A stereoselective synthesis of phosphinic acid phosphapeptides corresponding to glutamyl-γ-glutamate and incorporation into potent inhibitors of folylpoly-γ-glutamyl synthetase

A Stereoselective Synthesis of Phosphinic Acid Phosphapeptides

Corresponding to Glutamyl-γ-glutamate and Incorporation into

Potent Inhibitors of Folylpoly-γ-glutamyl Synthetase

David M. Bartley and James K. Coward*

Departments of Medicinal Chemistry and Chemistry, University of Michigan,

Ann Arbor, Michigan, 48109-1055

jkcoward@umich.edu

Received April 13, 2005

Radical addition of H₃PO₂to N-/C-protected vinyl glycine led to the corresponding H-phosphinic

acid in excellent yield. The non-nucleophilic H-phosphinic acid was converted to a nucleophilic P^III

species, RP(OTMS)₂, which was used in two approaches to the target phosphinic acid containing

pseudopeptide. New methodology was developed that led to excellent yields in the reaction of

RP(OTMS)₂with unactivated electrophiles, including an acyclic homoallylic bromide. However, en

route to the target pseudopeptide, Arbuzov reaction of RP(OTMS)₂with a cyclic homoallylic bromide,

(R)-3-(bromomethyl)-cyclopent-1-ene, led to a rearranged allylic phosphinic acid rather than the

desired homoallylic derivative, a putative glutarate surrogate. Conjugate addition of RP(OTMS)₂

to R-methylene glutarate containing a chiral auxiliary resulted in only modest diastereoselectivity.

Purification by flash chromatography provided protected derivatives of both diastereomers of the

pseudopeptide. Following global deprotection, coupling of (S)-H-Glu-γ-[Ψ(P(O)(OH)(CH₂))]-(S)-Glu-

OH and (S)-H-Glu-γ-[Ψ(P(O)(OH)(CH₂))]-(R)-Glu-OH to (4-amino-4-deoxy-10-methyl)pteroyl azide

led to the target compounds for biochemical study as inhibitors of the ATP-dependent ligase,

folylpoly-γ-glutamate synthetase.

Introduction

expected to be cytotoxic. In addition, one known mech-

anism of antifolate resistance is down regulation of FPGS

activity,^4,5and cells with this type of resistance should

be inherently sensitive to FPGS inhibition.⁶FPGS re-

quires a heterocycle for substrate binding. With the

glutamate ligation site being distal to the heterocycle,

we^7,8and others^9-16have shown it is possible to incor-

porate known heterocyclic inhibitors of other folate-

Folylpoly-γ-glutamate synthetase (FPGS, EC 6.3.2.17)

catalyzes the ATP-dependent polyglutamylation of di-

etary folates and antifolate cancer therapeutics (Scheme

1). The poly-γ-glutamyl metabolites are important cofac-

tors in one-carbon metabolism and are essential for cell

growth and survival.¹Folylpoly-γ-glutamyl derivatives

aid in cellular retention of reduced folate cofactors and

are better substrates for many of the folate-dependent

enzymes employed in one-carbon metabolism.^2,3

(4) Liani, E.; Rothem, L.; Bunni, M. A.; Smith, C. A.; Jansen, G.;

Assaraf, Y. G. Int. J. Cancer 2003, 103, 587-599.

(5) Roy, K.; Egan, M. G.; Sirlin, S.; Sirotnak, F. M. J. Biol. Chem.

1997, 272, 6903-6908.

FPGS is an attractive target for anticancer drug design

for many reasons. As noted above, FPGS is required for

cell viability, and therefore inhibition of this enzyme is

(6) McGuire, J. J.; Coward, J. K. Drugs Future 2003, 28, 967-974.

(7) Tsukamoto, T.; Haile, W. H.; McGuire, J. J.; Coward, J. K. Arch.

Biochem. Biophys. 1998, 355, 109-118.

* To whom correspondence should be addressed. Phone: 734-936-

2843. Fax: 734-647-4865.

(8) Valiaeva, N.; Bartley, D.; Konno, T.; Coward, J. K. J. Org. Chem.

2001, 66, 5146-5154.

(1) McBurney, M. W.; Whitmore, G. F. Cell 1974, 2, 173-182.

(2) McGuire, J. J.; Coward, J. K. In Folates and Pterins; Blakley,

R. L., Benkovic, S. J., Eds.; Wiley: New York, 1984; Vol. 1, pp 135-

190.

(9) Rosowsky, A.; Forsch, R. A.; Freisheim, J. H.; Moran, R. G.; Wick,

M. J. Med. Chem. 1984, 27, 600-604.

(10) Cichowicz, D. J.; Cook, J.; George, S.; Shane, B. In Proceedings

of the Second Workshop on Folyl- and Antifoyl-polyglutamates; Gold-

man, I. D., Ed.; Praeger Publishers: New York, 1985; pp 7-13.

(3) Shane, B. Vitam. Horm. 1989, 45, 263-335.

Published on Web 07/19/2005

J. Org. Chem. 2005, 70, 6757-6774

6757

Bartley and Coward

SCHEME 1

dependent enzymes into the design of FPGS inhibitors.

This inhibitor design may allow for the synergistic effect

of inhibiting two enzymes in the folate pathway with one

drug. Recently, Alimta, a multitargeted antifolate (MTA)

known to inhibit dihydrofolate reductase, thymidylate

synthetase, and glycineamide ribonucleotide formyltrans-

ferase,¹⁷has been approved for clinical use in the United

States. In addition to the clinical importance of FPGS

inhibitors, the design of cell-permeable compounds would

allow for investigations on the control of intracellular

folate cofactor levels by FPGS. Several other proteins are

involved in the uptake and/or regulation of intracellular

folate concentrations, such as γ-glutamyl hydrolase, the

enzyme that cleaves the poly-γ-glutamyl tail,¹⁸in addition

to the folate transporter and reduced folate carrier that

aid in cellular uptake of folates and antifolates.¹⁹Al-

though each of these proteins has been studied in great

detail individually, little is known about how the expres-

sion levels or levels of activity of these proteins are

interrelated or controlled in terms of normal metabolism

or in development of drug resistance to antifolates.

Our laboratory has previously reported the use of

phosphorus-containing pseudopeptides 1 as inhibitors of

FPGS.^7,8The phosphorus moiety mimics the tetrahedral

intermediate²⁰in FPGS catalysis (Scheme 1), and these

compounds are very potent inhibitors of FPGS. The

phosphonate analogue 1a (2′S,2′′S) inhibits human FPGS

with K_i) 46 nM.⁷More recently, we have published the

synthesis and biological evaluation of racemic phosphi-

nate analogues with the most potent being the MTX-

phosphinate 1b (K_i) 3.1 nM).^8,21This phosphinate is a

mixture of racemic diastereomers (2′RS,2′′RS) and on the

basis of the known stereospecificity of FPGS,²it is likely

that only the 2′S,2′′S diastereomer inhibits the enzyme

and therefore the actual inhibition constant may be

subnanomolar. To determine if this hypothesis is correct,

a stereoselective synthesis of the phosphinate pseudo-

peptide 2 has been investigated.

The most common methods for the generation of

phosphorus-carbon (P-C) bonds are often not feasible

with highly functionalized molecules.²²Previous efforts

to synthesize highly functional phosphorus-containing

pseudopeptides of type 2 led to the realization that the

best method for the synthesis of the P-C bonds was the

in situ generation of P^IIIspecies and their addition to

alkyl halides or Michael acceptors.⁸Unfortunately, the

(11) Rosowsky, A.; Freisheim, J. H.; Bader, H.; Forsch, R. A.; Susten,

S. S.; Cucchi, C. A.; Frei, E. J. Med. Chem. 1985, 28, 660-667.

(12) Rosowsky, A.; Freisheim, J. H.; Moran, R. G.; Solan, V. C.;

Bader, H.; Wright, J. E.; Radike-Smith, M. J. Med. Chem. 1986, 29,

655-660.

(13) Rosowsky, A.; Moran, R. G.; Freisheim, J. H.; Bader, H.; Forsch,

R. A.; Solan, V. C. NCI Monogr. 1987, 5, 145-152.

(14) Rosowsky, A.; Forsch, R. A.; Moran, R. G.; Kohler, W.; Fre-

isheim, J. H. J. Med. Chem. 1988, 31, 1326-1331.

(15) Patil, S. A.; Shane, B.; Freisheim, J. H.; Singh, S. K.; Hynes, J.

B. J. Med. Chem. 1989, 32, 1559-1565.

(16) Hynes, J. B.; Singh, S. K.; Fetzer, O.; Shane, B. J. Med. Chem.

1992, 35, 4078-4085.

(17) Shih, C.; Chen, V. J.; Gossett, L. S.; Gates, S. B.; Mackellar,

W. C.; Habeck, L. L.; Shackelford, K. A.; Mendelsohn, L. G.; Soose, D.

J.; Patel, V. F.; Andis, S. L.; Bewley, J. R.; Rayl, E. A.; Moroson, B. A.;

Beardsley, G. P.; Kohler, W.; Ratnam, M.; Schultz, R. M. Cancer Res.

1997, 57, 1116-1123.

(18) Galivan, J.; Ryan, T. J.; Chave, K.; Rhee, M.; Yao, R.; Yin, D.

Pharmacol. Ther. 2000, 85, 207-215.

(19) Jansen, G. In Antifolate Drugs in Cancer Therapy; Jackman,

A. L., Ed.; Humana Press: Totowa, New Jersey, 1999; pp 293-321.

(20) Banerjee, R. V.; Shane, B.; McGuire, J. J.; Coward, J. K.

Biochemistry 1988, 27, 9062-9070.

(21) McGuire, J. J.; Haile, W. H.; Valiaeva, N.; Bartley, D.; J., G.;

Coward, J. K. Biochem. Pharmacol. 2003, 65, 315-318.

(22) Collinsova, M.; Jiracek, J. Curr. Med. Chem. 2000, 7, 629-647.

6758 J. Org. Chem., Vol. 70, No. 17, 2005

Synthesis of Phosphinic Acid Phosphapeptides

SCHEME 2

SCHEME 3

current literature procedures for synthesizing P-C bonds

with P^IIIintermediates using alkyl halides are limited

to alkyl halides activated by neighboring electron-

withdrawing groups^8,23,24and are ineffective on unacti-

vated alkyl halides unless forcing conditions are used

(i.e., hexamethyl disilazane (HMDS), 110 °C). Recently,

Liu et al. have shown that HMDS conditions result in

racemization of amino acid stereocenters,²⁵and often

even these harsh conditions result in rather low

yields.^8,26-29Boyd and Regan reported that the reaction

proceeds in good yield at room temperature regardless

of the nature of the electrophile,³⁰but there are no other

examples of this in the literature. The lack of suitable

methodology for the introduction of carbon-phosphorus

bonds into highly functionalized molecules has led us to

investigate new protocols for the formation of these

bonds.

containing amino acids (eq 1, 7, X ) O), and we have

used it to set the N-terminal stereocenter in phosphonate

pseudopeptide 9 (Scheme 3),³⁸a precursor of 1a. Suc-

cessful use of the â-bromoalkyl phosphonate 8 as a

precursor of an R,â-unsaturated phosphonate (eq 1, 6, X

) O, R¹) R²) Et) was encouraging in terms of the

proposed strategy for synthesis of the corresponding

phosphinate pseudopeptide 2.

Results and Discussion

The installation of various P-O esters, R²(eq 1), either

prior to or following formation of the new C-terminal P-C

bond (X ) CH₂), and the ability to use a variety of alkyl

groups, R¹, most notably a glutarate surrogate such as 5

(Scheme 2), was appealing. As will be discussed below

in more detail, stereoselective synthesis of the R-isomer

of the homoallylic alcohol, 3-(hydroxymethyl)-cyclopen-

tene (ent-5, X ) OH), was effected by the method of

Maeda and Innouye.³⁹Conversion of the alcohol to

several cyclopentene-based homoallylic electrophiles (ent-

5, X ) OMs, OTs, etc.) could be effected by standard

methods (see Supporting Information for details). In

addition, the reaction of an acyclic homoallylic bromide

with BTSP was investigated.

Unfortunately, all efforts to effect formation of the

desired P-C bond using standard methodology involving

in situ generation of BTSP from HMDS and ammonium

hypophosphite (AHP)³⁰followed by reaction with any of

these homoallylic electrophiles led only to very low yields

of the desired product, the bis-TMS ester precursor of 6

(eq 1, X ) CH₂).³⁸Surprisingly, the homoallylic electro-

philes ent-5 (X ) OMs, OTs) were recovered unchanged

Initially, this research focused on the use of Schollko-

pf’s bis-lactim ether 3³¹in the synthesis of phosphinate

pseudopeptide 2 (Scheme 2). This approach provides a

stereoselective route to the N-terminal amino acid of the

pseudopeptide and involves phosphinic acid synthon 4,

containing both the N- and C-terminal P-C bonds of 2.

The phosphinic acid would, in turn, be synthesized via

reaction of the nucleophilic P^IIIreagent, (TMSO)₂PH, bis-

(trimethylsilyl)-phosphonite (BTSP), and a homoallylic

electrophile 5, containing the carbon backbone of the

C-terminal glutaric acid moiety of 2. Reaction of the

resulting H-phosphinic acid with 1,2-dibromoethane,

again via a bis-TMS derivative, followed by esterification

and dehydrohalogenation would provide 4.⁸

The bis-lactim heterocycle 3 has been used by several

research groups^32-37in the synthesis of phosphorus-

(23) Thottathil, J. K.; Przybyla, C. A.; Moniot, J. L. Tetrahedron Lett.

1984, 25, 4737-4740.

(24) Deprele, S.; Montchamp, J.-L. J. Org. Chem. 2001, 66, 6745-

6755.

(25) Liu, X.; Hu, E.; Tian, X.; Mazur, A.; Ebetino, F. H. J. Organomet.

Chem. 2002, 646, 212-222.

(26) Zeng, B.; Wong, K. K.; Pompliano, D. L.; Reddy, S.; Tanner, M.

E. J. Org. Chem. 1998, 63, 10081-10086.

(33) Shapiro, G.; Beuchler, D.; Ojea, V.; Pombo-Villar, E.; Ruiz, M.;

Weber, H. P. Tetrahedron Lett. 1993, 34, 6255-6258.

(34) Ojea, V.; Ruiz, M.; Shapiro, G.; Pombo-Villar, E. Tetrahedron

Lett. 1994, 35, 3273-3276.

(27) Jones, P. B.; Parrish, N. M.; Houston, T. A.; Stapon, A.; Bansal,

N. P.; Dick, J. D.; Townsend, C. A. J. Med. Chem. 2000, 43, 3304-

3314.

(28) An, H.; Wang, T.; Maier, M. A.; Manoharan, M.; Ross, B. S.;

Cook, P. D. J. Org. Chem. 2001, 66, 2789-2801.

(29) Miller, D. J.; Hammond, S. M.; Anderluzzi, D.; Bugg, T. D. H.

J. Chem. Soc., Perkins Trans. 1 1998, 1, 131-142.

(30) Boyd, E. A.; Regan, A. C. Tetrahedron Lett. 1994, 35, 4223-

4226.

(31) Schollkopf, U.; Busse, U.; Lonsky, R.; Hinrichs, R. Liebigs Ann.

Chem. 1986, 2150-2163.

(32) Schick, A.; Kolter, T.; Giannis, A.; Sandhoff, K. Tetrahedron

1995, 51, 11207-11218.

(35) Ruiz, M.; Ojea, V.; Shapiro, G.; Weber, H. P.; Pombo-Villar, E.

Tetrahedron Lett. 1994, 35, 4551-4554.

(36) Ojea, V.; Fernandez, M. C.; Ruiz, M.; Quintela, J. M. Tetrahe-

dron Lett. 1996, 37, 5801-5804.

(37) Ojea, V.; Conde, S.; Ruiz, M.; Fernandez, M. C.; Quintela, J.

M. Tetrahedron Lett. 1997, 38, 4311-4314.

(38) Bartley, D. M. Ph.D. Dissertation, University of Michigan, Ann

Arbor, MI, 2004.

(39) Maeda, K.; Inouye, Y. Bull. Chem. Soc. Jpn. 1994, 67, 2880-

2882.

J. Org. Chem, Vol. 70, No. 17, 2005 6759

Bartley and Coward

SCHEME 4

SCHEME 5

TABLE 1. Oxidation/Elimination of 14

and a yellow-orange solid was formed during the reaction.

A review of the literature found very few reports of this

byproduct.^24,40However, Voronkov et al.⁴⁰reported the

solid to be a polymer of “high phosphorus and silicon

content” presumably formed from the polymerization of

BTSP. The low yields observed in our research and

possibly those encountered in this type of reaction

throughout the organophosphorus literature may be due

to the complete consumption of BTSP as a result of a

competing polymerization reaction prior to completion of

P-C bond formation. To test this hypothesis, the acyclic

homoallylic bromide, 4-bromo-1-butene, was treated with

BTSP in toluene at reflux temperature (110 °C) for 60 h.

