Carboxylic acid to thioamide hydrogen bonding

Article abstract of DOI:10.1016/j.tet.2008.09.082

The lactam groups of dipyrrinones avidly engage in amide-amide hydrogen bonding to form dimeric association complexes in non-polar solvents (in CHCl₃, K_D ～25,000 M^-1 at 22 °C). The corresponding thioamides (dipyrrinthiones

Full text of DOI:10.1016/j.tet.2008.09.082

Tetrahedron 65 (2009) 77–82
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Carboxylic acid to thioamide hydrogen bonding
*
Suchitra Datta, David A. Lightner
Department of Chemistry, University of Nevada, Reno, NV 89557, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The lactam groups of dipyrrinones avidly engage in amide–amide hydrogen bonding to form dimeric
association complexes in non-polar solvents (in CHCl3, KD w25,000 Mꢀ1 at 22 ꢁC). The corresponding
thioamides (dipyrrinthiones), prepared from dipyrrinones by reaction with Lawesson’s reagent, also form
intermolecularly hydrogen-bonded dimers in non-polar solvents, albeit with much weaker association
constants (in CHCl3, KD w200 Mꢀ1 at 22 ꢁC). When a carboxylic acid group is tethered to C(9) of the
dipyrrinone, as in the hexanoic acid of [6]-semirubin, tight intramolecular hydrogen bonding between
the carboxylic acid group and the lactam moiety (intramolecular Kassoc [25,000) is found in CHCl3 with
no evidence of dimers. In contrast, the analogous dipyrrinthione, [6]-thiosemirubin, eschews intra-
molecular hydrogen bonds, as determined using NMR spectroscopy and vapor pressure osmometry,
preferring to form intermolecularly hydrogen-bonded dimers of the thioamide–thioamide type.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 21 August 2008
Received in revised form
24 September 2008
Accepted 26 September 2008
Available online 8 October 2008
Keywords:
Pyrroles
Dipyrrinthiones
NMR
VPO
1. Introduction
synthetic pigment model for one-half of bilirubin, [6]-semirubin (1)
(Fig. 1B),8 whose dipyrrinones, like those in bilirubin, have the
Although not as well studied as amide–amide and carboxylic
acid–carboxylic acid hydrogen bonding,1 amides are known to form
strong hydrogen bonds to carboxylic acids, as was first recognized
in bilirubin (Fig. 1A),2,3 the yellow pigment of jaundice.4 There,
nature designed a molecule whose lipophilic character and poor
aqueous solubility (Ksp w4ꢂ10ꢀ15 M at 37 ꢁC)5 may be understood
by considering pigment’s structure in which both propionic acids
are tucked inward and firmly intramolecularly hydrogen-bonded to
the opposing dipyrrinones,2,3,6 the key factor in bilirubin’s aqueous
solubility and metabolism. In further illustration of this attraction,
strong, selective intermolecular hydrogen bonding between
a carboxylic acid and a pyridone lactam has also been reported,7 to
the exclusion of carboxylic acid–carboxylic acid and amide–amide
hydrogen bonding. The unusual type of hydrogen bonding seen in
bilirubin is probably facilitated by the adjacency of the pyrrole
hydrogen and has been replicated in simpler analogs such as
[6]-semirubin (Fig. 1B)8 and related analogs where recent studies
have shown that the dipyrrinone moiety9 is an excellent receptor
for a carboxylic acid group10,11 and probably also a carboxylate
anion.3c,6b
syn-Z configuration,3,6,8 and which was shown to be a monomer in
chloroform with its CO2H group engaged tightly in intramolecular
hydrogen bonding (Fig 1A,B).8 In contrast, the ethyl ester (2) of 1
and simple dipyrrinones with an alkyl group at C(9) replacing the
carboxylic acid chain, e.g., kryptopyrromethenone (3), were shown
to form strongly intermolecularly hydrogen-bonded dimeric
structures (Fig. 1C) in CDCl3 with large association constants (KD
w25,000 Mꢀ1 at 22 ꢁC)12 d surprisingly, much larger than those of
simple amides, whose KD varies from w30 to 7000.13 The obser-
vation that [6]-semirubin (1) eschews dipyrrinone to dipyrrinone
dimer formation8 is consistent with the dipyrrinone unit strongly
favoring (Kassoc [25,000) complexation to the carboxylic acid
group over a second dipyrrinone molecule. As such,1 is an excellent
model for exploring amide to carboxylic hydrogen bonding.
Though less well studied, thioamides, like amides, are known
to dimerize in non-polar solvents by forming intermolecular
hydrogen bonds. The simple thiolactams,
g-butyrothiolactam and
d-valerothiolactam, were shown to exhibit dimerization equili-
brium constants (KD¼278 and 438 Mꢀ1 in CCl4 at 25 ꢁC)13a,b similar
to the parent lactams.13a,b But the thiolactam analogs of 2-pyridone
were found to have variable, but always lower KD values than the
parent: 4100 Mꢀ1 versus 7100 Mꢀ1 in CCl4,13c 28 versus 2200 Mꢀ1
in benzene,13d and 2.7ꢃ1.0ꢂ103 Mꢀ1 versus 2.5ꢃ1.0ꢂ104 Mꢀ1 in
CHCl3.14 To the best of our knowledge, hydrogen bonding between
a thioamide and a carboxylic acid remains unknown.
The essential components for the hydrogen bonding seen in
bilirubin are an amide (lactam) of a Z-configuration dipyrrinone
and a carboxylic acid chain of at least six carbons that is tethered at
C(9). This complementary combination was designed into the
In order to explore and evaluate the potential for such hydrogen
bonding, we took advantage of the special ability of dipyrrinones to
sequester a carboxylic acid group. Using [6]-semirubin (1) as
* Corresponding author. Tel.: þ1 775 784 4980; fax: þ1 775 784 6804.
E-mail address: lightner@scs.unr.edu (D.A. Lightner).
