Imploded bilirubins: Synthesis and properties of 10-nor-mesobilirubin- XIIIα and analogs

Article abstract of DOI:10.1007/s00706-008-0034-1

Six 10-nor-bilirubin analogs have been synthesized and investigated. Lacking the C(10) CH₂ group, these linear tetrapyrroles have a bipyrrole core rather than a dipyrrylmethane core and thus a different shape. Whereas the propionic acid groups

Full text of DOI:10.1007/s00706-008-0034-1

Monatsh Chem (2009) 140:97–110
DOI 10.1007/s00706-008-0034-1
ORIGINAL PAPER
Imploded bilirubins: synthesis and properties
of 10-nor-mesobilirubin-XIIIa and analogs
Edward B. Nikitin Æ Sanjeev K. Dey Æ
David A. Lightner
Received: 24 June 2008 / Accepted: 30 June 2008 / Published online: 27 September 2008
Ó Springer-Verlag 2008
Abstract Six 10-nor-bilirubin analogs have been synthe-
sized and investigated. Lacking the C(10) CH2 group, these
linear tetrapyrroles have a bipyrrole core rather than a
dipyrrylmethane core and thus a different shape. Whereas
the propionic acid groups of bilirubin are well engaged in
intramolecular hydrogen bonding to the dipyrrinones,
molecular modeling studies of the 10-nor-rubins predict that
propionic acid chains are too short to engage the CO2H
hydrogen fully in intramolecular hydrogen bonding with the
dipyrrinones. Butyric acid chains, however, can and do lead
to a stabilized conformation with a bipyrrole dihedral angle
of approximately 115°. Spectroscopic studies verify the
predictions and vapor pressure osmometry indicates that the
10-nor-rubins with butyric acids are monomeric in CHCl3.
less so, e.g., analogs such as the synthetic pigment, meso-
bilirubin-XIIIa (Fig. 1c) are intramolecularly hydrogen-
bonded. Such conformation-determining hydrogen bonding,
which depends on the unusually strong attraction of dipyrri-
none and carboxylic acid [3–6], dominates the pigment’s
shape, properties and metabolism [7–9]. The importance of
the C(10) CH2 group [10], and even a C(10)-S or Se [11–13],
to the pigment’s shape cannot be overstated: with two bond
rotational degrees of freedom between C(10) and the con-
joined approximately planar dipyrrinones, a large number of
conformations is implicitly possible; yet, in order to minimize
nonbonded steric interactions, few are low-energy, and those
are stabilized as a ridge-tile shape, with C(10) at the crease, by
intramolecular hydrogen bonding [10]. With more bond
rotational degrees of freedom—as in a homorubin, where
C(10)–CH2 is replaced by a –CH2–CH2– unit—the molecule
folds into a cup shape [14]. Also, in exploded bilirubins,
where the C(10) CH2 is replaced by–CꢁC– or –CꢁC–CꢁC–,
the pigments are linear in shape, even with intramolecular
hydrogen bonding [15]. Such shape-altered bilirubin analogs
are important to our understanding of the molecular mecha-
nisms at the various steps of hepatic elimination [7–9, 11–15].
In the current study, we have prepared bilirubin analogs
lacking a C(10) CH2: imploded rubins with a bipyrrole core
replacing the standard dipyrrylmethane. In the following,
we discuss the syntheses of 1–6, their stereochemistry and
properties (Structure 1).
Keywords Bipyrrole ꢀ Dipyrrinone ꢀ NMR
Introduction
Bilirubin (Fig. 1a), the yellow-orange pigment of jaundice
and mammalian bile [1, 2] is a linear tetrapyrrole [1] formed
in metabolism by opening the heme macrocycle. Along with
its analogs that have propionic acids located at C(8) and
C(12), bilirubin preferentially adopts a conformation that
resembles a half-opened book or ridge-tile that is greatly
stabilized by a network of intramolecular hydrogen bonds
linking propionic acid groups to the dipyrrinone units
(Fig. 1b). The location of the propionic acids is important,
and the nature and position of the remaining b-substituents
Results and discussion
Synthesis aspects
E. B. Nikitin ꢀ S. K. Dey ꢀ D. A. Lightner (&)
Department of Chemistry, University of Nevada,
Reno, NV 89557-0216, USA
Retrosynthetic analysis indicated two feasible routes
(Fig. 2a) toward the synthesis of 10-nor-rubins with a
e-mail: lightner@scs.unr.edu
123

98
E. B. Nikitin et al.
has already been reported in the literature [17, 18], and
(given the availability of 9-H dipyrrinones in our lab)
suggested an attractive, straightforward route. However,
both the 9-H dipyrrinone acid and ester (Fig. 2b) failed to
couple under a variety of established coupling conditions
[1, 17], and the starting material was recovered. Similarly,
the iodo-ester of Fig. 2b failed to couple under standard
Cu-catalyzed bipyrrole coupling methods [1, 19–21], and
the 9-H dipyrrinone was recovered. Apparently, the
presence of the propionic acid or ester group at C(8) is
involved in quenching the radical intermediate(s), because
coupling works well with a C(8)–CH3 group.
CO2H
HO2C
(A)
2
1
3
4
7
8
12
13
14
17
16
18
5
15
10
CH2
O
6
O
9
11
N
H
N
H
N
H
N 19
H
Dipyrrinone
Dipyrrinone
BILIRUBIN (Linear)
(B)
O
H
Given these unexpected difficulties, we focused on route
2 of Fig. 2a, the ‘‘1 ? 2?1’’ synthesis that expands from
the bipyrrole nucleus [22] and which required the prepa-
ration of the new bipyrroles 7, 9, 17 and 18 of Fig. 2a.
These bipyrroles thus became the core component targets
in the syntheses of 1–6, the final steps of which are outlined
in Scheme 1 showing: (a) reaction of tetraacid 7 or 17 with
5-bromomethylene-4-ethyl-3-methyl-3-pyrrolin-2-one [23]
to give dimethyl esters, e.g., 1e and 2e, which can be
saponified to 1 or 2—a standard dipyrrinone-forming pro-
cedure [16]; and (b) condensation of bipyrrole–dialdehyde
9 or 18 with 3,4-dimethyl-3-pyrrolin-2-one to give 3 or 4,
or with 3,4-diethyl-3-pyrrolin-2-one [24–27] to give 5 or
6—a well-developed dipyrrinone- and bilirubin-forming
reaction [1, 25, 28–30]. The condensations shown in (a)
above give dimethyl esters, which are saponified to diacids;
those of (b) gave diacids directly.
O
N
H
96-99°
15
O
H
N
8
10
11
12
N 9
H
4
O
NH
O
O
H
BILIRUBIN (Ridge-tile)
CO2H
HO2C
(C)
2
1
3
4
7
8
12
11
13
17
16
18
5
15
10
CH2
O
6
O
19
9
14
N
H
N
H
N
H
N
H
We first set out to prepare nor-mesobilirubin-XIIIa (1),
which required bipyrrole tetra acid intermediate 7, the
synthesis of which was designed from multifunctional
MESOBILIRUBIN-XIII
α
Fig. 1 a Bilirubin as a linear tetrapyrrole. b Bilirubin in its most
stable, intramolecularly hydrogen-bonded conformation. Only one of
two enantiomers is shown. c A synthetic bilirubin analog, mesobil-
irubin-XIIIa in a linear projection. The ridge-tile conformation is also
the most stable
2
R2
(A) R
Route
(1)
7
3
R1
R1
2
(CH2)nCO2H
O
(CH2)nCO2H
8
N
N
N
N
H
"2+2"
H
H
H
9 H
O
O
H
H
N
N
R1
HO2C(CH2)n
R2
R2
R2
Route
(2) "1+2+1"
R1
R1
CO2H
CO2H
O
N
N
N
N
H H
H H
O
X
O
O
H H
H H
N
N
(CH2)nCO2H
R1
N
N
R1
n
X
HO2C
N
HO2C
H
7
17
9
2
3
2
3
CO2H
CO2H
CHO
CHO
R2
R2
H
N
HO2C(CH2)n
1:
3:
5:
R1 = CH3, R2 = CH2CH3
R1 = R2 = CH3
R1 = R2= CH2CH3
2:
4:
6:
R1 = CH3, R2 = CH2CH3
R1 = R2 = CH3
R1 = R2 = CH2CH3
18
X
(B)
X = H, R = H
X = H, R = CH3
X = I, R = CH3
8
Structure 1
CO2R
N
H
N
H
9
X
O
2,20-bipyrrole core: (1) a ‘‘2 ? 2’’ approach [1, 16]
involving bipyrrole-formation coupling of 9-H dipyrri-
nones, and (2) a ‘‘1 ? 2?1’’ approach [16] wherein the
2,20-bipyrrole nucleus was extended at each end through
standard dipyrrinone-forming coupling reactions. The first
Fig. 2 a Possible retrosynthetic routes to 10-nor-bilirubin analogs:
(1) by coupling two 9-H dipyrrinones (R1 and R2 = CH3 or CH2CH3)
and (2) by dipyrrinone-forming reactions on a bipyrrole core (X = H
or CHO). b Dipyrrinones used in failed coupling attempts in route (1)
of a
123

