New opinions on the amidoalkylation of hydrophosphorylic compounds

Article abstract of DOI:10.1016/j.tetlet.2010.03.020

A new and milder version of the procedure for the synthesis of N-protected α-aminoalkylphosphorylic compounds by reaction of alkyl carbamates, aldehydes and hydrophosphorylic compounds in acetic anhydride/acetyl chloride and a new mechanism for this type of reaction are described. The isolation, for the first time, of N,N′-benzylidene- and N,N′-alkylidenebiscarbamates as intermediates from the reaction medium and studies of the direct reaction of pre-obtained biscarbamates and hydrophosphorylic compounds in acetic anhydride are reported. A new version of the mechanism for this reaction which includes an Arbuzov-type reaction is proposed.

Full text of DOI:10.1016/j.tetlet.2010.03.020

Tetrahedron Letters 51 (2010) 2613–2616
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
New opinions on the amidoalkylation of hydrophosphorylic compounds
*
Maxim E. Dmitriev, Valery V. Ragulin
Institute of Physiologically Active Compounds, Severny pr., 1, Chernogolovka, Moscow Region 142432, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
A new and milder version of the procedure for the synthesis of N-protected
a-aminoalkylphosphorylic
Received 13 January 2010
Revised 24 February 2010
Accepted 5 March 2010
Available online 11 March 2010
compounds by reaction of alkyl carbamates, aldehydes and hydrophosphorylic compounds in acetic
anhydride/acetyl chloride and a new mechanism for this type of reaction are described. The isolation,
for the first time, of N,N0-benzylidene- and N,N0-alkylidenebiscarbamates as intermediates from the reac-
tion medium and studies of the direct reaction of pre-obtained biscarbamates and hydrophosphorylic
compounds in acetic anhydride are reported. A new version of the mechanism for this reaction which
includes an Arbuzov-type reaction is proposed.
Keywords:
Oleksyszyn reaction
Alkylidenebiscarbamates
N-Alkoxycarbonylimine cation
Ó 2010 Elsevier Ltd. All rights reserved.
A convenient mimic of a substrate in the transition state of at
least two classes of hydrolytic enzymes, for example, Zn-metallo-
proteases and aspartyl proteases, involves substitution of the pep-
tide bond by a nonhydrolyzable phosphinate moiety (Fig. 1).1
Construction of the pseudo-peptide fragment (A) is usually accom-
plished by addition of the pre-obtained N-protected amin-
oalkylphosphonous component (B) of pseudo-peptide 1 to the
The procedure for the aminoalkylation of 3 by the addition of
phosphonous acids 3 or their silyl esters to the corresponding N-
protected Schiff bases would involve a series of protection–depro-
tection steps.9 An alternative amidoalkylation of trivalent phos-
phorus compounds could be a more convenient route to pseudo-
a,a-dipeptide building blocks 1 (Scheme 1).
The procedure for the Kabachnik–Fields-type reaction with
corresponding
a
-substituted acrylates.2 However, protection of
amides as the amino component10–12 and trivalent phosphorus
chlorides, aldehydes or ketones in acetic acid (Oleksyszyn reac-
tion)11 was modified by Yuan et al.12 for dialkyl phosphites with
carbonyl compounds and amides or carbamates in acetyl chloride.
the nitrogen and phosphorus atoms of aminoalkylphosphonous
building blocks increases the total number of steps for the pro-
cess3,4 and, moreover, attempts to add the aminoalkylphospho-
nous component to
a-R-substituted acrylates were unsuccessful
in some cases.4
A
We earlier proposed and now are continuing the development
of an alternative synthetic route to pseudo-dipeptides via reverse
construction of the desired molecules. The route consists of the
NH
O
R
R
O
OH
HN
C(O)NH
HOP(O)CH2
OH
P
R`
N
H
HO
B
addition of hypophosphite to
a-R-substituted acrylates and crea-
O
R`
O
tion of the first phosphorus–carbon bond with formation of phos-
phonous acids 3 followed by generation of the amino acid function
and the second phosphorus–carbon bond.5–7 We are developing
2
1
Figure 1. Structures of pseudo-peptide 1 and the peptide fragment 2.
this methodology for the synthesis of pseudo-
tides,5 pseudo- -aminobutanoyl peptides6 and pseudo-
tides.7 Elaboration of methods for the synthesis of phosphinic
pseudo- -dipeptides 1 includes the search for a convenient pro-
cedure for the step involving the construction of the -amin-
c-glutamyl pep-
c
a,a-dipep-
(A)
(A)
a,a
O
O
a
+
NH2
OAlk
O
R
O
R`
H
N
ophosphorylic function using the phosphonous acid component 3
which contains the isostere of the corresponding amino acid (A).