0040-4020/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2008.09.082

78
S. Datta, D.A. Lightner / Tetrahedron 65 (2009) 77–82
1 equiv of Lawesson’s reagent for 2 h. Under similar reaction con-
A
Z
ditions, the known methyl xanthobilirubinate (7)17 was converted
smoothly to the thioamide analog (5) in 83% yield. Synthesis of 4
and 8 (Scheme 1) required the known [6]-semirubin (1)7 and its
methyl ester (7),7 both obtained following Friedel–Crafts acylation
at carbon-9 of 2,3,7,8-tetramethyl-10H-dipyrrinone.8 Unlike the
formation of thiolactam 5 from 7, conversion of the amide of [6]-
semirubin methyl ester (2) to its thioamide (8) with Lawesson’s
reagent proceeded inefficiently (32% yield). The reason for this is
unclear and, disappointingly, methyl ester 8 resisted conversion to
the parent free acid (4). Numerous attempts at converting 8 to 4 led
to decomposition, as did direct thionation of dipyrrinone acid 1 to
4. In order to circumvent these problems, especially the difficulties
encountered in the methyl ester hydrolysis step (8/4), [6]-semi-
rubin (1) was converted to its more easily deprotected tert-butyl-
diphenylsilyl ester (9) in good yield. Successful thionation of 9
using Lawesson’s reagent, followed by fluoride ion deprotection of
the silyl ester group, afforded the desired [6]-thiosemirubin (4) in
32% isolated yield from 9.
C
O
O
2
H
O
N
N
H
H
9
CH2
O
O
H H
N
N
H
O
C
Z
Bilirubin
Z
B
2
N
N
H
H
9
CH2
O
O
H
O
C
[6]-Semirubin (1)
C
Z
2
N
N
H H
9
O
3
O
2
9
H
H
N
N
3
N
N
HH
Z
O
(CH2)5CO2H
1
a (71%) b (78%)
Kryptopyrromethenone Dimer
Figure 1. (A) Bilirubin, (B) its intramolecularly hydrogen-bonded dipyrrinone analog
[6]-semirubin (1), and (C) intermolecularly hydrogen-bonded kryptopyrromethenone (3).
N
N
N
H H
O
(CH2)5CO2R
a model, we perturbed the dipyrrinone by converting its lactam
moiety to thiolactam to give the dipyrrinthione, [6]-thiosemirubin
(4). For comparison and evaluation of thiolactam–thiolactam
association involving hydrogen bonding, we prepared the dipyr-
rinthiones, methyl thioxanthobilirubinate (5),15 and kryptopyrro-
methenethione (6)15 (Fig. 2), from which we could further
evaluate the thioamide–thioamide hydrogen bonding and whose
KD values might serve as a reference value to be exceeded if
intramolecular hydrogen bonding prevails in 4. In the following, we
describe the syntheses of 4, its methyl ester (8), and dipyrrinthiones
5 and 6 (Fig. 2), and we analyze their ability to engage in hydrogen
bonding in terms of inter- and intramolecular hydrogen-bonded
structures.
2: R=CH3
9: R=SiPh2t-Bu
c,d
c (32%)
(32%)
N
HH
S
(CH2)5CO2R
4: R=H
8: R=CH3
Scheme 1. at-BuPh2SiClþimidazole/dry THF, rt, 4 days; bCH3OH/H2SO4; cLawesson’s
reagent/dry THF, rt, 36 h; d(n-Bu)4NF/THF to 5% H2O, 12 h, rt.
2.2. Constitutional structures from NMR
2. Results and discussion
2.1. Synthesis
The constitutional structures of 4 and 8 follow from the struc-
tures of their precursors (1 and 2 of Fig. 1 and Scheme 1) and from
their 13C NMR spectra (Table 1). Thus, evidence for the loss of the
lactam C]O at w174 ppm and the appearance of the C]S carbon at
w187 ppm correlates with the C]S chemical shifts of dipyrrin-
thiones 5 and 615 and the expectation of a more strongly deshielded
carbon. Likewise, C(2) is deshielded to w132 ppm in 4 and 8 (and 5
and 6) relative to the lactam parents’ w123 ppm. Similar, but
smaller deshieldings at C(4) and C(5) are found in 4 and 8 (5 and 6)
relative to their parent dipyrrinones. Interestingly, the usual small
(0.1 ppm) deshielding of C(5) in the acid relative to the ester (1 vs 2)
is much smaller than the (w0.6 ppm) shielding in thiosemirubins 4
and 8.
The 4Z-configuration of dipyrrinones (Figs. 1 and 2, and
Schemes), as well as bilirubin, has been long established:2 in the
solid by X-ray crystallography;3,18,19 in solution by nuclear Over-
hauser effect (NOE) spectroscopy, where NOEs are found between
C(5)–H and the C(3) and C(7) CH3 groups, and between the two
NHs.2,18,20 Thus, it came as no surprise that NOE studies of 4, 6, and
8 indicated the favored syn-Z configuration (Fig. 3). E-Configuration
dipyrrinones are accessible only by photoirradiation.21
The syntheses of 4–6 and 8 involve thionation of the lactam
units of the corresponding dipyrrinones using Lawesson’s re-
agent.16 We had previously converted kryptopyrromethenone 3 to
its thioamide analog (6) in 82% yield15 by treatment in dry THF with
N
N
R
H H
N
N
HH
X
(CH2)5CO2R
X
X
O
S
S
O
X
R
R
1 :
2 :
4 :
8 :
O
O
S
S
H
CH3
H
3 :
5 :
6 :
7 :
CH2CH3
(CH2)2CO2CH3
CH2CH3
(CH2)2CO2CH3
CH3
Figure 2. The thiolactam analogs (4–6 and 8) of [6]-semirubin (1) and its methyl ester
(2), methyl xanthobilirubinate (7), and kryptopyrromethenone (3).