Synthesis and properties of 10-nor-mesobilirubin-XIIIa and analogs
99
Scheme 1
HO2C(CH2)n
H
CO2H
N
n = 2
n = 3
O
CHBr
1e (36%)
2e (26%)
1 (61%)
2 (51%)
N
H
1) NaOH
2) HCl
N
H
MeOH, cat HBr
∆, 8 h
HO2C
(CH2)nCO2H
7: n = 2
17: n = 3
RO2C(CH2)n
O
N
H
CHO
N
H
3: n = 2 (38%) or 4: n = 3 (29% from 1)
5: n = 2 (40%) or 6: n = 3 (31% from 1)
1) NaOH/EtOH-H2O
23°C/12 h
2) HCl
N
H
OHC
(CH2)nCO2R
)
9: n = 2 (R = Et
O
N
18: n = 3 (R = H)
H
monopyrrole 22 (R = CH3). The latter had been prepared
from diester 21 (R = CH3) (Scheme 2) and converted to an
a-iodopyrrole (23, R = CH3), which was viewed as suitable
for Cu-catalyzed coupling to bipyrrole 24 (R = CH3) [31].
Selective de-esterification (debenzylation) of 24 to 25 (R =
CH3) was planned, followed again by decarboxylative
iodination to give 26, the iodines of which would be
replaced by acrylate esters via a Heck coupling reaction
[32]. However, this plan presented difficulties because the
diacid intermediate 25, when R = CH3, was highly insol-
uble [31] in characterization and Heck reaction solvents. In
order to overcome these solubility problems, we success-
fully carried out the entire reaction sequence of Scheme 2
with R = CH2CH2CH3, which improved the solubility of
the pyrrolic compounds sufficiently in standard organic
solvents [31, 33] so that debenzylation of 24–25 was
smoothly achieved.
Bipyrrole-forming reactions involving Cu-catalyzed
coupling of iodopyrroles can be capricious and depend,
inter alia, on the nature of the Cu and the activity at its
surface [1, 17, 18, 20, 21]. The deiodinated monopyrrole,
e.g., 28 (R = CH2CH2CH3), is a typical by-product, and at
Scheme 2
R
CO2Bn
R
CO2Bn
R
CO2Bn
Br2/SO2Cl2
I2/KI
NaHCO3
(84%)
EtO2C
EtO2C
EtO2C
CO2H
I
N
H
N
H
N
H
21
22
23
R'O2C
H
R
CO2R'
R
CO2Et
R'
N
I2/KI
NaHCO3
Cu/DMF
+
(47%)
EtO2C
H
N
H
N
H
EtO2C
(63% from 25, R = CH2CH2
CO2R'
28: R' = Bn
29: R' = H
24: R' = Bn
25: R' = H
conc
H2SO4
conc.
H2SO4
CO2Et
I
H
N
H
N
R
R
CO2Et
CO2Et
CO2Et
Pd(OAc)2
(68%)
N
N
EtO2C
R
EtO2C
R
H
H
I
CO2Et
26
27: R = CH2CH2CH3
10: R = CH3
123

100
E. B. Nikitin et al.
times even the major product. It was through this by-
product that we serendipitously discovered a shortcut to 27
involving a new bipyrrole coupling reaction [33] that
considerably shortened the sequence and facilitated the
project by providing a route to 10 that avoided poorly
soluble 25 (R = CH3) (vide infra).
synthetic path from 25 to 27. Thus, by mistake, in one run
the monopyrrole by-product 28 (R = CH2CH2CH3) was run
through the same two steps, and we obtained 27 in good
yield. At first we failed to realize that anything unusual had
happened, until we discovered that 28 (and not the intended
24) had been mistakenly carried forward—and actually
gave 27. Apparently 28 had been converted to 29 and on to
its 4,5-diiodo monopyrrole analog, and it was in fact this
monopyrrole that had undergone a one-pot bipyrrole cou-
pling together with a Heck reaction [32], with both
coupling and Heck reactions using the same Pd(OAc)2
catalyst. Subsequent test reactions showed that treatment of
the 5-iodo derivative of 28 under the same reaction con-
ditions (either with or without ethyl acrylate) failed to give
a bipyrrole, which was not surprising, as the Pd(OAc)2
catalyst is not known to convert iodopyrroles to bipyrroles.
We thus suspect that first a Heck reaction occurred at C(4)
of the diiodo-pyrrole to introduce the ethyl acrylate group,
which when coordinated to the Pd directed a bipyrrole
coupling reaction. A similar reaction with 1,2-diiodoben-
zene failed to give a biphenyl.
As outlined in Scheme 2 (R = CH2CH2CH3), the con-
version of 21–27 became routine, with acceptable yields at
each step, except for the Cu-catalyzed bipyrrole coupling
reaction of 23–24. Here the yield of 24 was variable, with
the deiodinated monopyrrole 28 being a major by-product.
It was normally separated from 24 before the latter was
carried forward in 2–3 steps (debenzylation, decarboxyl-
ation, iodination) to form diiodide 26. Diiodo-bipyrrole 26
was then converted via a Heck coupling [32] with ethyl
acrylate smoothly to afford 27, which, after reduction of
the acrylate ester carbon–carbon double bonds, was to
become the core component of the ‘‘1 ? 2 ? 1’’ synthesis
of 10-nor-rubins. However, they would be rubins with
n-propyl groups in place of the typical methyl groups on
the middle pyrrole rings.
Serendipitously, in one run, not 24 but the separated,
deiodinated monopyrrole by-product 28 (R = CH2CH2CH3)
was accidentally carried forward with debenzylation to 29,
iodination, the Pd(OAc)2-catalyzed reaction with ethyl
acrylate, and this led to the fortuitous discovery of a new
bipyrrole coupling reaction [33]. At that time the conver-
sion of 24–27 (R = CH2CH2CH3) had become so routine
that we were isolating but no longer characterizing the two
bipyrrole intermediates (25 and 26) in the two-step
The conversion of a 4,5-diiodopyrrole in one step to a
4-substituted-a,a0-bipyrrole thus represents a new and ef-
ficient bipyrrole coupling reaction. This discovery bypasses
the stepwise procedure of Scheme 2 and led to our aban-
doning it, along with all of the n-propyl pyrrole chemistry,
because the diiodo-pyrrole coupling substitution circum-
vents the insolubility problem posed by intermediate 25
(R = CH3) and thus opened up a facile route (Scheme 3)
to core bipyrroles 8 and 9 used in the syntheses of
Scheme 3
CO2Et
CO2Et
CO2Et
CO2H
Br2, SO2Cl2,
AcOH
H2SO4 conc.,
100°C, 20 min
+
EtO2C
EtO2C
EtO2C
CHO
N
H
N
H
N
H
15
14
13
KMnO4, acetone, water
10 hours, r.t.
(70% from 15)
CO2Et
CO2H
I
CO2Et
H
N
CO2Et
Pd(OAc)2, Et3N,
I2, KI, NaHCO3,
EtO2C
EtO2C
CO2H
I
EtOH, water, 65°C
(62% from 15)
MeCN, 60°C, 4 h
N
H
N
H
N
H
Et O2C
CO2Et
(58%)
12
11
10
EtO2C
CO2R
H
N
H
N
CO2R
CHO
H2/PtO2
EtOH, 1.5 h
1) TFA / (EtO)3CH
8
2) H2O
N
H
N
H
RO2C
OHC
RO2C
EtO2C
9
1) NaOH/MeOH-H2O
3 h
8: R = Et
(92%)
∆
2) HNO3-NaNO3 /H2O
-10°C
7: R = H
(90%)
123

Synthesis and properties of 10-nor-mesobilirubin-XIIIa and analogs
101
10-nor-rubins 1–3. In the most direct path to the desired
core bipyrroles 7 and 9 (outlined in Scheme 3), multi-
functional 22 (R = CH3) of Scheme 2 was not required;
rather, the synthesis starts from the readily available di-
ethyl 3,5-dimethyl-1H-pyrrole-2,4-dicarboxylate 15 [23].
Conversion of its a-CH3 to a-CO2H using Br2/SO2Cl2 was
often incomplete and gave a mixture of a-CHO 14 and
a-CO2H 13. However, the former could be converted
smoothly to the latter by reaction with dilute KMnO4, and
acid 13 was readily and selectively de-esterified to diacid
12 using conc. H2SO4. The latter diacid afforded 4,5-di-
iodopyrrole 11 following reaction with I2/KI in aqueous
NaHCO3 [15], and 11 coupled to give bipyrrole 10 in good
yield in a one-pot reaction with ethyl acrylate in the
presence of the Pd(OAc)2 catalyst. Catalytic hydrogenation
of 10 reduced the acrylic ester groups to propionic (8), and
saponification of 8 gave 7, which could be converted
readily to dialdehyde 9. From 7, we prepared 1; from 9 we
prepared 3 and 5 (see Scheme 1).
20 in good yield. The latter was reduced to give bis-butyric
ester bipyrrole 19, which was converted to 17, and then to
18, as shown in Scheme 4. From 17, we were able to
prepare 2, and from 18 we prepared 4 and 6, as outlined in
Scheme 1.
Structures and 13C NMR spectroscopy
The constitutional structures of 1–6 follow from their
syntheses (Schemes 1, 2, 3, 4) and are confirmed by their
13C NMR spectra. Most notably, in comparing 1 to meso-
bilirubin-XIIIa (Fig. 1c), the unique resonance at 23.3 ppm
corresponding to the C(10) CH2 group is absent, as it is in
2–6 (Table 1). Similarly, in the 1H NMR spectra,
the unique resonance near 4 ppm corresponding to the
C(10) CH2 is also absent in 1–6. One may find all of
the corresponding dipyrrinone ring carbon resonances of
mesobilirubin-XIIIa in the bis-dipyrrinones 1–6, and the
13C resonances of the b-substituents of 1 match up well
with those of mesobilirubin-XIIIa [29] and our recent
assignments for xanthobilirubic acid [35]. Similar corre-
lations for 3 and 5 may be found. The 13C NMR signals
from 10-nor-butyric acid rubins 2, 4 and 6 also correlate
well to mesobilirubin-XIIIa and its bis-butyric acid analog.
Differences between certain chemical shifts from 1 to 6
versus mesobilirubin-XIIIa are found mainly in the pyrrole
ring resonances: C(9)/C(11) are 3–4 ppm more deshielded
in 1–6, C(8)/C(12) are *7 ppm more deshielded, and C(7)/
C(13) and C(6)/C(14) are *8 ppm more deshielded.
Smaller deshieldings (1–2 ppm) are found in the pyrroli-
none rings of 1–6. All are apparently due to absence of the
C(10) CH2.
Synthesis of 2, 4, and 6 was also desired because
molecular modeling studies indicated that a butyric acid
chain would be a better fit in intramolecular hydrogen
bonding to the opposing dipyrrinones. Preliminary studies
showed that we could introduce a terminal acetylene by a
Sonogashira reaction [34] of diiodo-bipyrrole 26 (R =
CH2CH2CH3) [31]. The discovery that 4,5-diiodo-mono-
pyrrole 11 would yield the bis-acrylate bipyrroles 27 (R =
CH2CH2CH3) and 10 (R = CH3) (Scheme 3) prompted us
to attempt, successfully, a Sonogashira reaction [33] of 11
with methyl 3-butynoate (Scheme 4). Indeed the combined
substitution-bipyrrole coupling reaction worked nicely
under Sonogashira conditions, giving diacetylene bipyrrole
Scheme 4
CO2Me
I
CO2Me
H
CO2Et
N
H2 / Pd(C)
HC
C
I
EtO2C
Pd(PPh3)2Cl2
Et3N, 60°C
10 h (68%)
MeOH, 90 min
(85%)
N
H
N
H
EtO2C
11
CO2Me
20
HO2C(CH2)3
MeO2C(CH2)3
H
H
CO2Et
R
N
N
1) NaOH / MeOH
2) HCl
(91%)
N
N
H
EtO2C
R
H
(CH2)3CO2Me
(CH2)3CO2H
19
1) TFA /(EtO)3CH
17: R = CO2H
18: R = CHO
2) H2O
123