The synthesis of phosphonous acids 3 was described earlier from
hypophosphorous acid salts.7,8
R
OAlk
O
O
O
P
H
P
- H2O
R`
OH
OH
(B)
3
1
* Corresponding author. Tel.: +7 496 524 97 30; fax: +7 496 524 95 08.
E-mail address: rvalery@dio.ru (V.V. Ragulin).
Scheme 1. Amidoalkylation of phosphonous acids
isostere of amino acid (A).
3
containing the structural
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.03.020

2614
M. E. Dmitriev, V. V. Ragulin / Tetrahedron Letters 51 (2010) 2613–2616
Table 1
N-Alkoxycarbonyl-
Usually N-acylated
a-aminoalkylphosphorylic compounds, with-
a-aminoalkylphosphorylic compounds (7a–k)
out isolation, are subjected to acidic hydrolysis followed by isola-
tion of the
a
-aminoalkylphosphonic acids.11 The amidoalkylation
Entry Compound
X
Y
R
Alk Yield (%)
of dialkyl phosphites in acetyl chloride can be accompanied by par-
tial dealkylation of the alkoxyphosphorylic fragment. The latter
version of the procedure of Yuan et al.12 in acetyl chloride was ex-
tended by Yiotakis13 using previously dealkylated phosphonous
carboxylic acids and Fmoc carbamates for the synthesis of phos-
phinic pseudo-peptides in an acetyl chloride/acetic acid mixture.
A milder procedure reported by Oleksyszyn14 consists of the
addition of an aldehyde to a pre-heated solution of the phospho-
rous acid and amide in acetic anhydride, however, the yields were
usually low, except for aromatic aldehydes.14 Nevertheless, this
procedure can be of interest for the synthesis of pseudo-peptide
blocks with retention of alkoxycarboxylic and alkoxyphosphorylic
fragments.
1
2
3
4
5
6
7
8
9
7a
7b
7c
7d
7e
7f
7g
7h
7i
MeO
BuO
Me
Me
Me
MeO Ph
BuO Ph
Et
Et
Et
70a; 67c; 63d
71c; 53d
OH
OH
OH
OH
OH
OH
Ph
Ph
77c; 73b; 59a; 52d
Me 79a; 61d
i-Bu Et
48c; 38b; 33a; 28d
Me
Me
i-Pr Me 31a
i-Bu Me 51c; 32a
HOCH2
EtOC(O)CH2CH2 OH
EtOC(O)CH2CH2 OH
Ph
Ph
Et
Et
54a
71c; 65b; 57a
10
11
7j
7k
i-Bu Me 32a
Et
Et
Ph
Et
54a
a
c
b
TFA (10 mol %) or TSA (2 mol %) as catalysts.
(AcCl/Ac2O = 1:4).
Ac2O.
d
Previously, bisamides 4 and 1-(acylamino)alkyl acetates 5 were
postulated as probable reaction intermediates of the amidoalkyla-
tion of PIII compounds (Scheme 2).14,15
the initial formation of insoluble N,N0-alkylidene- or N,N0-benzy-
lidenebiscarbamates 4 and simultaneous disappearance of the
crystalline intermediate on their reaction with the hydrophosph-
orylic component 6 were observed. We obtained a mixture of
N,N0-Arylidene- or N,N0-alkylidenebisamides 4 were proposed as
probable intermediates in the direct amidoalkylation of PIII com-
pounds by Oleksyszyn,14 because they react satisfactorily with PIII
nucleophiles11 and are easily formed from carbonyl compounds
and amides.16 However, these compounds were not isolated from
the reaction medium.11,12,14
biscarbamates (4, R0 = OAlk) and N-protected
a-aminophosphory-
lic compounds 7 in different ratios, when the three-component
reaction was stopped early. Silica gel chromatography allowed
the separation of the intermediate biscarbamates 4 and the de-
sired compounds 7.
