Preparation of vinylglycines by thermolysis of homocysteine sulfoxides

Article abstract of DOI:10.1016/j.tetlet.2009.06.082

The synthesis and efficacy of preparing Cbz-VG-OMe (1) by thermolysis of alkyl and aryl homocysteine sulfoxides were surveyed. This investigation determined that aryl sulfoxide analogs were more effective for the reaction and that the 2-nitrophenyl analog 10f possessed a unique ability to syn eliminate at temperatures as low as 100 °C. The thermolysis of sulfoxide 10f was additionally discovered to occur under toluene reflux and when sodium acetate was added, Cbz-VG-OMe (1) could be obtained in high purity by simple filtration of the precipitated sulfenic acid byproduct 12. This mild protocol which was also applied in the synthesis of VG dipeptide 13 would have utility in the general synthesis of olefins and alkenes from 2-nitrophenylsulfoxides.

Full text of DOI:10.1016/j.tetlet.2009.06.082

Tetrahedron Letters 50 (2009) 5067–5070
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Preparation of vinylglycines by thermolysis of homocysteine sulfoxides
Sravan Kumar Patel a, Timothy E. Long a,b,
*
a The Department of Pharmaceutical and Biomedical Sciences, University of Georgia, Athens, GA 30602-2352, USA
b Center for Drug Discovery, University of Georgia, Athens, GA 30602-2352, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis and efficacy of preparing Cbz-VG-OMe (1) by thermolysis of alkyl and aryl homocysteine
sulfoxides were surveyed. This investigation determined that aryl sulfoxide analogs were more effective
for the reaction and that the 2-nitrophenyl analog 10f possessed a unique ability to syn eliminate at tem-
peratures as low as 100 °C. The thermolysis of sulfoxide 10f was additionally discovered to occur under
toluene reflux and when sodium acetate was added, Cbz-VG-OMe (1) could be obtained in high purity by
simple filtration of the precipitated sulfenic acid byproduct 12. This mild protocol which was also applied
in the synthesis of VG dipeptide 13 would have utility in the general synthesis of olefins and alkenes from
2-nitrophenylsulfoxides.
Received 30 April 2009
Revised 3 June 2009
Accepted 16 June 2009
Available online 21 June 2009
Published by Elsevier Ltd.
R
X
1. Introduction
O
Vinylglycine1 (VG) is a natural, non-protein
a
-amino acid which
CbzHN
CO2Me
CbzHN
CO2Me
acts as an inhibitor of enzymes with the pyridoxal phosphate (PLP)
X=S, R=Me
X=Se, R=Ph
cofactor such as alanine racemase,2 aspartate aminotransferase,3
1
and
mechanism-based deactivation of pyridoxal enzymes,1 efforts have
been made to identify other b, -unsaturated amino acids in nature.
a
-ketoglutarate dehydrogenase.4 As a useful tool for studying
weak
base
+
c
In addition, protected forms of VG have had synthetic utility in the
preparation of metabotropic glutamate receptors agonists,5 poly-
6N HCl
H2N
CO2H
CbzHN
CO2Me
c
-glutamate synthetase inhibitors,6 and the antitumor antibiotic
(+)-FR900482.7
3
2
Traditionally, the methods of choice to synthesize L-VG have
(+)-L-vinylglycine
(VG)
been by pyrolysis of protected methionine sulfoxide (MetO)8 and
thermolysis of aryl selonoxides (Scheme 1) obtained from either
protected L L L
-glutamate,9 -homoserine,10 or -homoserine .11 For
Scheme 1. Reported preparations of L-vinylglycine (VG).
multi-gram syntheses, the MetO pyrolysis approach is most com-
monly implemented; however due to the high vacuum (63 mmHg)
and temperature (>150 °C) requirements, isomerization is a consis-
tent problem for the reaction.8 The migratory occurrence to the
more thermally stable b-methyldehydroalanine (i.e., 2) is further
approach, we wanted to identify alternative sulfinyl substituents
that would facilitate the syn elimination at temperatures below
150 °C. This Letter describes: (i) the preparation of unreported
(S)-homocysteine (Hcy) sulfides and sulfoxides (HcyO); (ii) the syn
elimination efficacy of S-alkyl- and S-aryl-substituted HcyO esters;
and (iii) a new protocol to generate olefins in high purity from
2-nitrophenyl sulfoxides under mild toluene reflux.
