Synthesis of p-aminophenyl aryl H-phosphinic acids and esters via cross-coupling reactions: Elaboration to phosphinic acid pseudopeptide analogues of pteroyl glutamic acid and related antifolates

Article abstract of DOI:10.1021/jo0707840

(Chemical Equation Presented) The synthesis of suitably protected p-aminophenyl H-phosphinic acids and esters from the corresponding para-substituted aryl halides has been accomplished via the Pd-catalyzed cross-coupling reaction of anilinium hypophosphite, either in the absence or presence of a tetraalkyl orthosilicate, to provide the free H-phosphinic acid or the corresponding ester, respectively. Subsequent conjugate addition of either a P^III species or phosphorus anion, generated in situ from either the free H-phosphinic acid or ester, to a 2-methylene glutaric acid ester provided the aryl phosphinic acid analogue of p-aminobenzoyl glutamic acid. Alkylation of these suitably protected p-aminophenyl phosphinic acid esters with a 6-(bromomethyl)-pteridine or the corresponding (bromomethyl)pyridopyrmidine, followed by hydrolytic removal of protecting groups, provided the target aryl phosphinic acid analogues of folic acid and related antifolates. Alternatively, for the synthesis of the folate or 5-deazafolate analogues on a slightly larger scale, reductive amination with either N²-acetyl or N ²-pivaloyl-6-formylpterin or the corresponding formylpyridopyrmidine and the same suitably protected p-aminophenyl phosphinic acid esters, followed by removal of protecting groups, is preferred. In the course of this research, it was observed that the nucleophilicity of both the aniline nitrogen and various P^III species derived from p-aminophenyl phosphinic acid derivatives is significantly reduced compared to that of the unsubstituted counterpart.

Full text of DOI:10.1021/jo0707840

Synthesis of p-Aminophenyl Aryl H-Phosphinic Acids and Esters via
Cross-Coupling Reactions: Elaboration to Phosphinic Acid
Pseudopeptide Analogues of Pteroyl Glutamic Acid and Related
Antifolates
Yonghong Yang and James K. Coward*
Departments of Medicinal Chemistry and Chemistry, UniVersity of Michigan,
Ann Arbor, Michigan 48109-1055
jkcoward@umich.edu
ReceiVed April 15, 2007
The synthesis of suitably protected p-aminophenyl H-phosphinic acids and esters from the corresponding
para-substituted aryl halides has been accomplished via the Pd-catalyzed cross-coupling reaction of
anilinium hypophosphite, either in the absence or presence of a tetraalkyl orthosilicate, to provide the
free H-phosphinic acid or the corresponding ester, respectively. Subsequent conjugate addition of either
a PIII species or phosphorus anion, generated in situ from either the free H-phosphinic acid or ester, to
a 2-methylene glutaric acid ester provided the aryl phosphinic acid analogue of p-aminobenzoyl glutamic
acid. Alkylation of these suitably protected p-aminophenyl phosphinic acid esters with a 6-(bromomethyl)-
pteridine or the corresponding (bromomethyl)pyridopyrmidine, followed by hydrolytic removal of
protecting groups, provided the target aryl phosphinic acid analogues of folic acid and related antifolates.
Alternatively, for the synthesis of the folate or 5-deazafolate analogues on a slightly larger scale, reductive
amination with either N2-acetyl or N2-pivaloyl-6-formylpterin or the corresponding formylpyridopyrmidine
and the same suitably protected p-aminophenyl phosphinic acid esters, followed by removal of protecting
groups, is preferred. In the course of this research, it was observed that the nucleophilicity of both the
aniline nitrogen and various PIII species derived from p-aminophenyl phosphinic acid derivatives is
significantly reduced compared to that of the unsubstituted counterpart.
Introduction
parasites, in the folic acid biosynthesis pathway.3-6 In mam-
malian cells, there is no pathway available to affect the de novo
synthesis of reduced folates. Therefore, the vitamin folic acid
and its reduced derivatives must be obtained from dietary intake.