Only 46% conversion to the bis-TMS precursor of 6 (R₁

) allyl) occurred with concomitant formation of a large

amount of orange solid. However, in a subsequent reac-

tion, when a second aliquot of BTSP was added to the

reaction mixture after 8 h at reflux temperature and the

reaction was allowed to proceed for an additional 18 h

at reflux temperature, the desired product was isolated

in 92% yield as its methyl ester after treatment with

CH₂N₂(see Supporting Information for details). These

results suggested that a more complex phosphite,

R₁CH₂-P(OTMS)₂, derived from the N-terminal portion

of the phosphinate-containing pseudopeptide 2, could be

employed under similar conditions (Scheme 4).

At the onset of this research, there were no useful

procedures in the literature that would allow for the

introduction of the N-terminal P-C as outlined in

Scheme 4.⁸Therefore, initial efforts described above

focused on the synthesis of 4, which contains both the

C- and N-terminal P-C bonds, for reaction with 3

(Scheme 2) to provide a stereoselective route to the

N-terminal amino acid. However, Montchamp’s group has

since described an extremely mild Et₃B-initiated radical

addition of AHP to alkenes²⁴and more recently a pal-

ladium-catalyzed cross-coupling of aqueous hypophos-

phorus acid to alkenes and alkynes.⁴¹We envisioned that

this chemistry could also be performed on protected

R-vinyl glycine 10 (Scheme 4).

entry^a

oxidant

solvent temp time, h

yield, %

23-60^b

1

2

3

4

5

6

7

30% H₂O₂

THF

rt

18

24

18

2

aq NaIO₄

29

NaIO₄/SiO₂

no reaction

30% H₂O₂, Et₃N THF

major decomp

30% H₂O₂, Pyr CH₂Cl₂reflux

83^b

68^b

30% H₂O₂

CH₂Cl₂reflux

THF 40 °C

5

^aSee Supporting Information for experimental details, entries

1, 2, 6, and 7. ^bYields based on g0.5 g (g1.2 mmol) scale reac-

tion.

occurred (Scheme 5).⁴³The extent of the rearrangement

could be limited by carrying out the elimination reaction

in solvent (mesitylene, bp 162-164 °C) rather than

neat,⁴²to give the desired product in 62% yield after

tedious chromatographic separation. The moderate yield

and difficult purification made this route undesirable for

large-scale preparation of protected R-vinyl glycine.

Ultimately, the synthesis of 10 from the corresponding

protected phenyl selenide 14⁴⁴proved to be the most

efficient.⁴⁵

Barton’s group, the first to report the use of 14 as a

precursor to R-vinyl glycine,⁴⁴used ozone to oxidize the

selenide prior to a selenoxide elimination to form the

olefin. They also reported the use of 30% H₂O₂as the

oxidant, but the yields were much lower than when ozone

was employed (56% vs 98%). Wishing to avoid the use of

ozone, we have developed a one-pot oxidation/elimination

reaction using H₂O₂as the oxidant. Several reaction

conditions and oxidants were investigated, and the

results are summarized in Table 1. When the solvent was

changed from THF (rt) to CH₂Cl₂(reflux, 40 °C), the

reaction rate and yield of 10 increased dramatically

(Table 1, entries 5 and 6). Higher temperatures (THF,

reflux temperature 66 °C) resulted in more rearrange-

ment, whereas use of THF at 40 °C resulted in a lower

yield (Table 1, entry 7). From this investigation, we

conclude that the optimum reaction conditions for the

successful one-pot conversion of 14 to 10 are oxidation

with H₂O₂and pyridine in refluxing CH₂Cl₂(Table 1,

There are numerous syntheses of R-vinyl glycine

reported in the literature; however, most suffer either

from low yields or low optical purity or are not amenable

to scale-up. Rapoport’s large-scale synthesis⁴²was used

initially, but during the elimination reaction (11 f 10),

significant double bond migration to form 12 and 13

(43) The checkers of the Organic Synthesis procedure also reported

seeing a large amount of rearrangement but attributed it to the silica

gel used during the purification; however, we observed the rearrange-

ment products in the NMR spectra of the crude reaction mixture prior

to purification.

(44) Barton, D. H. R.; Crich, D.; Herve, Y.; Potier, P.; Thiery, J.

Tetrahedron 1985, 41, 4347-4357.

(45) Commercially available S-2-amino-butyrolactone was converted

to S-2-amino-4-bromo-butyric acid using a slight modification of the

procedure of Koch and Buchardt (Synthesis 1993, 1065). Following

protection of the amino acid, phenyl selenide, 14, was formed in 98%

yield using the method of Barton.⁴⁴See Supporting Information for

details.

(40) Voronkov, M. G.; Marmur, L. Z.; Dolgov, O. N.; Pestunovich,

V. A.; Pokrovski, E. I.; Popel, Y. I. J. Gen. Chem. USSR 1971, 41, 2005-

2008.

(41) Deprele, S.; Montchamp, J.-L. J. Am. Chem. Soc. 2002, 124,

9386-9387.

(42) Carrasco, M.; Jones, R. J.; Kamel, S.; Rapoport, H.; Truong, T.

In Organic Syntheses; Wiley: New York, 1998; Collect. Vol. IX, pp 63-

66.

6760 J. Org. Chem., Vol. 70, No. 17, 2005

Synthesis of Phosphinic Acid Phosphapeptides

SCHEME 6

entry 5). The procedure can easily be scaled up to several

grams without appreciable loss of yields. The protected

R-vinyl glycine 10 was treated with Et₃B and AHP in

methanol at room temperature open to air to give 15 in

94% yield (Scheme 6). This demonstrates that the Mont-

champ procedure is useful for highly sensitive, function-

alized compounds that are prone to racemization and

rearrangement.

SCHEME 7

With the synthesis of 15 completed, the stereoselective

synthesis of a synthon that could be elaborated to the

C-terminal portion of 2 was required. As noted above

(Schemes 3 and 4), a suitably substituted cyclopentene

derivative could serve as a glutarate surrogate that could

be unmasked by ring-opening oxidation of the double

bond. Based upon our previous work,⁸masking the diacid

functionality should allow for a broader range of reaction

conditions to be employed for the synthesis of the key

P-C bonds. We envisioned that a variety of electrophiles

(e.g., 5, Scheme 4) could be prepared from the corre-

sponding 3-(hydroxymethyl)-cyclopent-1-ene 16 or 17.⁴⁶

The R-stereoisomer 16, which would lead to the (2R)

2-substituted glutarate moiety of 2 analogous to un-

natural (2R)-glutamic acid, has previously been prepared

by Maeda and Inouye.³⁹A route to the desired S-isomer,

17, was developed that allowed for the use of the same

commercially available starting material 18 in the syn-

theses of 16 and 17 (Scheme 7). Both isomers are

desirable for biochemical analysis to assess stereochem-

ical determinants for phosphinic acid inhibitors of FPGS.

Treatment of 18 with Ac₂O and pyridine furnished the

unstable enol acetate, which was immediately reduced

using the method of Rozzell⁴⁷(H₂, 5% Rh/C, 1 atm) to

afford the racemic mixture of 20 and 21. The desired

â-hydroxy ester 22 was obtained using Lipase AK “Amano”

to selectively hydrolyze only the (2R)-acetate 20, in 45%

yield. The original report of this selective hydrolysis by

Sakai and co-workers noted only that an unspecified P.

fluorescens lipase was used,⁴⁸but the microorganism was

subsequently identified as P. fluorescens Amano P (Amano

Pharmaceutical Co.).⁴⁹We have found that Lipase AK

“Amano” from the same supplier works equally well for

this reaction. The resulting alcohol 22 was converted to

xanthate 23 in 97% yield. Pyrolysis of 23 furnished ethyl

3-(S)-cyclopentenecarboxylate in 78% yield. Reduction of

the ester with LiAlH₄gave the desired (S)-(-)-3-(hy-

droxymethyl)-cyclopent-1-ene (17) in 71% yield.

The R-isomer 16 has been used for numerous model

reactions because it is available in larger quantities than

the S-isomer 17. This is due to the fact that the C-3

stereocenter of 16 is set via a one-step, enzyme-catalyzed

dynamic kinetic resolution, versus three steps required

to obtain enantiomerically pure 17. Given the fact that

an acyclic homoallylic bromide, 4-bromo-1-butene, re-

acted with BTSP in high yield under the optimized

conditions described above, synthesis of 3-(bromomethyl)-

cyclopent-1-ene was pursued as a key intermediate

(Scheme 4, 5, X ) Br). Although nonclassical carbocation

rearrangements of 3-(X-methyl)-cyclopent-1-enes have

been described,⁵⁰the synthesis and study of this class of

compounds is not well documented in the literature.

Several different methods for the synthesis of 24, the

R-isomer of the desired homoallylic bromide, were in-

vestigated and are summarized in Table 2. Initially,

treatment of 16 with CBr₄and PPh₃in CH₂Cl₂resulted

in yields higher than 100% because an impurity (CHBr₃)

could not be removed from the desired product (Table 2,

entry 1). The inability to remove CHBr₃proved detri-

mental to the P-C bond-forming reactions as will be

discussed below. In a previously reported synthesis of the

racemic form for use in the study of its possible rear-

rangement under radical⁵¹and Grignard⁵²conditions,

separation of the desired product from CHBr₃by GC was

effective; however we wished to synthesize 24 on a scale

that would not be amenable to GC purification.

(46) Conversion of the alcohol into a suitable leaving group for P^III

chemistry proved much more difficult than was anticipated. Compound

16 could easily be converted into the corresponding mesylate or tosylate

in good yield by treatment with the corresponding sulfonyl chloride

and pyridine. However, the triflate and iodide could be synthesized

only in low yield and the resulting product decomposed at room

temperature within minutes, presumable through a nonclassical

carbocation intermediate. Ultimately, conditions were optimized for

synthesis of the corresponding bromide (Table 2). See Supporting

Information for details.

Activation of the alcohol via Mitsunobu (or related)

conditions followed by bromide displacement (Table 2,

entries 2-6) resulted in low yields and/or formation of

byproduct(s) that were difficult to separate from 24. The

conversion of 16 to 24 with PPh₃Br₂was effective (Table

(47) Rozzell, J. D. Tetrahedron Lett. 1982, 23, 1767-1770.

(48) Xie, Z.; Suemune, H.; Sakai, K. J. Chem. Soc., Chem. Commun.

1987, 838-839.

(49) Fang, C.; Ogawa, T.; Suemune, H.; Sakai, K. Tetrahedron:

Asymmetry 1991, 2, 389-398.

(50) Hanack, M.; Schneider, H. J. Tetrahedron 1964, 20, 1863-1879.

(51) Friedrich, E. C.; Halomstead, R. L. J. Org. Chem. 1972, 37,

2550-2554.

(52) Maercker, A.; Geub, R. Chem. Ber. 1973, 106, 773-797.

J. Org. Chem, Vol. 70, No. 17, 2005 6761

Bartley and Coward

TABLE 2. Synthesis of 3-(Bromomethyl)-cyclopent-

1-ene

TABLE 3. BSA-Mediated S_N2 Displacement by

Phosphinic Acids

entry^a

conditions

CBr₄, PPh₃

solvent

yield, %

>100^b

60

1

CH₂Cl₂

THF

2

DIAD, PPh₃, ZnBr₂

DIAD, PPh₃, ⁿBu₄NBr

DDQ, PPh₃, ⁿBu₄NBr

NBS, PPh₃

3

CH₂Cl₂

benzene

CH₂Cl₂

0

4

5

50-87

75

6

7^d

8^d

9^d

PPh₃Br₂

56-80^c

75

PPh₃Br₂, PPh₃

PPh₃Br₂, Pyr

82

^aSee Supporting Information for experimental details, entries

1-8. ^bProduct contaminated with CHBr₃. ^cYield highly dependent

upon source of PPh₃Br₂. ^dPPh₃Br₂stoichiometry: entry 7, 1.12

equiv; entries 8 and 9, 1.32 equiv.

SCHEME 8

often cited for the formation of P-C bonds using P^III

intermediates, their methods suffer from extremely

diminished yields when unactivated alkyl halides are

used.⁸In our research, the reactions utilizing a P^III

coupling methodology were followed routinely by ³¹P

NMR, which allowed for the monitoring of all reaction

intermediates as shown in Scheme 9.

In a typical literature procedure, the three reagents,

an H-phosphinic acid, an alkyl halide, and N,O-bis-

(trimethylsilyl)acetamide (BSA) or TMSCl/DIEA (as the

TMS source) are added in one pot and the reaction is

stirred for several hours at room temperature. Using this

procedure, reaction of 26 and the simple phenylpropyl

H-phosphinic acid 27 to give 30 (Scheme 9) proceeded in

only 35% yield even at reflux temperature in CH₂Cl₂

(data not shown).⁵³However, by preforming the P^III

intermediate at slightly elevated temperatures prior to

addition of the alkyl halide, the yields can be greatly

improved. Initial conversion of 27 to its TMS ester 28

followed by formation of the P^IIIintermediate 29 and

finally formation of the TMS ester of the desired product

30 upon addition of 26 could be monitored by ³¹P NMR

and easily quantitated by integration of the correspond-

ing peaks. Monitoring P-C bond formation in this way

has allowed us to optimize the coupling of P^IIIspecies to

unactivated alkyl bromides in high yield (Table 3).

2, entry 7), but the yields varied when different sources

of PPh₃Br₂were used. Side products formed that lacked

the endocyclic double bond, presumably from attack of

the olefin by Br₂or HBr contaminating the reagent. This

problem could be overcome by the inclusion of additional

PPh₃in the reaction to consume any excess Br₂(Table 2,

entry 8) or by the addition of pyridine to the reaction

mixture (Table 2, entry 9). The latter conditions proved

to be more optimal for generation of the bromide and gave

the desired product in 82% yield in 1 h. Compound 24 is

relatively stable and can be stored for at least 1 month

at -20 °C and is stable for 1-2 days at room tempera-

ture. However, when left at room temperature for several

days, 24 decomposes to a hydrocarbon polymer with no

double bond or alkyl bromide evident by NMR. As a

result, the prudent course is to prepare 24 just prior to

use. The S-isomer 25 was also prepared using this

method (Scheme 8).

However, when 24 (prepared via CBr₄/PPh₃method,

vide supra) was substituted for 26 in the reaction with

15, the isolated yield of the desired phosphinic acid was

only 8%! This low yield appeared to be due to contamina-

tion of 24 with bromoform, which was shown to inhibit

the alkylation reaction.³⁸These dramatically differing

results led us to explore the other methods for the

To couple 24, 25, or an acyclic analogue, 4-bromo-1-

butene (26), to 15, a new method was developed to

accomplish the desired P-C bond formation. Although

the methods of Boyd and Regan³⁰and Thottathil²³are

SCHEME 9

6762 J. Org. Chem., Vol. 70, No. 17, 2005

Synthesis of Phosphinic Acid Phosphapeptides

SCHEME 10

synthesis of 24 described above and summarized in Table

2. The use of PPh₃Br₂and pyridine in CH₂Cl₂provided

the best method for the preparation of 24 without

contamination by CHBr₃, and this method was used to

synthesize all of the 3-(bromomethyl)-cyclopent-1-ene

derivatives described below. Unfortunately, when 24,

prepared using the PPh₃Br₂/pyridine method, was used

in the BSA-mediated P-C bond-forming reaction, the

reaction was very slow, ³¹P NMR indicated only a 23%

conversion after 22 h. An additional 3 equiv of 24 was

added in portions over time, but after 48 h only 30%

conversion to the desired phosphinic acid derivative was

observed by ³¹P NMR.

There are two possible explanations for the lower rate

of reaction observed with 24: either the 3-(bromomethyl)-

cyclopent-1-ene is not stable to the reaction conditions

or the additional steric encumbrance of the electrophile

may be retarding the reaction. ¹H NMR of the crude

reaction mixture indicated that 24 does not decompose

under these conditions. The slower reaction rate and

lower yield may be due to the extra substitution at the

â-position of 24 compared to that of 26. To test this

hypothesis, 3-(bromomethyl)-cyclopentane was prepared

from the commercially available 3-(hydroxymethyl)-cy-

clopentane using the PPh₃Br₂/pyridine procedure, and its

reaction with 27 was monitored by ³¹P NMR. After 24 h

only 6% conversion was observed, although prolonged

reaction (11 days) did result in a conversion of 70% to

the desired phosphinic acid. Other changes in reaction

conditions (solvent and temperature) did not result in

improved rates or yields. In contrast, reaction of 24 with

thiophenol (NaH, THF, rt) proceeded smoothly in 4 h to

provide the corresponding homoallylic phenylthioether

in 60% yield. Similarly, Michaelis-Becker reaction of

diethyl phosphite with 24 (Na, THF, reflux temperature)

provided the expected cyclopentene-containing phospho-

nate in 65% yield (M. Ferguson, unpublished results).

Therefore, the electrophilicity of 24 is adequate for

effective reaction with appropriate nucleophiles.

