensing 5 with the isomeric acid
Scheme 1
Scheme 2

Xanthobilirubinic Acid Regioisomers
1467
6b to afford the methyl ester 7b of 2 [16]. This preparation of 7b represents a new
synthesis using the more modern methodology. The compound had previously been
prepared only by Fischer in 1931 [12] using the method of Fig. 2B: condensation of
hemopyrrole carboxylic acid (2,3-dimethylpyrrole-4-propionic acid) with 2-formyl-
3-ethyl-4-methylpyrrole-5-carboxylic acid with CH3OH-48% HBr to afford a dipyr-
rylmethene that was brominated by Br2 in CH3CO2H to decarboxylate and replace
with Br, then treated with AgOAc or KOAc to give 2.
Dipyrrinone 4 had also been prepared in the early 1930s by Fischer and Adler
[13] from the appropriate dipyrrylmethene, as above and in Fig. 2B. We preferred a
more modern synthesis that would allow us to prepare both 3 and 4 and for these
two dipyrrinones, one new (3) and the other (4) not synthesized since 1931 [13], we
turned to the methodology developed by Jayasundera, Inomata, and Kinoshita for
preparing dipyrrinones used in their syntheses of (tetrapyrrole) phycocyanobilin
pigments [17]. As outlined in Scheme 2, this was accomplished by condensing the
known kryptopyrrole aldehyde (8) with an appropriate pyrrolinone, 9a or 9b. The
last two were synthesized by Barton-Zard pyrrole-forming reactions [18], borrow-
ing from the methodology developed by Kinoshita and Inomata [17, 19].
The syntheses of 3 and 4 share two common intermediates, pyrrole aldehyde 8 and
methyl 4-nitrobutanoate which is converted to pyrrolinones 9a or 9b (Scheme 2A).
To synthesize 9a, methyl 4-nitrobutanoate was converted to nitroalcohol 11a
(Scheme 2B) by condensation (Henry reaction) with acetaldehyde, catalyzed by
KF in 2-propanol. Acetylation of 11a to 12a in 95% yield, followed by a Barton-
Zard pyrrole-forming condensation with p-toluenesulfonylmethylisocyanide (TsMIC)
[20] in the presence of tetramethylguanidine (TMG) afforded crystalline 13a in
62% yield. Conversion of 13a to 9a was accomplished in 3 steps, as outlined in
Scheme 2B. The unsubstituted ꢁ-position of tosylpyrrole 13a is brominated using
N,N,N-trimethylanilinium perbromide (PTT) to afford a 96% yield of 14a, which
was converted to 15a in 46% yield in aqueous TFA. Detosylation of 15a with
NaBH4 in ethanol smoothly gave an 80% yield of 9a, which was condensed with
8 in hot aqueous-ethanolic KOH to give 3 in 71% yield. As is reported below, 9a
doubtless contains some ethyl ester due to trans-esterification, but the ester groups
are saponified under the reaction conditions above.
To prepare 9b, methyl 4-oxo-butanoate was condensed with nitroethane in the
presence of DBU to afford methyl 4-hydroxy-5-nitrohexanoate (11b, Scheme 2B),
which was acetylated (to 12b), and the nitroacetate submitted to a standard Barton-
Zard pyrrole forming reaction [18, 19] with TsMIC in the presence of TMG to give
a 45% overall yield of tosylpyrrole 13b. Conversion of 13b to 9b proceeded smoothly
in two steps: (1) bromination at the vacant pyrrole ꢁ-position using PTT to afford
a 67% isolated yield of 14b, then (2) reaction of 14b with TFA in aqueous DMSO,
followed by Zn=I2 to give the tosylpyrrolinone (15b) corresponding to 9b, and
finally detosylation with NaBH4 in ethanol.
The latter (9b) was mainly the methyl ester (85%), with 15% ethyl ester due to
trans-esterification in the NaBH4=ethanol [19] reductive detosylation of 15b. A
change of solvent, from ethanol to methanol, avoided trans-esterification, but the
reductive detosylation was much less effective and the yield of 9b was low. Appar-
ently in ethanol, the reducing agent is the more effective NaBH4, which reacts more
slowly with ethanol than methanol. In methanol, the NaBH4 is rapidly converted to

1468
S. E. Boiadjiev et al.
NaBH(OMe)3, a less effective reagent for cleaving the tosyl group. An acidic quench
before ethanol removal and immediate chromatographic purification cleanly gave
the methyl ester (9b). Condensing 9b with 8 under a fairly standard coupling pro-
cedure (refluxing aqueous-ethanolic KOH) failed. However, after much experimen-
tation, condensation using N-methylmorpholine in CH3CN afforded methyl ester
10b in 36% yield. A better yield (68%) of the free acid (4) was obtained instead of
10b using DBU in CH3OH to effect the condensation.
Structures and Spectroscopy
The structures of 1–4 follow logically from their synthetic precursors, the method
of synthesis, and from characterization by NMR spectroscopy. XBR (1) and its
dihydro analog, bilirubinic acid, are of historical interest as the first known dipyr-
rinones, both obtained by degradation of bilirubin (whose structure was unknown
at the time). Their structures were proved by Fischer, first by degradation [9, 10]
then by synthesis [12, 13]. The synthesis of 1 was improved from the original Fischer
synthesis [12, 13], as outlined in Scheme 1. The regioisomeric -XBR (2) was first
synthesized by Fischer in 1931 [13] and more recently by the XBR-type procedure
as outlined in Scheme 1 [16]. Like 1 and 2, the synthesis of 4 was first achieved by
Fischer [13], who condensed kryptopyrrole and 5-formyl-3-methyl-4-(2-carboxy-
ethyl)pyrrole-2-carboxylic acid to produce an isomer of the dipyrrylmethene
shown in Fig. 2B, with the C(3) ethyl and C(8) propionic acid interchanged. The
resultant dipyrrylmethene was then converted to 4. The new route to 4 (Scheme 2)
has many advantages and gave a product with properties identical to Fischer’s 4.