S. Datta, D.A. Lightner / Tetrahedron 65 (2009) 77–82
79
Table 1
Table 2
a
13C NMR chemical shifts (
d)
and assignments of 1–8 in CDCl3
Comparison of the pyrrole, lactam, and carboxylic acid 1H NMR chemical shiftsa of
1–8 in CDCl3 and (CD3)2SOb
X
R1
R2
R3
Compound
Lactam NH
Pyrrole NH
CO2H
1
2
3
4
5
6
7
8
O
O
O
S
S
S
O
S
(CH2)5CO2H
CH3
CH3
R3
5
1
10.42c
8.98c
13.22c
(CH2)5CO2CH3 CH3
CH2CH3
CH2CH3
CH3
7
4
6
3
9.81
10.12
11.95
CH3
CH2CH3
CH3
(CH2)2CO2CH3 CH2CH3
CH2CH3 CH2CH3
(CH2)2CO2CH3 CH2CH3
CH3
2
8
R2
(CH2)5CO2H
CH3
N
N
2
3
4
5
6
7
11.17c
9.81
10.11c
10.12
d
d
HH
1
9
R1
X
CH3
CH3
11.10
9.83
10.05
10.33
d
d
(CH2)5CO2CH3 CH3
11.05d
9.55d
13.06d
Carbon
1
2
3
4
5
6
7b
8
11.92
10.98
12.62
1
C]CX
]C–
CH3
]C–
CH2/CH3
CH3
]C–
]CH–
]C–
]C–
CH3
174.7 173.9 174.1 186.8 186.7 185.9 174.1 186.7
123.6 123.2 122.2 132.2 131.5 131.0 122.4 133.5
8.2
142.2 142.2 148.2 138.0 146.1 146.0 148.4 140.1
9.8
11.47
11.44
9.95
10.94
d
d
2
21
3
8.6
9.5
11.1
10.4
10.4
9.6
9.6
11.41e
9.94e
d
31
32
4
9.9
d
15.4
17.5
12.6
d
18.4
15.1
18.4
17.3
18.0
15.0
11.3
d
11.43
10.91
d
11.15
9.72
10.25
10.26
d
d
128.7 128.3 127.0 131.4 132.3 132.0 127.2 133.2
101.3 101.2 101.2 103.4 103.9 104.2 101.4 104.0
122.4 122.2 122.9 122.9 129.3 129.8 122.4 124.1
125.5 125.2 122.2 122.9 121.3 123.5 119.1 130.5
5
6
8
11.26f
11.59
9.71f
10.91
d
d
7
71
8
9.7
9.7
11.5
12.3
11.9
11.9
11.5
11.0
a
d
ppm downfield from (CH3)4Si for 3ꢂ10ꢀ3 M solutions in CDCl3 and (CD3)2SO at
]C–
135.8 135.8 124.6 138.0 123.6 123.5 124.6 140.6
8.8
25 ꢁC.
81
82
83
9
CH2/CH3
CH2/CH3
CO2R
9.1
d
18.0
15.1
d
12.0
d
19.7
34.7
173.5
18.0
15.1
d
19.9
35.2
174.1
10.6
d
b
The data from (CD3)2SO are shown in italics, concentration¼4ꢂ10ꢀ2 M.
d
c
Concentration¼1ꢂ10ꢀ3 M.
d
d
d
d
d
e
Concentration¼8ꢂ10ꢀ3 M at 3ꢂ10ꢀ2 M, dlactam¼11.77 and dpyrrole¼9.88 ppm.
Concentration¼1.33ꢂ10ꢀ2 M; at 3.09ꢂ10ꢀ3 M, dlactam¼10.27 and dpyrrole¼9.06
]C–
116.7 115.9 131.1 117.0 136.1 136.0 131.6 118.8
23.0 25.7 8.5 26.1 9.8 9.7 8.5 26.7
91
CH2/CH3
92–95 CH2
d
d
d
d
c
d
e
f
ppm.
f
Concentration¼1ꢂ10ꢀ2 M.
96
CO2H/CO2CH3 180.8 174.2
CO2CH3 51.4
d
d
178.7
d
d
d
d
d
174.2
52.4
d
51.7
51.2
a
In parts per million downfield from (CH3)4Si for 2ꢂ10ꢀ2 M solutions in (CD3)2SO
With dipyrrinthiones 5 and 6 as reference standards for in-
vestigating the pyrrole and lactam NH chemical shift, we note that
in CDCl3, consistent with those of the parent dipyrrinones (7 and 3),
the lactam NH (w11.5 ppm) is more deshielded than the pyrrole NH
(w10 ppm). And in fact the thiolactam and pyrrole NH chemical
shifts (11.47 and 11.41, and 9.95 and 9.94 ppm) of 5 and 6 are not
very different from the lactam (11.15 and 11.10 ppm) and pyrrole
(10.02 and 10.05 ppm) NH chemical shifts of 3 and 7. Remarkably,
the thiolactam NH chemical shifts of 5 and 6 are nearly identical in
CDCl3 and (CD3)2SO, which is not seen in the parent lactams, 3 and
7, yet the pyrrole NH chemical shifts of 5 and 6 in CDCl3 are shielded
by w1 ppm from those in (CD3)2SOda much larger difference than
that found in 3 and 7 (0.01–0.28 ppm). The reasons for this are
unclear. Comparing 4 and 8 to 5 and 6, we note that the NH
chemical shifts of both 4 and 8 in CDCl3 and (CD3)2SO are similar to
those of 5 and 6 in the corresponding solvent. These data suggest
similar environments for the dipyrrinthiones in (CD3)2SO, as is
typically found for this solvent. Unexpectedly, the data also suggest
similar environments in CDCl3da solvent where intramolecular
hydrogen bonding between carboxylic acid and dipyrrinthione
might have been expected in 4, and dipyrrinthione to dipyrrin-
thione hydrogen-bonded dimer formation expected only for 5, 6,
and 8danalogous to 1 versus 2, 3, and 7. However, these data alone
do not disprove dipyrrinthione to CO2H hydrogen bonding in 4.
at 25 ꢁC.
b
Values taken from Ref. 15.
c
d
e
f
R1: 21.4 (92dCH2), 26.2 (93dCH2), 28.2 (94dCH2), 32.6 (95dCH2) ppm.