Reasoning that the reaction with 24 was proceeding,

albeit at a slow rate due to steric constraints at the

electrophilic center and attenuated nucleophilicity of the

P^IIIintermediate, we postulated that both rate and yield

could perhaps be enhanced by increasing the equivalents

of the alkyl halide. Compound 15, containing the N-

terminal stereocenter, was combined with varying

amounts of 24 and BSA, and it was found that yields of

greater than 80% (³¹P NMR) could be obtained by

increasing the ratio of 24/BSA/15 from 2:3:1 to 10:10:1.

Using these optimized conditions, the reaction was

complete in 44 h. After purifying the phosphinate ester

by silica gel chromatography (85% yield), we discovered

that either during the course of the reaction or the

workup the double bond had isomerized to give 35

(Scheme 10). The structure of 35 is based on extensive

NMR data (¹H, ¹³C, ³¹P, COSY, APT, and HETCOR), the

key NMR signal being the quaternary carbon of the

olefin. The reaction was repeated three times and in each

instance 35 was the only product isolated.

Reaction of 24 with BSA in CH₂Cl₂at reflux temperature

(48h) evaluated stability of the double bond in the

starting material under the reaction conditions. Follow-

ing normal workup (2 M HCl, 1:1 dilution, 15-30 min)

to hydrolyze the remaining BSA and the TMS ester of

the crude phosphinic acid, 24 was obtained unchanged.

A similar reaction was carried out with 24 and 15 in the

absence of BSA and again 24 was recovered intact. It is

possible that the rearrangement occurs during the acidic

workup after the P-C bond is formed, resulting in

rearrangement of the desired product, 34, to the allylic

isomer, 35. To determine if the acidic workup was causing

the rearrangement of the initially formed product, the

P-C bond-forming reaction was repeated and the reac-

tion was quenched with TBAF to remove the phosphinic

acid TMS ester and decompose any remaining BSA in a

nonprotic environment. After treatment with TBAF, the

reaction was concentrated and NMR was taken of the

crude mixture. The NMR spectral data indicated that the

rearrangement had occurred even with the basic workup

procedure. These results led us to conclude that the

rearrangement was occurring during the reaction and

that the use of 3-(2-bromomethyl)cyclopent-1-ene (24 or

25) was not a viable option for setting the C-terminal

P-C bond.

To prevent the double bond migration described above,

temporary removal of the olefin was investigated. The

most logical choice for this “protection” was to convert

the double bond to a diol, which ultimately could be

oxidatively cleaved to furnish the desired diacid func-

tionality of the target pseudopeptide. However, the P^III

intermediate in the P-C bond-forming reaction is not

stable to protic reagents or solvents so the diol would

have to be further protected as an acetonide as shown

retrosynthetically in Scheme 11. Considering the dif-

ficulties associated with forming the second P-C bond

SCHEME 11

A series of control experiments was performed in an

attempt to understand the basis for the migration.

(53) After 18 h at room temperature only a very small amount of

product (2%) was observed.

J. Org. Chem, Vol. 70, No. 17, 2005 6763

Bartley and Coward

SCHEME 12

FIGURE 1. Stereochemical assignment of 39 and 40.

FIGURE 2. Stereochemical assignment of 41 and 42.

desired 2,3-trans product 39 via either route. On the basis

of these results, the OsO₄procedure is favored because

of a more facile workup and easier purification of the

product than with the KMnO₄-based oxidation. The 2,3-

cis and 2,3-trans diols were converted to the acetonides

by treatment with 2,2-dimethoxypropane, acetone, and

catalytic acid in 96% yield for the 2,3-trans isomer 41,

and 95% for the 2,3-cis isomer 42 (Scheme 12).

The stereochemistry of the two oxidation products 39

and 40 was assigned on the basis of their NOESY NMR

spectra (Figure 1). The stereochemical assignment is

supported by the NOE cross-peak observed between the

protons on C1 and C3 of 40. There was no NOE observed

for the equivalent protons of 39. The 2,3-cis and 2,3-trans

stereochemistry assigned using the NOESY spectrum of

39 and 40 was further supported by the coupling constant

for the C2 and C3 protons on 41 and 42. For compound

41 the J_2,3coupling constant is 0 Hz indicating a dihedral

angle of ∼90° while the coupling constants for the

corresponding protons of 42 is 7.2 Hz consistent with a

dihedral angle of ∼20° signifying that both protons are

on the same side of the ring (Figure 2).

SCHEME 13

Several methods for the deprotection of the trityl

group were investigated. Deprotection with BF₃‚Et₂O⁵⁷

or formic acid⁵⁸resulted in decomposition and low

yields. Hydrogenation with 10% Pd/C⁵⁹also proved un-

successful yielding quantitative recovery of starting

material. Finally, successful deprotection of the trityl

group was accomplished by hydrogenation (40 psi) in the

presences of Pearlman’s catalyst, Pd(OH)₂/C, at room

temperature. Conversion of the hydroxymethyl ace-

tonides to the corresponding bromomethyl derivatives

was carried out using PPh₃Br₂and pyridine to provide

36 in 79% yield and 37 in 82% yield (Scheme 12). The

2,3-trans (47) and 2,3-cis (48) isomers of the bromomethyl

acetonides, derived from the 3S-homoallylic alcohol 17,

were also prepared using the same methods (Scheme 13).

The absolute stereochemistry depicted for 46, deduced

via NMR analysis of this series as described above for

39-42, was confirmed via a crystal structure obtained for

the corresponding sulfone.⁶⁰

with 24, the increased steric encumbrance presented by

an acetonide on the same face of the ring as the leaving

group (hereafter referred to as the 2,3-cis isomer, e.g.,

37) could be problematic, and therefore the 2,3-trans

isomer (e.g., 36) was desired. The most common and

highly stereoselective method for the stereoselective

formation of diols from olefins is the Sharpless asym-

metric dihydroxylation.⁵⁴Unfortunately, the Sharpless

procedure does not provide good asymmetric induction

with disubstituted Z-olefins.⁵⁵

Reaction of the 3R alcohol 16 with trityl chloride in

pyridine provided the trityl ether in 94% yield. Oxidation

of 38 was first attempted using the classic Upjohn

procedure⁵⁶of catalytic OsO₄and NaIO₄, which provided

a 90% yield of 1.25:1 mixture of the trans and cis diols

39 and 40, respectively (Scheme 12). The use of the trityl

protecting group allowed for easy separation of the two

isomers. An alternative oxidant, KMnO₄, gave much

better selectivity (2,3-trans:2,3-cis (39:40), 6:1) than OsO₄

but the reaction yield was much lower, 64% vs 90%,

resulting in formation of about the same amount of the

(56) VanRheenen, V.; Kelly, R. C.; Cha, D. Y. Tetrahedron Lett. 1976,

23, 1973-1976.

(57) Cabaret, D.; Wakselman, M. Can. J. Chem. 1990, 68, 2253-

(54) Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. A.;

Hartung, J.; Jeong, K.; Kwong, H.; Morikawa, K.; Wang, Z.; Xu, D.;

Zhang, X. J. Org. Chem. 1992, 57, 2768-2771.

(55) Wang, L.; Sharpless, K. B. J. Am. Chem. Soc. 1992, 114, 7568-

7570.

2257.

(58) Bessodes, M.; Lomiotis, D.; Antonakis, K. Tetrahedron Lett.

1986, 27, 579-580.

(59) Mirrington, R. N.; Schmalzl, K. J. J. Org. Chem. 1972, 37,

2877-2881.

6764 J. Org. Chem., Vol. 70, No. 17, 2005

Synthesis of Phosphinic Acid Phosphapeptides

SCHEME 14

SCHEME 15

SCHEME 16

Formation of the second P-C bond using 15 and each

of the four bromomethyl acetonides 36, 37, 47, and 48,

proved to be even slower than was the case with the

corresponding free olefins 24 or 25 (Scheme 14). After 8

days of heating, the reaction of 15 and 36 went only to

15% completion as determined by ³¹P NMR. Similar

results were observed with 37. The 3S-2,3-trans isomer

47, with correct stereochemistry for elaboration to 2, was

then used with longer reaction times in an attempt to

force the reaction to completion. Unfortunately, even

after 1 month only 24% conversion was observed by ³¹P

NMR. As expected, the reaction was even slower on the

S-2,3-cis isomer 48 with less than 5% conversion after

several weeks. The low yields of the desired complex

phosphinic acids (e.g., 49 and 50) obtained in the reac-

tions of all isomers of 3-(bromomethyl)-cyclopentane 1,2-

acetonides, together with the double bond migration

observed with the corresponding cyclopentene (Scheme

10), indicated that (bromomethyl)-cyclopentene deriva-

tives would not be effective glutarate surrogates as

proposed in our retrosynthetic analysis for the stereose-

lective synthesis of the C-terminal C-P bond (Schemes

3 and 4). Therefore, an alternate route to the desired

compounds was explored.

Two possible compounds that could be appended to 15

to furnish the desired product 2 are an acyclic bromo-

methyl derivative of dimethyl glutarate such as 51 or a

methylene glutarate derivative containing a chiral aux-

iliary such as 52 (Scheme 15). The synthesis of 51 was

not pursued because this ring-open form would be as

sterically hindered at the â-position as 24 and also would

be much more prone to dehydrohalogenation. For this

reason, our attention focused on synthesizing the second

carbon-phosphorus bond through a stereoselective

Michael addition of 15 to 52. Recently, Liu et al. have

shown that the stereoselectivity of this type of Michael

addition can be greatly influenced by incorporation of a

chiral auxiliary on the R-carboxylate of simple acrylates.²⁵

Compound 53c was synthesized to investigate this route.

A tert-butyl ester was chosen as the γ-protecting group

to allow for ultimate synthesis of a more versatile

dipeptide analogue with N- and C-termini being differ-

entially protected.

Although synthetic routes to both 2-methyleneglutaric

acid and the corresponding di-esters are readily avail-

able,^61-68selective esterification of one of the two acids

has been reported only rarely. 2-Methyleneglutaric acid

5-methyl ester 53a has been prepared electrochemically

by the addition of CO₂to pent-4-ynoic acid methyl ester.

However the free acid was not isolated, and instead the

product was converted to the dimethyl ester for isolation

and purification (38% yield).⁶⁹As reported in the patent

literature, compound 53a has also been synthesized via

an iodine-mediated selective esterification of the non-

conjugated carboxylic acid of 2-methyleneglutaric acid.⁷⁰

A retrosynthetic strategy for the formation of 2-meth-

yleneglutaric acid 5-tert-butyl ester 53c is shown in

Scheme 16. Compound 54 can be prepared via a Michael

reaction of a suitably protected malonate and tert-butyl

acrylate.⁷¹We envisioned that 54 could be converted to

compound 53c via a tandem Mannich reaction and in situ

decarboxylation.^25,72The Michael addition proved to be

problematic as several attempted preparations of 54 led

also to the R,R-disubstituted malonate 57 (Table 4). Thus,

when the anion of diethyl malonate 55a was generated

(61) Shabtai, J.; Hey-Igner, E. J. Org. Chem. 1978, 43, 4086-4090.

(62) Amri, H.; Rambaud, M.; Villieras, J. Tetrahedron Lett. 1989,

30, 7381-7382.

(63) Drewes, S. E.; Emslie, N. D.; Karodia, N. Synth. Commun. 1990,

20, 1915-1921.

(64) Jackson, P. F.; Cole, D. C.; Slusher, B. S.; Stetz, S. L.; Ross, L.

E.; Donzanti, B. A.; Trainor, D. A. J. Med. Chem. 1996, 39, 619-622.

(65) Brase, S.; Waigell, B.; de Meijere, A. Synthesis 1998, 2, 148-

152.

(66) Yi, C. S.; Liu, N. J. Organomet. Chem. 1998, 553, 157-161.

(67) Jenner, G. Tetrahedron Lett. 2000, 41, 3091-3094.

(68) Jackson, P. F.; Tays, K. L.; Maclin, K. M.; Ko, Y.; Li, W.;

Vitharana, D.; Tsukamoto, T.; Stoermer, D.; Lu, X. M.; Wozniak, K.;

Slusher, B. S. J. Med. Chem. 2001, 44, 4170-4175.

(69) Dunach, E.; Perichon, J. Synlett 1990, 3, 143-145.

(70) Mitusbishi Petrochemical Co. Ltd., Japanese Patent 59152347,

1984.

(60) Alexander, M. D. Ph.D. Dissertation, University of Michigan,

Ann Arbor, MI, 2004.

(71) Makarevic, J.; Skaric, V. Heterocycles 1995, 41, 1207-1218.

J. Org. Chem, Vol. 70, No. 17, 2005 6765

Bartley and Coward

TABLE 4. Michael Reaction of Malonate Esters and

SCHEME 17

tert-Butyl Acrylate

entry^a

R

base

54:57^b

1

2

3

4

5

6

Et

NaOEt

mixture

4:1

c

K₂CO₃

KOtBu

NaH

Bn

mixture^d

2.3:1

Bn

TMSE

c

K₂CO₃

only 57

2.4:1

c

^aSee Supporting Information for experimental details, entries

4-6. ^bRatios of individual components based upon NMR integra-

tion. ^cPhase-transfer conditions: [(nBu₄)N⁺Br^-/benzene], reflux

^dMajor component appeared to be the R,γ-disubstituted product

based on NMR spectral analysis.

using stoichiometric amounts of NaH, NaOEt, KOt-Bu,

or K₂CO₃under phase-transfer conditions,⁷³a mixture

of products was obtained. The best ratio of desired

monosubstituted product 54 to undesired disubstituted

product 57 was 4:1 obtained with diethyl malonate (55a)

and tert-butyl acrylate (56) using phase-transfer condi-

tions (Table 4, entry 2). However, incomplete cleavage

of the ethyl esters was observed when either a stoichio-

metric amount or slight excess (2.1 equiv) of base was

used. The use of a large excess of base resulted in partial

cleavage of the tert-butyl ester. In an effort to synthesize

a substituted malonate containing more labile esters, the

reaction of dibenzyl malonate (55b) or bis(trimethylsi-

lylethyl) malonate (55c) and 56 was investigated. Un-

fortunately, under a variety of reactions conditions (Table

4, entries 3-6), only unsatisfactory mixtures were ob-

tained. Further transformation of the inseparable mix-

ture of 54a and 57a (Table 4, entry 2) was pursued. It

was found that, with only an extractive workup, the

mixture of hydrolysis products 58 and 59, could be

converted to the desired product 53c under conditions

involving Mannich base formation and decarboxylation,

in a respectable overall yield of 49% from 55a (Scheme

17, Path A). This method provides the desired γ-tert-butyl

ester 53c via a route that avoids chromatographic

purification of intermediate products.

Considering the significant amount of disubstitution

that was observed in all reactions of 55 and 56, the use

of stoichiometric base to deprotonate the malonate is not

necessary. Only one example was found in the literature

where a catalytic amount of base was used.⁷⁴Surpris-

ingly, in most literature procedures using stoichiometric

base, the reaction is run on large scale and the desired

product is purified by distillation.^75-77This approach

typically results in low yields for this seemingly simple

carbon-carbon bond-forming reaction. The addition of

55b to 56 was repeated using a catalytic amount of NaH

(Scheme 17, Path B). NaH was added to a solution of 55b

in THF, followed by the slow addition of 56 over 30 min.

The reaction yielded 83% of the monosubstituted mal-

onate 54b. The benzyl esters of compound 54b were

removed by catalytic hydrogenation over Pd/C to give 58

in quantitative yield. Sequential formation of a Mannich

base followed by decarboxylation gave the desired product

53c in an unoptimized yield of 40% and provided the

required building block for installation of a chiral aux-

iliary at the R-position.

On the basis of the work of Liu et al.,²⁵the diphenyl-

methyl derivative of Evans’ oxazolidinone was chosen as

the chiral auxiliary. Initially, coupling of the oxazolidi-

none to the methylene glutarate was attempted via an

acid chloride, but this method was not effective. Synthesis

of 52 by way of the mixed anhydride using the method

of Deslongchamp⁷⁸was also attempted but furnished the

desired product in only 40% yield. The most common

method for the synthesis of peptide bonds is the use of

carbodiimide coupling reagents in conjunction with HOBt

to form the activated ester of the carboxylic acid that is

being coupled. Although HOBt esters are usually formed

in situ and not isolated, research in our laboratory has

demonstrated that higher yields of coupling product can

be obtained if the HOBt ester of an aromatic carboxylic

acid is first isolated and then reacted with the desired

nucleophile.⁶⁰Yields of the chiral auxiliary coupling

perhaps could be increased by HOBt-mediated activation

of 53c and isolation of the HOBt ester. In fact, reaction

of 53c with HOBt and DCC in THF provided the

activated ester 60 in 89% yield following silica gel

(72) Mannich, C.; Ganz, E. Ber. 1922, 55, 3486-3504.

(73) Prabhu, K. R.; Pillarsetty, N.; Gali, H.; Katti, K. V. J. Am.

Chem. Soc. 2000, 122, 1554-1555.