Table 1. 13C NMR chemical shifts (ꢃ, ppm) of four isomeric dipyrrinones (1–4) in (CD3)2SO solvent
Carbona
1 (XBR)b
2 ( -XBR)
3 (exo- -XBR)
4 (endo- -XBR)
1
C¼O
171.9
122.6
147.2
127.3
97.6
121.7
122.3
118.7
129.4
8.1
171.9
122.6
147.3
127.3
97.8
121.4
126.2
114.5
129.1
8.0
171.3
125.9
142.0
128.1
98.5
121.9
122.4
121.5
128.9
19.1
32.6
9.3
171.6
123.6
144.4
127.0
98.2
121.9
122.4
121.6
128.8
8.2
2
–C¼
3
–C¼
4
–C¼
5
–CH¼
–C¼
6
7
–C¼
8
–C¼
9
–C¼
21
22
31
32
71
72
81
82
91
CH2=CH3
CH2
–
–
–
CH2=CH3
CH2=CH3
CH2=CH3
CH2
17.2
14.8
9.2
17.2
14.7
19.6
35.4
8.6
19.6
34.6
9.1
–
9.4
–
–
–
CH2=CH3
CH2=CH3
CH3
19.5
35.0
11.0
174.0
16.9
15.5
10.9
173.7
16.9
15.5
10.9
173.6
–
11.0
173.9
CO2H
a
b
For carbon numbering system, see Fig. 1; assignments are based on gHMBC experiments

Xanthobilirubinic Acid Regioisomers
1469
Fig. 3. HMBC of xanthobilirubinic acid (1) in (CD3)2SO
Regioisomeric dipyrrinone 3 was not prepared by Fischer and was apparently
unknown before this work. All dipyrrinones 1–4 analyzed correctly for the basic
formula, C17H22N2O3. Their 13C NMR spectral data (Table 1) correlate well with
each structure. For example, in Fig. 3 is shown the HMBC of 1, from which one
can make correlated assignments (Table 1) of 1H and 13C resonances. Thus, the lac-
tam ring carbons of 1 and 2 show, as expected, the same set of structure-correlated
resonances with essentially identical chemical shifts. The pyrrole ring carbon reso-
nances of 3 and 4 similarly show the same structure-correlated chemical shifts, as
expected. The regioisomeric differences between the pairs 1 and 2 and 3 and 4
show up in the pyrrole carbon signals of 1 and 2 and in the lactam carbon signals of
3 and 4. Interestingly, having a propionic acid group on the pyrrole ring causes the
C(5) resonance to appear 0.4–0.9 ppm more shielded than when the propionic acid
is on the lactam ring (in 3 and 4).
Fig. 4. Nuclear Overhauser effects (curved arrows: strong, solid curve; weak, dashed curve) of XBR
(1), -XBR (2), exo- -XBR (3), and endo- -XBR (4) in (CD3)2SO

1470
S. E. Boiadjiev et al.
Although its complete chemical structure is well-established in the literature,
1H{1H}-Nuclear Overhauser effect spectroscopy (NOE) in (CD3)2SO solvent recon-
firmed the constitution and syn-Z stereochemistry at the exocyclic C(4)–C(5) dou-
ble bond of XBR (1). Similarly, NOEs (Fig. 4) from 2, 3, and 4 in (CD3)2SO clearly
indicate the syn-Z conformation, as well as the spatial relationship of the methyl
and propionic acid groups on the lactam ring (of 3 and 4) and pyrrole ring (of 1 and
2) – in accord with the expectation from directed synthesis. With the structures and
stereochemistry of 1–4 firmly established, we began studies on their hepatic meta-
bolism, initial results of which are reported below.
Metabolism
The hepatic metabolism and biliary excretion of 1–4 was studied by injecting small
(ꢂ0.25 mg) bolus doses, dissolved in ꢂ1 cm3 of rat serum, intravenously in normal
(Sprague-Dawley) adult male rats and homozygous adult male Gunn rats. Gunn
rats are a mutant strain that is deficient in the UGT1A family of glucuronosyl
transferases, including the specific isozyme, UGT1A1, that catalyzes bilirubin glu-
curonidation [21]. Samples of the injectate and of bile collected before and after
injection of each dipyrrinone were analyzed by HPLC. Each compound was stu-
died in at least two normal rats and two Gunn rats.
