Synthesis of α1-(Cbz-aminoalkyl)-α2- (hydroxyalkyl)phosphinic esters

Article abstract of DOI:10.1016/j.tetlet.2005.03.083

A synthesis of α₁-(Cbz-aminoalkyl)-α₂- (hydroxyalkyl)phosphinic esters was achieved by the 1,2-addition of the appropriate aldehyde to Cbz-protected phosphinic analogues of amino acid esters in the presence of at least three equivale

Full text of DOI:10.1016/j.tetlet.2005.03.083

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 3359–3362
Synthesis of a1-(Cbz-aminoalkyl)-a2-(hydroxyalkyl)phosphinic
esters
*
Marcin Drag and Jozef Oleksyszyn
´
Institute of Organic Chemistry, Biochemistry and Biotechnology, Wroclaw University of Technology,
Wybrzeze Wyspianskiego 27, 50-370 Wroclaw, Poland
Received 25 January 2005; revised 8 March 2005; accepted 15 March 2005
Available online 2 April 2005
Abstract—A synthesis of a1-(Cbz-aminoalkyl)-a2-(hydroxyalkyl)phosphinic esters was achieved by the 1,2-addition of the appropri-
ate aldehyde to Cbz-protected phosphinic analogues of amino acid esters in the presence of at least three equivalents of trimethylsilyl
chloride and NEt3. The complete deprotection of the product esters could be achieved in one step using 35% HBr in acetic acid.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Phosphinic peptides and pseudopeptides are interesting
classes of compounds possessing broad biological activi-
ties. These compounds have attracted particular interest
as inhibitors of transition state analogues of Zn-
metalloproteases.1,2
atht e a2 carbon atom (Scheme 1). Such structures rep-
resent very close structural analogues of bestatine, where
the OH moiety is replaced by the more acidic P–OH
fragment. We suggest that this modification will
improve binding to the zinc ion in the active center of
aminopeptidase N.
In our investigations of the synthesis of new inhibitors
of aminopeptidase N (CD13), which, according to some
recentreports, may be implicated in the process of pro-
grammed cell death of cancer cells, we have focused on
analogues of bestatine, a potent inhibitor of aminopep-
tidase N and an apoptosis inducer (Scheme 1).3–5
There are no previous reports of the synthesis of a1-
aminoalkyl-a2-hydroxyalkylphosphinates. There is one
example of the preparation of a similar compound,
which possesses as an additional amino group at the
b2 position. This symmetrical pseudodipeptide was syn-
thesized in 47% yield by a classical condensation of the
Cbz-protected phosphinic analogue of phenylalanine
with Cbz-phenylalaninal in the presence of NEt3. After
appropriate derivatization, the product proved to be a
potent inhibitor of protease HIV.6,7
One group of such compounds, a1-aminoalkyl-a2-
hydroxyalkylphosphinates 6, possess an amino group
attached to the a1 carbon atom and a hydroxy group
Herein we report a general synthetic approach to the
synthesis of esters of a1-(Cbz-aminoalkyl)-a2-hydroxy-
alkyl phosphinic acids 5 starting from Cbz-protected
phosphinic analogues 4 of amino acids esters, prepared
according to well established procedures, in a few steps
O
OH
R1
R3
O
P
H2N
NH
COOH
NH2 OH
OH
from the appropriate aldehyde and benzhydrylamine
8,9
saltwiht hypophosphorous acid.
Deprotection of 1
Bestatine
6
in 40% HBr resulted in phosphinic amino acid analogue
2.8 Subsequent protection with a Cbz group gave phos-
phinic acid 3, which was converted into the desired ester
4 by action with the appropriate alcohol in the presence
of DCC and DMAP in THF (Scheme 2).8,9
R1, R3 = alkyl, aryl
Scheme 1. Structure of bestatine and its phosphinic analogues 6.
Keywords: Phosphinates; Phosphinic esters; Addition.