(74) Quast, H.; Janiak, R. Leibigs Ann. Chem. 1991, 1305-1308.

(75) Gresham, T. L.; Jansen, J. E.; Shaver, F. W.; Frederick, M. R.;

Beears, W. L. J. Am. Chem. Soc. 1951, 73, 2345-2347.

(76) Dubowchik, G. M.; Firestone, R. A. Tetrahedron Lett. 1994, 35,

4523-4526.

(77) Gomez-Bengoa, E.; Cuerva, J. M.; Mateo, C.; Echavarren, A.

M. J. Am. Chem. Soc. 1996, 118, 8553-8565.

(78) Phoenix, S.; Bourque, E.; Deslongchamps, P. Org. Lett. 2000,

2, 4149-4152.

6766 J. Org. Chem., Vol. 70, No. 17, 2005

Synthesis of Phosphinic Acid Phosphapeptides

TABLE 5. BSA-Mediated Conjugate Addition of

Phosphinic Acids

SCHEME 18

chromatography (Scheme 18). When 60 was treated with

the lithiated anion 61, compound 52 was isolated in 84%

yield. The use of isolated activated esters similar to 60

is rare in the literature, perhaps due to the incorrect

assumption that compounds of this type are not stable

to silica gel chromatography or to long-term storage.

However, we have found that both aromatic and aliphatic

HOBt esters are stable to silica gel chromatography and

are stable to prolonged (several months) storage at

-20 °C.

pseudopeptides 66 and 67 in high yield (Scheme 19). In

addition to clean deprotection, this method allows for

recovery of the chiral auxiliary in greater than 90%

yield by simple extraction of the crude reaction mix-

ture with diethyl ether. Compounds 66 and 67 have the

added advantage of being differentially protected at the

N- and C-termini, which allows for their future extension

into longer chain peptides at either the N-terminus or

C-terminus or both. The protected pseudopeptides, 66

and 67, were globally deprotected by heating at reflux

temperature for 12 h in 6 M HCl, leading to the pure

phosphinic acids as their HCl salts 68 and 69 in quan-

titative yield. To complete the synthesis of the target

FPGS inhibitors, phosphinic acid pseudopeptides 74 and

75 were coupled to the acyl azide, AMPte-N₃, derived

from 4-deoxy-4-amino-10-methylpteroic acid,^8,11using the

methodology described previously.^8,81Subsequent ion

exchange chromatography provided the desired products

as their triethylammonium salts 70 and 71 in excellent

yield.

In conclusion, during the course of investigating a

stereoselective synthesis of phosphinic acid based inhibi-

tors of FPGS, improved methodology for the synthesis of

P-C bonds between highly functionalized phosphinic

acids and unactivated alkyl bromides has been developed.

Unfortunately, this procedure led to an unexpected

rearrangement when applied to a cyclic homoallylic

bromide and was, therefore, ineffective for the synthesis

of the target pseudopeptides 70 and 71. An alternate

synthesis was pursued that relied upon the conjugate

addition of 15 to 52 and subsequent chromatographic

separation of stereochemically pure phosphinate pseudo-

peptides 64 and 65. Deprotection of 64 and 65 and the

ensuing coupling to 4-amino-4-deoxy-10-methylpteroyl

azide completed the synthesis of 70 and 71.

Using the procedure of Liu et al.,²⁵the Michael reaction

of 52 and a P^IIIspecies derived from 15 was investigated.

When these conditions were used to generate a P^III

species from 15 and effect subsequent Michael addition

to 52, a 1:1.5 ratio of two diastereomers was detected by

HPLC analysis of the crude reaction mixture. The

procedure described by Liu et al.²⁵used TMSCl and DIEA

to generate the BTSP nucleophile prior to addition of the

Michael acceptor. However, TMSCl and DIEA vapors

form copious amounts of a fluffy white solid, and the

vapor also reacts with rubber septa, thus rendering this

method inconvenient. Because of the low selectivity

observed and problems encountered with the use of

TMSCl and DIEA, reaction using BSA as the silylating

agent was investigated. Reactions of 15 and either methyl

acrylate, dimethyl methylene glutarate, or 52 were

carried out in the presence of BSA, leading to acceptable

yields of the desired products (Table 5).

The modest stereoselectivity of the Michael addition

was slightly lower when BSA was used as the silylating

agent than when TMSCl/DIEA was used (1.2:1 vs 1.5:1).

However, the BSA reaction was much cleaner by ³¹P

NMR and did not have any of the problems associated

with TMSCl and DIEA noted above. The BSA reaction

can also be performed in one step with all of the reagents

added at one time. There was no difference in selectivity

regardless of how the reaction was quenched (2 M HCl,

abs EtOH, or TFA) or at what temperature (rt, 0, or -10

°C). Based upon these results, the optimized reaction

conditions are reaction of 15, 52, and BSA at room

temperature in CH₂Cl₂for 24 h followed by quenching

with 2 M HCl at room temperature and esterification of

the resulting free acid with TMSCHN₂. The resulting

product, a 1.2:1 mixture of two diastereomers, 64 and

65, was obtained in 72% yield, and the two isomers were

separated by silica gel chromatography (A, higher R_f

isomer; B, lower R_fisomer).⁷⁹

The successful synthesis of phosphinate pseudopep-

tides 70 and 71, including the coupling to the 2,4-

diaminopteridine heterocycle, provides the first stereo-

chemically pure phosphinate analogues of Glu-γ-Glu for

biochemical study as FPGS inhibitors and as potential

ligands for crystallographic research. Preliminary bio-

chemical data indicate that 70, derived from 64 (isomer

Using LiOOH to cleave the chiral auxiliary⁸⁰and then

treating the free acid with CH₂N₂provided fully protected

(79) Subsequent reactions (Scheme 19) were carried out separately

on compounds derived from either 64 (A series, even numbers) or 65

(B series, odd numbers).

(80) Evans, D. A.; Britton, T. C.; Ellman, J. A. Tetrahedron Lett.

1987, 28, 6141-6144.

(81) Luo, J.; Smith, M. D.; Lantrip, D. A.; Wang, S.; Fuchs, P. L. J.

Am. Chem. Soc. 1997, 119, 10004-10013.

J. Org. Chem, Vol. 70, No. 17, 2005 6767

Bartley and Coward

SCHEME 19

concentrated and the crude oil was partitioned between 5%

aqueous KHSO₄(30 mL) and EtOAc (60 mL). The aqueous

layer was extracted with EtOAc (2 × 60 mL) and the combined

organic phase was dried over Na₂SO₄, filtered, and concen-

trated to give 0.6 g of 15 as a clear oil (85%): ¹H NMR (CDCl₃)

δ 7.04 (d, 1H, J_PH) 555 Hz), 7.33 (m, 5H), 5.91 (br, H), 5.10

(s, 2H), 4.40 (m, 1H), 3.73 (s, 3H), 2.13 (m, 1H), 2.0-1.8

(overlapping m, 3H). ¹³C NMR (CDCl₃) δ 172.2, 156.4, 136.4,

129.0, 128.7, 128.6, 67.6, 54.2 (d, J ) 16 Hz), 53.1, 25.7 (d, J

) 93 Hz), 24.3. ³¹P NMR (121 MHz, CDCl₃) δ 35.5 Hz. MS

(ESI) m/z 338.1 ([M + Na]⁺, 100). HRMS (ESI) calcd for C₁₃H₁₈-

NO₆PNa 338.0769 [M + Na]⁺, found 338.0767.

A, IC₅₀) ca. 5 nM) is the more active inhibitor of the

two diastereomers 70 and 71 (W. H. Haile and J. J.

McGuire, personal communication) and is the most

potent FPGS inhibitor synthesized to date. Although, the

configuration of the 2′′-stereocenter of 70 and 71 has not

been proven, these preliminary biochemical results and

the known requirement for ^L-(2S)glutamate in both the

acceptor and donor substrates reported to date strongly

suggest that the configuration of the more potent inhibi-

tor, isomer 70, is 2′S,2′′S.

Ethyl (1S,2R)-(-)-2-hydroxycyclopentanecarboxylate,

22. To a stirring mixture of 20/21 (0.5 g, 2.5 mmol) in 100 mL

of phosphate buffer (0.1M, ph 7.0) was added 0.3 g of lipase

(Lipase AK “Amano” from Psuedomonas fluorescens, Amano

Enzyme) and the reaction was stirred at 37°C for 18 h. The

reaction was extracted with Et₂O (3 × 150 mL). The organic

layer was dried over Na₂SO₄, filtered and concentrated. The

crude residue was purified by silica gel chromatography (7:3

pentane/Et₂O) to give 0.18 g of 22 as a clear, colorless oil (45%

(maximum yield 50%)): ¹H NMR (CDCl₃) δ 4.42 (q, 1H, J )

3.7 Hz), 4.17 (q, 2H, J ) 7.1 Hz), 2.66 (m, 1H), 2.07-1.6

(overlapping m, 7H), 1.27 (t, 3H, J ) 7.1 Hz). ¹³C NMR (CDCl₃)

δ 175.1, 73.8, 60.7, 49.5, 34.1, 26.5, 22.2, 14.3. [R]²⁴_D-14.1° (c

1.0, CHCl₃) (lit.⁴⁸-14.3° (c 0.63, CHCl₃)). MS (EI) m/z 101.1

(100), 159.2 ([M + H]⁺, 15). HRMS (EI) calcd for C₈H₁₅O₃

159.1021 [M + H]⁺, found 159.1028.

(1S,2R)-2-((Methylthio)-carboxyoxy)-cyclopentane-

carboxylic Acid Ethyl Ester, 23. To a solution of 22 (0.82

g, 5.0 mmol) in anhydrous DMSO (2.5 mL) was added DBU

(0.83 mL, 5.5 mmol) at room temperature and the reaction

was stirred for 1 h. Carbon disulfide (33.0 mL, 2.0 mmol) was

then added and the reaction was stirred for 1.5 h. Methyl

iodide (0.61 mL, 10.0 mmol) was added and the reaction was

stirred for 1 h. The reaction was concentrated by rotary

evaporation (40 °C, 10 mmHg) and the resulting solution was

diluted with CH₂Cl₂(80 mL) and washed with H₂O (80 mL).

The aqueous layer was extracted with CH₂Cl₂(80 mL). The

combined organic layers were washed with brine (40 mL),

dried over Na₂SO₄, filtered and concentrated. The crude

material was purified by silica gel chromatography (9:1

hexanes/EtOAc) to give 1.2 g of 23 as an orange oil (97%): ¹H

NMR (CDCl₃) δ 6.14 (m, 1H), 4.14 (q, 2H, J ) 7.1 Hz) 3.03 (m,

1H), 2.49 (s, 3H), 2.3-2.1 (m, 1H), 2.1-1.8 (m, 4H), 1.8-1.6

(m, 1H), 1.23 (t, 3H, J ) 7.1 Hz). ¹³C NMR (CDCl₃) δ 214.6,

171.4, 86.2, 60.9, 49.2, 32.3, 26.2, 224, 18.7, 14.4.

Experimental Section

General Experimental Procedures. See Supporting In-

formation.

N-Cbz-Vinyl Glycine-O-methyl Ester, 10.

A. Elimination from Sulfoxide 11 (modified Rapoport

procedure). A solution of 11 (1.6 g, 5.0 mmol) in mesitylene

(20 mL) was heated at reflux temperature for 24 h. After

cooling to room temperature, the entire reaction was loaded

onto a silica gel column and eluted with hexanes to remove

mesitylene and then 4:1 hexanes/ EtOAc to elute the product.

The fractions containing the desired product were pooled and

concentrated to give 0.74 g of 10 (62%): ¹H NMR (CDCl₃) δ

7.36 (m, 5H), 5.91 (ddd, 1H, J ) 16.9 Hz, 10.4 Hz, 5.4 Hz),

5.47 (d, 1H, J ) 6.0 Hz), 5.36 (dd, 1H, J ) 16.9, 1.2 Hz), 5.28

(dd, 1H, J ) 10.4, 1.2 Hz), 5.13 (s, 3H), 4.95 (m, 1H). ¹³C NMR

(CDCl₃) δ 171.0, 155.7, 136.3, 132.5, 128.7, 128.4, 128.3, 117.9,

67.3, 56.3, 52.9. [R]²⁴-12.3° (c 1.0, MeOH) (lit.⁴⁴-12.4° (c

D

0.5, MeOH)).

B. Oxidative Elimination from Selenide 14. (Table 1,

Entry 5) To

a solution of 14 (0.5 g, 1.2 mmol) in

CH₂Cl₂(50 mL) were added 30% H₂O₂(3 mL) and pyridine (3

mL) and the solution was heated at reflux temperature for 2

h. The reaction was cooled to room temperature and then

washed with H₂O, saturated aqueous CuSO₄, 2 M HCl,

saturated aqueous NaHCO₃, and brine (60 mL each). The

organic layer was dried over Na₂SO₄, filtered, and concen-

trated. The crude product was purified by CombiFlash chro-

matography to give 0.25 g of 10 as a clear oil (83%). ¹H NMR,

¹³C NMR, and optical rotation data were identical to that

reported above for 10 prepared by the modified Rapoport

procedure. MS (ESI) m/z 272.2 ([M + Na]⁺, 100). HRMS (ESI)

calcd for C₁₃H₁₅NO₄Na 272.0899 [M + Na]⁺, found 272.0895.

(S)-2-Benzyloxycarbonylamino-4-hydroxyphosphinoyl-

butyric Acid Methyl Ester, 15. To a solution of 10 (0.5 g,

2.0 mmol) and NH₄H₂PO₂(0.42 g, 5.0 mmol) in MeOH (2 mL)

was added Et₃B (2.0 mL of 1.0 M solution in hexanes, 2.0

mmol) at room temperature. The reaction was stirred vigor-

ously open to the air. The reaction was monitored by ³¹P NMR

(two drops of crude reaction mixture in 0.5 mL CD₃OD), and

after 2 h the reaction was complete. The reaction was

(S)-3-(hydroxymethyl)-cyclopent-1-ene, 17. Xanthate 23

(2.2 g, 9.0 mmol) was placed in a 100 mL round-bottom flask

equipped with a condenser and was heated at 190 °C for 3.5

h. The reaction was allowed to cool to room temperature. The

desired product was purified by Kugelrohr distillation (110 °C,

40 mmHg) to give 1.1 g of ethyl (S)-2-cyclopentene-1-carboxy-

late as a clear oil (85%): ¹H NMR (CDCl₃) δ 5.88 (m, 1H), 5.72

6768 J. Org. Chem., Vol. 70, No. 17, 2005

Synthesis of Phosphinic Acid Phosphapeptides

(m, 1H), 4.13 (q, 2H, J ) 7.1 Hz), 3.53 (m, 1H), 2.6-2.3 (over-

lapping m, 2H), 2.13 (m, 2H), 1.26 (t, 3H). ¹³C NMR (CDCl₃) δ

concentrated. The resulting oil was dissolved in MeOH (3 mL)

and treated with TMSCHN₂(2 M in hexanes) until a persistent

yellow color was observed. The reaction was stirred for 30 min

at room temperature, then quenched with AcOH and concen-

trated. The residue was purified by CombiFlash chromatog-

raphy (95:5 CHCl₃/MeOH) to give 0.12 g of 31 (80%): ¹H NMR

(CDCl₃) δ 7.30-7.16 (m, 5H), 5.80 (m, 1H), 5.00 (m, 2H), 3.67

(d, 3H, J ) 6.3 Hz), 2.70 (m, 2H), 2.28 (m, 2H), 1.90 (m, 2H),

1.80-1.70 (m, 4H). ¹³C NMR (CDCl₃) δ 140.9, 137.3, 137.2,

128.58, 128.55, 126.3, 115.4, 51.1, 51.0, 36.8, 36.7, 27.5, 26.5,

25.98, 25.96, 23.67, 23.65. ³¹P NMR (CDCl₃) δ 58.7.

Pentyl-(3-phenyl-propyl)-phosphinic Acid Methyl

Ester, 32. A 10 mL round-bottom flash was charged with 27

(0.1 g, 0.5 mmol) and CH₂Cl₂(2 mL) and equipped with a reflux

temperature condenser fitter with a rubber septum. The

apparatus was flushed with Ar and fitted with an Ar-filled

balloon. BSA (1.2 mL, 5.0 mmol) was added via syringe

through the septum. The reaction was heated at reflux

temperature for 4 h. 1-Bromopentane (0.31 mL, 5.0 mmol) was

added. The reaction was returned to reflux temperature and

heated for 72 h, then quenched by stirring vigorously with 2

M HCl (2 mL) for 15 min. The resulting mixture was diluted

with CH₂Cl₂(10 mL) and washed with 2 M HCl (2 × 5 mL)

and H₂O (5 mL). The organic layer was dried over Na₂SO₄,

filtered, and concentrated. The resulting oil was dissolved in

MeOH (3 mL) and treated with TMSCHN₂(2 M in hexanes)

until a persistent yellow color was observed. The reaction was

stirred for 30 min at room temperature, then quenched with

AcOH and concentrated. The residue was purified by prepara-

tive TLC (95:5 CHCl₃/MeOH) to give 0.12 g of 32 (92%): NMR

(CDCl₃) δ 7.31-7.15 (m, 5H), 3.69 (d, 3H, J ) 6.3 Hz), 2.70

(m, 2H), 1.90 (m, 2H), 1.72-1.62 (m, 4H), 1.62-1.50 (m, 2H),

1.35-1.30 (m, 4H), 0.88 (t, 3H, J ) 6.9 Hz). ¹³C NMR (CDCl₃)

δ 141.3, 128.9, 128.8, 126.5, 51.3, 51.2, 37.2, 37.0, 33.5, 33.3,

28.4, 27.8, 27.2, 26.6, 24.0, 22.6, 21.9, 21.8, 14.2. ³¹P NMR

(CDCl₃) δ 60.4.

