Thionation of Aminophthalimide Hindered Carbonyl Groups and Application to the Synthesis of 3,6′-Dithionated Pomalidomides

Article abstract of DOI:10.1055/s-0040-1720460

Herein, we present a new one-pot procedure for the 3,6′-dithionation of pomalidomide derivatives in which the key 3-position sulfur atom is preferentially installed at the desired (but sterically congested) carbonyl of the aminophthalimide system and with regiochemistry distinct from Lawesson's Reagent thionation methods. When heated in 1,4-dioxane with P ₄S ₁₀-pyridine complex, pomalidomides are smoothly and reproducibly converted into their 3,6′-dithionated analogues in roughly 30% isolated yield and at various scales. While detrimental to the desired 3,6′-type outcome when employing Lawesson's Reagent, we hypothesize that the pomalidomide aniline group instead facilitates P ₄S ₁₀-type thionation at the otherwise hindered 3-position carbonyl, contributing to the selectivity observed. When paired with classical methods of thionation, this approach offers an interesting and appealing addition to the synthetic toolbox, permitting facile late-stage access to complementary thionated pomalidomides in direct single-flask procedures.

Full text of DOI:10.1055/s-0040-1720460

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart
2020, 31, 917–922
letter
917
en
M. T. Scerba et al.
Letter
Synlett
Thionation of Aminophthalimide Hindered Carbonyl Groups and
Application to the Synthesis of 3,6′-Dithionated Pomalidomides
Michael T. Scerba*a
Maxime A. Sieglerb
Nigel H. Greiga
0
0
0
0
-
0
0
0
3
-
4
1
6
5
-
7
8
1
0
NH2
NH2
NH2
O
O
S
O
O
O
Lawesson’s
reagent
P4S10-pyridine
complex
NH
NH
NH
N
S
N
O
N
S
toluene, Δ
1,4-dioxane, Δ
a Drug Design & Development Section, Translational Gerontolo-
gy Branch, Intramural Research Program, National Institute on
Aging, National Institutes of Health, 251 Bayview Blvd, Balti-
more, MD 21224, USA
S
O
O
1,6'-dithiopomalidomide
pomalidomide
3,6'-dithiopomalidomide
major dithionated product
(accessible thionation)
previous work
major dithionated product
(hindered thionation)
this work
scerbamt@mail.nih.gov
b Small Molecule X-ray Facility, Department of Chemistry, Johns
Hopkins University, Baltimore, MD 21218, USA
MDBeMariluctahgiimClaDoseoecrlrresTeeir.g,sbSMnpacom&DenrtdD2b@ia1enm2vg2eaAl4iol,u.pnUtmihShoAe.grnotvSection, Translational Gerontology Branch, Intramural Research Program, National Institute on Aging, National Institutes of Health, 251 Bayview Blvd,
Received: 22.12.2020
Accepted after revision: 13.03.2021
Published online: 28.04.2021
NH2
4
O
O
O
O
NH
6'
NH
3
5
6
2'
3'
N
1
O
N
O
DOI: 10.1055/s-0040-1720460; Art ID: st-2020-v0642-l
4' 5'
7
O
O
Abstract Herein, we present a new one-pot procedure for the 3,6′-
dithionation of pomalidomide derivatives in which the key 3-position
sulfur atom is preferentially installed at the desired (but sterically con-
gested) carbonyl of the aminophthalimide system and with regiochem-
istry distinct from Lawesson’s Reagent thionation methods. When heat-
ed in 1,4-dioxane with P4S10–pyridine complex, pomalidomides are
smoothly and reproducibly converted into their 3,6′-dithionated ana-
logues in roughly 30% isolated yield and at various scales. While detri-
mental to the desired 3,6′-type outcome when employing Lawesson’s
Reagent, we hypothesize that the pomalidomide aniline group instead
facilitates P4S10-type thionation at the otherwise hindered 3-position
carbonyl, contributing to the selectivity observed. When paired with
classical methods of thionation, this approach offers an interesting and
appealing addition to the synthetic toolbox, permitting facile late-stage
access to complementary thionated pomalidomides in direct single-
flask procedures.
pomalidomide
thalidomide
1
2
NH2
NH2
S
O
O
O
NH
NH
N
S
N
S
O
S
3,6'-dithiopomalidomide
(3,6'-DT-pom)
1,6'-dithiopomalidomide
(1,6'-DT-pom)
3
4
Figure 1 Structures of pomalidomide, thalidomide, and dithionated
pomalidomides. The numbering pattern illustrates the locations of in-
terest, including the 3-, 1- and 6′-position carbonyls of pomalidomide
and related aminophthalimides.