When injected into Gunn rats and normal rats, XBR (1) was excreted rapidly
into bile in unchanged form. In normal rats, but not Gunn rats, the unchanged
pigment in bile was accompanied by a very small proportion of a metabolite that
Fig. 5. Representative HPLC traces of bile collected just before (t ¼ 0) and at the indicated times
after injection of regioisomers 1–4 (panels a–b, respectively) in normal (Sprague-Dawley) rats; the
insets show chromatograms of the solutions injected; peaks marked with an asterisk are presumptive
acyl glucuronide metabolites; chromatographic conditions for each panel were similar, but not
identical; BDG and BMG are the diglucuronide and monoglucuronides of bilirubin

Xanthobilirubinic Acid Regioisomers
1471
eluted faster on HPLC (Fig. 5a). This metabolite disappeared on hydrolytic treat-
ment of bile samples with sodium hydroxide and with ꢀ-glucuronidase. On this
basis we conclude that the metabolite is XBR acyl glucuronide. Unchanged isomer
2 was also excreted promptly in bile in both Gunn and normal rats after intravenous
injection. However, unexpectedly, in normal rats (Fig. 5b) a major fraction of the
injected dose was converted to a more polar metabolite, identified as the acyl glu-
curonide of 2 by NaOH and ꢀ-glucuronidase hydrolysis. In the Gunn rat slow for-
mation of unidentified more polar metabolites was also observed. As expected,
isomer 3 was cleared from the circulation and excreted rapidly through the liver
in unchanged form in normal rats (Fig. 5c). But surprisingly, in Gunn rats, the
unchanged form was accompanied by a comparable amount of a more polar me-
tabolite with an almost identical absorption spectrum. The identity of this meta-
bolite remains to be determined. The final isomer 4 appeared rapidly in bile in
unchanged form following injection in Gunn and normal rats. However, in both
strains extensive metabolism to more polar products was also seen (Fig. 5d). These
were stable to NaOH and ꢀ-glucuronidase hydrolysis and, therefore, are probably
not glucuronides.
Thus, all four dipyrrinone regioisomers are cholephilic and are excreted rapidly
in bile in unchanged form to different degrees following their intravenous injec-
tion in the rat. Isomers 1 and 2 also seem to form acyl glucuronides, the latter more
so than the former. Isomer 4 seems to differ from the other three in its extensive
conversion to unknown metabolites that do not appear to be acyl glucuronides.
Before beginning the metabolic studies, we had expected that the metabolism
and hepatobiliary excretion behavior of the four regioisomers would be essentially
identical and that they would all be excreted rapidly in Gunn rats and normal rats,
with little or no acyl glucuronidation in the latter. The subtle effects of propionate
positioning are somewhat surprising. Blanckaert et al. [22] studied the metabolism
of the ꢁ, ꢀ, ꢂ, and ꢃ isomers of bilirubin in Gunn rats and normal Wistar rats. They
found that the IXꢁ isomer, the only one in which the two carboxyl groups can
undergo intramolecular hydrogen-bonding to contralateral lactam and NH groups,
was excreted as glucuronides in normal rats and hardly at all as unchanged pigment
in Gunn rats. In contrast, the three other regioisomers were excreted rapidly in bile
in unchanged form in Gunn rats and partly as conjugates in normal rats. The ꢂ isomer,
which corresponds to regioisomer 3 of the XBR family, underwent the greatest
amount of conjugation. However, we were unable to detect significant amounts of
glucuronide in the bile of Sprague-Dawley rats following injection of 3. Thus the
metabolism of dipyrrinone carboxylates may not be predictive of the metabolism
of the corresponding bilirubins, even those that cannot undergo intramolecular
hydrogen bonding of the type shown by bilirubin IXꢁ and mesobilirubin IXꢁ.
Experimental
Nuclear magnetic resonance (NMR) spectra were obtained in CDCl3 solvent on a GE QE-300 spectrom-
eter operating at 300 MHz (for most of the 1H NMR data, unless otherwise indicated) and at
125 MHz on a Varian Unity Plus 500 MHz spectrometer for most of the 13C NMR data. NOE
NMR spectra were obtained at 500 MHz. Chemical shifts were reported in ꢃ=ppm referenced to the
residual CHCl3 1H signal at 7.26 ppm and 13C signal at 77.23ppm. Infrared spectra were recorded on a

1472
S. E. Boiadjiev et al.
Perkin-Elmer model 1610-FT IR instrument. Ultraviolet-visible spectra were recorded on a Perkin-
Elmer ꢄ-12 spectrophotometer. Melting points were taken on a Mel-Temp capillary apparatus. Satis-
factory combustion analyses for carbon, hydrogen, and nitrogen were carried out by Desert Analytics,
Tucson, AZ, and gave results within ꢃ0.4% of the theoretical values. Analytical thin layer chromatog-
raphy (TLC) was carried out on J.T. Baker silica gel IB-F plates (125 ꢅm layers). Radial chromatog-
raphy was carried out on Merck preparative layer grade silica gel PF254 with CaSO4 binder using a
Chromatotron (Harrison Research, Inc., Palo Alto, CA) with 1, 2, or 4 mm thick rotors. Commercial
reagents were used as received from Aldrich or Acros. Spectroscopic data were obtained in spec-
tral grade solvents from Fisher and Acros. Deuterated chloroform and dimethylsulfoxide were from
Cambridge Isotope Laboratories. Bromomethylenepyrrolinone 5 [15, 23]; pyrroles 6a [15, 23], 6b [16],
and 8 [24] and dipyrrinone esters 7a [15, 23] and 7b [16] were synthesized according to previously
published methods.
Methyl 4-nitro-5-hydroxyhexanoate (11a)
Acetaldehyde (30.0 g, 0.68 mol) and KF (0.80 g, 14 mmol) were stirred with 70 cm3 2-propanal in
an ice bath and had added to it dropwise a solution of methyl 4-nitrobutyrate (41.25 g, 0.28 mol) in
70 cm3 2-propanol. The solution was allowed to warm to room temperature and stirred overnight.