Unlike mammalian cells, many microorganisms such as Plas-
modium falciparum, the important parasite causing serious
malaria, have the ability to synthesize de novo the partially
reduced folate, 7,8-dihydrofolic acid, which is then converted
In contrast to widespread use of antibiotics and other drugs
for the treatment of bacterial diseases, generally via chemo-
therapy, the development of chemotherapy for the treatment of
parasitic diseases1,2 has been limited, due to the close similarities
between parasite and host metabolism. However, it has been
known for many years that a significant difference exists
between human and microorganisms, including the malaria
(3) Hyde, J. E. Acta Trop. 2005, 94, 191-206.
(4) Bermingham, A.; Derrick, J. P. BioEssays 2002, 24, 637-648.
(5) Lee, C. S.; Salcedo, E.; Wang, Q.; Wang, P.; Sims, P. F. G.; Hyde,
J. E. Parasitology 2001, 122, 1-13.
(6) Nzila, A.; Ward, S. A.; Marsh, K.; Sims, P. F. G.; Hyde, J. E. Trends
Parasitol. 2005, 21, 292-298.
* Address correspondence to this author. Phone: 734-936-2843. Fax: 734-
647-4865.
(1) Despommier, D. D.; Gwadz, R. W.; Hotez, P. J. Parasitic Diseases,
3rd ed.; Springer-Verlag New York, Inc.: New York, 1995.
(2) Schirmer, R. H.; Muller, J. G.; Krauthsiegel, R. L. Angew. Chem.,
Int. Ed. 1995, 34, 141-154.
10.1021/jo0707840 CCC: $37.00 © 2007 American Chemical Society
Published on Web 06/28/2007
5748
J. Org. Chem. 2007, 72, 5748-5758

p-Aminophenyl H-Phosphinic Acids and Esters
FIGURE 1. Partial folate biosynthesis pathway in microorganisms as targets for drug discovery.
to a variety of reduced folylmono- and poly-γ-glutamates via
normal host and/or parasite metabolism. Intermediates that play
important roles in folate biosynthesis, including 7,8-dihy-
dropteroate (DHP), 7,8-dihydrofolate (DHF), 5,6,7,8-tetrahy-
drofolate (THF), and 5,10-methylene-5,6,7,8-tetrahydrofolate
(5,10-CH2-THF), have been shown to be present in the malaria
parasite.6 On the basis of this significant difference between
mammalian cells and the malaria parasite, inhibition of the folate
biosynthesis pathway in the parasite has been extensively
investigated with the goal of developing drugs which could kill
the parasite while leaving the human host unaffected.
The folate biosynthesis pathway has provided a remarkably
rich source of targets for developing new drugs.7-9 Folate
antagonism has been widely developed for cancer treatment and
the use of antifolates has occupied an important position for
over 50 years. Folate antagonism has also proved to be a good
strategy for development of antibacterial drugs, including those
that target the malaria parasite. For example, sulfadoxine (4-
amino-N-(5,6-dimethoxypyrimidin-4-yl)benzenesulfonamide), an
inhibitor of dihydropteroate synthetase (DHPS, EC 2.5.1.15),
and pyrimethamine (5-(4-chlorophenyl)-6-ethylpyrimidine-2,4-
diamine), an inhibitor of dihydrofolate reductase (DHFR, EC
1.5.1.3), can be used in combination for the successful treatment
of certain drug-resistant strains of the malaria parasite (Figure
1).6,10 Other DHFR and DHPS inhibitor combinations have also
been developed as effective antimalarials.6,7
Less extensively studied as a potential target for new
antimalarial drugs is 7,8-dihydrofolate synthetase (DHFS, EC
6.3.2.12), which catalyzes the ATP-dependent conversion of
DHP to DHF by addition of a single L-glutamate moiety.11 We
have recently reported the synthesis of potent inhibitors of
folylpoly-γ-glutamate synthetase (FPGS, EC 6.3.2.17) based on
pseudopeptides containing complex alkyl phosphinic acids.12-14
Herein we describe the design of heretofore unknown pseudopep-
tides containing aryl phosphinic acids as potential DHFS
inhibitors. Interestingly, in both E. coli15 and P. falciparum,16
DHFS activity is contained in a bifunctional protein that also
harbors folylpoly-γ-glutamate synthetase (FPGS, EC 6.3.2.17)