Key words thionation, pomalidomides, dithionation, aminophthalim-
ides, regioselectivity
mide in both RAW 264.7 cellular studies and in rodents
challenged with lipopolysaccharide, but additionally low-
ered inflammation-induced COX-2 and iNOS.
Pomalidomide (1) is a well-known derivative of thalido-
mide, possessing a variety of immunoregulatory and tu-
moricidal properties. In this regard, it is currently being
marketed for the treatment of multiple myeloma.1 Research
in our laboratory has demonstrated advantageous conse-
quences of thionation on the efficacy and toxicity profiles
of various thalidomide analogues in the treatment of mild
traumatic brain injuries and other proinflammatory pro-
cesses.2 We recently identified 3,6′-dithiopomalidomide (3;
3,6′-DT-pom) (Figure 1) as showing particular promise in
lowering classical proinflammatory markers in response to
applied stress.3 In these studies, 3,6′-DT-pom (3) demon-
strated a similar anti-TNF- activity to that of pomalido-
Despite these advantages, our laboratory had yet to de-
velop even a modestly scalable method yielding 3,6′-DT-
pom and similar desirable materials. Our initial syntheses
(which combined pomalidomide with Lawesson’s Reagent
(LR) in refluxing toluene) were impeded by solubility lim-
itations and also generated multiple thionated isomers and
byproducts which were difficult to separate using standard
purification methods.4 The major dithionated product in
these cases proved to be the isomeric 1,6′-dithiopomalido-
mide (4; 1,6′-DT-pom). Further compounding the issue,
only minor amounts of the desired 3,6′-DT-pom were effi-
ciently isolable from these complex reaction mixtures. Even
in the best circumstances, the reactions were erratic and
not reliably repeatable owing to the aforementioned limita-
© 2021. Thieme. All rights reserved. Synlett 2021, 31, 917–922
Georg Thieme Verlag KG, Rüdigerstraße 14, 70469 Stuttgart, Germany

918
M. T. Scerba et al.
Letter
Synlett
-Br, -OMe, and -NO2 substitution proximal to a carbonyl can
direct thionation to the opposing side of the imide, suggest-
ing that a new paradigm could be required to achieve pref-
erential reactivity at our desired positions in pomalido-
mide.
O
OMe
O
NH2
O
O
MeO
N
S
N
S
NH
S
S
S
O
While seemingly detrimental to Lawesson’s-type reac-
tivity at the pomalidomide 3-position carbonyl, we hypoth-
esized that the phthalimide 4-amino group might instead
be utilized to our advantage if combined in tandem with
the proper thionation partner. We initially envisioned the
most direct use of this aniline could be as a molecular chap-
erone, guiding an appropriately compact non-Lawesson’s-
type reagent toward the desired 3-position (Figure 3a). Pre-
sumably, a thionation center substituted by an appropriate
leaving group could permit amine attack, thereby localizing
the sulfur-transferring source proximal to the object car-
bonyl. Alternatively, the aniline could facilitate the desired
thionation via hydrogen bonding7 or through inductive/res-
onance electronic effects (Figure 3b,c).8 In this regard, a va-
riety of conceivable pathways unfavored with Lawesson’s
Reagent or simply unavailable to the 1-position carbonyl
could prime the 3-position towards thionation, and espe-
cially when conditions might otherwise and independently
support a more accessible location.
5
6
7
Br
NO2
O
O
O
NH
N
N
N
S
S
S
8
9
Figure 2 Thionated cyclic imides generated using Lawesson’s Reagent
and showing thionation at the less-hindered carbonyl
tions. Needless to say, quantities of the valuable 3,6′-DT-
pom sufficient for advanced biological studies were not fea-
sible using the Lawesson’s-based method. These shortcom-
ings, combined with our desire to investigate complemen-
tary 3,6′-type materials, led us to explore new reagents and
conditions for the scalable production of 3,6′-DT-pom (3)
and similarly thionated aminophthalimides.