(2S)-2-Benzyloxycarbonylamino-4-(but-3-enyl-methoxy-

phosphinoyl)-butyric Acid Methyl Ester, 33. A solution

of 15 (0.10 g, 0.32 mmol) in CH₂Cl₂(2 mL) freshly dried over

MgSO₄was added to a flame-dried 10 mL round-bottom flask

equipped with a stir bar and condenser. BSA (0.32 mL, 1.3

mmol) was added and the reaction was heated at reflux

temperature for 4 h, after which 26 (0.1 mL, 0.96 mmol) was

added and the reaction was heated at reflux temperature for

22 h. ³¹P NMR analysis of a 0.1 mL aliquot of the reaction

indicated incomplete conversion to the desired product. An

additional 0.1 mL BSA was added and the reaction was heated

at reflux temperature for 18 h, at which time ³¹P NMR analysis

of a second 0.1 mL aliquot indicated ∼90% conversion, and

the reaction was allowed to warm to room temperature. 1 M

HCl (2 mL) was added and the reaction was stirred vigorously

for 30 min. The reaction was then diluted with CH₂Cl₂(5 mL),

and the aqueous layer was extracted with additional CH₂Cl₂

(5 mL). The combined organic layer was washed with 5 mL

each 1 M HCl, H₂O, and brine, then dried over Na₂SO₄,

filtered, and concentrated. The crude product was dissolved

in MeOH and treated with freshly prepared CH₂N₂until a

yellow color persisted. The reaction was stirred for 15 min at

room temperature before quenching with AcOH. The crude

methyl ester was purified by CombiFlash (EtOAc/2% MeOH,

10 g column, 30 mL/min flow rate) to give 0.08 g of 33 as a

clear oil (69% isolated yield, 77% corrected yield (10% of the

reaction mixture had been removed to monitor reaction

progress (³¹P NMR): ¹H NMR (CDCl₃) δ 7.33 (m, 5H), 5.96-

5.74 (overlapping m, 2H), 5.10 (s, 2H), 5.09 (m, 2H), 4.41 (m,

1H), 3.73 (s, 3H), 3.66 (m, 3H), 2.31 (m, 2H), 2.18 (m, 1H),

2.03 (m, 1H), 1.76 (m, 4H). ¹³C NMR (CDCl₃) δ 172.3, 156.4,

137.4, 137.3, 136.5, 128.9, 128.8, 128.7, 128.6, 128.5, 160.0,

67.5, 54.4, 53.1, 51.5, 27.70, 27.65, 26.98, 26.93, 26.26, 26.23,

25.47, 25.39, 24.45, 24.39, 23.71, 23.68. ³¹P NMR (CDCl₃) δ

57.4. MS (ESI) m/z 384.2 ([M + H]⁺, 100). HRMS (ESI) calcd

for C₁₈H₂₇NO₆P 384.1576 [M + H]⁺, found 384.1572.

175.4, 134.2, 129.0, 60.9, 53.8, 51.1, 32.6, 26.9, 14.6. [R]²³

D

-176.4° (c 1.0, CHCl₃). MS (EI) m/z 67.1 ([M-CO₂Et]⁺, 100),

140.1 ([M + H]⁺, 16). HRMS (EI) calcd for C₈H₁₂O₂140.0837

[M]⁺, found 140.0841.

A solution of Ethyl (S)-2-cyclopentene-1-carboxylate (1.1 g,

7.8 mmol) in Et₂O (5 mL) was added dropwise to a stirring

suspension of LiAlH₄(0.41 g, 10.9 mmol) in Et₂O (20 mL) at 0

°C under Ar. The reaction was allowed to warm to room

temperature and stirred for 4 h. The reaction was quenched

using standard workup conditions (0.41 mL H₂O was added,

followed by 0.41 mL 15% aqueous NaOH and finally by 1.23

mL H₂O all at 0 °C).⁸²The reaction was allowed to warm to

room temperature and stirred until all of the excess LiAlH₄

had reacted to form a solid, white precipitate of hydroxides.

The precipitate was removed by filtration and washed with

Et₂O. The filtrate was concentrated and the crude product was

purified by Kugelrohr distillation (20 mmHg, 80-85 °C) to give

0.70 g of 17 as a clear oil (92%): ¹H NMR (CDCl₃) δ 5.87 (m,

1H), 5.66 (m, 1H), 3.56 (t, 2H, J ) 5.7 Hz), 2.92 (m, 1H), 2.35

(m, 2H), 2.01(m, 1H), 1.64 (m, 1H), 1.40 (t, 1H, J ) 5.7 Hz).

¹³C NMR (CDCl₃) δ 133.5, 131.2, 66.6, 48.6, 32.3, 26.1. [R]²⁴

D

-152.8° (c 1.44, CHCl₃). MS (EI) m/z 98.1 ([M]⁺, 7), 67.0 ([M

- CH₂OH]⁺, 100). HRMS (EI) calcd for C₆H₁₀O 98.0732 [M]⁺,

found 98.0730.

(R)-3-(Bromomethyl)-cyclopent-1-ene, 24. PPh₃Br₂/Pyr

Method. (Table 2, Entry 9) To a suspension of PPh₃Br₂(1.4

g, 3.3 mmol) in anhydrous CH₂Cl₂(10 mL) at -10 °C was

added a solution of 16 (0.25 g, 2.5 mmol) and pyridine (0.33

mL, 4.0 mmol) in anhydrous CH₂Cl₂(1 mL). The reaction was

stirred at -10 °C for 10 min and then allowed to warm to room

temperature and stir for 1 h. The reaction mixture was filtered

through a silica gel plug (pentane) and the fractions containing

24 were pooled and concentrated to give 0.33 g of 24 as a clear,

colorless oil (82%): ¹H NMR (CDCl₃) δ 5.87 (m, 1H), 5.65 (m,

1H), 3.35 (m, 2H), 3.13 (m, 1H), 2.35 (overlapping m, 2H), 2.11

(m, 1H), 1.62 (m, 1H). ¹³C NMR (CDCl₃) δ 133.4, 132.1, 48.4,

38.6, 32.1, 28.9. MS (EI) m/z 160.0 and 162.0 ([M]⁺, 6), 67.1

([M - CH₂Br]⁺, 100). HRMS (EI) calcd for C₆H₉Br 159.9888

[M]⁺, found 159.9882.

(S)-3-(Bromomethyl)-cyclopent-1-ene, 25. To a suspen-

sion of PPh₃Br₂(1.4 g, 3.3 mmol) in anhydrous CH₂Cl₂(10 mL)

at -10 °C was added a solution of 17 (0.25 g, 2.5 mmol) and

pyridine (0.33 mL, 4.0 mmol) in anhydrous CH₂Cl₂(1 mL).

The reaction was stirred at -10 °C for 10 min and then allowed

to warm to room temperature and stir for 1 h. The reaction

mixture was absorbed onto silica gel (2 g) and loaded onto a

silica gel plug. The plug was eluted with pentane and the

fractions containing 25 were pooled and concentrated to give

0.31 g of 25 as a clear, colorless oil (79%): ¹H NMR (CDCl₃) δ

5.86 (m, 1H), 5.65 (m, 1H), 3.36 (m, 2H), 3.13 (m, 1H), 2.33

(overlapping m, 2H), 2.10 (m, 1H), 1.62 (m, 1H). ¹³C NMR

(CDCl₃) δ 133.3, 132.1, 48.5, 38.6, 32.0, 28.9. MS (EI) m/z 160.0

and 162.0 ([M]⁺, 5), 67.1 ([M - CH₂Br]⁺, 100). HRMS (EI) calcd

for C₆H₉Br 159.9888 [M]⁺, found 159.9883.

But-3-enyl-(3-phenyl-propyl)-phosphinic Acid Methyl

Ester, 31. A 10 mL round-bottom flash was charged with 27

(0.1 g, 0.5 mmol) and CH₂Cl₂(3 mL) and equipped with a

condenser fitter with a rubber septum. The apparatus was

flushed with Ar and fitted with an Ar-filled balloon. BSA (0.38

mL, 1.5 mmol) was added via syringe through the septum. The

reaction was heated at reflux temperature for 6 h, after which

4-bromo-1-butene (26, 0.1 mL, 1.0 mmol) was added. The

reaction was returned to reflux temperature and heated for

18 h, then quenched by stirring vigorously with 2 M HCl (2

mL) for 15 min. The resulting mixture was diluted with CH₂-

Cl₂(10 mL) and washed with 2 M HCl (2 × 5 mL) and H₂O (5

mL). The organic layer was dried over Na₂SO₄, filtered, and

(82) Fieser, L. F.; Fieser, M. In Reagents for Organic Synthesis; John

Wiley & Sons Inc.: New York, 1967; Vol. 1, p 584.

J. Org. Chem, Vol. 70, No. 17, 2005 6769

Bartley and Coward

(2S)-2-Benzyloxycarbonylamino-4-(cyclopent-1-enyl-

methyl-methoxy-phosphinoyl)-butyric Acid Methyl

Ester, 35. A solution of 15 (0.1 g, 0.32 mmol) and BSA (0.79

mL, 3.2 mmol) in anhydrous CH₂Cl₂(2 mL) was heated at

reflux temperature for 4 h. A solution of 24 (0.52 g, 3.2 mmol)

in anhydrous CH₂Cl₂(1 mL) was added and the reaction

continued at reflux temperature for 44 h. The reaction was

cooled to room temperature, 2 M HCl (2 mL) was added and

the reaction was vigorously stirred for 30 min. CH₂Cl₂(5 mL)

was added and the layers were separated. The aqueous layer

was extracted with CH₂Cl₂(5 mL) and the combined organic

layers were dried over Na₂SO₄and concentrated. The crude

oil was dissolved in MeOH (5 mL) and treated with CH₂N₂

until the reaction remained yellow. The reaction was stirred

for 15 min at room temperature, then quenched with a drop

of AcOH and concentrated. The crude product purified by silica

gel chromatography (EtOAc with 2% MeOH) to give 0.12 g of

35 as a clear colorless oil (85%): ¹H NMR (CDCl₃) δ 7.39-

7.29 (m, 5H), 5.80 (m, 1H), 5.55 (m, 1H), 5.11 (s, 2H), 4.40 (m,

1H), 3.74 (s, 3H), 3.67 (m, 3H), 2.65 (m, 2H), 2.34 (overlapping

m, 4H), 2.18 (m, 1H), 1.95 (m, 1H), 1.89 (m, 2H), 1.8-1.6 (m,

2H). ¹³C NMR (CDCl₃) δ 172.4, 156.4, 136.6, 134.22, 134.16,

130.2, 130.1, 128.9, 128.6, 128.5, 67.1, 54.46, 54.35, 53.0, 51.61,

51.58, 51.56, 51.53, 36.54, 36.52, 33.11, 33.10, 31.83, 31.79,

31.14, 31.09, 25.4, 23.9, 23.82, 23.77, 23.09, 23.03. ³¹P NMR

(CDCl₃) δ 54.1. MS (ESI) m/z 410.2 ([M + H]⁺, 100). HRMS

(ESI) calcd for C₂₀H₂₉NO₆P 410.1733 [M + H]⁺, found 410.1729.

(R)-3-[(phenylthio)methyl]-cyclopent-1-ene. To a sus-

pension of NaH (80% dispersion in mineral oil, 0.019 g, 0.62

mmol) in anhydrous THF (3.5 mL) at 0°C under Ar, PhSH

(0.063 mL, 0.62 mmol) was added dropwise over 5 min. The

suspension was stirred for 30 min. A solution of 24 (0.10 g,

0.62 mmol) in anhydrous THF (3.5 mL) was added dropwise

over 5 min. The reaction was warmed to room temperature

and stirred for 4 h, after which it was concentrated and the

residue partitioned between H₂O (20 mL) and CH₂Cl₂(30 mL).

The aqueous layer was washed with CH₂Cl₂(2 × 30 mL). The

combined organic layers were dried over Na₂SO₄, filtered and

concentrated to give 0.24 g of crude orange oil. The crude

product was purified using a silica gel plug and eluting with

pentane to give 0.07 g of (R)-3-((phenylthio)methyl)-cyclopent-

1-ene as a slightly yellow oil. (60%): ¹H NMR (CDCl₃) δ 7.27

(m, 5H), 5.80 (m, 1H), 5.76 (m, 1H), 2.94 (overlapping m, 3H),

2.36 (m, 2H, 2.13 (m, 1H), 1.62 (m, 1H). ¹³C NMR (CDCl₃) δ

137.2, 133.4, 132.1, 129.2, 129.0, 127.6, 125.9, 45.3, 39.7, 32.1,

29.9.

(R)-3-(Trityloxymethyl)-cyclopent-1-ene, 38. To a solu-

tion of 16 (0.35, 3.6 mmol) in anhydrous pyridine (4 mL) was

added trityl chloride (1.5 g, 5.4 mmol) under Ar and the

reaction was stirred at room temperature for 24 h. The reaction

was filtered through a silica gel plug (4:1 hexanes/EtOAc) and

the filtrate was concentrated to give a mixture of 38 and TrCl.

The crude product was purified by silica gel chromatography

(95:5 hexanes/EtOAc) to give 1.1 g of 38 (92%): ¹H NMR

(CDCl₃) δ 7.6-7.2 (m, 15H), 5.87 (overlapping m, 2H), 3.10

(overlapping m, 3H), 2.39 (m, 2H), 2.11 (m, 1H), 1.61 (m, 1H).

¹³C NMR (CDCl₃) δ 144.6, 132.6, 131.9, 128.9, 128.8, 127.9,

127.8, 127.0, 126.9, 86.3, 67.6, 46.5, 32.0, 26.9. MS (CI) m/z

243.1 ([Tr]⁺, 100), 341.2 ([M + H]⁺, 1).

(1R,2S,3S)-3-(Trityloxymethyl)-cyclopentane-1,2-di-

ol, 39, and (1S,2R,3S)-3-(Trityloxymethyl)-cyclopentane-

1,2-diol, 40. To a stirring solution of 38 (0.5 g, 1.5 mmol) in

7.5 mL THF, 7.5 mL acetone, and 1 mL H₂O at 0 °C was added

OsO₄(0.02 mL of 2.5 wt % in tBuOH) followed by NMO (0.35

g, 8.8 mmol). The reaction was stirred at room temperature

for 30 h. The reaction was diluted with CH₂Cl₂(150 mL) and

washed with 10% aqueous Na₂SO₃(100 mL) and NH₄Cl (100

mL). The combined aqueous phases were extracted with CH₂-

Cl₂(2 × 150 mL). The combined organic layers were dried over

Na₂SO₄, filtered, and concentrated. The crude residue was

purified by silica gel chromatography (2:1 hexanes/EtOAc) to

give 0.30 g of 2,3-trans isomer, 39 (53%) and 0.22 g of the 2,3-

cis isomer, 40 (40%).

Compound 39: ¹H NMR (CDCl₃) δ 7.6-7.2 (m, 15H), 4.12

(m, 1H), 3.80 (m, 1H), 3.40 (Overlapping m, 2H), 3.15 (Over-

lapping m, 2H), 2.41 (m, 1H), 2.10 (m, 1H), 1.97 (m, 1H), 1.86

(m, 1H), 1.32 (m, 1H). ¹³C NMR (CDCl₃) δ 143.9, 128.6, 128.5,

127.9, 127.1, 86.9, 78.3, 73.2, 66.5, 43.4, 30.0, 23.4.

Compound 40: ¹H NMR (CDCl₃) δ 7.5-7.2 (m, 15H), 4.15

(m, 1H), 4.07 (m, 1H), 3.33 (m, 1H), 3.24 (m, 1H), 2.90 (d, 1H,

J ) 3.2 Hz), 2.52 (d, 1H, J ) 6.7 Hz), 2.29 (m, 1H), 1.85 (m,

1H), 1.7-1.6 (overlapping m, 2H), 1.57 (m, 1H). ¹³C NMR

(CDCl₃) δ 143.9, 128.6, 128.1, 127.3, 78.2, 74.6, 74.4, 64.0, 41.4,

30.7, 23.9.