102
E. B. Nikitin et al.
Table 1 13C NMR chemical shifts of 10-nor-mesobilirubins 1–6 and mesobilirubin-XIIIa (MBR) in (CD3)2SO solvent
Carbona
1
3
5
2
4
6
MBRa
1/9
C=O
–C=
172.9
123.5
148.5
129.2
98.4
129.99
129.95
126.3
133.3
8.0
172.7
124.3
148.5
129.2
98.4
129.89
129.98
126.2
134.0
8.4
172.8
124.4
148.6
129.2
98.4
130.01
129.96
126.3
134.1
19.2
13.7
19.7
14.0
9.7
172.1
124.2
146.2
130.9
99.1
127.1
126.2
125.1
137.5
8.2
172.2
123.3
146.2
131.1
99.2
127.2
128.7
125.1
137.5
8.2
172.1
124.9
146.1
130.9
99.0
127.1
126.3
125.0
137.4
16.1
14.4
18.3
15.8
9.3
171.9
122.9
147.1
127.8
97.7
122.4
122.0
119.3
130.4
8.1
2/18
3/17
–C=
4/16
–C=
5/15
–CH=
–C=
6/14
7/13
–C=
8/12
–C=
9/11
–C=
21/181
22/182
31/171
32/172
71/131
81/121
82/122
83/123
83/123
10
CH2/CH3
CH3
–
–
–
–
–
CHCH3
CH3
19.6
15.0
9.6
19.6
–
17.2
14.9
9.3
9.4
17.2
14.8
9.1
–
CH3
9.6
9.5
CH2
21.2
33.7
–
21.2
33.8
–
21.2
33.8
–
20.8
23.9
34.6
174.2
–
21.0
24.0
34.7
174.3
–
20.9
24.0
34.6
174.2
–
19.2
34.6
–
CH2
CH2
CO2H
CH2
174.2
–
174.0
–
174.0
–
174.1
23.3
In d, ppm downfield from (CH3)4Si
a
Chemical shifts from [29] with assignments for 21/181 CH3 and 71/131 CH3 assigned according to our most recent studies [35]
Solution and chromatographic properties
nonpolar organic solvents such as CHCl3 and CH2Cl2. In
contrast, 2, 4 and 6 exhibit good solubility in CHCl3 and
CH2Cl2, an indication that the latter are less polar than the
former. Consistent with such polarity, on silica gel TLC
using 5% by vol CH3OH in CH2Cl2 as eluent, 2, 4 and 6
have an Rf * 0.75–0.78, and 1, 3 and 5 have an Rf * 0.29–
0.30, under the same conditions where mesobilirubin-XIIIa
exhibits an Rf * 0.89. As with bilirubin and mesobilirubin-
XIIIa, 2, 4 and 6 are not extracted into 5% (or saturated)
aqueous sodium bicarbonate from chloroform; however 1,
3 and 5 are extracted. Clearly, replacing the central –CH2–
of mesobilirubin-XIIIa by C–C has a major impact on the
chloroform/bicarbonate partition coefficient of 1, 3 and 5,
but it does not have a large effect when the chain lengths
are extended to butyric. Taken collectively, these data
suggest the presence of effective intramolecular hydrogen
bonding in 2, 4 and 6, but not in 1, 3 and 5, as predicted by
CPK molecular models, and as designed.
The 10-nor-rubins 1, 3 and 5 with propionic acid substit-
uents are more polar and less soluble in CHCl3 than the
corresponding 10-nor-rubins 2, 4 and 6 with butyric acid
substituents. Increased polarity in rubins typically signifies
decreased effectiveness of intramolecular hydrogen bond-
ing, which we attribute, at least in part, to the longer
hydrogen bonding distances in 1, 3 and 5 than in 2, 4 and 6.
In bilirubin and mesobilirubin-XIIIa, propionic acid chains
enjoy optimal intramolecular hydrogen bonding (Fig. 1c)
[36–40], but butyric acid chains are also found to possess
the right geometry for engaging the dipyrrinones in
hydrogen bonding [41–45]. When the bilirubin acid chains
are as short as acetic or as long as pentanoic and hexanoic,
rather counterintuitively the pigment polarity increases
[45]. Generally, when the acid chain lengths are mis-
matched for effective intramolecular hydrogen bonding, as
in bilirubins with pentanoic through octanoic acids at C(8)
and C(12), counterintuitively, the pigments typically
exhibit increased polarity (on TLC), decreased solubility
in CHCl3, and increased solubility in dilute aqueous
bicarbonate.
Intramolecular hydrogen bonding
In 10-nor-bilirubins, removal of the C(10) CH2 group
removes one rotational degree of freedom of the dipyrri-
nones relative to one another, allowing motion solely about
the bipyrrole sp2–sp2 C–C bond. Inspection of CPK models
indicates that an anticlinal (ac) conformation about the
Unlike their bright yellow parent, mesobilirubin-XIIIa,
10-nor-rubins 1–6 are yellow-orange solids that form yel-
lowish solutions. Analogs 1, 3 and 5 are insoluble in most
123

Synthesis and properties of 10-nor-mesobilirubin-XIIIa and analogs
103
C(9)–C(11) bond is the one most likely to lead to intra-
molecular hydrogen bonding with alkanoic acids situated at
C(8) and C(12), as in bilirubin. In such a conformation, the
C(8)/C(12) propionic CO2H of 1, 3 and 5 can engage in
hydrogen bonding with the opposing dipyrrinone’s NHs
and the lactam only if the dipyrrinones are twisted about
the C(5)–C(6) and C(14)–C(15) bonds. Otherwise the O=C
distance is too long for hydrogen bond formation. Fully
engaged hydrogen bonding between the lactam C=O and
the acid OH requires a longer acid chain than propionic.
The cost of twisting within the dipyrrinones destabilizes
the system by about 17–20 kJ/mol. Ideally, as in bilirubin,
the dipyrrinones would be planar. And for planar dipyrri-
nones, the butyric acids of 2, 4 and 6 appear to have the
ideal minimum chain length.
well as the deshielded COOH (*13.6 ppm) chemical shift
are representative examples. Similar chemical shifts have
been observed in a diverse array of other intramolecularly
hydrogen-bonded rubins [11–13, 24, 28–30, 46–50], even
in ‘‘exploded’’ rubins [15, 51], as well as in semirubins [3]
and hemirubins [52, 53], where only one dipyrrinone is
present. Table 2 shows the excellent coincidence of lactam,
pyrrole and carboxylic acid 1H NMR chemical shifts of 2, 4
and 6 in CDCl3 with those of mesobilirubin-XIIIa. On this
basis, one might conclude that these 10-nor-rubins are
intramolecularly hydrogen-bonded. Table 2 also clearly
shows differences between 1, 3 and 5 relative to 2, 4 and 6
in CDCl3. The NH chemical shifts of the former set are
very much like those seen for intermolecularly hydrogen-
bonded dipyrrinone dimers, thus suggesting more strained
dipyrrinone to carboxylic acid hydrogen bonding, and
possibly dipyrrinone-to-dipyrrinone dimers or oligomers,
rather than monomers.
Consistent with the predictions based on acid chain
1
length, the H NMR spectra of 2, 4 and 6 in CDCl3 reveal
NH and COOH chemical shifts (Table 2) very similar to
those of mesobilirubin-XIIIa [29] and characteristic of
intramolecular hydrogen bonding (Fig. 1b). In clear con-
trast, the corresponding chemical shifts of 1, 3 and 5
diverge significantly and suggest a different type of
hydrogen bonding. Dipyrrinones are known to participate
avidly in hydrogen bonding, especially to carboxylic acids,
but also to each other, with association constants of approx.
3 9 104 M-1 at 25 °C in CDCl3. At very high dilution,
dipyrrinone monomers show pyrrole and lactam NH
chemical shifts of *8 ppm, but more commonly seen
dipyrrinone dimers exhibit lactam and pyrrole NH signals
Thus, on the basis of these data, one might assume
carboxylic acid to dipyrrinone hydrogen bonding in 2, 4
and 6—as predicted by molecular modeling, but not in 1, 3
and 5. The hydrogen bonding observed in the butyric acid
O
H
O
N
H
O
H
H
N
15
13
7
1
at *11 and *10 ppm, respectively, in the H NMR. In
N
H
H
H
contrast, when the dipyrrinone is engaged in hydrogen
bonding to a COOH group, the lactam and pyrrole chem-
ical shifts are typically *10.5 and *9 ppm. The more
shielded pyrrole NH signal appears to be diagnostic of
intramolecularly hydrogen-bonded bilirubin-type mole-
cules. In bilirubin and mesobilirubin-XIIIa, the lactam NH
(*10.6 ppm) and pyrrole NH (*9.2) chemical shifts as
5
O
H
N
O
O
Fig. 3 Strong NOEs found for 4 in CDCl3 are shown by curved,
double-headed arrows. Weaker NOEs are shown by curved, dashed
arrows. Similar NOEs are seen in 2 and 6
1
Table 2 Comparison of carboxylic acid OH and lactam and pyrrole NH H NMR chemical shifts of 10-nor-bilirubins 1–6 and mesobilirubin-
XIIIa (MBR) in CDCl3 and (CD3)2SO
d (ppm) in (CD3)2SOa
a
Compound
d (ppm) in CDCl3
Lactam NH
Pyrrole NH
COOH
Lactam NH
Pyrrole NH
COOH
1
11.23
11.15
11.29
10.77
10.75
10.69
10.58
10.13
10.11
10.39
9.30
11.87
11.90
11.81
13.35
13.44
13.60
13.64
9.95
9.90
9.95
9.94
9.75
9.90
9.77
10.23
10.31
10.25
10.71
10.91
10.39
10.31
11.91
11.98
11.88
11.98
11.91
11.88
11.89
3
5
2
4
9.22
6
9.44
MBRb
9.15
a
d, ppm relative to (CH3)4Si at 23 °C for 10-3 to 10-4 M solutions
Values from [29]
b
123