activity. For this reason, it has been considered unlikely that it
would be possible to specifically inhibit only the DHFS locus
of this bifunctional protein. However, as reported recently, an
unexpected dihydropteroate binding site, distinct from the folate
site previously identified in FPGS,17 is revealed in the crystal
structure of folC, the bifunctional enzyme from E. coli.15 This
discovery suggests the design of potential specific inhibitors of
the DHFS locus of the bifunctional enzyme, including that from
P. falciparum.
(11) Nzila, A.; Ward, S. A.; Marsh, K.; Sims, P. F. G.; Hyde, J. E. Trends
Parasitol. 2005, 21, 334-339.
(12) Valiaeva, N.; Bartley, D.; Konno, T.; Coward, J. K. J. Org. Chem.
2001, 66, 5146-5154.
(13) Bartley, D. M.; Coward, J. K. J. Org. Chem. 2005, 70, 6757-6774.
(14) McGuire, J. J.; Haile, W. H.; Valiaeva, N.; Bartley, D.; Guo, J. X.;
Coward, J. K. Biochem. Pharmacol. 2003, 65, 315-318.
(15) Mathieu, M.; Debousker, G.; Vincent, S.; Viviani, F.; Bamas-
Jacques, N.; Mikol, V. J. Biol. Chem. 2005, 280, 18916-18922.
(16) Salcedo, E.; Cortese, J. F.; Plowe, C. V.; Sims, P. F. G.; Hyde, J.
E. Mol. Biochem. Parasitol. 2001, 112, 239-252.
(7) Biagini, G. A.; O’Neill, P. M.; Nzila, A.; Ward, S. A.; Bray, P. G.
Trends Parasitol. 2003, 19, 479-487.
(8) Kompis, I. M.; Islam, K.; Then, R. L. Chem. ReV. 2005, 105, 593-
620.
(9) McGuire, J. J. Curr. Pharm. Des. 2003, 9, 2593-2613.
(10) What is Malaria?; World Health Organization: 2006; http://
www.rollbackmalaria.org/malariaFAQ.html.
(17) Sun, X. L.; Cross, J. A.; Bognar, A. L.; Baker, E. N.; Smith, C. A.
J. Mol. Biol. 2001, 310, 1067-1078.
J. Org. Chem, Vol. 72, No. 15, 2007 5749

Yang and Coward
SCHEME 1
The proposed reaction mechanism of the DHFS-catalyzed
reaction is depicted in Scheme 1. In this mechanism, the
“accepting substrate”, DHP, is converted to an acyl phosphate
intermediate via an ATP-dependent process.18 Subsequent
reaction leads to a tetrahedral intermediate resulting from attack
of the activated carbonyl by the amine moiety of the “incoming
substrate”, L-glutamate. The tetrahedral intermediate then col-
lapses to form a new amide bond concomitant with the expulsion
of Pi. On the basis of this mechanism, the synthesis of stable
aryl phosphinic acid mimics, 1, of the proposed tetrahedral
intermediate as a new class of DHFS inhibitors has been
investigated.
formed during catalysis (Scheme 1). In the course of this
investigation, development of methods for the efficient synthesis
of phosphinic acid esters 7, which carry both aryl C-P and
alkyl C-P bonds, was pursued. The para disposition of the pho-
sphorus substituent and the amino group results in attenuation
of the reactivity of both nucleophiles: i.e., the aryl phosphinic
acid (or PIII species derived therefrom) and the aniline nitrogen
of 7 or its precursors. The consequences of this altered reactivity
as it affects the synthesis of the desired target compounds and
key intermediates have been investigated in detail.