The light yellow solution was diluted with 200 cm3 CH2Cl2, washed with H2O (2ꢄ200 cm3) and
brine (100 cm3), and then dried (MgSO4). The solvent was removed (rotovap), yielding a clear
yellow oil. Yield 50.87 g (95%) 11a [25]; IR (thin film): ꢆꢀ¼ 3463, 2955, 1736, 1551, 1440, 1371,
1206, 1176, 1008, 736 cmꢅ1; 1H NMR: ꢃ ¼ 1.28 and 1.31 (d, J ¼ 10.5 Hz, 3H), 2.18–2.55 (m, 4H),
3.70 (s, 3H), 4.12–4.28 (m, 1H), 4.51 (m, 1H) ppm; 13C NMR: ꢃ ¼ 19.1, 19.4, 23.4, 25.0, 29.8,
29.9, 51.9, 68.3, 91.8, 93.0, 172.4 ppm.
2-(p-Toluenesulfonyl)-3-methyl-4-(2-methoxycarbonylethyl)-1H-pyrrole (13a, C16H19NO4S)
Methyl 4-nitro-5-hydroxyhexanoate from above (50.9 g, 0.22 mol) and DMAP (50 mg) were dissolved
in 100 cm3 acetic anhydride and stirred at room temperature overnight [25]. The solution was then
cooled in an ice bath and water added to destroy the excess anhydride, and the mixture was then stirred
for an additional 3 h. The green solution was then diluted with 200cm3 CH2Cl2, washed with H2O
(2ꢄ500 cm3), aqueous saturated NaHCO3 (5ꢄ500 cm3), H2O (1ꢄ500 cm3), and brine (1ꢄ500 cm3),
and then dried (MgSO4). The solvent was removed (rotovap) to afford a green oily product (12a). It
was used directly in the next step. Yield 57g (95%); IR (thin film): ꢆꢀ¼ 2955, 1825, 1745, 1555, 1439,
1
1372, 1234, 1020, 956, 854 cmꢅ1; H NMR: ꢃ ¼ 1.31, 2.18 (d, J ¼ 10.5Hz, 3H), 2.20 (s, 3H), 2.18–
2.55 (m, 4H), 3.70 (s, 3H), 4.12–4.28 (m, 1H), 4.51 (m, 1H) ppm; 13C NMR: ꢃ ¼ 15.6, 16.2, 20.5, 20.6,
21.9, 23.7, 24.6, 29.4, 29.7, 51.7, 69.3, 69.7, 88.7, 89.6, 169.2, 171.7ppm.
p-Toluenesulfonylmethylisocyanide (TsMIC) (0.78 g, 4.0 mmol) [20] and 1,1,3,3-tetramethylgua-
nidine (1cm3, 2.2 mol eq) were stirred with 5 cm3 THF=i-PrOH (1=1 by volume) and had added to it
a solution of methyl 4-nitro-5-acetoxyhexanoate (1.00 g, 3.6 mmol) in 5 cm3 THF=i-PrOH (1=1 by
volume). The solution was then stirred at room temperature for 24h. The solution was transferred to an
Erlenmeyer flask and chilled in an ice bath. Water was slowly added, with stirring, precipitating the
crude product. The precipitate was then collected by vacuum filtration, air dried, and recrystallized
from CH2Cl2 and hexane to afford colorless crystals. Yield 0.72g (62%); mp 140–141ꢁC; IR (KBr):
ꢆꢀ¼ 3326, 2951, 1734, 1437, 1316, 1301, 1137, 1087, 813, 707, 686 cmꢅ1; 1H NMR: ꢃ ¼ 2.16 (s, 3H),
2.40 (s, 3H), 2.49 (t, J ¼ 7 Hz, 2H), 2.68 (t, J ¼ 7 Hz, 2H), 3.65 (s, 3H), 6.71 (d, J ¼ 3 Hz, 1H), 7.28 (d,
J ¼ 8 Hz, 2H), 7.75 (d, J ¼ 8 Hz, 2H), 8.86 (br.s, 1H) ppm; 13C NMR: ꢃ ¼ 9.3, 20.6, 21.7, 34.6, 51.8,
120.5, 124.0, 124.2, 124.5, 126.7, 129.9, 140.0, 143.7, 173.4 ppm.
2-(p-Toluenesulfonyl)-3-methyl-4-(2-methoxycarbonylethyl)-5-bromo-1H-pyrrole
(14a, C16H18BrNO4S)
2-(p-Tosyl)-3-methyl-4-(methoxycarbonylethyl)-1H-pyrrole (10.19 g, 31.72 mmol) and PTT (13.2 g,
35.13 mmol) were dissolved in CH2Cl2 (200cm3) and stirred at 0ꢁC for 30min. The solution was

Xanthobilirubinic Acid Regioisomers
1473
washed with aqueous sodium bisulfite (2ꢄ200 cm3) and brine (1ꢄ200 cm3), and then dried (MgSO4).
The solvent was removed (rotovap) to yield a tan powder. Yield 12g (96%); mp 136–137ꢁC; IR (KBr):
1
ꢆꢀ¼ 3290, 2951, 1737, 1550, 1437, 1301, 1148, 1093, 1062, 911, 816, 712 cmꢅ1; H NMR: ꢃ ¼ 2.18
(s, 3H), 2.41 (s, 3H), 2.41 (t, J ¼ 7 Hz, 2H), 2.66 (t, J ¼ 7 Hz, 2H), 3.64 (s, 3H), 7.30 (d, J ¼ 8 Hz, 2H),
7.77 (d, J ¼ 8 Hz, 2H), 8.97 (br.s, 1H) ppm; 13C NMR: ꢃ ¼ 9.7, 20.3, 21.5, 33.8, 51.6, 105.3, 123.1,
124.8, 125.2, 126.6, 129.8, 139.4, 143.8, 172.9 ppm.