104
E. B. Nikitin et al.
10-nor-rubins is further supported by NOEs between the
COOH and lactam NH (Fig. 3) and the additional, stronger
NOEs, between the lactam and pyrrole NHs and between
the C(5)/C(15) H and the C(7)/C(13) CH3 support the syn-
Z-conformation of the pigments. Pigments 1, 3 and 5 were
insufficiently soluble in CDCl3 for NOE studies.
coplanarity to minimize nonbonded steric interactions and
stabilized by intramolecular hydrogen bonding, given suf-
ficient length of the alkanoic acid substituents at C(8) and
C(12). Upon comparing 1, 3 or 5 and 2, 4 or 6, molecular
models show a better match for intramolecular hydrogen
bonding with butyric acids (2, 4, 6) than with propionic
acids (1, 3, 5): both CO2H groups of the first set are fully
engaged in intramolecular hydrogen bonding to an oppos-
ing dipyrrinone; in 1, 3 or 5 the CO2H group is less well
engaged. Conformational analysis by molecular dynamics
calculations using Sybyl (see the footnote near the start of
the ‘‘Experimental’’ section) gave an energy-minimum
conformation of 2, 4 and 6 in which the 114° interplanar
torsion angle (/ = N–C–C–N) of the anticlinal bipyrrole
unit is some 35° smaller than the / = 148° angle of 1, 3 and
5. In 2, 4 and 6, both butyric acid groups are fully engaged
in intramolecular hydrogen bonding to an opposing dip-
yrrinone (Fig. 4), but for 1, 3 and 5, while the propionic
acid C=O hydrogen bonds equally effectively [see d(b) and
d(c) of Table 3], in the latter, the CO2H hydrogen lies
Additional evidence in support of an intramolecularly
hydrogen-bonded structure for 2, 4 and 6 comes from vapor
pressure osmometry (VPO) determinations [3, 54] of
molecular weight in CHCl3 solution. For 4 (FW = 574.7)
and 6 (FW = 630.8), we determined molecular weights of
588
10 and 649
10, respectively. On this basis, we
take solutions of 2, 4, and 6 in CHCl3 to be monomeric.
Unfortunately, pigments 1, 3 and 5 were insufficiently
soluble in CHCl3 for VPO studies.
Conformation analysis from molecular dynamics
Independent rotations of the approximately planar, ther-
modynamically most stable syn-Z-dipyrrinones of 1–6
about the bipyrrole bond (/) lead to an infinite number of
conformations, including two high-energy limiting cases
[as determined by the Sybyl force field (see the footnote
near the start of the ‘‘Experimental’’ section)] where the
dipyrrinones lie coplanar: the syn and anti. Lying between
these extremes is a low-energy conformation twisted from
˚
*0.1 A further removed from the lactam C=O than in 2, 4,
6 and mesobilirubin [see d(a) of Table 3]. Note that the
computed hydrogen bond distances (Table 3) of 2 and 4
and 6 are nearly identical to those in mesobilirubin. It is
also important to note here that, as with other bipyrroles,
1–6 adopt equi-energetic enantiomeric conformations.
Fig. 4 Ball and stick (see the
footnote near the start of the
‘‘Experimental’’ section)
2
3
representations of the minimum
2
3
energy conformations of 3 (bis-
propionic acid), 4 (bis-butyric
acid) and 10-nor-rubins, as
determined by molecular
1
4
4
1
5
5
mechanics computations using
the Sybyl forcefield. The
N21
N21
N22
6
6
7
relevant torsion angles (/ and
w, in degrees) are: / = (22-9-
11-12), w1 = (4-5-6-22), and w2
= (23-24-15-16). In 3: / = 110°,
w1 = 20°, and w2 = 9°; in 4: /
= 118°, w1 = 19°, and w2 = 6°.
Significantly, the O–HꢀꢀꢀO
7
N22
8
8
9
9
11
11
12
13
(lactam) bond angle in 3 is 35°,
whereas in 4 it is 173°—
12
N23
indicating that the OH of 3, but
not 4, is turned away from
proper intramolecular hydrogen
bonding, and consistent with the
finding that the CO2HꢀꢀꢀO
N23
16
13
14
15
14
15
N24
N24
17
16
(lactam) hydrogen bond is long
in 3; whereas in 4 it is the same
as in bilirubin and mesobilirubin
(Table 3). The computed results
for 1 and 5, and 2 and 6 are
similar to those of 3 and 4,
respectively
19
19
17
18
18
3
4
123

Synthesis and properties of 10-nor-mesobilirubin-XIIIa and analogs
105
Table 3 Comparison of computed torsion angles and nonbonded distances of 10-nor-rubins 3–6 and mesobilirubin-XIIIa (MBR) in their most
stable conformations
R2
ψ1
C O
O
Z
(CH2)n
H
O
4
N
R1
N
H22
φ
21H
b
R1
H
N
H
N
c
Oy
O1
ψ2
a
H
Z
R2
(CH2)n
15
C
Ox
˚
Distances (A)
Torsion angles (°)
H-bond angles (°)
n
R1
R2
/
w1
w2
d(a)
d(b)
d(c)
OHꢀꢀꢀO
N22–HꢀꢀꢀO
N21–HꢀꢀꢀO
3
2
2
3
3
2
Me
Et
Me
Et
148
148
114
114
64
25
25
24
24
17
25
25
26
25
17
1.63
1.63
1.52
1.51
1.53
1.53
1.53
1.56
1.57
1.59
1.55
1.55
1.61
1.60
1.56
161
161
171
169
170
145
145
142
142
163
143
144
146
148
154
5
4
Me
Et
Me
Et
6
MBR
Me
Et
From Sybyl vers. 7.0 run on an SGI workstation. The 0° torsion angles are those where the relevant nitrogens are syn-periplanar
Table 4 Solvent-dependence of the UV–visible spectral data of 10-nor-rubins 1–6 with mesobilirubin-XIIIa (MBR)
Compound
emax (dm3 mol-1 cm-1) [kmax (nm)]
Benzene
CHCl3
(CH3)2CO
CH3OH
CH3CN
(CH3)2SO
1
39,200 (450)
38,200 (452)
37,000 (456)
38,900 (453)
35,950 (456)
40,000 (455)
49,900 (439)
38,400 (455)
40,100 (454)
39,800 (457)
38,700 (455)
39,450 (456)
37,800 (454)
50,300 (431)
37,900 (454)
39,800 (456)
38,000 (455)
37,500 (455)
36,950 (451)
38,500 (452)
49,700 (427)
38,200 (450)
37,100 (451)
36,600 (452)
37,200 (452)
39,300 (453)
38,900 (454)
50,600 (425)
37,600 (449)
37,000 (449)
39,900 (452)
40,100 (452)
39,150 (451)
36,500 (452)
49,000 (425)
36,400 (450)
36,000 (454)
34,800 (453)
37,700 (456)
30,250 (453)
37,250 (452)
52,500 (428)
3
5
2
4
6
MBR
Shoulders (or) inflections were determined by first- and second-derivative spectra at 22 °C in concentrations of *1.4 9 10-5
M
From the computed relative energies of the most stable
conformations and the syn and anti-periplanar bipyrrole
conformations (/ = 0° and / = 180°, respectively), one can
compute the lowest-energy barriers to racemization via the
ap bipyrrole conformers as *105 kJ/mol for 1, 3 and 5 and
*120 kJ/mol for 2, 4 and 6, suggesting that the global
minimum conformations of the former are less stable than
those of the latter by *15 kJ/mol.
chloroform, acetone, methanol, acetonitrile, and dimethyl-
sulfoxide), the UV–visible long wavelength absorbance,
i.e., kmax, of 1–6 show much less solvent-dependence than
of mesobilirubin-XIIIa (Table 4), indicating very little
conformational change over the range of solvent polarity.
The absorptivity of the long wavelength band is smaller
than in mesobilirubin-XIIIa.
In the UV–visible spectrum of mesobilirubin-XIIIa, the
long wavelength absorption near 430 nm originates from
exciton coupling [10, 55], from the electric dipole–
electric dipole transition moment interaction of the two
dipyrrinone chromophores. The dipyrrinone chromophores
of this study typically exhibit UV–visible kmax * 400–420
nm for their long-wavelength absorption band, with the
electric transition dipole moment lying along the long axis
of the chromophore [1, 55]. The origin of the UV–visible
band of 1–6 is not entirely clear. The *450 nm band is
somewhat red-shifted from the usual dipyrrinone kmax
(400–420 nm) or mesobilirubin-XIIIa; the intensity is less
Conformation from UV–visible spectra
Additional evidence regarding the conformation of 1–6
comes from solvent-dependent UV–visible spectra. Com-
pared with their ‘‘parent’’ rubin (mesobilirubin-XIIIa, with
broad absorption near 430 nm and a shoulder near 395 nm
in most solvents), the 10-nor-rubins showed (Table 4) a
broad band centered near 450 nm, thus accounting for their
more orange color. Over a wide range of solvents of
varying polarity and hydrogen bonding ability (benzene,
123