(1R,2S,3S)-3-(Trityloxymethyl)-cyclopentane-1,2-diol

Dimethyl Acetonide, 41. To a solution of 39 (0.69 g, 1.8

mmol) in acetone (20 mL) and dimethoxypropane (20 mL) was

added TsOH (0.03 g, 0.2 mmol) and the reaction was stirred

at room temperature for 10 min. The reaction was diluted with

EtOAc (150 mL) and washed with saturated aqueous NaHCO₃

(50 mL) and brine (50 mL). The organic layer was dried over

Na₂SO₄, filtered and concentrated. The crude product was

purified by silica gel chromatography (95:5 hexanes/EtOAc)

to give 0.73 g of 41 (97%): ¹H NMR (CDCl₃) δ 7.6-7.2 (m,

15H), 4.62 (dd, 1H, J ) 5.4, 5.5), 4.45 (d, 1H, J ) 5.7 Hz), 3.05

(overlapping m, 2H), 2.50 (m, 1H), 2.09 (m, 1H), 1.83 (m, 1H),

1.8-1.6 (overlapping m, 2H), 1.55 (s, 3H), 1.36 (s, 3H). ¹³C

NMR (CDCl₃) δ 144.2, 128.7, 127.8, 127.0, 109.5, 86.6, 83.6,

81.0, 63.7, 46.0, 31.5, 26.6, 25.8, 24.2.

(1S,2R,3S)-3-(Trityloxymethyl)-cyclopentane-1,2-diol

Dimethyl Acetonide, 42. To a solution of 40 (0.2 g, 0.53

mmol) in acetone (10 mL) and dimethoxypropane (10 mL) was

added TsOH (0.01 g, 0.05 mmol) and the reaction was stirred

at room temperature for 15 min. The reaction was diluted with

EtOAc (100 mL) and washed with saturated aqueous NaHCO₃

(50 mL) and brine (50 mL). The organic layer was dried over

Na₂SO₄, filtered and concentrated. The crude product was

purified by silica gel chromatography (95:5 hexanes/EtOAc)

to give 0.21 g of 42 (95%): ¹H NMR (CDCl₃) δ 7.6-7.2 (m,

15H), 4.67 (overlapping m, 2H), 3.47 (dd, 1H, J ) 8.6, 7.3),

3.15 (dd, 1H, J ) 8.6, 7.2 Hz), 1.92 (m, 1H), 1.85 (m, 1H), 1.69

(m, 1H), 1.5-1.4 (overlapping m, 2H), 1.37 (s, 3H), 1.33 (s, 3H).

¹³C NMR (CDCl₃) δ 144.6, 129.0, 127.9, 126.9, 109.1, 86.5, 81.0,

80.5, 62.8, 45.6, 32.4, 26.1, 26.0, 24.1.

(1R,2S,3S)-3-(Hydroxymethyl)-cyclopentane-1,2-diol Di-

methyl Acetonide, 43. To a solution of 41 (0.9 g, 2.2 mmol)

in EtOH (100 mL) was added 10% Pd(OH)₂on carbon (Pearl-

man’s catalyst) (1 g) and the reaction was shaken on a Parr

hydrogenator (40 psi H₂) for 8 h. The catalyst was removed

by filtering through silica gel using EtOAc as solvent and

concentrating the eluant. The crude product was purified by

silica gel chromatography (CHCl₃to elute triphenylmethane,

then EtOAc to elute the desired product) to give 0.32 g of 43

(85%): ¹H NMR (CDCl₃) δ 4.65 (dd, 1H, J ) 4.9, 5.5 Hz), 4.45

(d, 1H, J ) 5.7 Hz), 3.45 (m, 2H), 2.22 (m, 1H), 1.98 (m, 1H),

1.82 (m, 1H), 1.75(m, 1H), 1.48 (m, 2H), 1.45 (s, 3H), 1.30 (s,

3H). ¹³C NMR (CDCl₃) δ 110.1, 83.4, 81.0, 63.6, 48.4, 31.3, 26.7,

25.4, 24.3.

(1S,2R,3S)-3-(Hydroxymethyl)-cyclopentane-1,2-diol Di-

methyl Acetonide, 44. To a solution of 42 (0.21 g, 0.51 mmol)

in EtOH (20 mL) was added 10% Pd(OH)₂on carbon (Pearl-

man’s catalyst) (0.2 g) and the reaction was shaken on a Parr

hydrogenator (40 psi H₂) for 12 h. The catalyst was removed

by filtering through silica gel using EtOAc as solvent and

concentrating the eluant. The crude product was purified by

CombiFlash chromatography (1:1 hexanes/EtOAc) to give 0.08

g of 44 (94%): ¹H NMR (CDCl₃) δ 4.65 (overlapping m, 2H),

3.88 (dd, 1H, J ) 11.1, 4.2 Hz), 3.77 (dd, 1H, J ) 11.1, 6.8

Hz), 2.53 (br s, 1H), 1.90-1.75 (overlapping m, 2H), 1.71(m,

1H), 1.6-1.3 (overlapping m, 2H), 1.44 (s, 3H), 1.30 (s, 3H).

¹³C NMR (CDCl₃) δ 109.5, 81.6, 81.2, 62.1, 46.4, 32.1, 25.9,

24.5, 23.8.

6770 J. Org. Chem., Vol. 70, No. 17, 2005

Synthesis of Phosphinic Acid Phosphapeptides

(1R,2S,3R)-3-(Bromomethyl)-cyclopentane-1,2-diol Di-

methyl Acetonide, 36. To a suspension of PPh₃Br₂(0.32 g,

0.75 mmol) in anhydrous CH₂Cl₂(4 mL) at -10 °C was added

a solution of 43 (0.10 g, 0.58 mmol) and pyridine (0.08 mL,

0.93 mmol) in anhydrous CH₂Cl₂(3 mL). The reaction was

stirred at -10 °C for 10 min and then allowed to warm to room

temperature and stir for 2 h. Silica gel (2 g) was added and

the reaction was concentrated by rotary evaporation. The dry

silica containing the crude reaction mixture was loaded onto

a silica gel column. Elution of PPh₃(98:2 hexanes/EtOAc)

followed by the desired product (9:1 hexanes/EtOAc), pooling

of product-containing fractions, and removal of solvents in

vacuo gave 0.14 g of 36 as a clear oil (79%): ¹H NMR (CDCl₃)

δ 4.64 (d, 1H, J ) 5.4 Hz), 4.41 (d, 1H, J ) 5.6 Hz), 3.25 (m,

2H), 2.42 (m, 1H), 2.05 (m, 1H), 1.79 (overlapping m, 2H), 1.56

(m, 1H), 1.42 (s, 3H), 1.28 (s, 3H). ¹³C NMR (CDCl₃) δ 110.1,

83.4, 81.0, 63.6, 48.4, 31.3, 26.7, 25.4, 24.3. MS (CI) m/z 97

([M - Br - acetone]⁺, 100), 235.1 ([M + H]⁺, 8). HRMS (CI)

calcd for C₉H₁₆BrO₂235.0334 [M + H]⁺, found 235.0327.

(1S,2R,3R)-3-(Bromomethyl)-cyclopentane-1,2-diol Di-

methyl Acetonide, 37. To a suspension of PPh₃Br₂(0.20 g,

0.46 mmol) in anhydrous CH₂Cl₂(3 mL) at -10 °C was added

a solution of 44 (0.06 g, 0.35 mmol) and pyridine (0.04 mL,

0.56 mmol) in anhydrous CH₂Cl₂(2 mL). The reaction was

stirred at -10 °C for 10 min and then allowed to warm to room

temperature and stir for 2 h. Silica gel (1 g) was added and

the reaction was concentrated by rotary evaporation. The dry

silica containing the crude reaction mixture was loaded onto

a silica gel column. Elution of PPh₃(98:2 hexanes/EtOAc)

followed by the desired product (9:1 hexanes/EtOAc), pooling

of product-containing fractions, and removal of solvents in

vacuo gave 0.07 g of 37 as a clear oil (82%): ¹H NMR (CDCl₃)

δ 4.57 (m, 1H), 4.45 (m, 1H), 1.93 (m, 1H), 1.76 (m, 1H), 1.61

(m, 1H), 1.4-1.3 (overlapping m, 6H), 1.20 (s, 3H). ¹³C NMR

(CDCl₃) δ 109.6, 80.9, 80.2, 48.1, 32.4, 32.0, 31.7, 27.9, 26.0,

24.0, 22.8, 14.3. MS (CI) m/z 97 ([M - Br - acetone]⁺, 100),

235.1 ([M + H]⁺, 15). HRMS (CI) calcd for C₉H₁₆BrO₂235.0334

[M + H]⁺, found 235.0328.

(1S,2R,3R)-3-(Trityloxymethyl)-cyclopentane-1,2-di-

ol, 45, and (1R,2S,3R)-3-(Trityloxymethyl)-cyclopentane-

1,2-diol, 46. To a solution of 17 (0.5, 5.1 mmol) in anhydrous

pyridine (5 mL) was added trityl chloride (2.1 g, 7.7 mmol)

under Ar and the reaction was stirred at room temperature

for 24 h. The reaction was filtered through a silica gel plug

(4:1 hexanes/EtOAc) and the filtrate was concentrated to give

a mixture of the desired product and TrCl. The crude product

was purified by silica gel chromatography (95:5 hexanes/

EtOAc) to give 1.6 g of (S)-3-(trityloxymethyl)-cyclopent-1-ene

(94%): ¹H NMR (CDCl₃) δ 7.6-7.2 (m, 15H), 5.87 (overlapping

m, 2H), 2.99 (overlapping m, 3H), 2.39 (m, 2H), 2.00 (m, 1H),

1.52 (m, 1H). ¹³C NMR (CDCl₃) δ 144.6, 132.6, 131.9, 128.9,

128.8, 127.9, 127.8, 127.0, 126.9, 86.3, 67.6, 46.5, 32.0,

26.9.

To a stirring solution of (S)-3-(trityloxymethyl)-cyclopentene

(1.5 g, 4.4 mmol) in 20 mL THF, 20 mL acetone, and 2 mL

H₂O at 0 °C was added OsO₄(0.05 mL of 2.5 wt % in tBuOH)

followed by NMO (1.0 g, 8.8 mmol). The reaction was stirred

at room temperature for 36 h. The reaction was diluted with

CH₂Cl₂(300 mL) and washed with 10% aqueous Na₂SO₃(300

mL) and NH₄Cl (300 mL). The combined aqueous phases were

extracted with CH₂Cl₂(3 × 200 mL). The combined organic

layers were dried over Na₂SO₄, filtered, and concentrated The

crude residue was purified by silica gel chromatography (2:1

hexanes/EtOAc) to give 0.90 g of 2,3-trans isomer, 45 (56%),

0.45 g of the 2,3-cis isomer, 46 (28%), and 0.29 g overlapping

fractions containing a mixture of 45 and 46 (18%).

Compound 46: ¹H NMR (CDCl₃) δ 7.5-7.2 (m, 15H), 4.15

(m, 1H), 4.07 (m, 1H), 3.33 (m, 1H), 3.24 (m, 1H), 2.90 (d, 1H,

J ) 3.2 Hz), 2.52 (d, 1H, J ) 6.7 Hz), 2.29 (m, 1H), 1.85 (m,

1H), 1.7-1.6 (overlapping m, 2H), 1.57 (m, 1H). ¹³C NMR

(CDCl₃) δ 143.9, 128.6, 128.1, 127.3, 87.2, 74.6, 74.4, 64.0, 41.4,

30.7, 23.9.

(1S,2R,3S)-3-(Bromomethyl)-cyclopentane-1,2-diol Di-

methyl Acetonide, 47. To a solution of 45 (0.87 g, 2.3 mmol)

in acetone (25 mL) and dimethoxypropane (25 mL) was added

TsOH (0.04 g, 0.2 mmol) and the reaction was stirred at room

temperature for 15 min. The reaction was diluted with EtOAc

(200 mL) and washed with saturated aqueous NaHCO₃(50

mL) and brine (50 mL). The organic layer was dried over Na₂-

SO₄, filtered and concentrated. The crude product was purified

by silica gel chromatography (95:5 hexanes/EtOAc) to give 0.91

g of (1S,2R,3R)-3-(trityloxymethyl)-cyclopentane-1,2-diol di-

methyl acetonide (96%): ¹H NMR (CDCl₃) δ 4.65 (dd, 1H, J )

4.9, 5.5 Hz), 4.45 (d, 1H, J ) 5.7 Hz), 3.45 (m, 2H), 2.22 (m,

1H), 1.98 (m, 1H), 1.82 (m, 1H), 1.75(m, 1H), 1.48 (m, 2H),

1.45 (s, 3H), 1.30 (s, 3H). ¹³C NMR (CDCl₃) δ 110.1, 83.4, 81.0,

63.6, 48.4, 31.3, 26.7, 25.4, 24.3.

To a solution of (1S,2R,3R)-3-(trityloxymethyl)-cyclopentane-

1,2-diol dimethyl acetonide (0.9 g, 2.2 mmol) in EtOH (100 mL)

was added 10% Pd(OH)₂on carbon (Pearlman’s catalyst) (1 g)

and the reaction was shaken on a Parr hydrogenator (40 psi

H₂) for 8 h. The catalyst was removed by filtering through silica

gel using EtOAc as solvent and concentrating the eluant. The

crude product was purified by CombiFlash chromatography

(1:1 hexanes/EtOAc) to give 0.32 g of (1S,2R,3R)-3-(hydroxy-

methyl)-cyclopentane-1,2-diol dimethyl acetonide (85%): ¹H

NMR (CDCl₃) δ 4.65 (dd, 1H, J ) 4.9, 5.5 Hz), 4.45 (d, 1H, J

) 5.7 Hz), 3.45 (m, 2H,), 2.22 (m, 1H), 1.98 (m, 1H), 1.82 (m,

1H), 1.75(m, 1H), 1.48 (m, 2H), 1.45 (s, 3H), 1.30 (s, 3H). ¹³C

NMR (CDCl₃) δ 110.1, 83.4, 81.0, 63.6, 48.4, 31.3, 26.7, 25.4,

24.3.

To a suspension of PPh₃Br₂(0.72 g, 1.7 mmol) in anhydrous

CH₂Cl₂(7 mL) at -10 °C was added a solution of (1S,2R,3R)-

3-(hydroxymethyl)-cyclopentane-1,2-diol dimethyl acetonide

(0.22 g, 1.3 mmol) and pyridine (0.72 mL, 1.7 mmol) in

anhydrous CH₂Cl₂(3 mL). The reaction was stirred at -10 °C

for 10 min and then allowed to warm to room temperature

and stir for 2 h. Silica gel (2 g) was added and the reaction

was concentrated by rotary evaporation. The dry silica con-

taining the crude reaction mixture was loaded onto a silica

gel column. Elution of PPh₃(98:2 hexanes/EtOAc) followed by

the desired product (9:1 hexanes/EtOAc), pooling of product-

containing fractions, and removal of solvents in vacuo gave

0.27 g of 47 as a clear oil (87%; 60% overall yield from 45): ¹H

NMR (CDCl₃) δ 4.67 (m, 1H), 4.44 (d, 1H, J ) 5.7 Hz), 3.28

(m, 2H), 2.46 (m, 1H), 2.08 (m, 1H), 1.9-1.7 (overlapping m,

2H), 1.55 (m, 1H), 1.46 (s, 3H), 1.31 (s, 3H). ¹³C NMR (CDCl₃)

δ 110.4, 84.6, 80.8, 48.3, 34.7, 31.0, 28.0, 26.7, 24.4. MS (CI)

m/z 97 ([M - Br - acetone]⁺, 100), 235.1 ([M + H]⁺, 11). HRMS

(CI) calcd for C₉H₁₆BrO₂235.0334 [M + H]⁺, found 235.0324.

(1R,2S,3S)-3-(Bromomethyl)-cyclopentane-1,2-diol Di-

methyl Acetonide, 48. To a solution of 46 (0.4 g, 1.1 mmol)

in acetone (15 mL) and dimethoxypropane (15 mL) was added

TsOH (0.02 g, 0.1 mmol) and the reaction was stirred at room

temperature for 20 min. The reaction was diluted with EtOAc

(150 mL) and washed with saturated aqueous NaHCO₃(50

mL) and brine (50 mL). The organic layer was dried over Na₂-

SO₄, filtered and concentrated. The crude product was purified

by silica gel chromatography (95:5 hexanes/EtOAc) to give 0.54

g of (1R,2S,3R)-3-(trityloxymethyl)-cyclopentane-1,2-diol di-

methyl acetonide (74%): ¹H NMR (CDCl₃) δ 7.6-7.2 (m, 15H),

4.67 (overlapping m, 2H), 3.47 (dd, 1H, J ) 8.6, 7.3 Hz), 3.15

(dd 1H, J ) 8.6, 7.2 Hz), 1.92 (m, 1H), 1.85 (m, 1H), 1.69 (m,

1H), 1.5-1.4 (overlapping m, 2H), 1.37 (s, 3H), 1.33 (s, 3H).