Our experience with thalidomide reactivity patterns5
led us to reason that the pomalidomide glutarimide 6′-posi-
tion would be easily thionated with almost any of the re-
agents we would consider for our new method. Therefore,
the difficult transformation (and the main hurdle needed to
be cleared) would be preferential thionation at the sterically
encumbered 3-position carbonyl concurrent with abated
reactivity at the readily available 1-position. Previous stud-
ies on Lawesson’s-type thionations of cyclic imides demon-
strated that the reactions occur preferentially at the less
crowded carbonyl, and are largely independent of simple
electronic effects (Figure 2). In these regards, -NH2, -CH3,
Table 1 Screen of Pomalidomide Thionation Conditions Highlighting
the Multithionated Product Distributions
NH2
O
O
NH
N
S
NH2
S
O
O
4
NH
thionation
N
O
+
NH2
S
O
O
S
NH
1
S
N
S
P
S
O
R
NH2
N
HN
NH2
LG
P
S
R
a)
O
S
O
3
N
N
O
O
O
Entry
Reagent
LR
Solvent
Time (h) Ratio 4/3a
1
2
3
4
5
toluene
18 h
2 h
2 h
2 h
2 h
4:1b
H
H
NH2
N
N
1:10b
3 onlyb
3 onlyc
H
H
P4S10–pyr
P4S10–pyr
P4S10–pyr
P4S10
pyridine
MeCN
c)
b)
O
O
O
N
1,4-dioxane
1,4-dioxane
N
N
O
O
O
3 detected with indi-
cation of trithionated
materials
Figure 3 Hypothetical reaction pathways for thionation of the poma-
lidomide aminophthalimide moiety with a focus on the desired 3-po-
sition carbonyl. (a) Interaction of the 4-position amino group with a
sterically accessible phosphorus-based thionating agent. The thionat-
ing center locates in proximity to the desired 3-position carbonyl. (b)
Distorted six-membered hydrogen bonding. (c) Increasing local car-
bonyl electron density via resonance/inductive effects. LG = leaving
group.
6
P4S10 + py
1,4-dioxane
2 h
1:44b
a Approximated using LC/MS extrapolated peak areas.
b Reaction not fully complete; starting pomalidomide material and/or
monothionated intermediates remain clearly detectable.
c The desired material 3 is the overwhelmingly major peak by far; starting
or monothionated materials are trace components.
© 2021. Thieme. All rights reserved. Synlett 2021, 31, 917–922

919
M. T. Scerba et al.
Letter
Synlett
the major dithionated isomer, with only trace amounts of
undesired 1,6′-DT-pom detected in certain cases (Table 1).10
The reaction was comparatively sluggish in pyridine, with
significant amounts of pomalidomide starting material and
presumed monothionated intermediates present after two
hours. A very minor amount of the 1,6′-isomer was present
as well. During the same timeframe, the reactions in both
acetonitrile and dioxane were absent of detectable 1,6-DT-
pom, with the MeCN iteration showing minimal remaining
monothionated adducts, while the dioxane attempt was
driven essentially to completion. Given the structural simi-
larities between P2S5 (generated during dissociation of 12)
and the P4S10–pyr complex 10, we wondered if traditional
P4S10 would be a viable thionation reagent in our system.11
When pomalidomide was screened with P4S10 in 1,4-diox-
ane, we noticed a rapid consumption of the starting materi-
al and significant production of 3,6′-DT-pom. Unfortunate-
ly, the reaction was quite indiscriminate, being complicated
by the production of 1,6′-DT-pom and materials consistent
with higher trithionated masses12 even after 15 minutes; by
two hours of reaction time, it was clear that the reaction
was being driven towards the trithionated congeners.13 In
an attempt to attenuate this harsh reactivity, we performed
S
S
S
P
S
S
P
S
MeO
S
P
S
N
N
P
OMe
S
S
P4S10–pyridine complex
(P4S10–pyr)
Lawesson’s reagent
P
S
S
P
11
10
S
P
S
P
S
S
S
S
S
tetraphosphorus decasulfide
(P4S10
)
12
Figure 4 Structures of relevant thionating reagents
When searching for appropriate reagents, we noticed
the recently investigated P4S10–pyridine complex 10 (P4S10–
pyr; Figure 4).9 It offered many desirable traits compared
with LR, including ease of preparation, simplified workup,
amenability to a variety of solvents, and, most interestingly,
a potential leaving group in the form of a pyridinium ligand.