*
In the final step, the desired esters 5 were obtained by a
modified procedure, originally described by Thottathil
Corresponding author. Tel.: +48 71 320 24 55; fax: +48 71 328 40
64; e-mail: marcin.drag@pwr.wroc.pl
0040-4039/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.03.083

3360
M. Drag, J. Oleksyszyn / Tetrahedron Letters 46 (2005) 3359–3362
O
O
P
R1
Ph
OH
P
O
H
Ph
Ph
R1
OH
R1
H
EtOH
1. 40% HBr, reflux, 1 h
2. EtOH, pH=6
Cbz-Cl, pH=10
water/dioxane, 24 h
H2PO2 H3N
H
reflux, 5 h
NH
NH2
Ph
1
2
O
O
OR2
O
OR2
R1
OH
R2OH, DCC, DMAP
THF, 4 h
R3CHO, Si(CH3)3Cl, NEt3
CH2Cl2, 24 h
R1
P
R1
R3
P
P
H
H
NH
OR4
NH
NH
Cbz
Cbz
Cbz
3
4
5
R1=Ph, CH2CH(CH3)2, CH(CH3)2; R2=Et, CH2Ph; R3=Ph, (CH2)3CH3, (CH2)2SCH3, (CH2)2Ph; R4=H, Si(CH3)3
Scheme 2. Preparation of compounds 1–5.
et al. for the 1,2-addition of crotonaldehyde to the ethyl
ester of 4-phenylbutylphosphinic acid.10 In our case the
1,2-addition of the appropriate aldehyde to the Cbz-pro-
tected phosphinic analogues 4 in the presence of at least
three equivalents of trimethylsilyl chloride and NEt3 in
dichloromethane afforded the desired products in 70–
86% yields.11 The application of a large excess of
trimethylsilyl chloride and NEt3 resulted in complete
conversion of the phosphinic substrate (Scheme 2), which
could not be achieved by application of the conditions
described in the original publication. Conversion to 5
was about 70–90%, and the reaction was accompanied
by a side product, namely Cbz-protected phosphinic
acid (about 10–30% according to the 31P NMR spectra
of the crude reaction mixture), which could be easily re-
moved by washing with 5% aqueous NaHCO3. Purifica-
tion by column chromatography gave the desired
product 5.12 Interestingly, in some cases (Table 1) t he
isolated product had a trimethylsilyl protected 5a and
5d derivates of protected hydroxy group, as confirmed
prising considering the aqueous work-up and the ex-
pected low stability of the silyl ethers derivatives 5a
and 5d to those conditions. This observation gave addi-
tional confirmation for the formation of a hydroxy
group during the 1,2-addition of aldehydes to the N-
Cbz-protected phosphinic esters 4 and further confirmed
the structure of 5.14
Full deprotection of compounds 5 could be easily
accomplished by the action of 35% hydrogen bromide
in acetic acid during 1 h at room temperature giving
a1-(aminoalkyl)-a2-(hydroxyalkyl)phosphinic acids 6 as
a mixture of two pairs of diastereomers, as exemplified
for 5c (Scheme 3).15,16
Recently, we have synthesized a large collection of com-
pounds 6, which were tested for their inhibitory activity
toward aminopeptidase N (APN). Some of the tested
compounds showed a significanteffectotwards APN
as well as inducing apoptosis in vitro in some cancer cell
lines. Currently we are studying the stereoselective
1
13
by H NMR, C NMR and ESI analysis.13 This is sur-
Table 1. Structural and synthetic data for compounds 5
Entry
R1
R2
R3
R4
Yield (%)
5a
5b
5c
5d
5e
5f
(CH3)2CHCH2
(CH3)2CHCH2
(CH3)2CH
C6H5
CH2Ph
CH2CH3
CH2Ph
CH2CH3
CH2Ph
CH2Ph
CH2Ph
CH2CH2Ph
CH2CH2SCH3
CH2CH2Ph
CH3
Si(CH3)3
H
71
80
75
73
86
86
85
H
Si(CH3)3
C6H5
C6H5
C6H5
C6H5
CH2CH2Ph
CH2CH2CH2CH3
H
H
H
5g
O
O
O
OH
P
P
1. 35% HBr / AcOH / 1 h, r.t.
2. MeOH / propylene oxide
NH OH
NH2 OH
Cbz
5c
6c
Scheme 3. An example of the deprotection of compounds 5.