¹³C NMR (CDCl₃) δ 144.6, 129.0, 127.9, 126.9, 109.1, 86.5, 81.0,

80.5, 62.8, 45.6, 32.4, 26.1, 26.0, 24.1.

Compound 45: ¹H NMR (CDCl₃) δ 7.5-7.2 (m, 15H), 4.15

(m, 1H), 4.07 (m, 1H), 3.33 (m, 1H), 3.24 (m, 1H), 2.90 (d, 1H,

J ) 3.2 Hz), 2.52 (d, 1H, J ) 6.7 Hz), 2.29 (m, 1H), 1.85 (m,

1H), 1.7-1.6 (overlapping m, 2H), 1.57 (m, 1H). ¹³C NMR

(CDCl₃) δ 143.9, 128.6, 128.1, 127.3, 87.2, 74.6, 74.4, 64.0, 41.4,

30.7, 23.9.

To a solution of (1R,2S,3R)-3-(trityloxymethyl)-cyclopentane-

1,2-diol dimethyl acetonide (0.44 g, 1.1 mmol) in EtOH (75 mL)

J. Org. Chem, Vol. 70, No. 17, 2005 6771

Bartley and Coward

was added 10% Pd(OH)₂on carbon (Pearlman’s catalyst) (0.5

g) and the reaction was shaken on a Parr hydrogenator (40

psi H₂) for 12 h. The catalyst was removed by filtering through

silica gel using EtOAc as solvent and concentrating the eluant.

The crude product was purified by silica gel chromatography

(CHCl₃to elute Ph₃CH then EtOAc to elute the desired

product) to give 0.09 g of (1R,2S,3R)-3-(hydroxymethyl)-

cyclopentane-1,2-diol dimethyl acetonide (50%): ¹H NMR

(CDCl₃) δ 4.65 (overlapping m, 2H), 3.88 (dd, 1H, J ) 11.1,

4.2 Hz), 3.77 (dd, 1H, J ) 11.1, 6.8 Hz), 2.53 (br s, 1H), 1.90-

1.75 (overlapping m), 1.71(m, 1H), 1.6-1.3 (overlapping m,

2H), 1.44 (s, 3H), 1.30 (s, 3H). ¹H NMR (CDCl₃) δ 109.5, 81.6,

81.2, 62.1, 46.4, 32.1, 25.9, 24.5, 23.8.

To a suspension of PPh₃Br₂(0.46 g, 1.4 mmol) in anhydrous

CH₂Cl₂(6 mL) at -10 °C was added a solution of (1R,2S,3R)-

3-(hydroxymethyl)-cyclopentane-1,2-diol dimethyl acetonide

(0.15 g, 0.9 mmol) and pyridine (0.46 mL, 1.1 mmol) in

anhydrous CH₂Cl₂(4 mL). The reaction was stirred at -10 °C

for 10 min and then allowed to warm to room temperature

and stir for 2 h. Silica gel (2 g) was added and the reaction

was concentrated by rotary evaporation. The dry silica con-

taining the crude reaction mixture was loaded onto a silica

gel column. Elution of PPh₃(98:2 hexanes/EtOAc) followed by

the desired product (9:1 hexanes/EtOAc), pooling of product-

containing fractions, and removal of solvents in vacuo gave

0.27 g of 48 as a clear oil (65%; 24% overall yield from 46): ¹H

NMR (CDCl₃) δ 4.67 (m, 1H), 4.56 (m, 1H), 3.59 (m, 1H), 3.41

(m, 1H), 2.06 (m, 1H), 1.87 (m, 1H), 1.73 (m, 1H), 1.6-1.45

(overlapping m, 2H), 1.43 (s, 3H), 1.31 (s, 3H). MS (CI) m/z 97

([M - Br - acetone]⁺, 100), 235.1 ([M + H]⁺, 25). HRMS (CI)

calcd for C₉H₁₆BrO₂235.0334 [M + H]⁺, found 235.0323.

2-Methylene-pentanedioic Acid 5-tert-Butyl Ester, 53c.

A. From 54a/57a. A mixture of 55a (30.3 g, 189 mmol), 56

(27.7 mL, 189 mmol), K₂CO₃(26.12 g, 189 mmol) and tetrabu-

tylammonium bromide (0.7 g, 1.9 mmol) in anhydrous benzene

(100 mL) was heated at reflux temperature for 12 h. The

reaction was filtered and the solvent concentrated to give 53.2

g of crude mixture of 54a and 57a. A portion of 54a/57a was

carried on without further purification. ¹H NMR (CDCl₃) δ 4.18

(m, 4H), 3.41 (t, J ) 7.4 Hz, 1H), 2.35-2.1 (overlapping m,

4H), 1.43 (s, 9H), 1.24 (m, 6H). ¹³C NMR (CDCl₃) δ 171.9, 169.2,

80.8, 61.6, 51.0, 32.9, 28.2, 24.1, 14.2.

To a solution of 54a/57a (10.0 g) in EtOH (8 mL) was added

a solution of KOH (3.6 g, 69.4 mmol) in H₂O (4 mL). The

reaction was stirred at room temperature for 18 h. The reaction

was extracted with Et₂O (2 × 20 mL). The aqueous layer was

acidified to pH 1.0 by the dropwise addition of 6 M HCl and

then extracted with Et₂O (3 × 20 mL). The combined organic

layer was dried over Na₂SO₄, filtered, and concentrated to give

7.0 g of crude mixture of 58 and 59, which was carried on

without further purification. ¹H NMR (CDCl₃) δ 3.74 (m, 1H),

3.49 (m, 1.4H), 2.37 (m, 2H), 2.19 (m, 2H), 1.43 (s, 9H), 1.22

(m, 3.2H).

2.47 (t, 2H, J ) 7.6), 1.45 (s, 9H). ¹³C NMR (CDCl₃) δ 172.4,

172.2, 138.7, 128.1, 80.7, 34.3, 28.2, 27.2. MS (ESI) m/z 223.2

([M + Na]⁺, 100). HRMS (ESI) calcd for C₁₀H₁₆NaO₄223.0944

[M + Na]⁺, found 223.0946.

B. From 58. To a suspension of 58 (0.19 g, 0.8 mmol),

prepared as described below, in formaldehyde (37 wt % in H₂O,

1 mL) was added Et₂NH (0.11 mL, 1.1 mmol) at room

temperature. The reaction was heated at reflux temperature

for 2 h and allowed to cool to room temperature and stir

overnight. The reaction became homogeneous during reflux

and remained so throughout the reaction. The reaction was

partitioned between EtOAc (10 mL) and 1 M HCl (5 mL). The

aqueous layer was extracted with EtOAc (10 mL) and CHCl₃

(2 × 20 mL). The combined organic layer dried over Na₂SO₄,

filtered, and concentrated. The crude product was purified by

silica gel chromatography (1:1 hexanes/EtOAc) to give 65 mg

of 53c as a white solid (40%): mp 61-63 °C. ¹H NMR and ¹³

C

NMR spectral data are identical to those reported above for

the sample of 53c synthesized from the mixture of 54a and

57a. HRMS (ESI) calcd for C₁₀H₁₆NaO₄223.0944 [M + Na]⁺,

found 223.0943.

2-Benzyloxycarbonyl-pentanedioic Acid 1-Benzyl Es-

ter 5-tert-Butyl Ester, 54b. To a solution of 55b (0.57, 2.0

mmol) in anhydrous THF (1 mL) under Ar at room tempera-

ture was added NaH (2 mg, 0.1 mmol) in one portion. After

the evolution of H₂gas ceased, 56 (0.15 mL, 1.0 mmol) was

added dropwise over a period of 30 min. The reaction was

stirred at room temperature under Ar for 3.5 h. The reaction

was concentrated and the crude residue was partitioned

between EtOAc (10 mL) and saturated aqueous NH₄Cl (5 mL).

The organic layer was washed with brine (5 mL), dried over

Na₂SO₄, filtered and concentrated. The crude product was

purified by CombiFlash chromatography (9:1 hexanes/EtOAc,

120 g column, 80 mL/min flow rate) to give 0.34 g of 55b as a

clear, colorless oil (83%): ¹H NMR (CDCl₃) δ 7.30 (m, 10H),

5.15 (s, 4H), 3.56 (t, 1H, J ) 6.9 Hz), 2.3-2.1 (overlapping m,

4H), 1.42 (s, 9H). MS (ESI) m/z 435.2 [M + Na]⁺.

2-Carboxy-pentanedioic Acid 5-tert-Butyl Ester, 58. To

a solution of 54b (0.34 g, 0.82 mmol) in anhydrous THF (10

mL) was added 10% Pd/C and the suspension was purged with

H₂from a balloon for 30 s. The reaction was stirred under H₂

for 5 h. The reaction was filtered through a plug of Celite with

EtOAc. The filtrate was concentrated to give 0.19 g of 58 as a

1

clear oil (100%). H NMR (CDCl₃) δ 11.4 (br s, 2H), 4.13 (br

m, 1H), 2.4-2.0 (overlapping m, 4H), 1.43 (s, 9H).

2-Methylene-pentanedioic Acid 1-benzotriazol-1-yl Es-

ter 5-tert-Butyl Ester, 60. To a solution of 53c (0.25 g, 1.2

mmol) in anhydrous THF (10 mL) under Ar was added HOBt

(0.16 g, 1.2 mmol) followed by DCC (0.25 g, 1.2 mmol) and the

reaction was stirred at room temperature for 4 h. The entire

reaction was filtered through a column of silica gel with CHCl₃

as solvent. Tubes containing the desired product were pooled

and concentrated to give 0.34 g of 60 as a clear oil (89%): ¹H

NMR (CDCl₃) δ 8.08 (m, 1H), 7.55 (m, 1H), 7.45 (m, 2H), 6.7

(s, 1H), 6.10 (s, 1H), 2.79 (t, 2H, J ) 7.3 Hz), 2.57 (t, 2H, J )

7.3 Hz), 1.46 (s, 9H). ¹³C NMR (CDCl₃) δ 171.4, 162.7, 143.7,

135.2, 131.4, 128.8, 125.0, 1120.7, 108.5, 33.9, 28.3, 27.6. MS

(ESI) m/z 283.1([M - tBu + Na]⁺, 100), 318.1 ([M + H⁺, 58).

HRMS (ESI) calcd for C₁₆H₁₉N₃NaO₄340.1273 [M + Na]⁺,

found 340.1270.

4-(4-Benzhydryl-2-oxo-oxazolidine-3-carbonyl)-pent-4-

enoic Acid tert-Butyl Ester, 52. To a solution of (S)-(-)-4-

(diphenylmethyl)-2-oxazolidinone (0.25 g, 1.0 mmol) in anhy-

drous THF (5 mL) at -78 °C under Ar was added nBuLi (1.6

M in hexanes, 0.63 mL, 1.0 mmol). The reaction was stirred

at -78 °C for 30 min. This solution was added via syringe to

a solution of 60 (0.32 g, 1.0 mmol) in anhydrous THF (5 mL)

cooled to -78 °C. The reaction was stirred at -78 °C for 20

min and then allowed to warm to room temperature and stir

for 1 h. The reaction was quenched with saturated aqueous

NH₄Cl (3 mL). The organic layer was removed by rotary

evaporation and resulting aqueous layer was diluted with H₂O

To a solution of 58/59 (7.0 g) in formaldehyde (37 wt % in

H₂O, 11 mL) was added Et₂NH (4.2 mL, 40.4 mmol) at room

temperature. The reaction was heated at reflux temperature

for 5 h. The reaction was allowed to cool to room temperature

and was diluted with CHCl₃(50 mL) and saturated aqueous

NaHCO₃(25 mL). The aqueous layer was extracted with CHCl₃

(50 mL). The NaHCO₃layer was acidified to pH 1.0 with 6 M

HCl and extracted with CHCl₃(100 mL then 50 mL). The

organic layer was washed with brine (50 mL), dried over Na₂-

SO₄, filtered and concentrated to give 0.57 g of crude product.

The original organic washes were acidified to pH 1.0 with 1

M HCl. The aqueous layer was extracted with CHCl₃(50 mL).

The combined organic layer was washed with H₂O (50 mL)

and brine (50 mL), dried over Na₂SO₄, filtered and concen-

trated to give an additional 3.4 g of crude product. The crude

product was combined and purified by silica gel chromatog-

raphy (8:1 hexanes/EtOAc with 2% AcOH) to give 3.4 g of 53c

as a colorless solid (49% overall from 55a): mp 61-63 °C. ¹H

NMR (CDCl₃) δ 6.34 (s, 1H), 5.73 (s, 1H), 2.61 (t, 2H, J ) 7.5),

6772 J. Org. Chem., Vol. 70, No. 17, 2005

Synthesis of Phosphinic Acid Phosphapeptides

(5 mL) and extracted with CH₂Cl₂(2 × 10 mL). The organic

layer was dried over Na₂SO₄, filtered, and concentrated. The

crude product was purified by silica gel chromatography (2:1

hexanes/EtOAc) to give 0.36 g of 52 as a clear oil (84%): ¹H

NMR (CDCl₃) δ 7.4-7.1 (m, 10H), 5.40 (m, 1H), 5.29 (s, 1H),

5.11 (s, 1H), 4.66 (d, 1H, J ) 7.2 Hz), 4.43 (t, 1H, J ) 9.2 Hz),

4.36 (dd, 1H, J ) 3.5, 9.2 Hz), 2.48 (m, 2H), 2.29 (m, 2H), 1.43

(s, 9H). ¹³C NMR (CDCl₃) δ 172.2, 166.7, 146.6, 141.6, 139.3,

138.8, 129.8, 128.4, 128.3, 128.0, 127.9, 127.8, 125.8, 122.4,

80.6, 63.0, 34.3, 28.2, 27.6. MS (ESI) m/z 402.2 ([M - tBu +

Na]⁺, 100), 458.2 ([M + Na]⁺, 56). HRMS (ESI) calcd for C₂₆H₂₉-

NNaO₅458.1943 [M + Na]⁺, found 458.1941.

stirred at room temperature for 15 h. The reaction was

quenched with abs EtOH (0.27 mL), which did not fully cleave

the TMS esters as indicated by ³¹P NMR. Two drops of TFA

were added and the reaction was stirred at room temperature

for 25 min to cleave the remaining esters. The reaction was

diluted with CH₂Cl₂(20 mL) and H₂O (5 mL). The organic

layer was washed with brine (5 mL), dried over Na₂SO₄,

filtered, and concentrated. The crude material was dissolved

in MeOH (5 mL) treated with TMSCHN₂until a persistent

yellow color was observed. The reaction was stirred at room

temperature for 30 min and then quenched with AcOH. The

reaction was concentrated and the crude material was purified

by silica gel chromatography (4:1 hexanes/EtOAc) to give 90

mg of 64 (higher R_f, isomer A) and 85 mg of 65, (lower R_f,

isomer B (combined yield, 72%).

Compound 64: analytical HPLC t_R) 22.4 min. ¹H NMR

(CDCl₃) δ 7.31 (m, 15H), 5.90 (m, 1H), 5.29 (m, 1H), 5.11 (s,

2H), 4.74 (m, 1H), 4.43 (m, 2H), 4.12 (m, 1H), 3.8-3.6

(overlapping m, 6H), 2.3-1.6 (overlapping m, 11H), 1.42 (s,

9H). ¹³C NMR (CDCl₃) δ 173.8, 173.7, 173.6, 172.06, 172.04,

131.2, 156.1, 153.3, 139.8, 138.21, 138.16, 136.3, 129.47,

129.39, 128.97, 128.88, 128.77, 128.73, 128.55, 128.49, 128.41,

128.17, 128.15, 127.84, 127.81, 127.2, 80.7, 67.1, 65.3, 65.0,

60.4, 56.74, 56.67, 54.1, 52.67, 52.63, 51.51, 51.46, 51.22, 51.17,

51.14, 50.81, 36.6, 36.4, 32.37, 32.32, 30.9, 29.6, 29.4, 28.9, 28.1,

27.6, 27.3, 25.0, 24.6, 24.2, 23.8. ³¹P NMR (CDCl₃) δ 55.2 and

54.7. MS (ESI) m/z 730.2 ([M - tBu + Na]⁺, 100), 787.3 ([M +

Na]⁺, 8). HRMS (ESI) calcd for C₄₀H₄₉N₂NaO₁₁P 787.2972 [M

+ Na]⁺, found 787.2967.

(S)-2-Benzyloxycarbonylamino-4-[methoxy-(2-meth-

oxycarbonyl-ethyl)-phosphinoyl]-butyric Acid Methyl

Ester, 62. To a solution of 15 (57 mg, 0.18 mmol) in anhyd

CH₂Cl₂(0.8 mL) was added methyl acrylate (24 µL, 0.27 mmol)

and BSA (130 µL, 0.54 mmol) at room temperature. The

reaction was stirred under Ar for 20 h. Reaction progress was

monitored by ³¹P NMR. The reaction was quenched with 1 M

HCl (1 mL). CH₂Cl₂(10 mL) was added and the reaction was

washed with 1 M HCl (2 × 10 mL) and H₂O (10 mL). The

organic layer was dried over Na₂SO₄, filtered and concentrated.