Initially, we screened P4S10–pyr with pomalidomide in
pyridine, acetonitrile, or 1,4-dioxane in a series of LC/MS
experiments and were pleased to discover 3,6′-DT-pom as
Table 2 Regioselective Mono- and Dithionation of Aminophthalimides with Associated Yieldsa
Entry
1
Substrate
Product
Scale
25 mg (0.09 mmol)
100 mg (0.36 mmol)
274 mg (1 mmol)
Yield, Avg.b
8.0 mg (29%)
35 mg (31%)
96 mg (31%)
NH2
NH2
O
S
O
O
NH
NH
N
O
N
S
O
O
O
S
1
3
2
3
25 mg (0.09 mmol)
8.6 mg (29%)
NH2
NH2
O
O
NH
NH
N
S
N
O
O
O
F
F
14
13
30 mg (0.09 mmol)c
12 mg (36%)
NH
NH
O
O
S
O
NH
NH
N
O
N
S
O
O
F
F
15
16
4
89 mg (0.36 mmol)d
244 mg (1 mmol)
45 mg (47%)
178 mg (68%)
NH2
NH2
S
O
N
N
O
O
18
17
a Reaction conditions: pomalidomide (1.0 equiv), P4S10–pyr (1.5 equiv), 1,4-dioxane, 100 °C, under N2. The reaction was monitored by LC/MS, and the product
was purified by chromatography.
b Average yield of two or three iterations, see Supporting Information.
c P4S10–pyr (2 equiv), 40 h.
d P4S10–pyr (0.75 equiv) was used (based upon the carbonyl stoichiometry relative to pomalidomide).
© 2021. Thieme. All rights reserved. Synlett 2021, 31, 917–922

920
M. T. Scerba et al.
Letter
Synlett
O
O
a)
b)
P4S10–pyr
H2N
H2N
Lawesson’s reagent
N
N
1,4-dioxane
toluene or 1,4-dioxane
O
O
19
19
(19)
1,4-dioxane
toluene
NH2
NH2O
c)
O
d)
P4S10–pyr
Lawesson’s reagent
N
N
1,4-dioxane
toluene or 1,4-dioxane
O
O
17
17
NH2
18
NH2
20
S
O
N
N
O
S
1,4-dioxane
toluene
(17)
Figure 5 Representative HPLC traces demonstrating amine electronic influence on reactivity patterns of isomeric aminophthalimides with differing
thionation systems at 1 h of reaction time. Materials were thionated with either P4S10–pyridine complex or Lawesson’s Reagent and using the indicated
solvent (1,4-dioxane or toluene) (see the Supporting Information for complete details). In systems (a) and (b), starting material 19 elutes at 4.5 min,
monothionated adducts appear at 10.9 min and 11.4 min (inferred by mass). In these instances, the regiochemical differences between thionation
protocols are seemingly negligible. In systems (c) and (d), starting material 17 elutes at 6.5 min, and the regiochemical bias is clear, with the indicated
major monothionated species eluting at 11.8 min (c) and at 12.8 min (d) (toluene or 1,4-dioxane). Spectra were obtained at 280 nm.
a similar reaction but with added pyridine to crudely ap-
proximate the in situ production of the P4S10–pyr reagent.14
We were cautiously optimistic to find that this iteration
proceeded in a much more controlled fashion, especially
given the sensitivity of multicomponent P4S10-derived thi-
onations to reagent quality, reaction conditions, and the
choice of the imide-type substrate.15 Whereas this method
was clearly superior to using P4S10 alone, there was still a
noticeable amount of undesired 1,6′-DT-pom (4) produced
in addition to other thionated isomers. Ultimately, the
P4S10–pyridine complex 10/1,4-dioxane motif afforded a
much cleaner LC/MS profile in our screen, demonstrated
consistent reproducibility, and was selected as the method
of choice for all further synthetic reactions.