3-(2-Methoxycarbonylethyl)-4-methyl-5-(p-toluenesulfonyl)-3-pyrrolin-2-one (15a, C16H19NO5S)
2-(p-Tosyl)-3-methyl-4-(2-methoxycarbonylethyl)-5-bromo-1H-pyrrole (14a) (0.29 g, 0.72mmol) was
dissolved in 6 cm3 TFA, and 1.2 cm3 H2O were added. The solution was then stirred for two days. The
solution was diluted with 50cm3 CH2Cl2 and washed with H2O (2ꢄ50cm3), aqueous sodium bicar-
bonate (2ꢄ50 cm3), and brine (1ꢄ50cm3). After drying (MgSO4), the solvent was removed (rotovap)
to afford the crude product. Recrystallization from methanol and water yielded light grey crystals.
Yield 0.11 g (46%); mp 130–131ꢁC; IR (KBr): ꢆꢀ¼ 3333, 2952, 1699, 1436, 1318, 1132, 1080, 814,
667, 627 cmꢅ1; 1H NMR: ꢃ ¼ 2.18 (s, 3H), 2.41 (s, 3H), 1.88–2.61 (m, 4H), 3.61 (s, 3H), 5.04 (s, 1H),
6.14 (s, 1H), 7.29 (d, J ¼ 8 Hz, 2H), 7.64 (d, J ¼ 8 Hz, 2H) ppm; 13C NMR: ꢃ ¼ 13.3, 19.0, 21.8, 31.4,
51.8, 79.3, 129.6, 129.8, 130.6, 136.2, 144.7, 146.2, 172.7, 172.7ppm.
3-(2-Methoxycarbonylethyl)-4-methyl-3-pyrrolin-2-one (9a)
3-(2-Methoxycarbonylethyl)-4-methyl-5-(p-tosyl)-3-pyrrolin-2-one (15a) (0.39 g, 1.16mmol) was dis-
solved in 35 cm3 abs ethanol that had added to it a suspension of sodium borohydride (0.055g,
1.45mmol). The mixture was stirred at room temperature for 30min, then diluted with 50cm3 CH2Cl2,
and washed with dilute aqueous acetic acid (1ꢄ50 cm3), H2O (2ꢄ50cm3), and brine (1ꢄ50cm3).
After drying (MgSO4), the solvent was removed (rotovap) to afford a colorless powder. Yield 0.17g
(80%); mp 74–75ꢁC (Ref. [26] 71–76ꢁC); IR (KBr): ꢆꢀ¼ 3233, 2954, 1735, 1675, 1570, 1437, 1174,
1066, 969 cmꢅ1; 1H NMR: ꢃ ¼ 2.01 (s, 3H), 2.57 (s, 3H), 3.64 (s, 3H), 3.81 (s, 2H), 7.13 (s, 1H) ppm;
13C NMR: ꢃ ¼ 13.1, 19.0, 32.1, 50.3, 51.4, 130.4, 150.6, 173.2, 175.8ppm.
2-(2-Carboxyethyl)-8-ethyl-3,7,9-trimethyl-(10H)-dipyrrin-1-one (3, C17H22N2O3)
3-(2-Methoxycarbonylethyl)-4-methyl-3-pyrrolin-2-one (9a) (0.16 g, 0.87 mmol) and 2-formylkrypto-
pyrrole (8) (0.14 g, 0.96 mmol) were dissolved in 10 cm3 methanol and 10cm3 6 M aqueous KOH. The
solution was heated to reflux and stirred overnight. It was then diluted with H2O (50 cm3) and acidified
with acetic acid. The desired product (3) was collected by vacuum filtration, affording a green=yellow
powder. Yield 0.26g (71%); mp 298–300ꢁC; IR (KBr): ꢆꢀ¼ 3356, 2924, 1664, 1631, 1461, 1451, 1259,
1
1159, 964, 668 cmꢅ1; H NMR (DMSO-d6): ꢃ ¼ 0.98 (t, J ¼ 7.3 Hz, 3H ), 2.03 (s, 3H), 2.07 (s, 3H),
2.17 (s, 3H), 2.30 (q, J ¼ 7.3 Hz, 2H), 2.40 (t, J ¼ 6.6 Hz, 2H), 2.46 (t, J ¼ 6.6 Hz, 2H), 5.95 (s, 1H),
9.78 (s, 1H), 10.24 (s, 1H), 12.09 (br.s, 1H) ppm; 13C NMR in Table 1.
3-(2-Methoxycarbonylethyl)-4-methyl-2-p-toluenesulfonyl-1H-pyrrole (13b, C16H19NO4S)
To a solution of 11.6g (100 mmol) methyl 4-oxobutanoate and 8.3 cm3 (115mmol) nitroethane in
10cm3 anhydrous CH3CN kept under N2 at 0ꢁC were added 1.5 cm3 (10 mmol) DBU during 30 min,
and the mixture was stirred at room temperature for 20h. The mixture was diluted with 200 cm3 ethyl
acetate:diethyl ether (1:1) and washed with 2% aqueous HCl (50 cm3) and H2O (3ꢄ50 cm3). After
drying (Na2SO4) and filtration, the solvents were evaporated under vacuum, the residue (11b) was kept
under 0.5 mm Hg vacuum at 40ꢁC for 3 h, and then used immediately in the following step.