These results were unexpected, as 1-(acylamino)alkyl acetates 5
would appear to be more preferable intermediates, if we accept the
reaction mechanism including nucleophilic attack of the phospho-
rus atom of the P–H component on the electrophilic carbon atom of
1-(acylamino)alkyl acetate 5 containing MeC(O)O as a good leaving
group.
These data formed a good basis for the studies on the direct
reaction of hydrophosphorylic compounds 6 and pre-obtained
N,N0-alkylidenebiscarbamates 4 in acetic anhydride (Scheme 4).
We found that biscarbamates 4 and dimethyl phosphite (6,
X = Y = MeO) or methylphosphonous acid (6, X = Me, Y = OH) in
acetic anhydride at room temperature reacted to give dimethyl
Soroka15 ruled out bisamides 4 as the main reaction intermedi-
ates and proposed 1-(acylamino)alkyl acetates 5 as possible inter-
mediates but again these compounds were not isolated from the
reaction medium (Scheme 2). In this connection, we report the re-
sults of our investigations on a milder procedure for the amidoal-
kylation of P–H compounds.
In this Letter, we report new data on the amidoalkylation of
hydrophosphorylic compounds 6 in acetic anhydride.
We found that milder conditions for the three-component con-
densation reaction of dialkyl phosphites, diethylphosphinous acid,
different alkylphosphonous acids, methyl and ethyl carbamates
and aldehydes in acetic anhydride at room temperature allow N-
alkoxycarbonyl-a-aminoalkylphosphorylic compounds 7 to be ob-
tained (Scheme 3, Table 1). A mixture of acetyl chloride and acetic
anhydride (1:4) proved to be a good medium for the amidoalkyla-
tion of hydrophosphorylic compounds, but the search for an opti-
mum mixture was not carried out.
Also, we found that the milder condensation procedure made
it possible to isolate N,N0-benzylidene- and N,N0-alkylidenebiscar-
bamates (4, R0 = OAlk) as intermediates from the reaction. Often
N-alkyloxycarbonyl-a-aminoalkyl phosphonates or N-alkyloxycar-
bonyl-
a
-aminoalkyl methylphosphinic acids 7 in 45–72% yields.
Moreover, we found that N,N0-benzylidenebiscarbamates (4,
R = Ph) reacted with hydrophosphorylic compounds 6 to give
a-aminoalkylphosphorylic compounds 7 in higher yields than
N,N0-isoamylidenebiscarbamates (4, R = i-Bu). Therefore, N,N0-ben-
zylidenebiscarbamates (4, R = Ph) are more reactive than N,N0-iso-
amylidenebiscarbamates (4, R = i-Bu). These results confirm the
electrophilic character of the biscarbamates. Note that P–H com-
pounds with different structures, namely, dialkyl phosphites, diet-
hylphosphinous acid and alkylphosphonous acids, were studied in
the three-component version of the condensation in acetic anhy-
dride. However, the hydrophosphorylic compounds showed no
apparent nucleophilic nature.
O
O
H
N
H
N
RCH(O) + R`C(O)NH2
AcOH / AcCl / Ac2O
R
R`
R
R`
O
NH
O
Me
R`
O
These data confirm the multistep character of the reaction
mechanism, and that the nucleophilic attack of the phosphorus
atom of the hydrophosphorylic compound at the electrophilic
4
5
Scheme 2. Proposed intermediates 4 and 5 in the amidoalkylation of trivalent
phosphorus compounds.
Ac2O or
Ac2O + H+
O
P
Y
O
R
O
R
O
P
Y
X
O
+
X
H
O
AlkO
N
H
N
H
OAlk
O
P
Y
Ac2O or Ac2O/AcCl
or Ac2O(+ H+)
R
O
N
H
X
P
Y
H
+
+
RCH(O)
X
O
OAlk
4
6
7
H2N
OAlk
N
H
R = i-Bu, Ph; Alk = Me, Et
X = Me, MeO; Y = MeO, OH
OAlk
7
6
Scheme 4. Synthesis of N-protected
a-aminoalkylphosphorylic compounds 7 via
Scheme 3. Synthesis of N-protected a-aminoalkylphosphorylic compounds 7.
reaction of biscarbamates 4 and hydrophosphorylic compounds 6.