sition thionations occurred cleanly and without significant
competition from the 1-position carbonyl. Pomalidomide
(1) afforded the key 3,6′-dithiopomalidomide (3) in 29%
isolated yield; this reaction worked equally well at all scales
examined. Halogen substitution para to the amine ap-
peared to have little effect on the reactivity pattern, as fluo-
rinated analogue 14 was generated in 29% yield. Notably,
the bulky N-isopropyl analogue 15 provided hindered 16 as
the major dithionated product in 36% isolated yield. Finally,
cyclohexyl-substituted analogue 18 was readily obtained,
particularly at the 1 mmol scale.
To further probe this reactivity and highlight the re-
giochemical preference for the 3-position carbonyl, we sub-
jected isomeric cyclohexyl aminophthalimides 17 and 19 to
a series of thionation conditions (Figure 5). In substituting
the glutarimide ring of the pomalidomide backbone with a
simple cyclohexane, we simplified our subsequent analyses
by eliminating any possible thionation event at the uninter-
Next, a small family of aminophthalimides reacted
smoothly in one-pot procedures to produce the desired thi-
onated pomalidomide derivatives in moderate but consis-
tent yields (Table 2).16 Most importantly, the relevant 3-po-
© 2021. Thieme. All rights reserved. Synlett 2021, 31, 917–922

921
M. T. Scerba et al.
Letter
Synlett
esting 6′-position, while simultaneously relieving some of
the potential electronic influence and steric congestion
neighboring the key imide moiety. With the amine re-
moved from direct proximity to the 3-position carbonyl,
thionations of 5-aminophthalimide 19 using P4S10–pyridine
or LR (Figure 5a,b) yielded nearly identical HPLC profiles af-
ter a one hour reaction time, with two resulting monothi-
onated products12 generated in practically equal ratios and
largely independent of the thionation reagent or solvent
employed (Figure 5; product peaks 10.9 and 11.4 min).
With the adjacent amine reestablished in 4-aminoph-
thalimide 17, the expected regiochemical preferences re-
surfaced, with LR providing a distinct major 1-thionated
species 20 (12.8 min) versus the 3-thionated product 18
(11.8 min) obtained with the P4S10–pyr/dioxane system
(Figure 5d,c). It is evident that in this simplified system, any
inherent electronic activation17 afforded by the amine alone
is not the sole driving force behind the observed re-
giochemical bias; if this were the case, one would expect
the product distributions in Figure 5c,d to mirror those in
Figure 5a,b. Although such activation may certainly con-
tribute to 4-aminophthalimide 3-position thionation and
reactivity in general, it is perhaps not significant enough to
overcome the regiochemical implications associated with
LR.
Additionally, a prominent solvent effect disparity be-
tween 1,4-dioxane and toluene influencing 3-thionation is
unlikely, as one would presume a more marked discrepancy
in the corresponding traces shown in Figure 5b,d. However,
the modest selectivity differences seen in pomalidomide
thionation (Table 1) may help to illuminate the importance
of aniline hydrogen bonding on the associated regiochemi-
cal bias, as the notably stronger hydrogen-bond-accepting
pyridine deteriorates the reaction rate and selectivity rela-
tive to weaker acceptors like 1,4-dioxane.18 While a full and
thorough mechanistic investigation is beyond the scope of
this preliminary (and largely empirical) study, the selectivi-
ty afforded here is quite striking and likely due to a complex
amalgamation of steric and electronic properties coupled
with the peculiarities of the specific thionating species in-
volved. Further mechanistic analyses and new thionated
materials are currently being explored.
Acknowledgment
N.G. has a patent on compounds 3 and 4, which is assigned to the Na-
tional Institute on Aging, NIH: Thalidomide Analogs and Methods of
Use. US 10,730,835 B2, 2020. The authors wish to thank Dr. Shelley
Jackson from the NIH NIDA Structural Biology Core for assistance
with the high-resolution mass spectrometry analyses.
Supporting Information
Supporting information for this article is available online at
https://doi.org/10.1055/s-0040-1720460.