The crude nitroalcohol 11b (17.6 g) was acetylated with 18.8cm3 (200mmol) acetic anhydride in
60cm3 anhydrous CH2Cl2 in the presence of 25mg DMAP for 18h at room temperature. Excess acetic
anhydride was destroyed with methanol (20 cm3) added during 20min, and after 1 h the mixture was
diluted with 100 cm3 CH2Cl2. The solution was poured carefully into 200 cm3 5% aqueous NaHCO3,
then washed with H2O (3ꢄ100 cm3), dried (anhydrous MgSO4), filtered, and the solvent was evapo-
rated under vacuum. The product (12b) was used without further purification in the next step.

1474
S. E. Boiadjiev et al.
To a solution of 16.6g (85 mmol) TsMIC [20] and 21.1cm3 (170 mmol) 1,1,3,3-tetramethylguani-
dine in 70 cm3 anhydrous THF=i-PrOH (1=1 by volume) kept under N2 at 0ꢁC was added a solution of
the crude nitroacetate 12b in 15cm3 THF=i-PrOH (1=1 by volume) during 45min. Then the mixture was
stirred for 20 h at room temperature. After dilution with 300 cm3 CH2Cl2, it was washed with 2% aque-
ous HCl (2ꢄ80cm3), and H2O (3ꢄ100 cm3). The solution was dried over anhydrous MgSO4, filtered,
and the solvents were evaporated under vacuum. The residue was purified by radial chromatography and
recrystallized from ethyl acetate-hexane to afford 13b. Yield 12.3g (45%); mp 108–109ꢁC; 1H NMR:
ꢃ ¼ 1.98 (d, 4J ¼ 0.8 Hz, 3H), 2.40 (s, 3H), 2.42 (m, 2H), 2.87 (m, 2H), 3.68 (s, 3H), 6.70 (dq, 3J ¼ 1.2 Hz,
4J ¼ 0.8 Hz, 1H), 7.28 (d, 3J ¼ 8.3Hz, 2H), 7.76 (d, 3J ¼ 8.3 Hz, 2H), 8.87 (br.s, 1H) ppm; 13C NMR:
ꢃ ¼ 9.8, 19.7, 21.5, 34.6, 51.6, 120.8, 121.1, 123.7, 126.6, 127.1, 129.8, 139.8, 143.7, 173.3ppm.
5-Bromo-3-(2-methoxycarbonylethyl)-4-methyl-2-p-toluenesulfonyl-1H-pyrrole
(14b, C16H18BrNO4S)
To a solution of 3.21g (10.0 mmol) pyrrole 13b in 80 cm3 CH2Cl2 cooled in an ice bath was added a
solution of 7.50 g (20.0 mmol) phenyl-N,N,N-trimethylammonium tribromide in 80cm3 CH2Cl2 during
15 min. After the addition was completed, stirring was continued for 15min more. Then the mixture
was diluted with 150 cm3 CH2Cl2, washed with 3% aqueous sodium thiosulfate (2ꢄ100 cm3) and H2O
(3ꢄ100 cm3), dried (anhydrous MgSO4), filtered, and the solvent was evaporated under vacuum. The
residue was purified by radial chromatography and the combined fractions containing product were
recrystallized from ethyl acetate-hexane. Yield 2.69g (67%); mp 193–194ꢁC (dec); 1H NMR: ꢃ ¼ 1.92
3
(s, 3H), 2.40 (s, 3H), 2.41 (m, 2H), 2.89 (m, 2H), 3.69 (s, 3H), 7.29 (d, J ¼ 8.3 Hz, 2H), 7.78 (d,
3J ¼ 8.3 Hz, 2H), 9.11 (br.s, 1H) ppm; 13C NMR: ꢃ ¼ 9.7, 20.3, 21.6, 34.4, 51.7, 105.1, 120.8, 124.9,
126.7, 128.0, 130.0 139.3, 144.1, 173.0 ppm.
4-(2-Methoxycarbonylethyl)-3-methyl-5-p-toluenesulfonyl-3-pyrrolin-2-one (15b, C16H19NO5S)
To a solution of 2.0g (5.0mmol) bromopyrrole 14b in 20 cm3 TFA under N2 was added a solution of
1.1 cm3 (15.0 mmol) anhydrous dimethylsulfoxide in 12cm3 TFA during 10 min. After stirring for
2.5 h, iodine (127mg, 0.5 mmol) followed by 1.31 g (20 mol) of zinc were added and stirring was
continued for 2 h. The solids were removed by filtration, the filtrate was diluted with 150 cm3 of H2O,
and the product was extracted with CHCl3 (4ꢄ50cm3). The combined extracts were washed with 2%
aqueous NaHCO3 (2ꢄ50cm3), and then with H2O (3ꢄ50cm3). After drying over anhydrous Na2SO4,
filtration, and evaporation of solvent under vacuum, the residue was purified by radial chromatography
and recrystallization from ethyl acetate-hexane. Yield 1.52 g (90%); mp 131–132ꢁC; 1H NMR:
ꢃ ¼ 1.61 (br.s, 3H), 2.43 (s, 3H), 2.61 (m, 1H), 2.71 (m, 1H), 2.87 (m, 1H), 2.98 (m, 1H), 3.68 (s,
3
3
3H), 5.21 (br.s, 1H), 6.42 (br.s, 1H), 7.31 (d, J ¼ 8.3 Hz, 2H), 7.65 (d, J ¼ 8.3 Hz, 2H) ppm; 13C
NMR: ꢃ ¼ 8.5, 21.6, 22.2, 32.2, 51.8, 77.1, 129.3, 129.6, 130.6, 135.1, 145.0, 145.8, 172.3, 173.3ppm.