M. Drag, J. Oleksyszyn / Tetrahedron Letters 46 (2005) 3359–3362
3361
MW = 581.8; MS (ESI) = 582.7 (M)+, 604.8 (M+Na)+,
620.8 (M+K)+.
synthesis of diastereomers of 6 and any results will be
published separately together with the biological studies,
in due course.17
Compound 5b: (beige semi-solid); 31P NMR (CDCl3,
121.5 MHz) d: 49.68 (s, 42%), 49.84 (s, 35%), 50.72 (s,
11%) and 51.25 (s, 12%); H NMR (CDCl3, 300.13 MHz)
1
d: 0.93 (m, 6H, 2 · CH3), 1.26 (m, 3H, OCH2CH3), 1.34–
2.00 (m, 5H, CH2, CH2, CH), 2.068, 2.089, 2.095 (3 · s,
3H, SCH3), 2.63–2.76 (m, 2H, SCH2), 4.08–4.25 (m, 5H,
NCHP, OCHP, OCH2CH3, NH), 5.09–5.23 (m, 3H,
CH2OCO, OH), 7.28–7.36 (m, 5H, ArH). MW = 417.4;
MS (ESI) = 417.8 (M)+, 439.9 (M+Na)+, 455.9 (M+K)+.
Compound 5c: (white semi-solid); 31P NMR (CDCl3,
121.5 MHz) d: 50.75 (s, 39%), 50.92 (s, 41%), 51.71 (s,
14%) and 52.41 (s, 6%); 1H NMR (CDCl3, 300.13 MHz) d:
0.91–1.27 (m, 6H, 2 · CH3), 1.87–2.05 (m, 3H, CH2, CH),
2.61–294 (m, 2H, CH2), 3.76–4.19 (m, 2H, NCHP,
OCHP), 5.00–5.16 (m, 5H, CH2OCO, OCH2 Ph, NH),
7.13–7.35 (m, 15H, ArH). MW = 495.5; MS (ESI) = 495.6
(M)+, 517.7 (M+Na)+, 533.6 (M+K)+.
References and notes
1. Dive, V.; Georgiadis, D.; Matziari, M.; Makaritis, A.;
Beau, F.; Cuniasse, P.; Yiotakis, A. Cell. Mol. Life Sci.
2004, 61, 1–10.
2. Yiotakis, A.; Georgiadis, D.; Matziari, M.; Makaritis, A.;
Dive, V. Curr. Org. Chem. 2004, 8, 1135–1158.
3. Scornik, O. A.; Botbol, V. Curr. Drug Metabolism 2001, 2,
67–85.
4. Ishi, K.; Usui, S.; Sugimura, Y.; Yoshida, S.; Hioki, T.;
Tatematsu, M.; Yamamoto, H.; Hirano, K. Int. J. Cancer
2001, 92, 49–54.
5. Sekine, K.; Fujii, H.; Abe, F. Leukemia 1999, 13, 729–734.
6. Stowasser, B.; Budt, K.-H.; Jian-Qi, L.; Peyman, A.;
Ruppert, D. Tetrahedron Lett. 1992, 33, 6625.
7. Our preliminary experiments on the condensation of Cbz-
Leu-P(O)(OEt)H with benzaldehyde in the presence of
NEt3 gave unsatisfactory results, repetition of the reaction
in the presence of CH2Cl2 also failed. Improved results
were achieved by condensation in the presence of KF in
DMF/CHCl3, but this method resulted in tedious separa-
tion of the desired product from the mixture.
Compound 5d: (yellow oil); 31P NMR (CDCl3,
121.5 MHz) d: 46.06 (s, 15%), 46.27 (s, 42%), 47.39 (s,
25%) and 47.87 (s, 18%); H NMR (CDCl3, 300.13 MHz)
1
d: 0.16 (m, 9H, 3 · CH3), 1.02 (t, 3H, CH3 (one diaste-
reomer), J = 7.1 Hz), 1.19–1.38 (m, 6H, 2 · CH3, remain-
ing three diastereomers (3 · t) and CH3 (4 · dd)), 3.86–
4.18 (m, 3H, OCHP, OCH2CH3), 5.02–5.33 (m, 3H,
CH2OCO, NCHP), 6.32 (m, 1H, NH), 7.28–7.47 (m, 10H,
ArH). MW = 449.5; MS (ESI) = 450.1 (M)+, 472.1
(M+Na)+, 488.2 (M+K)+.