106
E. B. Nikitin et al.
than that of the latter but larger than that of typical dip-
yrrinones [1, 3–6, 10, 46–50]. The *450 nm band, found
also in simpler 10-nor-rubins without alkanoic acid groups
[17, 18], might be viewed as originating in conjugation
rather than from exciton coupling [56]. In the simpler 10-
nor-rubins, the bipyrrole twist angle (/) is computed (PPP
molecular orbital calculation) to be 65° [1, 17]. The X-ray
structure indicates 38° [1]—both reveal a synclinal ste-
reochemistry. In contrast, with limited (1, 3, 5) or full (2, 4, 6)
intramolecular hydrogen bonding, the 10-nor-rubins are
predicted to have an anticlinal stereochemisty about C(9)–
C(11). Just how such a different stereochemistry may affect
the UV–visible spectrum is unclear. Alternatively, one
might view the broad 450 nm band(s) as having exciton
coupling between the two dipyrrinones, whose UV–visible
characteristics remain unknown. Irrespective of its origin,
the 450 nm band in the UV–visible spectra of 1–6 remains
essentially wavelength- and absorption-invariant over a
wide range of solvent polarity (Table 4), suggesting that
the relative orientation of the dipyrrinones is unchanged, as
is the relative orientation of electric dipole transition
moments. Further computational studies are required to
sort out the origin of the 450 nm band.
were performed by Desert Analytics (Tucson, AZ, USA)
and were within 0.4% of theoretical values. Fast atom
bombardment high-resolution mass measurements from
the Nebraska Center for Mass Spectrometry gave
molecular ions with exact masses within 0.7–2.8 ppm of
theoretical values. Infrared spectra were recorded on a
PerkinElmer (Waltham, MA, USA) Spectrum 2000
FT-IR infrared spectrophotometer. All UV–visible spectra
were recorded on a PerkinElmer k-12 spectrophotometer.
Flash column chromatography was carried out using
silica gel, 60–200 mesh (M. Woelm, Eschwege,
Germany). Radial chromatography was carried out on 1
or 2 mm thick rotors of Merck (Darmstadt, Germany)
silica gel PF254-gypsum, preparative layer grade, using a
Chromatotron (Harrison Research, Inc., Palo Alto, CA,
USA). The ball and stick drawings were created from the
atomic coordinates of the molecular dynamics structures
using Mu¨ller and Falk’s program (Cherwell Scientific,
Oxford, UK) for Macintosh.1 Molecular dynamics
calculations were carried out on a Silicon Graphics
workstation using Sybyl 7.0 with a Gasteiger–Hu¨ckel
calculation of electronic charge.
All solvents were reagent grade, obtained from Fisher
(Pittsburgh, PA, USA) and Acros (Geel, Belgium).
Deuterated chloroform, dichloromethane and dimethyl-
sulfoxide were from Cambridge Isotope Laboratories
(Andover, MA, USA); (trimethylsilyl)acetylene, propargyl
alcohol, 3-butyn-1-ol and phenylacetylene were from GFS
Chemicals (Powell, OH, USA). Pd(C)-10%, Cu powder
(99%, for organic synthesis, batch #16920 AB), palladium
acetate and Pd(PPh3)2Cl2 were from Sigma-Aldrich (St.
Louis, MO, USA. Copper(I) iodide (Cu2I2, stock #26110,
lot #060674) was from Alfa Products (Ward Hill, MA,
USA). Diethyl 3,5-dimethyl-1H-pyrrole-2,4-dicarboxylate
(15) [57] and 5-bromomethylene-4-ethyl-3-methyl-2-oxo-
2,5-dihydropyrrole [23], 3,4-dimethyl-3-pyrrolin-2-one
[24–26] and 3,4-diethyl-3-pyrrolin-2-one [27] were pre-
pared according to previously described procedures.
Concluding comments
Unlike bilirubin, the 10-nor-rubins of this work cannot be
folded into ridge-tile shapes such as in Fig. 1; however,
with the propionic acids extended to butanoic, full intra-
molecular hydrogen bonding (Fig. 3) is possible and can
preserve a twisted, anticlinal, rotated conformation about
the bipyrrole core. With propionic acid chains, the lactam
C=O to CO2H hydrogen bond is longer, and the system is
more polar and somewhat less conformationally stable.
Interestingly, molecular modeling shows that with acetic
acid chains, full hydrogen bonding (acid to dipyrrinone) is
possible, but the bipyrrole is planar.
10-nor-Mesobilirubin-XIIIa (1, C32H38N4O6)
Tetrapyrrole dimethyl ester (1e) (30 mg, 0.05 mmol) was
suspended in nitrogen-purged methanol (5 cm3) and heated
to reflux. Nitrogen-purged 1 M aqueous sodium hydroxide
(2 cm3) was then added, and reflux was continued for 6 h.
The mixture was cooled to room temperature, water
(5 cm3) was added, then the methanol was removed under
Experimental
1
All H and 13C NMR spectra were obtained on a General
Electric (Fairfield, CT, USA) QE-300 or GN-300 spec-
trometer at 300 and 75 MHz, respectively, or on a
Varian (Palo Alto, CA, USA) Unity 400 MHz spec-
trometer. Chemical shifts in CDCl3 are reported in ppm,
referenced to the residual chloroform proton signal at
7.26 ppm and its 13C signal at 77.23 ppm, unless
otherwise noted. All GC–MS spectra were obtained from
a Varian CP-3800 mass spectrometer. Melting points
were taken on a Mel-Temp apparatus. Satisfactory
combustion analyses for carbon, hydrogen, and nitrogen
1
Molecular mechanics and dynamics calculations employed to find
the global energy minimum conformation of 1 were run on an SGI
Octane Workstation using version 7.0 of the Sybyl forcefield, as
described in [10, 46–50]; the ball and stick drawings were created
from the atomic coordinates using Mu¨ller and Falk’s ‘‘Ball and Stick’’
program for the Macintosh (http://www.orc.uni-Linz.ac.at/mueller/
ball_and_stick.shtml).
123