4-(2-Methoxycarbonylethyl)-3-methyl-3-pyrrolin-2-one (9b)
To a solution of 1.01 g (3.0mmol) tosylpyrrolinone 15b in 50cm3 abs ethanol kept at 10ꢁC was added
339 mg (9.0mmol) sodium borohydride during 10 min. After 10min stirring, the excess reagent was
quenched with 2 cm3 10% aqueous HCl and then the ethanol solvent was evaporated almost comple-
tely under vacuum. The residue was partitioned between 100cm3 H2O and 100cm3 CHCl3. The
organic layer was washed with H2O (2ꢄ20cm3), dried over anhydrous MgSO4, filtered, and the
solvent was evaporated under vacuum. The residue was purified by radial chromatography and recrys-
1
tallized from hexane-ethyl acetate. Yield 385 mg (70%); mp 85–86ꢁC (Ref. [27] mp 82–83ꢁC); H
NMR: ꢃ ¼ 1.79 (m, 3H), 2.48 (t, J ¼ 7.5 Hz, 2H), 2.68 (t, J ¼ 7.5 Hz, 2H), 3.65 (s, 3H), 3.81 (br.s, 2H),
7.56 (br.s, 1H) ppm; 13C NMR: ꢃ ¼ 8.3, 23.0, 32.4, 48.3, 51.8, 129.6, 150.7, 172.6, 176.1 ppm.
8-Ethyl-3-(2-methoxycarbonylethyl)-2,7,9-trimethyl-(10H)-dipyrrin-1-one (10b, C18H24N2O3)
Pyrrolinone 9b (366mg, 2.0 mmol) was dissolved in 5 cm3 anhydrous CH3CN and transferred into a
thick wall tube. Aldehyde 8 (907mg, 6.0 mmol) and 2.3 cm3 (20.0 mmol) N-methylmorpholine were

Xanthobilirubinic Acid Regioisomers
1475
added under Ar, the tube was sealed and heated at 95–100ꢁC for 84h. After cooling, the mixture was
diluted with 200 cm3 CHCl3, washed with 1% aqueous HCl (100 cm3) and H2O (3ꢄ100 cm3). After
drying over anhydrous Na2SO4, filtration, and evaporation of the solvent, the residue was purified
by radial chromatography and recrystallization from CH2Cl2–CH3OH. Yield 228 mg (36%); mp
1
224–226ꢁC; H NMR: ꢃ ¼ 1.07 (t, J ¼ 7.6Hz, 3H), 1.97 (s, 3H), 2.15 (s, 3H), 2.40 (s, 3H), 2.41
(q, J ¼ 7.6 Hz, 2H), 2.55 (t, J ¼ 7.8 Hz, 2H), 2.89 (t, J ¼ 7.8 Hz, 2H), 3.69 (s, 3H), 6.16 (s, 1H), 10.31
(br.s, 1H), 11.36 (br.s, 1H) ppm; 13C NMR: ꢃ ¼ 8.7, 9.5, 11.5, 15.4, 17.4, 20.0, 34.5, 51.8, 101.8, 122.1,
123.1, 123.5, 125.4, 126.4, 131.6, 144.2, 172.9, 173.5ppm.
3-(2-Carboxyethyl)-8-ethyl-2,7,9-trimethyl-(10H)-dipyrrin-1-one (4, C17H22N2O3)
A mixture of 158 mg (0.5mmol) 10b, 12 cm3 ethanol, and 4 cm3 10% aqueous NaOH was heated at
reflux for 3 h. After cooling, the ethanol solvent was removed under vacuum. The residue was diluted
with 5 cm3 H2O and slowly acidified at 0ꢁC with 10% aqueous HCl. After stirring for 30 min, the
precipitated product was collected by filtration, washed with H2O (3ꢄ5 cm3), and dried at 80ꢁC under
vacuum for 24h (P2O5) to give 4. Yield 123 mg (81%); mp 271–273ꢁC (dec); 1H NMR ((CD3)2SO):
ꢃ ¼ 0.97 (t, J ¼ 7.5 Hz, 3H), 1.77 (s, 3H), 2.01 (s, 3H), 2.16 (s, 3H), 2.29 (q, J ¼ 7.5 Hz, 2H), 2.37
(t, J ¼ 7.6 Hz, 2H), 2.73 (t, J ¼ 7.6 Hz, 2H), 5.97 (s, 1H), 9.79 (s, 1H), 10.25 (s, 1H), 12.46 (v.br.s, 1H)
ppm; 13C NMR in Table 1.
Metabolism Studies
The methods used have been described in detail elsewhere [4, 28]. For ꢀ-glucuronidase hydrolysis of
glucuronides in bile 40 mm3 glucuronidase solution (prepared by adding 1 cm3 water to 1000 units
Sigma bacterial E. coli Type II or VIIa glucuronidase) was mixed with 20mm3 bile, incubated at 37ꢁC
in the dark for 60 min, diluted with 140 mm3 HPLC eluent and microfuged to remove precipitated
protein. The supernate (20 mm3) was chromatographed at once. For NaOH hydrolysis, 20 mm3 samples
of bile were mixed with 10 mm3 1.0 M NaOH, followed after 3 min by 10mm3 1.0 M HCl and 80mm3
of HPLC eluent; the final mixture was microfuged and of the supernate an HPLC was carried out
without delay. The isocratic HPLC eluent was 0.1 M di-n-octylammonium acetate (prepared from
Aldrich di-n-octylamine and glacial acetic acid) in MeOH containing from 2–8% water and the column
an Ultrasphere-IP 5 ꢅm C-18 ODS column (25 cm, ID 0.46cm) fitted with a similar precolumn
(4.5cm, ID 0.46 cm). The column flow rate was 0.75–1.0cm3=min and the detector was set at the
absorption maximum of the injected regioisomer in the HPLC solvent.