Results and Discussion
A series of aryl phosphinic acids, 1a-e, has been designed
to incorporate both the “tetrahedral mimic” and a heterocyclic
portion analogous to the pterin moiety of the natural substrate,
7,8-dihydrofolate. The stable aryl phosphinic mimic could
The proposed DHFS inhibitor 1 is composed of three frag-
ments including the heterocycle, a p-aminophenylphosphinic
acid portion similar to p-aminobenzoic acid (pABA), and a
glutarate portion similar to glutamate (Scheme 1). Under phy-
siological conditions, the phosphinic acid would be deprotonated
and, as such, mimic the tetrahedral intermediate postulated to
form during DHFS-catalyzed synthesis of DHF. Retrosynthetic
analysis (Scheme 2) suggests two approaches to coupling the
p-aminophenylphosphinic acid ester moieties, 7, to the hetero-
cycle. Thus, an SN2 alkylation of 7 by 825-27 or a reductive
amination of 9 by 728-30 is proposed for the C-N bond
formation. The key intermediate 7 is an unusual phosphinic acid
ester that includes both an aryl C-P bond and an alkyl P-C
bond. As shown in Scheme 2, a retrosynthetic analysis indicates
possibly be phosphorylated in the presence of ATP by a DHFS-
catalyzed reaction, leading to a phosphorylated phosphinate,19-24
thus leading to an even closer mimic of the transient intermediate
(24) Pai, C. H.; Chiang, B. Y.; Ko, T. P.; Chou, C. C.; Chong, C. M.;
Yen, F. J.; Chen, S. J.; Coward, J. K.; Wang, A. H. J.; Lin, C. H. EMBO
J. 2006, 25, 5970-5982.
(25) Piper, J. R.; Montgomery, J. A. J. Org. Chem. 1977, 42, 208-211.
(26) Piper, J. R.; McCaleb, G. S.; Montgomery, J. A. J. Heterocycl. Chem.
1987, 24, 279-282.
(18) Banerjee, R. V.; Shane, B.; McGuire, J. J.; Coward, J. K.
Biochemistry 1988, 27, 9062-9070.
(19) Abell, L. M.; Villafranca, J. J. Biochemistry 1991, 30, 6135-6141.
(20) Fan, C.; Moews, P. C.; Walsh, C. T.; Knox, J. R. Science 1994,
266, 439-443.
(27) Feng, Y.; Coward, J. K. J. Med. Chem. 2006, 49, 770-788.
(28) Maunder, P.; Finglas, P. M.; Mallet, A. I.; Mellon, F. A.; Razzaque,
M. A.; Ridge, B.; Vahteristo, L.; Witthoft, C. J. Chem. Soc., Perkin Trans.
1 1999, 1311-1323.
(29) Taylor, E. C.; Palmer, D. C.; George, T. J.; Fletcher, S. R.; Tseng,
C. P.; Harrington, P. J.; Beardsley, G. P. J. Org. Chem. 1983, 48, 4852-
4860.
(21) Fan, C.; Park, I. S.; Walsh, C. T.; Knox, J. R. Biochemistry 1997,
36, 2531-2538.
(22) Chen, S. J.; Lin, C. H.; Walsh, C. T.; Coward, J. K. Bioorg. Med.
Chem. Lett. 1997, 7, 505-510.
(23) Lin, C. H.; Chen, S. J.; Kwon, D. S.; Coward, J. K.; Walsh, C. T.
Chem. Biol. 1997, 4, 859-866.
(30) Taylor, E. C.; Hamby, J. M.; Shih, C.; Grindey, G. B.; Rinzel, S.