R1: 24.8 (92dCH2), 28.7 (93dCH2), 29.9 (94dCH2), 34.0 (95dCH2) ppm.
R1: 27.6 (92dCH2), 30.3 (93dCH2), 30.9 (94dCH2), 35.9 (95dCH2) ppm.
R1: 25.1 (92dCH2), 29.2 (93dCH2), 30.0 (94dCH2), 24.5 (95dCH2) ppm.
2.3. Hydrogen bonding
The NH chemical shifts of dipyrrinones have frequently been
used in 1H NMR to detect hydrogen bonding: intermolecular
between two dipyrrinones20 and between a dipyrrinone and
a carboxylic acid.6,8–10 It seemed reasonable to assume that this
structure-distinguishing probe might carry over to dipyrrinthiones.
Dipyrrinones that engage exclusively in intermolecular hydrogen
bonding of the dipyrrinone to dipyrrinone type in CDCl3, as in 2, 3,
and 7 (Table 2), show lactam NH resonances near 11 ppm and
pyrrole NHs near 10 ppm whereas, with the presence of a carbo-
xylic acid group, as in [6]-semirubin, the pyrrole NH moves upfield
by w1 ppm (to w9 ppm), and the lactam NH also moves upfield (to
w10.4 ppm). In (CD3)2SO a leveling effect is exerted and the lactam
NH signals of 1–3 and 7 are more shielded (to w9.8 ppm) than the
pyrrole NH (w10.2 ppm).
R1
H
2.4. NOE measurements
R1
One indicator of dimer formation lies with NOE data from 1H
NMR spectroscopy in CDCl3. Thus, when dipyrrinones form
intermolecularly hydrogen-bonded dimers, an NOE is observed
between the C(21) hydrogens and those at C(91), as has been noted
for 3, 7, and even 1. In contrast, when the dipyrrinone sequesters
a CO2H group through hydrogen bonds, an NOE is observed be-
tween the CO2H hydrogen and the dipyrrinone lactam NHdas is
seen in [6]-semirubin (1), bilirubin, etc. For dipyrrinthiones 5, 6,
and 8, we observe NOEs (Fig. 3) of the former type. For 4, we
also see the former type and not the latter. These data suggest
N
N
HH
S
R2
R2
S
R1
H
H
N
N
R1
H
Figure 3. Key NOEs observed for
4
(R1¼H, R2¼CH2CH2CH2CH2CO2H),
8
(R1¼H,
R2¼CH2CH2CH2CH2CO2CH3), and 6 (R1¼CH3, R2¼H) in CDCl3 are shown by double-
headed arrows. The dashed arrows indicate weak NOEs. No NOEs were observed
between the CO2H and the lactam or pyrrole NH.

80
S. Datta, D.A. Lightner / Tetrahedron 65 (2009) 77–82
Table 3
that all of the dipyrrinthiones of this study form predominantly
intermolecularly hydrogen-bonded dimers in CDCl3. Again, how-
ever, although the failure to find a CO2H to dipyrrinthione NOE does
not support carboxylic acid to dipyrrinthione hydrogen bonding, it
does exclude its possibility.
Molecular weights of dipyrrinthiones and dipyrrinones 1–8 determined by vapor
pressure osmometrya at 45 ꢁC in CHCl3 solution
Dipyrrinthione
Formula weight
(g/mol)
Molecular weight
by VPO (g/mol)
Concentration
range (mol/kg)
1
2
3
4
5
6
6
7
8
330
344
258
346
384
274
274
316
360
337 ꢃ 20b
584 ꢃ 30b
509ꢃ 20c
689 ꢃ 7
2.1–6.6ꢂ10ꢀ3
1.7–6.1ꢂ10ꢀ3
2.0–12.8ꢂ10ꢀ3
1.0–6.5ꢂ10ꢀ3
0.72–2.2ꢂ10ꢀ3
4–7ꢂ10ꢀ3
2.5. Beer’s law behavior
654 ꢃ 20
535 ꢃ 12
416 ꢃ 21
579 ꢃ 25c
700 ꢃ 30
Further evidence for a monomer$dimer equilibrium may be
found in the Beer’s law plots (Fig. 4) of the dipyrrinthiones. The
parent dipyrrinone, [6]-semirubin (1), was found to obey Beer’s
law, whereas its ester (2) diverges.8 Just as dipyrrinones 2, 3 and 7
(but not 1) diverge from ideality with increasing concentration in
CHCl3,7 so do their thiolactam analogs 5, 6, and 8das might be
expected. What was not originally expected was that 4, like 1,
would not diverge, yet, its Beer’s law behavior, like that of 8, is
found to be consistent with dimer formation.
4–7ꢂ10ꢀ4
1.5–6.5ꢂ10ꢀ3
1.6–5.8ꢂ10ꢀ3
a
Calibrated with benzil (FW¼210, measured MW¼208ꢃ5).
b
c
Data from Ref. 7.
Data from Ref. 15.
unexpected observation that 4, unlike 1, also was a dimer in
CHCl3dat least at the concentrations of the VPO measurement.