Synthesis and properties of 10-nor-mesobilirubin-XIIIa and analogs
107
reduced pressure (rotovap). The resulting mixture was
cooled to 5 °C and carefully acidified with concentrated
HCl until a precipitate formed. The precipitate was
collected by filtration, washed with water, and dried in
vacuo to produce diacid (1). Yield: 17 mg (61%); mp 228–
230 °C; 1H NMR: d = 1.14 (6H, t, J = 7.5 Hz), 1.84 (6H, s),
2.45 (6H, s), 2.47 (6H, t, J = 7.3 Hz), 2.50 (4H, q, J = 7.5
Hz), 2.63 (4H, t, J = 7.3 Hz), 6.34 (2H, s), 10.13 (2H, brs),
11.23 (2H, brs), 11.87 (2H, brs) ppm; 13C NMR data are in
Table 1; HRMS (FAB, glycerol): calcd for M ? H?,
C32H39N4O6, 575.2870; found 575.2864.
(Na2SO4). After the solvent was removed, dialdehyde 9 was
obtained as a greenish solid and used in the next step without
further purification. Dialdehyde 9 and 3,4-dimethyl-3-pyrr-
olin-2-one [24–26] (0.31 g, 2.84 mmol) were placed in a 50
cm3 round-bottom flask containing a magnetic stir bar. To
the mixture of solids was added 3 cm3 of methanol and 10
cm3 of 4 M aqueous KOH, and the solution was stirred at
room temperature for 12 h, then warmed briefly (over 30
min) almost to reflux. The cooled solution was then acidified
to pH 3 with concentrated HCl. A bright yellow solid (3) was
separated, repeatedly washed with water, collected by
centrifugation and dried. Yield: 0.15 g (38% from 9); mp
232–235 °C; 1H NMR: d = 1.77 (6H, s), 2.25 (6H, s), 2.45
(6H, s), 2.47 (4H, t, J = 7.4 Hz), 2.63 (4H, t, J = 7.4 Hz), 6.68
(2H, s), 10.11 (2H, brs), 11.15 (2H, brs), 11.90 (2H, brs)
ppm; 13C NMR data are in Table 1.
10-nor-Mesobilirubin-XIIIa dimethyl ester
(1e, C34H42N4O6)
To a 50 cm3 round-bottom flask equipped with a magnetic
stirrer and
a
heating mantle were added 3,30-bis
-(2-carboxyethyl)-4,40-dimethyl-1H,10H-2,20-bipyrrole-5,
50-dicarboxylic acid (7) (230 mg, 0.55 mmol), 5-bromom-
ethylene 4-ethyl-3-methyl-3-pyrrolin-2-one (250 mg, 1.16
mmol), methanol (15 cm3), and five drops 40% aqueous
HBr. The mixture was heated at reflux under a blanket of
nitrogen, and after 8 h the mixture was cooled. Methanol
was removed under reduced pressure (rotovap), and the
dark green residue dissolved in a small portion of CH2Cl2.
Hexane (50 cm3) was added, and the resulting precipitate
was collected by filtration. (The orange-red hexane filtrate
contained primarily unreacted bromomethylenepyrroli-
none.) The precipitate was redissolved in CH2Cl2, and
the resulting solution was passed through a short column of
silica gel, eluting with a mixture of CH2Cl2 and CH3OH
(97:3 vol/vol). The solvent was removed under reduced
pressure, and tetrapyrrole dimethyl ester 1e was purified by
radial chromatography using a 97:3 vol/vol mixture of
CH2Cl2 and CH3OH as eluent. The purified diester was
crystallized from CH2Cl2–hexane, collected by filtration
and washed with hexane to give pure 1e. Yield: 0.12 g
(36%); mp 230–232 °C; 1H NMR: d = 1.13 (6H, t, J = 7.5
Hz), 1.82 (6H, s), 2.51 (4H, q, J = 7.5 Hz), 2.53 (3H, s),
2.62 (4H, q, J = 7.3 Hz), 10.47 (2H, brs) ppm.
2,18-Des-methyl-2,18-diethyl-10-nor-mesobilirubin-XIIIa
(5, C34H42N4O6)
Dialdehyde 9, prepared from 3,30-bis-(2-carboxyethyl)-
4,40-dimethyl-1H,10H-2,20-bipyrrole-5,50-dicarboxylic acid
(7) (30 mg, 0.71 mmol) as described above, and 3,4-
diethyl-1H-pyrrolin-2-one [27] (0.39 g, 2.84 mmol) were
treated as above for the synthesis of 5. The desired yellow
solid diacid (5) was isolated as above. Yield: 0.17 g, 40%
(from 7); mp 135–138 °C; 1H NMR: d = 1.11 (6H, t, J = 7.5
Hz), 1.12 (6H, t, J = 7.4 Hz), 2.25 (4H, q, J = 7.5 Hz), 2.40
(4H, q, J = 7.4 Hz), 2.45 (6H, s), 2.47 (4H, t, J = 7.5 Hz),
2.63 (4H, t, J = 7.5 Hz), 6.49 (2H, s), 10.39 (2H, brs), 11.29
(2H, brs), 11.81 (2H, brs) ppm; 13C NMR data are in
Table 1.
8,12-Des-propionic acid-8,12-bis-butyric acid-10-nor-
mesobilirubin-XIIIa (2, C34H42N4O6)
Using the procedure for the preparation of 1 from 1e,
tetrapyrrole dimethyl ester 2e (82 mg, 0.13 mmol) was
suspended in nitrogen-purged methanol (4 cm3) and heated
to reflux. Nitrogen-purged aqueous sodium hydroxide (1.5
cm3, 1 M) was then added, and reflux was continued for 6
h. Work-up as for 1 afforded diacid 2. Yield: 38 mg (51%);
1
3,17-Des-ethyl-3,17-dimethyl-10-nor-mesobilirubin-XIIIa
(3, C30H34N4O6)
mp 153–156 °C; H NMR: d = 1.13 (6H, t, J = 7.5 Hz),
1.70 (4H, m), 1.82 (6H, s), 2.36 (4H, t, J = 7.4 Hz), 2.51
(8H, m), 2.63 (6H, s), 6.12 (2H, s), 9.30 (2H, brs), 10.77
(2H, brs), 13.35 (2H, brs) ppm; 13C NMR: d = 8.9 (q), 10.3
(q), 15.2 (q), 20.0 (t), 22.0 (t), 24.0 (t), 34.0 (t), 99.6 (d),
124.6 (s), 125.4 (s), 130.0 (s), 130.5 (s), 130.8 (s), 132.1
(s), 137.7 (s), 173.6 (s), 177.9 (s) ppm; 13C NMR data in
(CD3)2SO are in Table 1.
3,30-Bis-(2-carboxyethyl)-4,40-dimethyl-1H,10H-2,20-bipyr-
role-5,50-dicarboxylic acid (7) (30 mg, 0.71 mmol) was
placed in a 100 cm3, round-bottom flask previously cooled in
an ice/water bath. TFA (0.7 cm3) was then added, and the
resulting dark brown solution was stirred in the ice/water
bath for 30 min. Anhydrous triethyl orthoformate (0.27 cm3)
was added while the reaction was still being cooled. The
resulting dark red solution was stirred for an additional 2 h at
0 °C, then water (50 cm3) was added to precipitate a green
solid (9). The solid was removed by extraction into CH2Cl2,
and the green organic layer was washed once with 1 M
aqueous NaOH solution, then with brine, before being dried
8,12-Des-(2-methoxycarbonylethyl)-8,12-(3-carboxy-
methyl)propyl-10-nor-mesobilirubin-XIIIa
(2e, C36H40N4O6)
Following the procedure for the synthesis of 1e, 3,30-bis-
(3-carboxypropyl)-4,40-dimethyl-1H,10H-2,20-bipyrrole-5,50-
123

108
E. B. Nikitin et al.
dicarboxylic acid 17 (210 mg, 0.5 mmol) and 5-bromom-
ethylene-4-ethyl-3-methyl-3-pyrrolin-2-one (237 mg, 1.1
mmol) in 12 cm3 methanol with four drops aqueous HBr
(40%) gave dimethyl ester 2e. Yield: 82 mg (26%); mp
3,30-Bis-(2-carboxyethyl)-4,40-dimethyl-1H,10H-2,20-bipyr-
role-5,50-dicarboxylic acid (7, C18H20N2O8)
A mixture of tetraester 8 (1.00 g, 1.97 mmol), NaOH (0.67
g, 0.017 mol), 95% ethanol (9 cm3) and H2O (3 cm3) was
heated at reflux for 3 h. The ethanol was removed by
distillation, and the residue was cooled. Then a solution of
NaNO2 (6.5 g, 0.076 mol) in water (10 cm3) was added,
and the flask was cooled to -10 °C in a dry ice/acetone
bath. Concentrated HNO3 (1.5 cm3) (at -20 °C) was added
very slowly to keep the temperature at -10 °C, and the
mixture was stirred an additional 30 min after the addition
was complete. The colorless product was collected by
filtration, washed with cold H2O and dried at 0.7 mm Hg in
a dessicator over P2O5 overnight to afford a pink solid.
Yield: 0.69 g (92%); 13C NMR ((CD3)2SO): d = 10.5, 19.6,
34.0, 119.3, 122.0, 124.4, 125.2, 162.4, 174.8 ppm; HRMS
(FAB, glycerol): calcd for M ? H?, C22H28N2O8,
393.1298; found 393.1317.
1
154–155 °C; H NMR: d = 1.13 (6H, t, J = 7.5 Hz), 1.82
(10H, m), 2.30 (4H, t, J = 7.4 Hz), 2.51 (4H, q, J = 7.5 Hz),
2.63 (10H, m), 3.60 (6H, s), 6.73 (2H, s), 10.9 (2H, brs),
10.41 (2H, brs); 13C NMR: d = 8.5 (q), 10.8 (q), 15.0 (q),
17.2 (t), 25.6 (t), 26.2 (t), 33.2 (t), 51.6 (q), 98.1 (d), 121.3
(s), 123.4 (s), 124.8 (s), 127.3 (s), 130.3 (s), 134.6 (s),
149.4 (s), 172.4 (s), 172.8 (s) ppm.
3,17-Des-ethyl-3,17-dimethyl-8,12-des-propionic acid-
8,12-bis-butyric acid-10-nor-mesobilirubin-XIIIa
(4, C32H38N4O6)
Using the procedure for the synthesis of 3, 3,30-bis-(3-
carboxypropyl)-4,40-dimethyl-1H,10H-2,20-bipyrrole-5,50-
dicarboxylic acid (17) (596 mg, 1.42 mmol) was placed in
a 100 cm3 round-bottom flask previously cooled in an ice/
water bath, TFA (1.4 cm3) was then added, and the
resulting dark brown solution stirred in the ice/water bath
for 30 min. Anhydrous triethyl orthoformate (0.52 cm3)
was added while the reaction was still being cooled, and
the resulting dark red solution of aldehyde 18 was treated
as in 3. 3,4-Dimethyl-3-pyrrolin-2-one [24–26] (790 mg,
7.1 mmol) was used in the NaOH-catalyzed condensation
to afford 4. Yield: 237 mg (29% from 24); mp 142–144 °C;
1H NMR: d = 1.73 (10H, m), 2.25 (6H, s), 2.33 (6H, t, J =
7.5 Hz), 2.64 (6H, t, J = 7.4 Hz), 6.60 (2H, s), 9.22 (2H,
brs), 10.75 (2H, brs), 13.44 (2H, brs) ppm; 13C NMR: d =
9.0 (q), 10.5 (q), 20.1 (t), 22.1 (t), 24.2 (q), 34.3 (t), 99.7
(d), 124.3 (s), 124.8 (s), 125.5 (s), 130.2 (s), 130.5 (s),
Diethyl 3,30-bis-(2-ethoxycarbonylethyl)-4,40-dimethyl-2,20-
bipyrrole-5,50 dicarboxylate (8, C26H36N2O8)
Diethyl 3,30-bis-(2-ethoxycarbonylethenyl)-4,40-dimethyl-
2,20-bipyrrole-5,50-dicarboxylate (10) (40 mg, 0.072 mmol)
was dissolved in 12 cm3 absol. ethanol and hydrogenated at
room temperature and atmospheric pressure over 10% PtO2
(1.6 mg, 0.0072 mmol) for 1.5 h. After filtration through a
pad of silica gel, the solution was evaporated to yield the
reduced product (8), which was purified by radial chroma-
tography (CH2Cl2 eluent). Yield: 0.037 g (92%); mp 151–
152 °C; 1H NMR: d = 1.29 (6H, t, J = 7.4 Hz), 2.26 (6H, s),
2.40 (2H, t, J = 7.5 Hz), 2.70 (2H, t, J = 7.5 Hz), 4.14 (4H,
q, J = 7.5 Hz), 4.10 (4H, q, J = 7.4 Hz), 9.0 (2H, brs) ppm;
HRMS (FAB, glycerol): calcd for M ? H?, C26H37N2O8,
505.2549; found 505.2553.
130.9 (s), 132.1 (s), 137.7 (s), 173.8 (s), 178.0 (s) ppm; 13
C
NMR data in (CD3)2SO are in Table 1.
2,18-Des-methyl-2,18-diethyl-8,12-des-propionic acid-
8,12-bis-butyric acid-10-nor-mesobilirubin-XIIIa
(6, C36H48N4O6)
Diethyl 3,30-bis-(2-ethoxycarbonylethenyl)-4,40-dimethyl-2,20-
bipyrrole-5,50-dicarboxylate (10, C26H36N2O8)
As in the synthesis of 5, 3,30-bis-(3-carboxypropyl)-4,40-
dimethyl-1H,10H-2,20-bipyrrole-5,50-dicarboxylic acid (17)
(25 mg, 0.6 mmol), TFA (0.7 cm3) and anhydrous triethyl
orthoformate (0.25 cm3) were used to generate aldehyde 18.
Upon treatment with 3,4-diethyl-1H-pyrrolin-2-one [27]
(417 mg, 3 mmol) in an NaOH-catalyzed condensation, the
desired diacid 6 was obtained. Yield: 117 mg (31% from
24); mp 146–148 °C; 1H NMR: d = 1.10 (12H, m), 1.70 (4H,
m), 2.25 (4H, q, J = 7.5 Hz), 2.33 (4H, t, J = 7.6 Hz), 2.40
(4H, q, J = 7.5 Hz), 2.69 (10H, m), 6.32 (2H, s), 9.44 (2H,
brs), 10.69 (2H, brs), 13.60 (2H, brs) ppm; 13C NMR: d =
9.0 (q), 10.4 (s), 14.1 (q), 15.5 (q), 20.2 (t), 22.1 (t), 22.1 (t),
34.3 (t), 100.0 (d), 124.0 (s), 124.7 (s), 125.5 (s), 130.2 (s),
130.6 (s), 130.8 (s), 132.2 (s), 137.7 (s), 173.8 (s), 176.9 (s)
ppm; 13C NMR data in (CD3)2SO are in Table 1.
A mixture of ethyl 4,5-diiodo-3-methyl-1H-pyrrole-2-car-
boyxlate (11) (0.37 g, 0.9 mmol), freshly distilled triethyl
amine (0.25 cm3, 1.8 mmol), ethyl acrylate (0.2 cm3, 1.8
mmol), PdCl2(PPh3)2 (0.014 g, 0.02 mmol), and acetoni-
trile (50 cm3) was heated to 60 °C and stirred under N2 for
12 h. The reaction mixture was cooled to room temperature
and taken up in CH2Cl2 (25 cm3). The solution was washed
with cold dilute 0.5 M aqueous HCl, saturated NaCl, and
dried (MgSO4). Evaporation of the solvent (rotovap) gave
the crude product, which was purified by radial chroma-
tography (hexane/Et2O, 4:1, vol/vol). Yield: 0.26 g (58%);
mp 140–141 °C; 13C NMR: d = 14.3, 14.5, 27.2, 60.5, 60.9,
118.0, 120.5, 121.7, 126.3, 133.2, 136.2, 161.5, 167.7 ppm;
HRMS (FAB, 3-NBA): calcd for C26H36N2O8, 504.2472;
found 504.2491.
123