M.; Beardsley, G. P.; Moran, R. G. J. Med. Chem. 1989, 32, 1517-1522.
5750 J. Org. Chem., Vol. 72, No. 15, 2007

p-Aminophenyl H-Phosphinic Acids and Esters
SCHEME 2
that formation of an alkyl C-P bond leading to the key inter-
mediate 7 could be achieved by Michael addition between pho-
sphinic acid 4, either as the free acid 4a-c or the corresponding
phosphinic acid ester 4d,e, and 2-methyleneglutarate 5.13,27 The
aryl C-P bond could be established via a palladium-catalyzed
cross-coupling reaction between 2 and 3 to yield 4a-c,31,32 or
alternatively, the aryl C-P bond could be installed via a similar
palladium-catalyzed cross-coupling reaction between 2 and 3
in the presence of (EtO)4Si or (MeO)4Si to form the phosphinic
acid esters 4d or 4e in a single operation.33,34
between benzyl 4-bromophenyl(methyl)carbamate (2a) and
anilinium hypophosphite (3) in the presence of Pd(PPh3)4. 31P
NMR indicated that the crude H-phosphinic acid 4a was
afforded in good yield with sufficient purity for use in the next
reaction. Michael addition of the corresponding ArP(OTMS)2
PIII species, formed in situ by reaction of 4a with N,O-
(bistrimethylsilyl)acetamide (BSA), to dimethyl 2-methylene-
glutarate (5a) followed by methylation of the resulting phos-
phinic acid (TMSCHN2) and purification by silica gel chromato-
graphy led to 6a in 45% yield over 3 steps (Table 1, Method
A). Although 4b and 4c could be obtained in high yield (31P
NMR) by using the same Pd-catalyzed cross-coupling reaction,
subsequent Michael addition to 5a to provide 6b or 6c was
unsuccessful in the case of 4b and proceeded in only very low
yield (6%) in the case of 4c. It should be noted that generation
of the requisite PIII intermediate could be observed in the BSA-
mediated reaction of 4a (31P NMR, δ 140 ppm) but not in the
identical reaction of 4b.
In pursuit of this synthetic strategy, H-phosphinic acid 4a
was obtained via a palladium-catalyzed coupling reaction
(31) Montchamp, J.-L.; Dumond, Y. R. J. Am. Chem. Soc. 2001, 123,
510-511.
(32) Schwan, A. L. Chem. Soc. ReV. 2004, 33, 218-224.
(33) Deprele, S.; Montchamp, J. L. J. Organomet. Chem. 2002, 643,
154-163.
(34) (a) Huang, Z.; Bravo-Altamirano, K.; Montchamp, J.-L. C. R. Chim.
2004, 7, 763-768. (b) Bravo-Altamirano, K.; Huang, Z.; Montchamp, J.-
L. Tetrahedron 2005, 61, 6315-6329.
J. Org. Chem, Vol. 72, No. 15, 2007 5751

Yang and Coward
TABLE 1. Conversion of Substituted Aryl H-Phosphinic Acids or Esters (4) to the Corresponding Aryl-Alkyl Phosphinic Acid Esters (6)
a Overall yield from 4 (two steps). b Reaction of PIII species with 5b in DCM (entry 2) or THF (entry 9). c Overall yield from 2 (three steps). d Also
observed no reaction with methyl acrylate. e Poor reaction with methyl acrylate, yield 7%. f Reaction of the ethyl esters of 4d and 4e (R4 ) Et) under
identical conditions led to 6b and 6c (-P(O)(OEt)-) in 53% and 47% yield, respectively. g Yields are for single-step reaction from 4. h Microwave irradialtion,