The crude product was dissolved in MeOH (2 mL) and

TMSCHN₂(2.0 M in hexanes) was added until a persistent

yellow color was observed. The reaction was stirred at room

temperature for 30 min, then quenched with AcOH and the

solution concentrated. The crude material was purified by

CombiFlash chromatography (EtOAc w/ 2% MeOH, 4 g col-

umn, 18 mL/min flow rate) to give 50 mg of 62 as clear oil.

1

Compound 65: analytical HPLC t_R) 23.1 min. ¹H NMR

(CDCl₃) δ 7.27 (m, 15H), 5.90 (m, 1H), 5.42 (m, 1H), 5.10 (s,

2H), 4.54 (m, 1H), 4.43 (m, 2H), 4.12 (m, 1H), 3.8-3.6

(overlapping m, 6H), 2.3-1.6 (overlapping m, 11H), 1.41 (s,

9H, tBu). ³¹P NMR (CDCl₃) δ 55.4 and 54.9. MS (ESI) m/z 730.2

([M - tBu + Na]⁺, 100), 787.3 ([M + Na]⁺, 5). HRMS (ESI)

calcd for C₄₀H₄₉N₂NaO₁₁P 787.2972 [M + Na]⁺, found 787.2965.

(67%) H NMR (CDCl₃) δ 7.28 (m, 5H), 7.75 (m, 1H), 5.10 (s,

2H), 4.38 (m, 1H), 3.74-3.63 (overlapping, 9H), 2.57 (m, 2H),

2.25-1.80 (overlapping m, 4H), 1.80-1.70 (m, 2H). ¹³C NMR

(CDCl₃) δ 172.82, 172.78, 172.64, 172.60, 172.0, 156.1, 136.2,

128.6, 128.5, 128.4, 128.3, 67.2, 54.18 (d, J ) 3.1 Hz), 53.97

(d, J ) 3.7 Hz), 53.8, 52.3, 51.40 (d, J ) 6.6 Hz), 51.37 (d, J )

6.6 Hz), 26.6-26.5 (overlapping), 25.17-25.06 (overlapping),

24.5, 23.4, 23.29, 23.26, 22,2, 22.0. ³¹P NMR (CDCl₃) 56.2 and

56.1. MS (ESI) m/z 416.1 ([M + H]⁺, 100). HRMS (ESI) calcd

for C₁₈H₂₇NO₈P 416.1474 [M + H]⁺, found 416.1469.

(3′S)-2-[(3′-Benzyloxycarbonylamino-3′-methoxy-

carbonyl-propyl)-methoxy-phosphinoylmethyl]-pentane-

1,5-dioic Acid 5-tert-Butyl Ester 1-Methyl Ester, 66. To a

solution of 64 (0.28 g, 0.37 mmol) in 3:1 THF/H₂O (4 mL) at 0

°C was added H₂O₂(0.17 mL, 1.5 mmol) followed by solid LiOH

(31 mg, 0.74 mmol). The reaction was stirred at 0 °C for 30

min. The reaction was quenched by the addition of Na₂SO₃

(0.28 g) in H₂O (1 mL). The reaction was extracted with Et₂O

(2 × 20 mL) to recover the chiral auxiliary. The aqueous layer

was acidified to pH 1.0 with 2 M HCl and extracted with Et₂O

(3 × 30 mL). The organic layer was concentrated and the crude

material dissolved in MeOH (5 mL) and treated with CH₂N₂

until a persistent yellow color was observed. The reaction was

stirred at room temperature for 15 min, then quenched with

AcOH and concentrated to give 0.17 g of 66 as a clear oil

(84%): ¹H NMR (CDCl₃) δ 7.32 (m, 5H), 5.93 (d, 1H, J ) 6.1),

5.10 (s, 2H), 4.38 (m, 1H), 3.8-3.6 (overlapping m, 9H), 2.84

(m, 1H), 2.3-2.1 (overlapping m, 4H), 1.90 (m, 3H), 1.76 (m,

3H), 1.43 (s, 9H). ¹³C NMR (CDCl₃) δ 174.71, 174.69, 174.67,

174.65, 172.02, 171.98, 171.78, 171.73, 156.1, 136.3, 128.8,

128.7, 128.5, 128.21, 128.15, 80.70, 80.68, 67.1, 54.2, 54.12,

54.07, 54.0, 52.6, 52.1, 51.34, 51.29, 51.22, 51.17, 51.12, 38.63,

38.61, 38.44, 38.41, 32.63, 32.57, 31.8, 31.2, 30.3, 30.0, 29.5,

29.3, 28.1, 25.6, 24.95, 24.93, 24.86, 24.77, 24.5, 24.4, 24.1, 23.8.

³¹P NMR (CDCl₃) δ 55.4 and 54.9. MS (ESI) m/z 509.1 ([M -

tBu + Na]⁺, 100), 566.2 ([M + Na]⁺, 10). HRMS (ESI) calcd

for C₂₅H₃₈NNaO₁₀P 566.2131 [M + Na]⁺, found 566.2127.

(S)-2-[(3-Benzyloxycarbonylamino-3-methoxycarbonyl-

propyl)-methoxy-phosphinoylmethyl]-pentanedioic Acid

Dimethyl Ester, 63. To a solution of 15 (93 mg, 0.29 mmol)

in anhydrous CH₂Cl₂(2 mL) were added 2-methyleneglutarate

(76 mg, 0.44 mmol) and BSA (220 µL, 0.87 mmol) at room

temperature. The reaction was stirred under Ar for 20 h.

Reaction progress was monitored by ³¹P NMR. The reaction

was quenched with 1 M HCl (1 mL). CH₂Cl₂(8 mL) was added

and the reaction was washed with 1 M HCl (2 × 5 mL) and

H₂O (5 mL). The organic layer was dried over Na₂SO₄, filtered,

and concentrated. The crude product was dissolved in MeOH

(2 mL) and TMSCHN₂(2.0 M in hexanes) was added until a

persistent yellow color was observed. The solution was stirred

for 30 min at room temperature, then quenched with AcOH,

and the solution concentrated. The crude material was purified

by silica gel chromatography (EtOAc/ 2% MeOH) to give 100

mg of 68 as clear oil, an inseparable mixture of two diaster-

eomers (70%, 1:1 by HPLC): ¹H NMR (CDCl₃) δ 7.23 (m, 5H),

5.80 (m, 1H), 5.08 (s, 2H), 4.37 (m, 1H), 3.7-3.5 (overlapping,

12 H), 2.79 (m, 1H), 2.35-2.05 (overlapping m, 4H), 2.00-

1.60 (overlapping m, 3H). ¹³C NMR (CDCl₃) δ 174.58, 174.55,

172.94, 172.88, 172.85, 171.0, 156.1, 136.2, 128.5, 128.2, 128.1,

67.0, 54.1, 54.0, 53.9, 52.6, 52.2, 51.7, 51.35, 51.31, 51.21, 51.16,

51.11, 38.6, 38.4, 31.16, 31.10, 30.2, 30.0, 29.7, 29.5, 29.2, 29.0,

28.8, 28.7, 24.8, 24.5, 24.2, 24.1, 23.7. ³¹P NMR (CDCl₃) δ 55.2

and 54.6. MS (ESI) m/z 502.2 ([M + H]⁺, 100). HRMS (ESI)

calcd for C₂₂H₃₃NO₁₀P 502.1842 [M + H]⁺, found 502.1839.

(3′S)-2-[(3′-Benzyloxycarbonylamino-3′-methoxy-

carbonyl-propyl)-methoxy-phosphinoylmethyl]-pentane-

1,5-dioic Acid 5-tert-Butyl Ester 1-Methyl Ester, 67. To a

solution of 65 (0.26 g, 0.34 mmol) in 3:1 THF/H₂O (4 mL) at 0

°C was added H₂O₂(0.16 mL, 1.5 mmol) followed by solid LiOH

(29 mg, 0.69 mmol). The reaction was stirred at 0 °C for 30

min. The reaction was quenched by the addition of Na₂SO₃

(0.28 g) in H₂O (1 mL). The reaction was extracted with Et₂O

(2 × 20 mL) to recover the chiral auxiliary. The aqueous layer

(3′S)-5-(4-Benzhydryl-2-oxo-oxazolidin-3-yl)-4-[(3′-

benzyloxycarbonylamino-3′-methoxy-carbonyl-propyl)-

methoxy-phosphinoylmethyl]-5-oxo-pentanoic Acid tert-

Butyl Ester, 64 and 65. To a solution of 15 (0.1 g, 0.32 mmol)

and 52 (0.15 g, 0.35 mmol) in anhydrous CH₂Cl₂(4 mL) under

Ar was added BSA (0.24 mL, 0.96 mmol) and the reaction was

J. Org. Chem, Vol. 70, No. 17, 2005 6773

Bartley and Coward

was acidified to pH 1.0 with 2 M HCl and extracted with Et₂O

(3 × 30 mL). The organic layer was concentrated and the crude

material dissolved in MeOH (5 mL) and treated with CH₂N₂

until a persistent yellow color was observed. The reaction was

stirred at room temperature for 15 min, then quenched with

AcOH and concentrated to give 0.15 g of 67 as a clear oil

(83%): ¹H NMR (CDCl₃) δ 7.32 (m, 5H), 5.94 (m, 1H), 5.10 (s,

2H), 4.38 (m, 1H), 3.8-3.6 (overlapping m, 9H), 2.80 (m, 1H),

2.3-2.1 (overlapping m, 4H), 1.91 (m, 3H), 1.78 (m, 3H), 1.43

(s, 9H). ¹³C NMR (CDCl₃) δ 174.81, 172.03, 171.75, 171.72,

156.1, 136.3, 128.8, 128.6, 128.5, 128.24, 128.18, 128.05, 126.7,

80.7, 67.1, 54.19, 54.11, 53.99, 52.8, 52.6, 52.19, 52.17, 51.38,

51.34, 51.21, 51.15, 38.62, 38.59, 31.9, 30.3, 29.8, 29.6, 29.14,

29.07, 29.04, 28.1, 24.92, 24.6, 24.2, 23.8. ³¹P NMR (CDCl₃) δ

55.4 and 54.8. MS (ESI) m/z 509.1 ([M - tBu + Na]⁺, 100),

566.2 ([M + Na]⁺, 7). HRMS (ESI) calcd for C₂₅H₃₈NNaO₁₀P

566.2131 [M + Na]⁺, found 566.2127.

for 12 h. The reaction was extracted with Et₂O (2 mL) and

the aqueous layer was freeze-dried to give 30 mg of 69 as

hydroscopic white solid (97%): ¹H NMR (D₂O) δ 4.14 (m, 1H),

2.77 (m, 1H), 2.43 (m, 3H), 2.19 (m, 3H), 1.95 (m, 5H). ³¹P NMR

(D₂O) δ 50.7.

To a solution of AMPte-N₃(20 mg, 0.06 mmol) and 69 (10

mg, 0.03 mmol) in anhydrous DMSO (2 mL) was added

tetramethylguanidine (0.022 mL, 0.18 mmol). The reaction was

stirred at room temperature for 18 h. The desired product was

precipitated by the addition of CH₃CN (10 mL) and collected

by centrifugation (9000 rpm × 30 min). The pellet was dried

under a stream of N₂and then purified by ion exchange

chromatography (DEAE-cellulose, 0.01 to 1.0 M Et₃NH⁺HCO₃^-).

Fractions containing the desired product were pooled and

lyophilized to give 19 mg of 71 as the tri-Et₃N salt (79%).

Analytical HPLC: t_R) 9.8 min. UV-vis (HPLC diode array,

phosphate buffer, pH 7.0/acetonitrile) λ_max260, 305, 375 nm;

(0.1 N HCl) 306 nm; (0.1 N NaOH) 258, 301, 270 nm). ¹H NMR

(D₂O) δ 8.60 (s, 1H), 7.66 (d, 2H, J ) 8.7 Hz), 6.78 (d, 2H, J )

8.7 Hz), 4.29 (m, 1H), 3.17 (m, 18H), 2.48 (m, 1H), 2.2-2.0

(overlapping m, 3H), 1.95 (m, 2H), 1.82 (m, 1H), 1.74 (m, 1H),

1.60 (m, 3H), 1.24 (m, 18H). ³¹P NMR (D₂O) δ 43.0. MS (ESI)

m/z 619.2 ([M + H]⁺, 100). HRMS (ESI) calcd for C₂₅H₃₂N₈O₉P

619.2030 [M + H]⁺, found 619.2037.

(3′S)-2-[[[3′-Carboxy-3′-[[4”-[[(2′′′,

4′′′-diamino-6′′′-

pteridinyl)methyl]methylamino]benzoyl]amino]propyl]-

hydroxyphosphinyl]methyl]pentane-1,5-dioic Acid Bis-

(triethylamine) Salt, 70. A solution of 66 (45 mg, 0.083

mmol) in 6 M HCl (2 mL) was heated at reflux temperature

for 12 h. The reaction was extracted with Et₂O (2 mL) and

the aqueous layer was freeze-dried to give 27 mg of 68 as

hydroscopic white solid (93%): ¹H NMR (D₂O) δ 4.13 (m, 1H),

2.77 (m, 1H), 2.45 (m, 3H), 2.19 (m, 3H), 1.95 (m, 5H). ³¹P NMR

(D₂O) δ 50.3.

Acknowledgment. This research was supported by

a grant from the National Cancer Institute (CA28097).

D.M.B. is a trainee in the Pharmacological Sciences

Training Program supported by a grant from the

National Institutes of General Medical Sciences

(GM07767). We are most grateful to Dr. John McGuire

and Mr. Bill Haile, Roswell Park Cancer Institute,

Buffalo, NY, for their gift of FPGS and for testing of 70

and 71. We thank Dr. Jean-Luc Montchamp, Texan

Christian University, Fort Worth, TX, for helpful dis-

cussions on P-C bond formation; and Dr Shoujun Chen,

Synta Pharmaceuticals Corp., Lexington, MA, for help-

ful advice on organophosphorus chemistry. We are also

grateful to Amano Pharmaceutical Co. for the generous

gift of Lipase AK “Amano”.

To a solution of AMPte-N₃(20 mg, 0.06 mmol) and 68 (10

mg, 0.03 mmol) in anhydrous DMSO (1 mL) was added

tetramethylguanidine (0.022 mL, 0.18 mmol). The reaction was

stirred at room temperature for 18 h. The desired product was

precipitated by the addition of CH₃CN (10 mL) and collected

by centrifugation (9000 rpm × 30 min). The pellet was dried

under a stream of N₂and then purified by ion exchange

chromatography (DEAE-cellulose, 0.01 to 1.0 M Et₃NH⁺HCO₃^-).

Fractions containing the desired product were pooled and

lyophilized to give 22 mg of 70 as the di-Et₃N salt (83%):

analytical HPLC t_R) 11.7 min. UV-vis (HPLC diode array,

phosphate buffer, pH 7.0/acetonitrile) λ_max260, 305, 375 nm;

(0.1 N HCl) 306 nm; (0.1 N NaOH) 257, 301, 270 nm). ¹H NMR

(D₂O) δ 8.61 (s, 1H), 7.66 (d, 2H, J ) 8.6 Hz), 6.78 (d, 2H, J )

8.6 Hz), 4.31 (m, 1H), 3.17 (m, 12H), 2.54 (m, 1H), 2.3-2.0

(overlapping m, 3H), 2.0-1.7 (overlapping m, 4H), 1.60 (m,

3H), 1.24 (m, 18H). ³¹P NMR (D₂O) δ 42.6. MS (ESI) m/z 619.2

([M + H]⁺, 100). HRMS (ESI) calcd for C₂₅H₃₂N₈O₉P 619.2030

[M + H]⁺, found 619.2024.

Supporting Information Available: Experimental pro-

cedures and characterization of compounds ent-5 (X ) Ms, Ts,

I), 10 (Table 1), 14, 24 (Table 2), 54b/57b, 57b, and 54c/57c;

¹H and ¹³C spectra for all new compounds described in

Experimental Section. This material is available free of charge

via the Internet at http://pubs.acs.org.

(3′S)-2-[[[3′-Carboxy-3′-[[4”-[[(2′′′,

4′′′-diamino-6′′′-

pteridinyl)methyl]methylamino]benzoyl]amino]propyl]-

hydroxyphosphinyl]methyl]pentane-1,5-dioic Acid Tris-

(triethylamine) Salt, 71. A solution of 67 (48 mg, 0.088

mmol) in 6 M HCl (2 mL) was heated at reflux temperature

JO0507439

6774 J. Org. Chem., Vol. 70, No. 17, 2005

Article Doi

A stereoselective synthesis of phosphinic acid phosphapeptides corresponding to glutamyl-γ-glutamate and incorporation into potent inhibitors of folylpoly-γ-glutamyl synthetase

DOI: 10.1021/jo0507439

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jo0507439

Full text of DOI:10.1021/jo0507439

Products guided by the article

R&D Labs maybe for 864682-82-6

Relevant to this article

Hot Product