M. E. Dmitriev, V. V. Ragulin / Tetrahedron Letters 51 (2010) 2613–2616
2615
carbon atom of the biscarbamate does not determine the general
reaction rate of the amidoalkylation process.
O
O
P
Y
+ AcCl (or Ac2O)
- HCl (or HOAc)
+ AcCl
- Ac2O
X
Y
O
X
H
P
Cl
It is known that both the formation of amidoalkylation reagents
Me
P
Y
X
from aldehydes and amides and a-amidoalkylation of carbon com-
10
6
pounds require acid catalysis.16 In this connection, we report that
9
biscarbamates 4 and hydrophosphorylic compounds 6 react to af-
Scheme 7. Proposed reactive P–OAc intermediates 9 formed from P–H compounds
6 in Ac2O/AcCl.
ford N-alkyloxycarbonyl-a-aminoalkylphosphorylic compounds 7
in yields usually lower than those obtained by the addition of cat-
alytic quantities of trifluoroacetic (TFA) or p-toluenesulfonic (TSA)
acids in contrast to the data of Oleksyszyn.14 These authors earlier
reported that the addition of TSA to the reaction mixture contain-
ing an aldehyde, an amide and phosphorous acid did not increase
We propose that P–OAc derivatives 9 of trivalent phosphorus
compounds are intermediate structures in equilibrium between
P–H compounds 6 and chlorides 10 in AcOH/Ac2O/AcCl medium
(Scheme 7). Two signals due to trivalent phosphorus compounds
were observed in the 31P NMR spectrum of methylphosphonous
acid in acetyl chloride or methyldichlorophosphine in acetic anhy-
dride. One signal at about d 198 corresponds to MePCl2. We pro-
pose that another signal at about d 185 corresponds to the
P–OAc derivative of methylphosphonous acid 9 (Supplementary
data, part 3).
the yield of the desired a
-aminoalkylphosphonic acids.14 However,
Yuan12c showed the positive influence of acidification on the three-
component condensation of 2-haloalkaneamides, benzaldehydes
and dialkyl phosphites. Moreover, we found that the positive catal-
ysis by TSA is more powerful than that by TFA. Also we observed
the positive influence of the addition of acetyl chloride on the
interaction of pre-obtained biscarbamate 4 and methylphospho-
nous acid in acetic anhydride.
The proposed intermediate 9 contains both a nucleophilic
phosphorus atom and an electrophilic carbon atom (acetyl
fragment). Therefore, compound 9 can readily participate in the
Arbuzov-type reaction.17 The formation of intermediate P–OAc
derivatives 9 can explain the unique role of acetic anhydride in
this reaction. We propose that nucleophilic attack of the phos-
phorus atom of the intermediate P–OAc compound 9 on the pos-
itively charged carbon atom of the N-alkoxycarbonylimine cation
8 occurs followed by the formation of the phosphorus–carbon
bond and simultaneous attack of anion Z on the electrophilic
carbon atom of the acylic fragment of the P–OAc derivative with
subsequent isolation of AcZ according to the Arbuzov-type reac-
tion (Scheme 8).
Our data suggest that the protonation of the oxygen (C@O) or
nitrogen atom of intermediate biscarbamates 4 followed by the
formation of the N-alkoxycarbonylimine cation (and salt) 8 and al-
kyl carbamate is the key step and is the rate-determining step of
the reaction (Scheme 5). Interestingly, compound 8 (Z = OAc) rep-
resents an ionic analogue of intermediates 5, which were proposed
earlier by Soroka.15
We propose that all the following steps are very fast and exert
no effect on the general rate of the reaction. Our data is in
agreement with the formation of proposed bisamides 4(R0=Alk)
(Scheme 2) followed by the formation of the N-acylimine cation
(analogue of N-alkoxycarbonylimine cation 8) (Scheme 5).11e
Surprisingly, the reaction of hydrophosphorylic compounds 6
and biscarbamates 4 does not occur at room temperature in acetic
acid. However, biscarbamates 4 and dimethyl phosphite or meth-
ylphosphonous acid 6 react to give the corresponding N-alkoxycar-
Additional studies are required for a more comprehensive
understanding of the role for proposed P–OAc intermediates 9 in
the mechanism for this reaction.