250 W. i R4 ) Me, NaOMe was used as the base.
An alternative approach to 6b or 6c involved conjugate
addition of the H-phosphinic acid alkyl ester, 4d or 4e, to
2-methyleneglutarate 5a. A similar palladium-catalyzed cross-
coupling of 2b or 2c and 3, but in the presence of (EtO)4Si or
(MeO)4Si, afforded ester 4d or 4e, respectively, in a single step
with moderate yield after purification by column chromatog-
raphy (eq 1). Initially, the use of elevated temperature (80 °C)
led to a low yield (26%) of 4d and a large amount of a diaryl
phosphinic acid ester 11a (31P NMR, δ 32-34 ppm). Ultimately,
a decrease in temperature (50 °C) and high dilution (0.05-
0.11 M of 2b or 2c) led to an improved yield of the desired
monoaryl phosphinic acid esters, 4d (64%) and 4e (55%) (31P
NMR, δ 25-28 ppm). Unfortunately, generation of 11 was
difficult to exclude completely under these conditions. However,
when Pd(OAc)2 and 1,3-bis(diphenylphosphino)propane (dppp)
were utilized, compounds 4d and 4e were furnished exclusively
in good yield. The bidentate phosphine ligand, dppp, could
promote and facilitate concerted reductive elimination by
occupying only two coordination sites cis to each other, forcing
the carbon group and phosphorus group to be also cis to each
other.35 In addition, the use of Pd(PPh3)4 as catalyst results in
the formation of a small amount of Ph3PdO, which is difficult
(35) Diederich, F.; Stang, P. J. In Metal-catalyzed Cross-coupling
Reactions; Wiley-VCH: Weinheim , Germany, 1998; pp 26-27.
5752 J. Org. Chem., Vol. 72, No. 15, 2007

p-Aminophenyl H-Phosphinic Acids and Esters
SCHEME 3
SCHEME 4
to remove completely during purification of 4d or 4e. The results
obtained for the Pd-catalyzed cross-coupling reactions in the
synthesis of aryl H-phosphinic acids or esters 4a-e are
summarized in Scheme 3.
a ArP(OMe)(OTMS) PIII intermediate are less reactive and
require longer reaction times at elevated temperatures to provide
acceptable yields of 6 than is the case with the corresponding
free H-phosphinic acid 4a involving a ArP(OTMS)2 PIII
intermediate. However, of the latter group of H-phosphinic acids,
only the N-(Me)Cbz H-phosphinic acid 4a led to acceptable
yields. N(Cbz)2 H-phosphinic acid resulted in very low yield,
which was detected by LC-MS, and N(Boc)2 H-phosphinic acid
gave 22% yield.
A third alternative to the synthesis of aryl-alkyl phosphinic
acids 6 involved the Michaelis-Becker (Pudovik) reaction,
which was examined with either NaOMe41 or KOt-Bu as the
base (Scheme 4; Table 1, Method C). Use of NaOMe as the
base in the reaction of 4e gave the expected product 6c (57%).
However, the identical reaction of 4d yielded the desired product
6b (25%) in addition to a methyl carbamate byproduct 6e (33%).
A similar methanolysis was reported recently by Shieh et al.,42
in which a heteroaromatic Cbz carbamate was converted to the
corresponding methyl carbamate. An N-Cbz protected aliphatic
amine elsewhere on the molecule was unaffected. In the current
research, when KOt-Bu was used as a less nucleophilic, bulky
base, 6b and 6d were obtained in modest yields. Acid hydrolysis
of 6c and 6e yielded 12, the free aryl phosphinic acid analogue
of p-aminobenzoylglutamic acid.
In summary, three methods to effect the synthesis of protected
aryl phosphinic acid esters, 6, from p-aminophenyl H-phosphinic
acids 4a-c, or esters 4d,e were investigated. These involved
reaction of the free acids or the corresponding esters via BSA-
mediated formation of nucleophilic PIII intermediates (Methods
A and B, respectively) or by a base-catalyzed Michaelis-Becker
reaction of the H-phosphinic acid esters (Method C). The results
of representative experiments are given in Table 1. These
methods provide access to the key intermediates 6, which were
then further elaborated to 1, the desired aryl phosphinic acid
analogues of folic acid.