2.6. VPO studies of self-association
2.7. Dimeric association
In order to establish the presence of dimers in non-polar sol-
vents, we used vapor pressure osmometry (VPO) to analyze the
molecular weights of 4–6 and 8 in CHCl3 solution (Table 3). Pre-
vious studies showed that dipyrrinones such as 3 and 7 were
dimeric, that [6]-semirubin ester (2) was also a dimer, but
[6]-semirubin (1) itself was a monomer. The data were part of the
argument in favor of an intramolecularly hydrogen-bonded struc-
ture for 1.8 When VPO was applied to dipyrrinthione esters 5, 6, and
8, we found that these compounds were dimers in CDCl3, consistent
with the behavior of their dipyrrinone ester parents (2, 3, and 7),
and showing that their sulfur analogs were capable of in-
termolecular hydrogen bonding. Just how strong this preference
might be remained to be determined. Especially revealing was the
In an earlier study of dimerization of 3 and 7, we determined the
dimeric association constant (KD) in CDCl3 from the concentration
dependence of the 1H NMR chemical shifts of the lactam and pyr-
role NHs. The KD values were large (KD w25,000 M at 25 ꢁC). Using
the same method, in order to compare the tightness of hydrogen
bonding in dipyrrinones with their thiolactam analogs, we de-
termined KD for 5 and 6 in CDCl3 and found the pyrrole and thio-
lactam NH chemical shifts to be more concentration-sensitive than
the pyrrole and lactam NHs of the parent dipyrrinones, 3 and 7. For
example, the thiolactam (dL) and pyrrole (dP) chemical shifts were
observed at 11.41 and 9.94 ppm, respectively, at 1.33ꢂ10ꢀ2 M
concentration of 6, and at 10.27 and 9.06 ppm, respectively, at
2.06ꢂ10ꢀ3 Mda difference (Dd) of 1.14 and 0.88 ppm, respectively.
In contrast, dL and dP were observed at 11.15 and 10.25 ppm,
respectively, for 1.41ꢂ10ꢀ2 M of 3 and at 10.84 and 10.04 ppm, re-
spectively, at 2.10ꢂ10ꢀ3 Mda Dd of 0.31 and 0.21 ppm, respectively.
This was a tip-off that the KD of 6, for example, might be
substantially lower than that of 3. This was realized from the KD
determinations, using the method described previously for 3 and
7,10 where the upper plateau (Fig. 5, upper) and lower foot (Fig. 5,
lower) of plots of NH chemical shifts versus pigment concentration
give dD and dM for dimer and monomer, respectively, for both the
thiolactam and pyrrole NHs. Using the equations developed earlier
70
A
Experimental Absorbance
60
50
40
30
20
10
0
of Thiosemirubin acid
Theoretical Absorbance of
Thiosemirubin acid
that relate KD to dD, dM, dobserved, and initial concentration of pig-
0
0.0005
0.001
0.0015
0.002
0.0025
ment, [MO], one extrapolates to KD values that are a factor of 100
smaller than those of the dipyrrinones (Table 4)da clear indication
that the thiolactam unit, unlike the lactam, in cooperation with the
pyrrole behaves more or less like an ordinary thiolactam. The rel-
ative failing in this regard of the dipyrrinthione may be due to the
longer C]S (vs C]O) bond that (in the dimer) increases the pyrrole
NH/S]C distance relative to the NH/O]C distance.
Concentration
70
60
50
40
30
20
10
0
B
Experimental Absorbance of
Thiosemirubin methyl ester
Theoretical Absorbance of
Thiosemirubin methyl ester
Although the 1H NMR study of the concentration dependence of
the pyrrole and lactam chemical shifts in CDCl3, VPO measure-
ments, and Beer’s law plots of [6]-semirubin (1) differed uniquely
from its methyl ester (2) as well as dipyrrinones 3 and 7, we failed to
detect a similar difference in its thiolactam derivative 4 relative to
thiolactams 5, 6, and 8. For 1, VPO indicated monomers in CHCl3; its
NH chemical shifts were invariant over the concentration range and
Beer’s law was obeyeddall of which are consistent with an intra-
molecular hydrogen-bonded monomer: dipyrrinone to CO2H. We
expected 4 to behave like 1 in these studies, but its behavior was
much more like its ester (8), 5 and 6, which may be considered to be
intermolecularly hydrogen-bonded dimers in chloroform.
0
0.0004
0.0008
Concentration
0.0012
0.0016
Figure 4. Beer’s law plot for (A) thiosemirubin (4) and (B) its methyl ester (8) showing
deviation from ideality. The actual absorbance measured for the highest concen-
tration¼2.0 (see Section 4). For the contrasting plots of 1, which obeys Beer’s law, and
2, which does not, see Ref. 8.

S. Datta, D.A. Lightner / Tetrahedron 65 (2009) 77–82
81
A
B
13
12
11
10
9
13
Lactam N-H
Pyrrole N-H
Lactam NH
Pyrrole NH
12
11
10
9
8
8
7
7
-0.01
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1
Concentration [M]
-0.01 0.01 0.03 0.05 0.07 0.09 0.11 0.13 0.15 0.17 0.19
Concentration [M]
0
11
10.5
10
9.5
9
8.5
8
7.5
7
25 C Lactam N-H
25 C Pyrrole N-H
56 C Lactam N-H
56 C pyrrole N-H
11
25 C Lactam N-H
25 C Pyrrole N-H
56 C Lactam N-H
56 C pyrrole N-H
10.5
10
9.5
9
8.5
8
7.5
-4.5
-4
-3.5
-3
-2.5
-2
-4
-3.5
-3
Log [C]
-2.5
-2
Log [C]
Figure 5. (A) Plots of pyrrole and lactam 1H NMR chemical shifts versus concentration in CDCl3 of thiosemirubin (4) (A, upper) and methyl thioxanthobilirubinate (5) (B, upper), and
comparison of data from the lowconcentration of 4 (A, lower) and 8 (B, lower) plotted as NH chemical shift versus log concentration at 25 and 56 ꢁC. Similar plots were found for 6 and 8.
2.8. Solution properties and UV–vis spectra
3. Concluding comments
Dipyrrinthiones (4–6 and 8) are red solids that form reddish
solutions with long-wavelength UV–vis absorption bands centered
near 490 nm. In contrast, the parent dipyrrinones are yellow with
typical intense long-wavelength UV–vis absorption near 410 nm.
Although bathochromically shifted, the absorption bands of the
thiolactam analogs have molar absorptivity constants nearly un-
changed from those of the parents, and their solutions are more
noticeably solvatochromic (Table 5) than those of the parent
dipyrrinones.