enhanced by the acidity of the
a-proton in N,O-protected forms
of VG (i.e., 1). The isomer will form quantitatively in the presence
of triethylamine or N-methylmorpholine8 and it is likewise
believed that this decomposition during purification on silica con-
tributes to an optimized yield of 60%.12
Due to the difficult chromographic separation of the a,b-isomers
from protected vinylglycines and our desire to find a non-pyrolytic
2. Results and discussion
The investigation began by examining the thermolysis of MetO
analogs with longer alkyl chains. The sulfoxides were synthesized
* Corresponding author. Tel.: +1 706 542 8597; fax: +1 706 542 5358.
E-mail address: tlong@rx.uga.edu (T.E. Long).
via the thiolation and oxidation of (S)-bromoethylglycine
6
0040-4039/$ - see front matter Published by Elsevier Ltd.
doi:10.1016/j.tetlet.2009.06.082

5068
S. K. Patel, T. E. Long / Tetrahedron Letters 50 (2009) 5067–5070
NH2 HCl
CH3
Br
S
HCl/Br
Cl
CO2H
33% HBr
1.
1. SOCl2, MeOH
5oC
O
aq iPrOH, AcOH
H2N
CO2H
H2N
CO2H
AcOH
sealed tube
60oC, 90.1%
RT, 78.8%
O
90oC
50
2. CbzCl, NaHCO3
DCM/H2O, 98.2%
2. HCl (g), 58.2%
4
L-Met
5
R
R
O
145oC
1
Br
CO2Me
S
S
NaIO4
RSH
neat
aq. MeOH
NaI, K2CO3
Me2CO, 90oC
CbzHN
CbzHN
CO2Me
CbzHN
CO2Me
6
7a R=Et (76.3%)
7b n-Pr (81.8%)
7c i-Pr (52.3%)
7d n-Bu (66.7%)
7e t-Bu (0%)
8a R=Et (92.6%) 8e n-Hex (99.6%)
7f n-Hex (88.1%)
7g n-Oct (81.9%)
7h n-Dec (70.2%)
8b n-Pr (74.2%)
8c i-Pr (11.4%)
8d n-Bu (87.4%)
8f n-Oct (68.1%)
8g n-Dec (83.6%)
Scheme 2. Preparation of HcyO(alkyl) esters 8.
(Scheme 2). Its precursor, the mixed bromide salt 5,13 was obtained
in 2 steps by the hydrobromination of (S)-homoserine lactone HCl
(4) followed by N,O-protection with –Cbz and –Me, respectively.
The Hcy(alkyl) esters 7 were next prepared by nucleophilic
thiolation of bromide 6 under Finkelstein conditions.14 Reactions
were performed in a sealed tube to prevent thiol evaporation and
alkyl sulfides 7 were obtained in 52–88% yield with the exception
of protected Hcy(tert-butyl) 7e which failed to form by this meth-
od. Oxidation of the sulfides by aqueous sodium periodate pro-
vided HcyO(alkyl) esters 8a–g as a mixture of diastereomers for
the elimination studies.
Attention was turned next to examining the elimination reaction
of aryl sulfoxides 10. The (S)-Hcy(aryl) analogs were prepared from
bromide 6 in the same manner described followed by oxidation with
m-CPBA which afforded higher yields than sodium periodate
(Scheme 3). The elimination studies (Table 2) revealed a greater
syn thermolysis rate for the phenyl sulfoxides than their HcyO(alkyl)
counterpart. First evaluated was HcyO(Ph) 10b which after 18 h
decomposed to the desired olefin 1 along with small amounts of
a,b-isomer 2 and deoxygenated Hcy(Ph) 9b. Chromatography puri-
fication on silica however yielded only 54.1% of pure (S)-Cbz-VG-
OMe. Additional reactions were performed on HcyO(Ph) 10b to
screen alternative conditions and each were found to increase re-
duced side product 9b formation including: reduced pressure
(3 mmHg); reflux in DMF; microwave; and higher temperature
(190 °C/0.25 h). At temperatures less than 140 °C, the elimination
rate diminished substantially and deoxygenated side product in-
creased proportionally to longer durations of heat exposure.