8. Baylis, E. K.; Campbell, C. D.; Dingwall, J. G. J. Chem.
Soc., Perkin Trans. 1 1984, 2845–2853.
9. Karanewsky, D. S.; Badia, M. C. Tetrahedron Lett. 1986,
27, 1751–1754.
Compound 5e: (beige semi-solid), 31P NMR (CDCl3,
121.5 MHz) d: 43.88 (s, 45%), 44.49 (s, 15%), 45.39 (s, 8%)
and 45.55 (s, 32%); 1H NMR (CDCl3, 300.13 MHz) d:
4.64–5.10 (m, 7H, OCHP, OH, CH2OCO, OCH2Ph,
NCHP), 7.03–7.50 (m, 20H, ArH). MW = 501.5; MS
(ESI) = 502.5 (M)+, 524.5 (M+Na)+, 540.4 (M+K)+.
Compound 5f: (white semi-solid); 31P NMR (CDCl3,
121.5 MHz) d: 47.78 (s, 47%, two tightly overlapping
10. Thottathil, J. K.; Ryono, D. E.; Przybyla, C. A.; Moniot,
J. L.; Neubeck, R. Tetrahedron Lett. 1984, 25, 4741–4744.
11. General procedure for the synthesis of compounds 5. In a
round bottom flask equipped with a magnetic stirrer was
dissolved 2 mmol of the appropriate Cbz-protected phos-
phinic acid ester 4 in 15 mL of dry CH2Cl2. After cooling
the mixture in an ice bath to 0 ꢁC, dry triethylamine
(6 mmol, 0.83 mL) and trimethylsilyl chloride (6 mmol,
0.75 mL) were added. After 0.5 h, 4 mmol of the appro-
priate aldehyde was added to the mixture. The reaction
was left tightly closed for 24 h. After completion of the
reaction, 3 mL of water was added. After 5 min, 20 mL of
dichloromethane was added and the organic phase was
separated, washed with 5% NH4Claq (10 mL), water
(10 mL), brineaq (10 mL), then dried (MgSO4) and evapo-
rated in vacuo. Column chromatography on silica gel (70–
200 mesh: 70 g) using hexane:AcOEt(1:1) gave the desired
compound 5.
1
resonance signals), 48.06 (s, 34%) and 48.76 (s, 19%); H
NMR (CDCl3, 300.13 MHz) d: 1.82–2.00 (m, 2H, CH2),
2.58–2.98 (m, 2H, CH2), 3.76–4.40 (m, 2H, OCHP, OH),
4.72–5.11 (m, 4H, CH2OCO, OCH2Ph), 5.40 (m, 1H,
NCHP), 7.05–7.35 (m, 20H, ArH); MW = 529.6; MS
(ESI) = 530.5 (M)+, 568.5 (M+K)+.
Compound 5g: (beige semi-solid); 31P NMR (CDCl3,
121.5 MHz) d: 47.67 (s, 51%), 47.82 (s, 11%), 48.27 (s,
22%) and 48.72 (s, 16%); H NMR (CDCl3, 300.13 MHz)
1
d: 0.85 (m, 3H, CH3), 1.18–1.96 (m, 6H, 3 · CH2), 3.44–
4.35 (m, 3H, OCHP, OH), 4.73–5.15 (m, 4H, CH2OCO,
OCH2Ph), 5.37 (m, 1H, NCHP), 7.28–7.36 (m, 15H, ArH).
MW = 481.5; MS (ESI) = 482.2 (M)+, 504.4 (M+Na)+,
520.4 (M+K)+.
12. The isolated compounds 4and 5 always contained up to 1–
3% by mass of DCU (N,N-dicyclohexyl urea—omitted
from the 1H NMR data), which remained from the
esterification of 3. Our attempts to eliminate this side
product by several re-crystallizations from dichlorometh-
ane or hexane/AcOEt(1:1) and column chromaot graphy
were only partially successful.