[6]-Thiosemirubin 4 is a dimer in CHCl3, as measured by VPO. Its
pyrrole and lactam NH chemical shifts are concentration-dependent
and it fails to obey Beer’s law. We conclude that the thiolactam unit of
4, and indeed the entire dipyrrinthione, offers no special attraction to
the carboxylic acid that the unique ability of a dipyrrinone to se-
quester a CO2H group through hydrogen bonding is lost in dipyrrin-
thiones. Since the available evidence suggests a weaker KD (w300 M)
for the monomer$dimer equilibrium in dipyrrinthiones relative to
dipyrrinones (KD w25,000 M), we predict that any hydrogen bonding
between the dipyrrinthione and CO2H units of 4 would have an in-
termolecular association constant (Kassoc) <w200–300 M.
Table 4
Extrapolated dimerization association constants (KD, M) and dimer (dD, ppm) and
monomer (dM, ppm) NH chemical shifts of dipyrrinthiones (4–6 and 8) at 25 ꢁ
compared with dipyrrinones 3, 7, and 8 at 25 ꢁC
C
4. Experimental section
4.1. General procedures
3
4
5
6
7
8
a
KD
KD
22,000
23,000
11.42
10.44
7.75
160
103
11.78
9.98
8.21
7.61
315
292
11.78
10.17
8.65
7.81
238
198
11.83
10.21
8.65
7.73
28,600
24,800
11.39
10.43
7.75
208
162
11.52
10.02
8.25
b
dDc (HN–C]X)
dDd (pyrrole)
dMc (HN–C]X)
dMd (pyrrole)
Nuclear magnetic resonance (NMR) spectra, NOEs, and T1 mea-
surements were obtained on a Varian Unity Plus 500 MHz spec-
trometer in CDCl3 solvent (unless otherwise specified) at 25 ꢁC.
8.10
8.10
7.52
Chemical shifts were reported in
d ppm referenced to the residual
a
From thiolactam or lactam NH data.
From pyrrole NH data.
Thiolactam (X¼S) or lactam (X¼O) NH NMR chemical shift of monomer (dM) and
CHCl3 1H signal at 7.26 ppm and 13C signal at 77.0 ppm. A combi-
nation of heteronuclear multiple bond correlation (HMBC) spectra
and 1H{1H} NOE data were used to assign 1H and 13C NMR spectra.
UV–vis spectra were recorded on a Perkin–Elmer Lambda-12 spec-
trophotometer. Melting points were taken on a Mel Temp capillary
apparatus and are corrected. Combustion analyses were carried out
by Desert Analytics, Tucson, AZ. High resolution mass spectra were
determined at the Nebraska Center for Mass Spectroscopy, Univer-
sity of Nebraska, Lincoln. Analytical thin layer chromatography was
b
c
dimer (dD).
d
Pyrrole NH NMR chemical shift of monomer (dM) and dimer (dD).
Table 5
Comparison of UV–vis spectral data of 1–8a
Pigment 3max max, nm)
(l
C6H6
CHCl3
CH3CN
CH3OH
(CH3)2SO
carried out on J.T. Baker silica gel IB-F plates (125 mm layers). Radial
1
2
3
4
27,200 (426) 28,200 (421) 26,600 (411) 31,500 (416) 30,400 (413)
33,200 (411) 28,800 (407) 33,600 (413) 33,500 (415) 33,200 (411)
36,100 (412) 32,200 (408) 32,000 (406) 39,400 (417) 35,600 (415)
28,400 (492) 29,100 (495) 28,000 (490) 29,400 (490) 28,300 (493)
26,200 (513)sh 27,800 (515)sh
26,200 (484) 32,500 (489) 28,500 (480) 32,000 (481) 31,100 (482)
34,800 (484) 32,200 (493) 31,200 (484) 34,500 (485) 33,700 (484)
26,500 (412) 34,000 (408) 28,900 (400) 37,700 (411) 34,600 (410)
38,500 (487) 38,300 (492) 42,350 (485) 46,000 (488) 42,600 (494)
chromatography was carried out on Merck silica gel PF254 with
gypsum preparative layer grade, using a Chromatotron (Harrison
Research, Palo Alto, CA). Spectral data were obtained in spectral
grade solvents (Aldrich or Fisher). THF used was distilled from Na/
benzophenone. Deuterated chloroform and dimethylsulfoxide were
from Cambridge Isotope Laboratories. 2,3,7,8-Tetramethyl-(10H)-
dipyrrin-1-one,8,10 [6]-semirubin (1),8 [6]-semirubin methyl ester
(2),8 kryptopyrromethenone (3),22 kryptopyrromethenethione
5
6
7
8
Solutions were w1ꢂ10ꢀ5 M at 22 ꢁC;
3 .
in L mol cmꢀ1
a

82
S. Datta, D.A. Lightner / Tetrahedron 65 (2009) 77–82
(6),15 methyl xanthobilirubinate (7),17 and methyl thioxanthobilir-
ubinate (5)15 were prepared as described in the literature.
Beer’s law measurements were carried out as follows:8 (1)
(rotovap), the residue was dissolved in CH2Cl2, and the solution was
passed through a short column of silica gel (filtration chromato-
graphy) using CH2Cl2 as eluent. The solutionwas evaporated and the
residue was purified by radial chromatography on silica gel using
CH2Cl2 as eluent. The deep red-colored solid (4) (32% yield), mp 215–
217 ꢁC, was obtained after evaporation of CH2Cl2. It was stored in the
measure a standard UV–vis spectrum and calculate 3 l
max at max. (2)
Using 3max calculate the highest concentration possible for the
shortest cell path length available, usually 0.5–1.0 mm, to give an
absorbance¼2. (3) Prepare
a
solution at the concentration
dark under N2 at 0 ꢁC. It had 1H NMR
d: 1.22–1.28 (m, 2H), 1.49–1.54
determined in (2) and measure the spectrum. (4) Perform serial
dilutions, e.g., 1:2, etc., until the lower limit of detection of the in-
strument is reached. (5) Normalize the absorbance data to a path
length of 1 cm and plot. This procedure gave the plots of Figure 4.