Comparable results were observed for substituted phenyl ana-
logs with the exception of HcyO(p-MeOPh) 10c which appeared
to have the greatest susceptibility to side product formation (Table
2). This was not unexpected as preceding research on the pyrolysis
of substituted aryl sulfoxides deduced a correlation between elim-
ination rate and phenyl ring substituents. Emerson15 established
that para-situated electron-donating groups (i.e., MeO, Me) slowed
the pyrolysis of aryl n-propyl sulfoxides while electron-withdraw-
ing moieties (i.e., Cl, NO2) enhanced the rate. The substitution
effect was indeed apparent for the HcyO(aryl) 10c–f with
HcyO(p-MeOPh) 10c as the lone analog with unconverted sulfoxide
remaining after 18 h. The thermolysis of HcyO(p-ClPh) and
HcyO(p-NO2Ph) esters 10d and 10e, respectively, was complete
within this period with only traces of isomer present. It was also
determined that the time of thermolysis could be reduced to
15 min by applying a temperature of 190 °C with limited side prod-
uct formation.
The influence of electron-withdrawing groups on the syn elim-
ination of aryl selenoxides was similarly examined by Sharpless
and Young.16 Their studies revealed that a nitro group located ortho
substantially accelerated the reaction causing decomposition to o-
nitrophenyl selenenic acid and 1-dodecene at 25 °C. This prompted
us to evaluate the thermolysis rate of HcyO(o-NO2Ph) 10f which
required preparation of 2-nitrothiophenol (11) via Ph3P-mediated
reduction of its commercial disulfide. Upon heating at 145 °C,
decomposition of sulfoxide 10f was observed within minutes and
after 1 h the thermolysis was complete. NMR analysis of the crude
product revealed complete conversion to the desired VG ester 1
and absence of both isomer and deoxygenated side products. The
The reactions were conducted for each sulfoxide at 145 °C in or-
der to minimize isomerization. 1H NMRs were taken daily over a 3-
day period to monitor the progress of the solventless experiments
and to provide ratio estimates of (S)-Cbz-VG-OMe (1), a,b-unsatu-
rated isomer (2), and starting materials (8a–g). As anticipated, the
syn elimination rate of Cbz-MetO-OMe8 was found to be apprecia-
bly less than that for the multi-carbon chain analogs (Table 1). The
highest yield of these was HcyO(n-Bu) ester 8d; however a com-
mon theme among all the alkyl sulfoxides was the slow elimina-
tion at 145 °C. In addition, the exposure of VG 1 to heat over
72 h appeared to increase isomer formation and the duration of
the thermolysis needed to be substantially reduced in order to
optimize the reaction. This could be accomplished by raising the
temperature to 190 °C as observed for HcyO(n-Bu) 8d although iso-
lated yields were typically lower presumably due to decomposition
on the silica.
Table 1
Syn elimination results of HcyO(alkyl) esters 8
R=
°C
Time (h)
mmHg
1:2:8a
Yieldb (%)
Me8
145
190
145
145
145
145
145
145
190
145
145
145
72
2
760
760
760
760
760
760
760
760
3
0.3:0:1
12.3c
1.2c
—
—
45.7
—
Me8
1:0.7:0.7
0.8:.03:1
0.6:0.1:1
0.5:0.4:1
0.2:0:1
0.8:0.1:1
0.7:0.3:1
1:0:0.2
Et (8a)
72
42
72
72
42
72
5
72
72
72
n-Pro (8b)
n-Pro (8b)
i-Pro (8c)
n-Bu (8d)
n-Bu (8d)
n-Bu (8d)
n-Hex (8e)
n-Oct (8f)
n-Dec (8g)
—
49.8
30.3
35.1
18.2
41.3
760
760
760
1:0:0.8
1:0.1:0.7
1:0.1:1.4
a
Est. based on integrations in crude 1H NMR.
Isolated yield.
Contained isomer 2.
b
c

S. K. Patel, T. E. Long / Tetrahedron Letters 50 (2009) 5067–5070
5069
R
R
O
S
S
RSH
m-CPBA
CH2Cl2
1
6
neat
NaI, K2CO3
Me2CO, 90oC
CbzHN
9a R=Bn
CO2Me
CbzHN
CO2Me
10a R=Bn (54.1%, 2 steps)
10b Ph (89.1%)
9b Ph (89.0%)
10c p-MeOPh (89.6%)
10d p-ClPh (95.8%)
10e p-NO2Ph (62.7%)
10f o-NO2Ph (95.3%)
9c p-MeOPh (45.6%)
9d p-ClPh (77.8%)
9e p-NO2Ph (62.2%)
9f o-NO2Ph (94.9%)
Scheme 3. Preparation of HcyO(aryl) analogs 10.