Synthesis and properties of 10-nor-mesobilirubin-XIIIa and analogs
109
Ethyl 4,5-diiodo-3-methyl-1H-pyrrole-2-carboxylate
(11, C8H9I2NO2)
and washed with water. The precipitate was dissolved in
saturated sodium bicarbonate. The undissolved aldehyde
was collected by vacuum filtration and the filtrate was
acidified with dilute HCl. The resulting precipitate was
collected by vacuum filtration, washed with water, and
dried. Acid 13 was recrystallized from 2-propanol to afford
9.23 g. Aldehyde 14 (the side product of this reaction) was
dissolved in 150 cm3 of reagent grade acetone, and a
solution of KMnO4 [5 g in 200 cm3 of acetone–water (1:1
by volume)] was added over 1 h. After 10 h of stirring at
room temperature, the purple solution was poured into 250
cm3 of a solution of 10% NaHSO3, the precipitate was
removed by vacuum filtration, and the yellow aqueous
solution was acidified with dilute HCl. The resulting white
precipitate was collected by vacuum filtration, washed with
water, and dried. Acid 13 was recrystallized from 2-
propanol. Yield: 11.2 g (70% combined yield); mp 271–
272 °C ([57], mp 273–274 °C); 13C NMR ((CD3)2SO): d =
10.7 (q), 13.9 (q), 14.1 (q), 60.3 (t), 60.6 (t), 118.9 (s),
121.4 (s), 127.3 (s), 127.5 (s), 160.2 (s), 160.8 (s), 165.0 (s)
ppm.
In a 1,000 cm3 beaker, 10 g (41.46 mmol) of crude
5-(ethoxycarbonyl)-4-methyl-1H-pyrrole-2,3-dicarbox-
ylic acid (12) was dissolved in 70 cm3 of hot ethanol.
Sodium bicarbonate (17.8 g, 207.3 mmol) in 100 cm3 of
water was added carefully, in portions, to the mixture. The
mixture was then heated to 65 °C on a hot plate. Potassium
iodide (40.4 g, 240 mmol) and iodine (13.16 g, 103.7 mmol)
were dissolved in 350 cm3 of water at 50–55 °C, and the
mixture was then added dropwise, as quickly as it was
decolorized in order to minimize the gas evolution. The
endpoint was colored slightly purple. The contents of the
flask were cooled to 50 °C, and precipitated crude 11 was
collected by vacuum filtration. Solid 11 was dissolved in
CH2Cl2 and dried over sodium sulfate. Sodium sulfate was
removed by gravity filtration, and the solvent was removed
under reduced pressure (rotovap). The product was crystal-
lized from 2-propanol:water (50/50, vol/vol) to give 11 as a
gray solid. Yield: 14.8 g (65% from 15); mp 141–142 °C;
13C NMR ((CD3)2SO): d = 14.4 (q), 15.6 (q), 61.0 (t), 81.9
(s), 84.9 (s), 124.8 (s), 132.3 (s), 160.2 (s) ppm.
3,30-Bis-(3-carboxypropyl)-4,40-dimethyl-1H,10H-2,20-
bipyrrole-5,50-dicarboxylic acid (17, C20H24N2O8)
5-(Ethoxycarbonyl)-4-methyl-1H-pyrrole-2,3-dicarboxylic
acid (12, C10H11NO6)
A mixture of dimethyl ester 19 (0.99 g, 1.97 mmol), NaOH
(0.67 g, 0.017 mol), 95% ethanol (9 cm3) and H2O (3 cm3)
was heated at reflux for 3 h, after which the ethanol was
removed by distillation. To the cooled residue was added
10 cm3 of an aqueous solution of NaNO3 (6.5 g, 0.076
mol), and the contents were cooled at -10 °C in a dry ice/
acetone bath. Concentrated HNO3 (1.5 cm3) (cooled to
-20 °C) was added very slowly, with stirring, to keep the
temperature at -10 °C, and the mixture was stirred an
additional 30 min after the addition was complete. The
precipitate was collected by filtration, washed with cold
H2O and dried at 0.7 mm Hg in a dessicator over P2O5
overnight to afford a pink solid. Yield: 0.75 g (91%); mp
185.6 °C; 13C NMR ((CD3)2SO): d = 11.5, 19.6, 20.3, 35.0,
118.3, 121.8, 124.2, 124.6, 162.4, 174.9 ppm; HRMS
(FAB, glycerol): calcd for M ? H?, C20H25N2O8,
421.1611; found 421.1623.
3,5-Bis-(ethoxycarbonyl)-4-methyl-1H-pyrrole-2-carbox-
ylic acid, 13 [23] (9.9 g, 37.14 mmol) was carefully added
in portions to 100 cm3 of concentrated sulfuric acid at 100
°C and stirred for 20 min. The solution was then poured
into 700 cm3 of ice and water and stirred vigorously for 10
min. The precipitated crude product, known since 1925
[58], was then collected by vacuum filtration and used
directly in the preparation of 11. 13C NMR ((CD3)2SO): d =
10.78 (q), 13.9 (q), 60.6 (t), 118.8 (s), 122.2 (s), 126.9 (s),
127.0 (s), 160.8 (s), 161.6 (s), 165.1 (s) ppm.
3,5-Bis-(ethoxycarbonyl)-4-methyl-1H-pyrrole-2-carbox-
ylic acid (13, C12H15NO6)
2,4-Diethyl-3,5-dimethyl-1H-pyrrole-2,4-dicarboxylate (15)
(15.8 g, 66.3 mmol) was added to 150 cm3 of glacial acetic
acid in a three-neck round-bottom flask. The contents were
stirred mechanically while being cooled to 10 °C in an ice
bath. Bromine (3.5 cm3) was added to the solution all at
once, then freshly distilled sulfuryl chloride (16 cm3) was
added dropwise over 1.5 h. During the addition, care was
taken to ensure that the mixture was stirred vigorously (the
contents of the flask will precipitate massively during
addition). After all of the sulfuryl chloride was added, the
solution was allowed to reach room temperature while
being stirred for 4.5 h. Water (40 cm3) was then added
dropwise, and after the addition was complete, the solution
was heated on a hot water bath at 60 °C for 30 min. The
solution was then poured into 1,000 cm3 ice and water with
stirring. The precipitate was collected by vacuum filtration
Diethyl 3,30-bis-(3-methoxycarbonyl-1-propyl)-4,40-
dimethyl-2,20-1H,10H-bipyrrole-5,50-dicarboxylate
(19, C26H36N2O8)
Diethyl 3,30-bis-(3-methoxycarbonyl-1-propynyl)-4,40-
dimethyl-2,20-1H,10H-bipyrrole-5,50-dicarboxylate (20)
(0.36 g, 0.72 mmol) was dissolved in 12 cm3 methanol
and hydrogenated at room temperature and atmospheric
pressure over 10% PtO2 (0.016 g, 0.072 mmol) for 1.5 h.
The resulting solution was vacuum-filtered through a pad
of silica gel, and the solvent was evaporated to give crude
19, which was purified by radial chromatography (CH2Cl2,
eluent). Yield: 0.31 g (85%); mp 158–159 °C; 1H NMR: d
123