15. 1 mmol of the appropriate compound 5 was treated with
1 mL of 35% HBr in acetic acid at rt for 1 h. The volatile
products were removed in vacuo and the residue treated
twice with dry Et2O, which was decanted from the
mixture. The remaining oil was dissolved in 2 mL of
MeOH and the pH was adjusted to 6 by addition of
propylene oxide. Precipitation started after cooling the
mixture in a refrigerator. If the compound did not
precipitate after 3 h, to the mixture was added 2 mL of
dry Et2O and the mixture was placed in the refrigerator
again. Filtration and washing with dry Et2O gave the
desired compound 6 in almost quantitative yield.
13. The 31P NMR spectra of non-silylated compounds 5
usually revealed the presence of small (1–2%) quantities of
silylated derivatives (omitted from the 31P NMR data).
The susceptibility of these compounds to silylation and the
stability of the silyl derivatives are unclear.
14. Compound 5a: (yellow oil); 31P NMR (CDCl3,
121.5 MHz) d: 49.69 (s, 34%), 50.20 (s, 13%), 50.16 (s,
15%) and 50.88 (s, 38%); 1H NMR (CDCl3, 300, 13 MHz)
d: 0.17 (m, 9H, 3 · CH3), 0.95 (m, 6H, 2 · CH3), 1.53–2.26
(m, 5H, 2 · CH2, CH), 2.52–2.87 (m, 2H, CH2 Ph), 3.87–
4.54 (m, 2H, NCHP, OCHP), 5.02–5.24 (m, 5H,
CH2OCO, OCH2Ph, NH), 7.20–7.37 (m, 15H, ArH).
Compound 6c: (white powder): yield 94%, mp = 233.5–
235.5 ꢁC, 31P NMR (D2O, 121.5 MHz) d: 39.26 (s, 49%)
1
and 39.66 (s, 51%); H NMR (D2O, 300.13 MHz) d: 0.91
(t, 6H, 2 · CH3, J = 6.8 Hz, diastereomeric), 0.93 (t, 6H,
2 · CH3, J = 6.75 Hz, diastereomeric), 1.78–1.94 (m, 2H,
CH2 CH2Ph), 2.11–2.19 (m, 1H, CH), 2.50–2.60 and

3362
M. Drag, J. Oleksyszyn / Tetrahedron Letters 46 (2005) 3359–3362
1
2.71–2.78 (m, 2H, CH2CH2Ph), 3.19 (dd, 1H, NCHP,
meric), 69.31 (d, JCP = 73.0 Hz, OCHP, diastereomeric),
126.3 (s, arom.), 128.63 (s, ArH), 128.72 (s, ArH), 141.30
(s, ArH); IR (KBr, m [cmÀ1]) 1018.75 (s, P–OH), 1052.0 (s,
C–O), 1161 (s, P@O), 1613.9 (m, N–H), 2872.1–3026.8
(ms, C–H), 3183 (ms, O–H).
3
2JHP = 9.9 Hz, JHH = 6.1 Hz, diastereomeric), 3.33 (dd,
2
3
1H, NCHP, JHP = 4.9 Hz, JHH = 4.9 Hz, diastereo-
meric), 3.70–3.79 (m, 1H, OCHP), 7.07–7.23 (m, 5H,
ArH); 13C NMR (D2O, 75.46 MHz) d: 17.20 (d,
J = 3.9 Hz, CH3, rotamer), 17.84 (d, J = 5.6 Hz, CH3,
rotamer), 19.47 (d, J = 8.2 Hz, CH3, rotamer), 19.67 (d,
J = 6.4 Hz, CH3, rotamer), 26.7 (d, J = 7.1, CH2CH2Ph),
30.64 (s, CH2CH2Ph), 30.76 (s, CH2CH2Ph), 30.93 (s,
16. The deprotection of 5 did notresultin exchange of hte
hydroxy group by bromine, as confirmed by H and 13C
1
NMR analysis. Our efforts to deprotect 5 in boiling 5 M
HCl failed because of partial cleavage of the C–P
bond.
1
CH), 31.12 (s, CH), 53.15 (d, JCP = 91.1 Hz, NCHP,
diastereomeric), 53.49 (d, JCP = 90.7 Hz, NCHP, diaste-
1
17. This work is the subject of patent applications—P370190
and P372057.
reomeric), 67.78 (d,1JCP = 72.5 Hz, OCHP, diastereo-