It is thought that complex 11 is a possible intermediate struc-
ture for the formation of the desired N-alkyloxycarbonyl-a-amin-
bonyl-a-aminoalkylphosphorylic compounds 7 in acetic anhydride
at room temperature (Scheme 6). In the latter case, acetic acid
oalkylphosphorylic compounds 7. Thus, we propose that the
three-component Oleksyszyn reaction with hydrophosphorylic
compounds, aldehydes and alkyl carbamates includes an Arbu-
zov-type reaction step (Scheme 8).
Also, we suggest a milder version of the undeservedly forgotten
procedure of Oleksyszyn14 with some minor modifications as an
improved method for the amidoalkylation of various hydropho-
sphorylic compounds in acetic anhydride/acetyl chloride. This
milder procedure may be of interest for the synthesis of N-pro-
formed during the reaction was the acidic catalyst.
These data suggest the formation of a more reactive intermedi-
ate (towards the alkoxycarbonylimine cation 8) than the starting
hydrophosphorylic compounds 6. In this connection, we propose
P–OAc derivatives 9 of trivalent phosphorus compounds as possi-
ble intermediates (Scheme 7) in the discussed three-component
reaction of alkyl carbamates, aldehydes and hydrophosphorylic
compounds 6 in acetic anhydride.
tected
a-aminophosphorylic compounds 7, which are promising
as substrates in combinatorial peptide synthesis.
R
Ac2O or AcCl
O
In conclusion, we have presented a new, milder version of the
H
N
NH2
+H+ from P-H compounds
OAlk
O
H
OAlk
procedure for the synthesis of N-protected
a-aminoalkylphosph-
+
O
C+
N
H
N
H
orylic compounds 7 by reaction of ethyl and methyl carbamates,
aldehydes and hydrophosphorylic compounds in acetic anhy-
dride/acetyl chloride, and new data on the mechanism of this
OAlk
O
AlkO
R
Z = OAc, Cl
Z-
4
8
Scheme 5. Formation of alkoxycarbonylimine salt 8 as a possible intermediate
from biscarbamate 4.
R
C+
X
H
N
R
OAlk
H
N
P
X
Y
Y
P+
O
Z--
O
H
AlkO
O
AlkO
R
O
H
XYPH(O) 6
OAlk
O
H
N
XYPH(O) 6
- AcZ
O
X
P
Y
O
O
9
7
N
H
O
N
H
8
Ac2O, r.t.
AcOH, r.t.
R
OAlk
4
7
Z = Cl, OAc
11
R = i-Bu, Ph; Alk = Me, Et ; X = Me, MeO; Y = MeO, OH
Scheme 6. Reactions of biscarbamates 4 and P–H compounds 6 in Ac2O.
Scheme 8. Proposed scheme for the Arbuzov-type reaction of acyl derivatives of PIII
compounds 9 and alkoxycarbonylimine salt 8.

2616
M. E. Dmitriev, V. V. Ragulin / Tetrahedron Letters 51 (2010) 2613–2616
reaction. We also isolated for the first time, N,N0-benzylidene- and
N,N0-alkylidenebiscarbamates (4, R0 = OAlk) as intermediates from
the reaction medium and have proposed a new version of the
mechanism for this reaction involving an Arbuzov-type reaction
step.
References and notes
1. Collinsova, M.; Jiracek, J. Curr. Med. Chem. 2000, 7, 629.
2. (a) Burchardt, J.; Meldal, M. J. J. Chem. Soc., Perkin Trans. 1 2000, 3306; (b)
Georgiadis, D.; Matziari, M.; Yiotakis, A. Tetrahedron 2001, 57, 3471.
3. Burchardt, J.; Ferreras, M.; Krog-Jensen, C.; Delaisse, J.-M.; Foged, N. T.; Meldal,
M. J. Chem. Eur. J. 1999, 5, 2877.
4. Georgiadis, D.; Matziari, M.; Vassiliou, S.; Dive, V.; Yiotakis, A. Tetrahedron
1999, 55, 14635.