110
E. B. Nikitin et al.
14. Pfeiffer WP, Lightner DA (1994) Tetrahedron Lett 35:9673
15. Tu B, Ghosh B, Lightner DA (2003) J Org Chem 68:8950
16. Boiadjiev SE, Lightner DA (1994) Synlett 777
= 1.35 (6H, t, J = 7.5 Hz), 1.82 (4H, m), 2.29 (10H, m),
2.70 (4H, m), 3.56 (6H, s), 4.28 (4H, t, J = 7.6 Hz), 8.50
(2H, brs) ppm; HRMS (FAB, 3-NBA): calcd for
C26H36N2O8, 504.2472; found 504.2482.
¨
17. Falk H, Flodl H, Wagner WG (1988) Monatsh Chem 119:739
¨
18. Falk H, Flodl H (1988) Monatsh Chem 119:1155
19. Guilard R, Aukaloo MA, Tardieux C, Vogel E (1995) Synthesis
Diethyl 3,30-bis-(3-methoxycarbonyl-1-propynyl)-4,
40-dimethyl-2,20-1H,10H-bipyrrole-5,50-dicarboxylate
(20, C26H28N2O8)
1480
20. Skowronek P, Lightner DA (2003) Monatsh Chem 134:889
21. Jiao L, Hao E, Vicente MGH, Smith KM (2007) J Org Chem
72:8119
Ethyl 4,5-diiodo-3-methyl-1H-pyrrole-2-carboxylate, 11
(0.28 g, 0.7 mmol), PdCl2(PPh3)2 (0.05 g, 0.07 mmol),
CuI (0.03 g, 0.14 mmol) and methyl 3-butynoate [56] (0.27
g, 2.8 mmol) were dissolved in 15 cm3 freshly distilled
triethyl amine. The mixture was heated at 60 °C and stirred
under nitrogen overnight. After cooling, the solvent was
evaporated at reduced pressure, and the crude product was
redissolved in CH2Cl2 and vacuum filtered through a pad of
silica gel. The CH2Cl2 was evaporated (rotovap) to yield
crude 20, which was purified by radial chromatography
(CH2Cl2, eluent). Yield: 0.24 g (68%); mp 154–155 °C; 1H
NMR: d = 1.35 (6H, t, J = 7.6 Hz), 2.45 (6H, s), 3.67 (10H,
m), 4.28 (8H, m), 9.38 (2H, brs) ppm. The compound was
hydrogenated directly to 19.
¨
22. Falk H, Flodl H (1989) Monatsh Chem 120:45
23. Shrout DP, Lightner DA (1990) Synthesis 1062
24. Atkinson JH, Atkinson RS, Johnson AW (1964) J Chem Soc 5999
25. Byun YS, Lightner DA (1991) J Heterocycl Chem 28:1683
26. Xie M, Lightner DA (1993) Tetrahedron 49:2185
27. Bonnett R, Buckley DG, Hamzetash DJ (1981) J Chem Soc
Perkin Trans 1 322
28. Chen Q, Huggins MT, Lightner DA, Norona W, McDonagh AF
(1999) J Am Chem Soc 121:9253
29. Brower JO, Lightner DA, McDonagh AF (2000) Tetrahedron
56:7869
30. Brower JO, Lightner DA, McDonagh AF (2001) Tetrahedron
57:7813
31. Nelson MJ (2003) Toward optically active bipyrroles. MS thesis,
University of Nevada, Reno, NV
32. Hagio H, Sugiura M, Kobayashi S (2006) Org Lett 8:375
33. Nikitin EB, Nelson MJ, Lightner DA (2007) J Heterocycl Chem
44:739
Acknowledgments We thank the National Institutes of Health (HD-
17779) for generous support of this work and the National Science
Foundation (CHE-0521191) for funds to acquire the 400 MHz NMR
used in this work. Edward B. Nikitin was an RC Fuson Graduate
Fellow. We thank Professor Thomas Bell for use of the VPO
apparatus.
34. Sonogashira K, Tohda Y, Hagihara N (1975) Tetrahedron Lett
4467
35. Boiadjiev SE, Conley BA, Brower JO, McDonagh AF, Lightner
DA (2006) Monatsh Chem 137:1463
36. Sheldrick WS (1983) Isr J Chem 23:55
37. Sheldrick WS (1976) J Chem Soc Perkin 2:1457
38. Bonnett R, Davies JE, Hursthouse NB, Sheldrick GM (1978) Proc
R Soc Lond Ser B 202:249
39. LeBas G, Allegret A, Mauguen Y, DeRango C, Bailly M (1980)
Acta Crystallogr Sect B B36:3007
References
40. Becker W, Sheldrick WS (1978) Acta Crystallogr Sect B 34:1298
41. Lightner DA, Person RV, Peterson BR, Puzicha G, Pu Y-M,
Bojadziev S (1991) Biomolecular spectroscopy II. In: Birge RR,
Nafie LA (eds) Proceedings of SPIE, vol 1432. SPIE, Bellingham
WA, p 2
42. Shrout DP, Puzicha G, Lightner DA (1992) Synthesis 328
43. Trull FR, Person RV, Lightner DA (1997) J Chem Soc Perkin
Trans 2:1241
1. Falk H (1989) The chemistry of linear oligopyrroles and bile
pigments. Springer, Wien
2. Chowdhury JR, Wolkoff AW, Chowdhury NR, Arias IM (2001)
Hereditary jaundice and disorders of bilirubin metabolism. In:
Scriver CF, Beaudet AL, Sly WS, Valle D (eds) The metabolic
and molecular bases of inherited disease. McGraw-Hill, New
York, Chap 125, pp 3063–3101
3. Huggins MT, Lightner DA (2000) J Org Chem 65:6001
4. Huggins MT, Lightner DA (2001) Tetrahedron 57:2279
5. Boiadjiev SE, Anstine DT, Lightner DA (1995) J Am Chem Soc
117:8727
6. Boiadjiev SE, Anstine DT, Maverick E, Lightner DA (1995)
Tetrahedron Asymm 6:2253
7. Lightner DA, McDonagh AF (2001) J Perinatol 21:S13
8. McDonagh AF, Lightner DA (1994) Cell Mol Biol 40:965
9. McDonagh AF, Lightner DA (1991) The importance of molecular
structure in bilirubin metabolism and excretion. In: Bock KW,
Gerok W, Matern S (eds) Hepatic metabolism and disposition of
endo and xenobiotics (Falk Symp no. 57). Kluwer, Dordrecht,
Chap 5, pp 47–59
10. Person RV, Peterson BR, Lightner DA (1994) J Am Chem Soc
116:42
11. Ghosh B, Lightner DA, McDonagh AF (2004) Monatsh Chem
135:1189
12. Ghosh B, Lightner DA (2003) J Heterocycl Chem 40:1113
13. Tipton AK, Lightner DA, McDonagh AF (2001) J Org Chem
66:1832
44. Boiadjiev SE, Lightner DA (1997) Chirality 9:604
¨
45. Dorner T, Knipp B, Lightner DA (1997) Tetrahedron 53:2697
46. Boiadjiev SE, Lightner DA (2003) J Org Chem 68:7591
47. Kar A, Lightner DA (1998) Tetrahedron 54:12671
48. Boiadjiev S, Lightner DA (1998) J Org Chem 63:6220
49. Boiadjiev SE, Holmes DL, Anstine DT, Lightner DA (1995)
Tetrahedron 51:10663
50. Boiadjiev SE, Person RV, Puzicha G, Knobler C, Maverick E,
Trueblood KN, Lightner DA (1992) J Am Chem Soc 114:10123
51. Tu B, Ghosh B, Lightner DA (2004) Monatsh Chem 135:519
52. Huggins MT, Lightner DA (2000) Tetrahedron 56:1797
53. Chen Q, Lightner DA (1998) J Org Chem 63:2665
54. Brower JO, Huggins MT, Boiadjiev SE, Lightner DA (2000)
Monatsh Chem 131:1047
´
55. Lightner DA, Gawronski JK, Wijekoon WMD (1987) J Am
Chem Soc 109:6354
56. Kasha M, El-Bayoumi MA, Rhodes W (1961) J Chim Phys Chim
Biol 58:916
57. Fischer H (1935) Org Synth 14:17
58. Fischer H, Walach B (1925) Chem Ber 58:2818
123