(m, 4H),1.86 (s, 3H), 2.02 (s, 3H), 2.03 (s, 3H), 2.06 (s, 3H), 2.24 (t, 2H,
J¼7.9 Hz), 2.67 (t, 2H, J¼7.5 Hz), 3.59 (s, 3H), 6.19 (s, 1H), 9.48 (br s,
1H), 10.98 (br s, 1H) ppm; 13C NMR data are in Table 1; UV–vis
spectral data may be found in Table 5. Anal. Calcd for C20H28N2O2S
(360.5): C, 66.63; H, 7.83; N, 7.77. Found: C, 67.03; H, 7.88; N, 7.42.
4.2. 9-[5-Carbo(tert-butyldiphenylsilyl)pentyl]-2,3,7,8-tetra-
methyl-(10H)-dipyrrin-1-one ([6]-semirubin tert-butyldi-
phenylsilyl ester) (9)
Acknowledgements
This work was supported by the U.S. National Institutes of
Health (HD17779). S.D. is an R.C. Fuson Graduate Fellow. We thank
the U.S. National Science Foundation (CHE-0521191) for matching
funds to acquire a 400 MHz NMR spectrophotometer and upgrade
our 500 MHz NMR. We also thank Professor T.W. Bell for use of the
vapor pressure osmometer.
[6]-Semirubin (1) (1 equiv, 100 mg, 0.303 mmol) and 3 equiv of
imidazole (61.8 mg, 0.909 mmol) were mixed in dry THF (40 mL).
Then2 equivof tert-butyldiphenylsilyl chloride (166.5 mgh0.16 mL,
0.606 mmol) was added to the reaction mixture and stirred at room
temperature. After4days,15 mLofwaterwasaddedandthereaction
solution was extracted with CH2Cl2. The organic layer was removed
and dried over anhyd Na2SO4, and the solvent was evaporated
(rotovap) to give a yellow solid. This was purified by radial chro-
matography on silica gel using CH3OH/CH2Cl2 (2:98). The isolated
yellow solid (71% yield), mp 144–146 ꢁC, was obtained after evapo-
ration of CH2Cl2/CH3OH solution and was treated directly with
References and notes
1. (a) Pimentel, G. C.; McClellan, A. L. The Hydrogen Bond; W.H. Freeman: San
Francisco, CA, 1960; (b) Joelsten, M. D.; Schaad, L. J. Hydrogen Bonding; Marcel
Dekker: New York, NY, 1974.
2. (a) Falk, H. The Chemistry of Linear Oligopyrroles and Bile Pigments; Springer:
Wien, 1989; (b) Lightner, D. A. In Bilirubin; Heirwegh, K. P. M., Brown, S. B., Eds.;
CRC: Boca Raton, FL, 1982; Vol. 1, pp 1–58.
3. (a) Bonnett, R.; Davies, J. E.; Hursthouse, M. B.; Sheldrick, G. M. Proc. R. Soc.
London, Ser. B 1978, 202, 249–268; (b) LeBas, G.; Allegret, A.; Mauguen, Y.; De
Rango, C.; Bailly, M. Acta Crystallogr. 1980, B36, 3007–3011; (c) Mugnoli, A.;
Manitto, P.; Monti, D. Acta Crystallogr. 1983, C39, 1287–1291.
4. Chowdhury, J. R.; Wolkoff, A. W.; Chowdhury, N. R.; Arias, I. M. In The Metabolic
and Molecular Bases of Inherited Disease; Scriver, C. R., Beaudet, A. L., Sly, W. S.,
Valle, D., Eds.; McGraw-Hill: New York, NY, 2001; Vol. II, pp 3063–3101.
5. Brodersen, R. In Bilirubin; Heirwegh, K. P. M., Brown, S. B., Eds.; CRC: Boca
Raton, FL, 1982; Vol. 1, of Ref. 1b, pp 75–123.
Lawesson’s reagent. It had 1H NMR
d: 1.37–1.42 (m, 2H), 1.42 (s, 9H),
1.65–1.72 (m, 4H), 1.91 (s, 3H), 1.95 (s, 3H), 2.11 (s, 6H), 2.45 (t, 2H,
J¼7.6 Hz), 2.75 (t, 2H, J¼7.2 Hz), 5.92 (s, 1H), 6.13 (s, 1H), 7.36 (t, 4H),
7.41 (d, 2H), 7.65 (d, 4H, J¼7.5 Hz),10.15 (s,1H),11.30 (s,1H) ppm; 13C
NMR
d
: 9.3, 9.8, 10.4, 10.6, 19.4, 25.8, 26.9, 27.5 (ꢂ3), 29.5, 30.8, 36.8,
101.2,116.6,122.9,123.8,126.0,128.0,128.9,130.7,132.7,136.0,136.6,
142.9, 173.6, 174.6 ppm. Anal. Calcd for C35H44N2O3Si (568.8): C,
73.90; H, 7.80; N, 4.92. Found: C, 73.48; H, 7.73; N, 5.32.
6. (a) Person, R. V.; Peterson, B. R.; Lightner, D. A. J. Am. Chem. Soc. 1994, 116, 42–
59; (b) Nogales, D.; Lightner, D. A. J. Biol. Chem. 1995, 270, 73–77; (c) Boiadjiev,
S. E.; Person, R. V.; Puzicha, G.; Knobler, C.; Maverick, E.; Trueblood, K. N.;
Lightner, D. A. J. Am. Chem. Soc. 1992, 114, 10123–10133; (d) Puzicha, G.; Pu,
Y.-M.; Lightner, D. A. J. Am. Chem. Soc. 1991, 113, 3583–3592.
7. Wash, P. L.; Maverick, E.; Chiefari, J.; Lightner, D. A. J. Am. Chem. Soc. 1997, 119,
3802–3806.
8. Huggins, M. T.; Lightner, D. A. J. Org. Chem. 2000, 65, 6001–6008.
9. For a review of dipyrrinones, see: Boiadjiev, S. E.; Lightner, D. A. Org. Prep.
Proced. Int. 2006, 38, 347–349.