Table 2
Syn elimination results of HcyO(aryl) esters 10
Methods to neutralize sulfenic acid 12 and prevent its decom-
position were evaluated using various bases. Reactions supple-
mented with pyridine, NaHCO3, and K2CO3 were effective but
each catalyzed partial isomerization. Conversely, the addition of
NaOAc (Scheme 4) negated impurity formation while permitting
facile removal of the sulfenic acid precipitate by filtration through
Celite. 1H NMR analysis of the toluene filtrate revealed >99%
conversion to VG ester 1 and the absence of side products. Substi-
tution with higher boiling chlorobenzene was also found to re-
duce reaction time to under 6 h; however prolonged reflux
(>10 h) resulted in minor amounts of isomer in the filtered
product.
R=
°C
Time (h)
mmHg
1:9:10a
Yieldb (%)
Bn (10a)
Bn (10a)
Ph (10b)
Ph (10b)
145
145
145
145
190
145
190
145
190
145
190
145
145
100
72
10
18
10
0.25
18
0.25
18
0.25
19
0.25
0.25
1
760
3
760
3
0.6:0:1
0.1:0:1
1:0.1:0
1:0.3:0.3
1:0:0.3c
1:0.4:0.2
1:0.4:0.2c
1:0:0c
29.9
—
54.1
57.9
25.0
35.5
9.2
41.0
67.0
53.9
62.9
—
Ph (10b)
760
760
760
760
760
760
760
760
760
760
p-MeOPh (10c)
p-MeOPh (10c)
p-ClPh (10d)
p-ClPh (10d)
p-NO2Ph (10e)
p-NO2Ph (10e)
o-NO2Ph (10f)
o-NO2Ph (10f)
o-NO2Ph (10f)
1:0.1:0c
1:0:0c
1:0:0
1:0:1.3
1:0:0
1:0:2.8
A utility of this protocol was next applied toward the synthesis of
the dipeptide
saponification of sulfide 9f followed by an EDC-mediated coupling
of the resulting acid with -Phe-OMe. S-Oxidation with m-CPBA
L-Cbz-VG-Phe-OMe (13). The preparation began with
35.5
—
18
L
a
Est. based on relative 1H peak areas in crude NMR.
Isolated yield.
Contained isomer 2.
b
c
and thermolysis of the sulfoxide under toluene reflux resulted in
precipitation of the acid 12 byproduct while providing vinyl dipep-
tide 13 in 85.2% isolated yield.
crude material was chromatographed on silica to provide pure
Cbz-VG-OMe in 35.5% yield.
3. Conclusion
Temperatures as low as 100 °C were additionally found to cata-
lyze the thermolysis of the o-nitrophenyl analog 10f and the reac-
tion was further evaluated under reflux conditions. Different
solvents were screened including chloroform and benzene; how-
ever higher boiling non-polar, aprotic solvents such as toluene
were required to complete the transformation. A serendipitous
finding of the reaction came with precipitation of the sulfenic acid
12 byproduct which was water-soluble and could be removed by
filtration. Unfortunately, a small portion of this acid decomposed
into several toluene-soluble impurities and we began exploring
conditions that avoided this outcome.
The described study found that the 2-nitrophenylsulfoxide 10f
can serve as an effective alternative to aryl selenoxides in the
thermolytic preparation of VG 1. With this precursor, it is like-
wise possible to avoid problems of isomerization without sele-
nide use while the ability to obtain essentially pure compound
via filtration of the sulfenic acid 12 byproduct is an added advan-
tage. This protocol is also being applied in our laboratory to gen-
erate other types of olefins and alkenes in high purity from 2-
nitrophenylsulfoxides. Typically, elimination reactions utilizing
sulfoxides involve the use of strong base and/or heat that can re-
sult in low yields. This first report on the unique eliminating
O N
O N
2
2
S
S
OH
OX
NaOAc
NaOAc
S
S
+
+
O
O
CbzHN
CbzHN
CO Me
CO Me
2
2
PhX, rfx
PhX, rfx
NO
NO
2
2
CbzHN
CbzHN
CO Me
CO Me
2
2
XX==MMee,,CCll
10f
10f
1
1
12
12
X=H,Na
X=H,Na
1. 5% HCl:AcOH, rfx
1. 5% HCl:AcOH, rfx
2. L-Phe-OMe-HCl
2. L-Phe-OMe-HCl
O N
O22N
O
O
Before reflux
After reflux
H
H
HOBt, EDC
HOBt, EDC
N
N
S
S
CbzHN
CbzHN
OMe
OMe
NaOAc suspended in a
PhMe solution of the
2-nitrosulfoxide.Yellow
coloration of the
solution is due to the
nitrophenyl group.