S
u
p
p
orti
n
gI
n
f
orm
a
ti
o
n
S
u
p
p
orti
n
gInform
a
ti
o
n
References and Notes
(1) (a) Hoy, S. M. Drugs 2017, 77, 1897. (b) Chanan-Khan, A. A.;
Swaika, A.; Paulus, A.; Kumar, S. K.; Mikhael, J. R.; Rajkumar, S.
V.; Dispenzieri, A.; Lacy, M. Q. Blood Cancer J. 2013, 3, e143.
(2) (a) Park, E.; Levis, W. R.; Greig, N. H.; Euisun, J.; Schuller-Levis,
G. J. Drugs Dermatol. 2010, 9, 330. (b) Tweedie, D.; Frankola, K.
A.; Luo, W.; Li, Y.; Greig, N. H. Open Biochem. J. 2011, 5, 37.
(3) (a) Greig, N. H.; Luo, W.; Tweedie, D.; Holloway, H. W.; Yu, Q.-S.;
Goetzl, E. J. US 10220028B2, 2019. (b) Lin, C.-T.; Lecca, D.; Yang,
L.-Y.; Luo, W. L.; Scerba, M. T.; Tweedie, D.; Huang, P.-S.; Jung, Y.-
J.; Kim, D. S.; Yang, C.-H.; Hoffer, B. J.; Wang, J.-Y.; Greig, N. H.
eLife 2020, 9, e54726.
(4) See the Supporting Information for LC/MS profiles of a typical
reaction with Lawesson’s Reagent compared with our new
method.
(5) Zhu, X.; Giordano, T.; Yu, Q.-S.; Holloway, H. W.; Perry, T. A.;
Lahiri, D. K.; Brossi, A.; Greig, N. H. J. Med. Chem. 2003, 46, 5222.
(6) (a) Huang, J.; Chen, B.; Zhou, B.; Han, Y. New J. Chem. 2018, 42,
1181. (b) Geneste, H.; Ochse, M.; Drescher, K.; Dinges, J.; Jakob,
C. WO 2014041175, 2014. (c) Milewska, M. J.; Bytner, T.;
Polonski, T. Synthesis 1996, 1485. (d) Fang, F. G.; Danishefsky, S.
J. Tetrahedron Lett. 1989, 30, 2747. (e) Greig, N. H.; Luo, W.;
Tweedie, D.; Holloway, H. W.; Yu, Q.-S.; Goetzl, E. J. US
8927725B2, 2015.
(7) (a) Seo, M-S.; Jang, S.; Jung, H.; Kim, H. J. Org. Chem. 2018, 83,
14300. (b) Furet, P.; Bold, G.; Hofmann, F.; Manley, P.; Meyer, T.;
Altmann, K.-H. Bioorg. Med. Chem. Lett. 2003, 13, 2967. (c) Kuhn,
B.; Mohr, P.; Stahl, M. J. Med. Chem. 2010, 53, 2601. (d) Takeda,
T.; Suzuki, Y.; Kawamata, J.; Noro, S.-i.; Nakamurac, T.;
Akutagawa, T. Phys. Chem. Chem. Phys. 2017, 19, 23905.
(8) (a) Vijayakanth,
T.;
Ram,
F.;
Praveenkumar,
B.;
Shanmuganathan, K.; Boomishankar, R. Chem. Mater. 2019, 31,
5964. (b) Kolodiazhna, A. O.; Kolodiazhnyi, O. I. Symmetry 2020,
12, 108. (c) Longobardi, L. E.; Wolter, W.; Stephan, D. W. Angew.
Chem. Int. Ed. 2015, 54, 809.
Conflict of Interest
(9) Bergman, J.; Pettersson, B.; Hasimbegovic, V.; Svensson, P. H.
J. Org. Chem. 2011, 76, 1546.
The authors declare no conflict of interest
(10) Thionation of Pomalidomide: General Procedure
Representative conditions: A vial was charged with 25 mg of
pomalidomide (1 equiv) and the appropriate thionating reagent
(1.5 equiv of P4S10-pyridine, or 0.75 equiv of P4S10). The vial was
sealed and placed under N2. The indicated solvent (2.5 mL) was
injected and the mixture was stirred vigorously while the con-
tents were briefly degassed. Then, the reaction was heated to
100 °C and monitored by LC/MS. See the Supporting Informa-
tion for complete details.