10. (a) Chen, Q.; Lightner, D. A. J. Org. Chem. 1998, 63, 2665–2675; (b) Tipton, A. K.;
Lightner, D. A. Monatsh. Chem. 1999, 130, 425–440; (c) Huggins, M. T.; Lightner,
D. A. Tetrahedron 2000, 56, 1797–1810.
11. (a) Boiadjiev, S. E.; Anstine, D. T.; Lightner, D. A. J. Am. Chem. Soc. 1995, 117,
8727–8736; (b) Boiadjiev, S. E.; Holmes, D. L.; Anstine, D. T.; Lightner, D. A.
Tetrahedron 1995, 51, 10663–10678.
4.3. 9-(5-Carboxypentyl)-2,3,7,8-tetramethyl-(10H)-dipyrrin-
1-thione ([6]-thiosemirubin) (4)
tert-Butyldiphenylsilyl ester (9, 1 equiv, 100 mg, 0.183 mmol)
from above and 3 equiv of Lawesson’s reagent (222.6 mg,
0.550 mmol) were mixed together in dry tetrahydrofuran (20 mL),
and the reaction mixture was stirred at room temperature for 2
days under N2 atmosphere. Then 2 mL (2.0 mmol) of a 1 M solution
of tetrabutylammonium fluoride in THF was added to deprotect the
silyl ester, and the reaction mixture was stirred for 12 h. THF sol-
vent was removed (rotovap), the residue was dissolved in CH2Cl2,
and the solution was passed through a short column of silica gel
(filtration chromatography), using CH2Cl2 as eluent. The product
was further purified by radial chromatography using CH2Cl2 as el-
uent. The deep red-colored solid (1) (32% yield), mp 178–180 ꢁC,
was obtained after evaporation of CH2Cl2. It was stored in the dark
12. Nogales, D. F.; Ma, J.-S.; Lightner, D. A. Tetrahedron 1993, 49, 2361–2372.
13. (a) For
d
-valerolactam and
g
-butyrolactam, KD¼277 and 287 Mꢀ1 at 25 ꢁC in
CCl4 by IR: Kulevsky, N.; Froehlich, P. M. J. Am. Chem. Soc. 1967, 89, 4839–4841;
(b) Lower values are reported (82 and 27 Mꢀ1) from VPO studies in benzene:
Montando, G.; Caccamuse, S.; Recca, A. J. Phys. Chem. 1975, 79, 1554; (c) For 2-
pyridone and 2-thiopyridone, KD¼7100 and 4100 Mꢀ1 in CCl4 at 25 ꢁC by IR
(Kulevsky, N.; Reinecke, W. J. Phys. Chem. 1968, 78, 3339–3340); (d) For 2-
pyridone and 2-thiopyridone in benzene, KD¼2200 and 28 Mꢀ1 at 37 ꢁC by VPO
(Krackov, M. H.; Lee, C. M.; Mautner, H. G. J. Am. Chem. Soc. 1965, 87, 892–896).
14. For thiopyridone, KD¼2.7ꢃ1.0ꢂ103 Mꢀ1 in CHCl3 at ambient T, and pyridone
under N2 at 0 ꢁC. It had 1H NMR
d: 1.38–1.43 (m, 2H), 1.57–1.66 (m,
2H), 1.71–1.74 (m, 2H), 1.90 (s, 3H), 1.94 (s, 3H), 2.09 (s, 3H), 2.11 (s,
3H), 2.48 (t, 2H, J¼7.2 Hz), 2.72 (t, 2H, J¼7.2 Hz), 6.11 (s, 1H), 8.96 (s,
1H), 10.49 (s, 1H), 13.5 (br s, 1H) ppm; 13C NMR data in Table 1; UV–
vis spectral data are in Table 5; HRMS (FAB, glycerol); calcd for
[MþHþ] C19H27N2O2S: 347.1793; found: 347.1784.
KD¼2.5ꢃ1.0ꢂ104
M
ꢀ1 in CHCl3 at 27 ꢁC by VPO (Beak, P.; Covington, J. B.; Smith,
S. G.; White, J. M.; Ziegler, J. M. J. Org. Chem. 1980, 45, 1354–1361).
15. Xie, M.; Lightner, D. A. J. Heterocycl. Chem. 1991, 28, 1753–1756.
16. Lawesson’s reagent: 2,4-bis-(4-methoxyphenyl)-1,3-dithia-2,4-diphosphetane
2,4-disulfide, used as available from Aldrich Chemicals.
17. (a) Grunewald, J. O.; Cullen, R.; Bredfeldt, J.; Strope, E. R. Org. Prep. Proced. Int.
1975, 7, 103–110; (b) Shrout, D. P.; Lightner, D. A. Synthesis 1990, 1062–1065.
4.4. 9-(5-Carbomethoxypentyl)-2,3,7,8-tetramethyl-(10H)-
dipyrrin-1-thione ([6]-thiosemirubin methyl ester) (8)
`
18. Cullen, D. L.; Pepe, G.; Meyer, E. F., Jr.; Falk, H.; Grubmayr, K. J. Chem. Soc., Perkin
Trans. 2 1979, 999–1004.
19. Kaplan, D.; Navon, G. Isr. J. Chem. 1983, 23, 177–186.
20. Huggins, M. T.; Lightner, D. A. Monatsh. Chem. 2001, 132, 203–221.
21. McDonagh, A. F.; Agati, G.; Lightner, D. A. Monatsh. Chem. 1998, 129, 649–660.
22. Fischer, H.; Joshioka, T.; Hartmann, P. Hoppe-Seyler’s Z. Physiol. Chem. 1932, 212,
146–156.
[6]-Semirubin methyl ester (2, 1 equiv, 100 mg, 0.291 mmol) and
3 equiv of Lawesson’s reagent (352 mg, 0.872 mmol) were mixed
together in dry THF (40 mL), and the reaction mixture was stirred at
room temperature for 36 h under N2. THF solvent was removed