Acid 12 absorbs onto the
NaOAc turning it brown.
The solution containing the
olefin is colorless due to
precipitation of the
Et N, MeCN
E3t3N, MeCN
3. m-CPBA, DCM
3. m-CPBA, DCM
O
O
CbzHN
CbzHN
CO Me
CO22Me
Ph
Ph
then 10 eq. NaOAc
then 10 eq. NaOAc
13
13
9f
9f
PhMe, rfx, 85.2%
PhMe, rfx, 85.2%
nitrophenyl by product 12.
Scheme 4. Preparation of vinylglycines from 2-nitrophenylsulfoxides.

5070
S. K. Patel, T. E. Long / Tetrahedron Letters 50 (2009) 5067–5070
properties of 2-nitrophenylsulfoxides has demonstrated that they
can be employed as an efficient replacement to selenoxides in the
preparation of molecules containing a heat- or oxidant-sensitive
olefin.
References and notes
1. Berkowitz, D. B.; Charette, B. D.; Karukurichi, K. R.; McFadden, J. M. Tetrahedron:
Asymmetry 2006, 17, 869.
2. Lacoste, A. M.; Darriet, M.; Neuzil, E.; Le Goffic, F. Biochem. Soc. Trans. 1988, 16,
606.
3. Rando, R. R. Biochemistry 1974, 13, 3859.
Acknowledgments
4. Lai, J. C. K.; Cooper, A. J. L. J. Neurochem. 1986, 47, 1376.
5. Selvam, C.; Goudet, C.; Oueslati, N.; Pin, J.-P.; Acher, F. C. J. Med. Chem. 2007, 50,
4656.
6. Valiaeva, N.; Bartley, D.; Konno, T.; Coward, J. K. J. Org. Chem. 2001, 66, 5146.
7. Paleo, M. R.; Aurrecoechea, N.; Jung, K. Y.; Rapoport, H. J. Org. Chem. 2003, 68,
130.
8. Afzali-Ardakani, A.; Rapoport, H. J. Org. Chem. 1980, 45, 4817.
9. Hanessian, S.; Sahoo, S. P. Tetrahedron Lett. 1984, 25, 1425.
10. Pellicciari, R.; Natalini, B.; Marinozzi, M. Synth. Commun. 1988, 69, 1715.
11. Berkowitz, D. B.; Smith, M. K. Synthesis 1996, 39.
12. Carrasco, M.; Jones, R. J.; Kamel, S.; Rapoport, H.; Truong, T. Org. Synth. 1992, 70,
29.
Financial support of this work was provided by the University of
Georgia and the Hayes’ endowment fund. Special thanks are given
to the Department of Pharmaceutical and Biomedical Sciences
NMR facilities, Dr. Dennis Phillips for mass spectroscopy analyses,
and Dr. Warren Beach for helpful discussion.
Supplementary data
13. (a) Persson, T.; Hansen, T. H.; Rasmussen, T. B.; Skindersø, M. E.; Givskov, M.;
Nielsen, J. Org. Biomol. Chem. 2005, 3, 253; (b) Bartley, D. M.; Coward, J. K. J. Org.
Chem. 2005, 70, 6757.
14. Finkelstein, H. Ber. Dtsch. Chem. Ges. 1910, 43, 1528.
15. Emerson, D. W.; Korniski, T. J. J. Org. Chem. 1969, 34, 4115.
16. Sharpless, K. B.; Young, M. K. J. Org. Chem. 1975, 40, 947.
Supplementary data (experimental procedures, characteriza-
tion data, and copies of the 1H NMR, 13C NMR, and HRMS spectra)
associated with this article can be found, in the online version, at
doi:10.1016/j.tetlet.2009.06.082.