5. (a) Ragulin, V. V. Russ. J. Gen. Chem. 2001, 71, 1823; (b) Ragulin, V. V. Russ. J. Gen.
Chem. 2004, 74, 1177; (c) Ragulin, V. V. Russ. J. Gen. Chem. 2007, 77, 861.
6. Ragulin, V. V. Russ. J. Gen. Chem. 2008, 78, 1655.
Acknowledgement
This study was supported by the Russian Foundation for Basic
Research (Grant No. 09-03-12157).
7. Rozhko, L. F.; Ragulin, V. V. Amino Acids 2005, 29, 139.
8. (a) Boyd, E. A.; Regan, A. C.; James, K. Tetrahedron Lett. 1992, 33, 813; (b)
Kurdyumova, N. R.; Rozhko, L. F.; Ragulin, V. V.; Tsvetkov, E. N. Russ. J. Gen.
Chem. 1997, 67, 1852.
9. (a) Baylis, E. K.; Campbell, C. D.; Dingwall, J. D. J. Chem. Soc., Perkin Trans. 1 1984,
2845; (b) Soroka, M.; Zygmunt, J. Synthesis 1988, 370; (c) Jiao, X.-Y.;
Verbruggen, C.; Borloo, M.; Bollaert, W.; De Groot, A.; Dommisse, R.;
Haemers, A. Synthesis 1994, 23.
Supplementary data
Detailed experimental procedures for the three-component
reaction of alkyl carbamates, hydrophosphorylic compounds and
aldehydes in acetic anhydride, isolation of intermediate biscarba-
mates from the reaction medium, synthesis of N,N0-alkylidenebisc-
arbamates in acetic anhydride, and investigation of the reaction of
pre-obtained N,N0-alkylidene- and N,N0-benzylidenebiscarbamates
4 and hydrophosphorylic compounds 6 in acetic anhydride with
addition of an acidic catalyst, as well as spectral and analytical
data, copies of 1H, 13C and 31P NMR and MS spectra for intermedi-
ate biscarbamates 4a–c and amidoalkylphosphorylic compounds
7a–k are available as Supplementary data, Parts I, II and III. Supple-
mentary data associated with this article can be found, in the on-
line version, at doi:10.1016/j.tetlet.2010.03.020.
10. Birum, G. H. J. Org. Chem. 1974, 39, 209.
11. (a) Oleksyszyn, J.; Tyka, R.; Mastalerz, P. Synthesis 1978, 479; (b) Oleksyszyn, J.;
Subotkowska, L.; Mastalerz, P. Synthesis 1979, 985; (c) Oleksyszyn, J. Synthesis
1980, 722; (d) Oleksyszyn, J.; Gruszecka, E.; Kafarski, P.; Mastalerz, P. Monatsh.
Chem. 1982, 113, 59; (e) Oleksyszyn, J. J. Prakt. Chem. 1987, 329, 19.
12. (a) Yuan, C.; Wang, G.; Chen, S. Synthesis 1990, 522; (b) Yuan, C.; Chen, S.;
Wang, G. Synthesis 1991, 490; (c) Yuan, C.; Chen, S. Synthesis 1992, 1124; (d)
Chen, S.; Coward, J. K. Tetrahedron Lett. 1996, 37, 4335; (e) Chung, S.-K.; Kang,
D.-H. Tetrahedron: Asymmetry 1996, 7, 21.
13. Matziari, M.; Yiotakis, A. Org. Lett. 2005, 7, 4049.
14. Oleksyszyn, J.; Gruszecka, E. Tetrahedron Lett. 1981, 22, 3537.
15. Soroka, M. Liebigs Ann. Chem. 1990, 1, 331.
16. (a) Zaugg, H. E.; Martin, W. B. Org. React. 1965, 14, 52; (b) Zaugg, H. E. Synthesis
1984, 181; (c) Gilbert, E. E. Synthesis 1972, 30.
17. Nifantiev, E. E.; Fursenko, I. V. Russ. Chem. Rev. 1970, 39, 1050.