Funding Information
This research was supported by the Intramural Research Program of
the NIH, National Institute on Aging (#AG000311).
N
a
ti
o
n
a
l
I
nstiutte
o
n
A
g
i
n
g
(A
G
0
0
0
3
1
1)
© 2021. Thieme. All rights reserved. Synlett 2021, 31, 917–922

922
M. T. Scerba et al.
Letter
Synlett
(11) (a) Ozturk, T.; Ertas, E.; Mert, O. Chem. Rev. 2010, 110, 3419.
(b) Lecher, H. Z.; Greenwood, R. A.; Whitehouse, K. C.; Chao, T.
H. J. Am. Chem. Soc. 1956, 78, 5018.
was cooled to r.t., and the red reaction liquor was decanted,
evaporated to dryness, and dissolved in EtOAc. Then, the dark
oily residue remaining in the original reaction vial was stirred
gently with EtOAc and sat. aq NaHCO3. The resulting two-phase
mixture was transferred to a separatory funnel and combined
with the previous EtOAc solution. The aqueous phase was care-
fully drained off, and the organic phase was washed succes-
sively with water, 0.5 M HCl, and brine, then dried (Na2SO4) and
concentrated. The crude material was purified by HPLC to give a
bright-orange powder; yield: 8.0 mg (29%). 1H NMR (400 MHz,
DMSO-d6):  = 12.63 (s, 1 H), 7.65 (br s, 2 H), 7.50–7.40 (m, 1 H),
7.18 (d, J = 8.5 Hz, 1 H), 7.03 (br s, 1 H), 5.64–5.40 (br m, 1 H*),
3.26–2.78 (br m, 2.5 H*), 2.02 (m, 1 H) (* difficult to resolve).
(12) We did not isolate these materials or characterize their struc-
tures. However, given the nature of the reaction, the observed
masses seemed plausible for the incorporation of the indicated
amount of sulfur. See the Supporting Information for examples.
(13) Fortunately, even after 8 hours of reaction time, the P4S10-pyri-
dine complex/1,4-dioxane system demonstrated 3,6′-DT-pom 3
as the overwhelmingly favored dithionated product and with
only traces of other thionated isomers. As was expected, reac-
tions conducted solely with P4S10 were driven significantly
towards presumed trithionated materials in the same period of
time. See the Supporting Information for examples.
HRMS (MALDI): m/z [M
+
Na]+ calcd for C13H11N3NaO2S2:
(14) Klingsberg, E.; Papa, D. J. Am. Chem. Soc. 1951, 73, 4988.
(15) (a) Kaneko, K.; Katayama, H.; Wakabayashi, T.; Kumonoka, T.
Synthesis 1988, 152. (b) Łapucha, A. R. Synthesis 1987, 256.
(16) 4-Amino-2-(2-oxo-6-thioxopiperidin-3-yl)-3-thioxoisoindo-
lin-1-one (3): Typical Procedure
328.01849; found: 328.01880. Structural assignments are
further supported and confirmed by single-crystal X-ray analy-
sis (see ref. 19 and Supporting Information).
(17) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165.
(18) Spencer, J. N.; Campanella, C. L.; Harris, E. M.; Wolbach, W. S.
J. Phys. Chem. 1985, 89, 1888.
(19) CCDC 2074092 (3), 2074088 (4), 2074087 (14), 2074089 (16),
2074090 (18), and 2074091 (20) contain the supplementary
crystallographic data for the compounds listed. The data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/structures.
A vial was charged with pomalidomide (1; 25 mg, 0.091 mmol,
1.0 equiv), P4S10–pyr (52 mg, 0.14 mmol, 1.5 equiv), then sealed
and placed under N2. 1,4-Dioxane (2.5 mL) was added by injec-
tion and the mixture was briefly degassed, then stirred and
heated at 100 °C. After 8 hours, the reaction changed from
cloudy yellow through clear orange to deep red. The mixture
© 2021. Thieme. All rights reserved. Synlett 2021, 31, 917–922