Acknowledgements
This work was supported by the U.S. National Institutes of Health (HD17779 and DK26307). BAC was
an NSF REU Undergraduate Research Fellow. JOB was an R.C. Fuson Graduate Research Fellow. SEB
is on leave from the Institute of Organic Chemistry, Sofia, Bulgaria.
References
[1] a) Falk H (1989) The Chemistry of Linear Oligopyrroles and Bile Pigments. Springer-Verlag,
Wien; b) Lightner DA (1982) Structure, Photochemistry and Organic Chemistry of Bilirubin. In:
Heirwegh KPM, Brown SB (eds) Bilirubin, Vol 1. CRC Press, Boca Raton, FL, pp 1–58
[2] Chowdhury JR, Wolkoff AW, Chowdhury NR, Arias IM (2001) Hereditary Jaundice and
Disorders of Bilirubin Metabolism. In: Sly AL, Valle D (eds) The Metabolic and Molecular
Bases of Inherited Disease, Vol II. McGraw-Hill Inc, NY, chap 125, pp 3063–3101
´
[3] Bobal P, Lightner DA (2000) Synthesis, p 1835
[4] Woydziak ZR, Boiadjiev SE, Norona WS, McDonagh AF, Lightner DA (2005) J Org Chem 70:
8417

1476
S. E. Boiadjiev et al.: Xanthobilirubinic Acid Regioisomers
[5] a) McDonagh AF, Lightner DA, Agati G (1998) Monatsh Chem 129: 649; b) Landen GL, Park Y-T,
Lightner DA (1983) Tetrahedron (Symposium-in-Print on Linear Tetrapyrroles) 39: 1893;
c) Lamola AA, Braslavsky SE, Schaffner K, Lightner DA (1983) Photochem Photobiol 37: 263
[6] McDonagh AF, Lightner DA, Boiadjiev SE, Brower JO, Norona WS (2002) Bioorg Med Chem
Lett 12: 2483
[7] Blanckaert N, Kabra PM, Farina FA, Stafford BE, Marton LJ, Schmid R (1980) J Lab Clin Med
96: 198
[8] a) Piloty O, Thannhauser SJ (1912) Liebig’s Ann Chem 390: 191; (b) Piloty O, Thannhauser SJ
(1913) Chem Ber 45: 2393
€
€
[9] a) Fischer H, Rose H (1912) Chem Ber 45: 1579; b) Fischer H, Rose H (1912) Chem Ber 45:
€
€
3274; c) Fischer H, Rose H (1913) Hoppe Seyler’s Z Physiol Chem 82: 391; d) Fischer H, Rose H
(1913) Chem Ber 46: 439
€
€
[10] a)FischerH, Bartholomaus E, RoseH (1913) Hoppe Seyler’s Z PhysiolChem84: 262; b)FischerH,
€
Rose H (1914) Hoppe Seyler’s Z Physiol Chem 89: 255
[11] Fischer H, Hess R (1931) Hoppe Seyler’s Z Physiol Chem 194: 193
[12] Fischer H, Adler E (1931) Hoppe Seyler’s Z Physiol Chem 197: 237
[13] Fischer H, Adler E (1931) Hoppe Seyler’s Z Physiol Chem 200: 209
[14] Fischer H, Hartmann P (1934) Hoppe Seyler’s Z Physiol Chem 226: 116
[15] Grunewald JO, Cullen R, Bredfeldt J, Strope ER (1975) Org Prep Proc Intl 7: 103
[16] Trull FR, Franklin RW, Lightner DA (1987) J Heterocyclic Chem 24: 1573
[17] Jayasundera KP, Kinoshita H, Inomata K (2000) Bull Chem Soc Jpn 73: 497
[18] Barton DHR, Kervagoret J, Zard SZ (1990) Tetrahedron 46: 7587
[19] Kinoshita H, Hayeshi Y, Murata Y, Inomata K (1993) Chem Lett, p 1437
[20] Chen Q, Huggins MT, Lightner DA, Norona W, McDonagh AF (1999) JAm Chem Soc 121: 9253
[21] Clarke DJ, Keen JN, Burchell B (1992) FEBS Lett 299: 183
[22] Blanckaert N, Heirwegh PM, Zaman Z (1977) Biochem J 164: 229–236
[23] Shrout DP, Lightner DA (1990) Synthesis, p 1062
[24] Lightner DA, Quistad GB (1973) J Heterocyclic Chem 10: 273
´
[25] Bobal P, Lightner DA (2001) J Heterocyclic Chem 38(2): 527
[26] Battersby AR, Dutton CJ, Fookes CJ (1988) J Chem Soc Perkin Trans 1: 1569
[27] Plieninger H, Kurze J (1964) Liebig’s Ann Chem 680: 60
[28] Ghosh B, Lightner DA, McDonagh AF (2004) Monatsh Chem 135: 1189