Compounds 7a-c, destined for SN2 alkylation by 6-(bromo-
methyl)pteridines 8a,b and reductive amination with 6-formylp-
teridines 9a-c, were obtained from 6a,b,d by hydrogenolysis
of the Cbz protecting group. Optimization of the SN2 alkylation
reaction conditions was investigated by varying the ratio of
electrophile to nucleophile, elevating the temperature, prolonging
the reaction time, or using microwave irradiation (Table 2). It
There is considerable literature precedent, including from this
laboratory,13,27 for the synthesis of phosphinic acids via addition
of PIII species, generated in situ from alkyl H-phosphinic acids,
to electrophilic olefins. However, to our knowledge, very few
examples involved a PIII species derived from an aryl H-
phosphinic acid36-38 or ester. None involved a PIII species
derived from a substituted aryl H-phosphinic acid or ester. Thus,
two routes involving a Michael reaction between 4d,e and 5a,b
were pursued (Table 1, Method B). First, BSA-mediated
formation of mono-TMS esters (31P NMR, δ 146 ppm) and
subsequent addition of the PIII species generated in situ was
investigated. Under varying reaction conditions, long reaction
times were required to obtain reasonable yields of the desired
product. Although the use of microwave conditions to increase
reaction rates39 led to a marked reduction in reaction time, it
was not effective in leading to enhanced yields (Supporting
Information, Table S1).
This research indicates that the nucleophilicity of the phos-
phorus PIII species derived from 4 is attenuated, apparently due
to the p-acyloxyamino (carbamate) substituent. Thus, BSA-
mediated Michael addition of N-Cbz- (4d) or N-Boc- (4e)
p-amino phenyl H-phosphinic methyl esters, respectively, to
2-methyleneglutarate esters 5a or 5b required elevated temper-
atures and extended reaction times to obtain moderate to good
yields of the desired phosphinic acids, 6 (Table 1, entries 6, 8,
9; see also Table S1, Supporting Information). In contrast,
reaction of a simple, unsubstituted phenylphosphinic acid at
room temperature for 24 h proceeded in good yield (Table 1,
entries 1 and 2). It should be recalled that the non-esterified
N-(Me)Cbz H-phosphinic acid, 4a, reacted at room temperature
to provide the N-methyl derivative, 6a, in 45% yield (vide
supra). A major benefit of the TMS group in the conjugate
addition reaction involves stabilization of developing positive
charge on phosphorus by silicon â to phosphorus.40 Therefore,
it is not surprising that the P-OMe esters 4d and 4e involving
(36) Enders, D.; Saint-Dizier, A.; Lannou, M.-I.; Lenzen, A. Eur. J. Org.
Chem. 2006, 29-49.
(37) Boyd, E. A.; Boyd, M. E. K.; Loh, V. M. Tetrahedron Lett. 1996,
37, 1651-1654.
(38) Wehbe, J.; Rolland, V.; Fruchier, A.; Roumestant, M. L.; Martinez,
J. Tetrahedron: Asymmetry 2004, 15, 851-858.
(41) Schoen, W. R.; Parsons, W. H. Tetrahedron Lett. 1988, 29, 5201-
5204.
(42) Shieh, W. C.; Xue, S.; McKenna, J.; Prasad, K.; Repic, O.;
Blacklock, T. Tetrahedron Lett. 2006, 47, 5645-5648.
(39) Kappe, C. O. Angew. Chem., Int. Ed. 2004, 43, 6250-6284.
(40) Lambert, J. B.; Zhao, Y.; Emblidge, R. W.; Salvador, L. A.; Liu,
X. Y.; So, J. H.; Chelius, E. C. Acc. Chem. Res. 1999, 32, 183-190.
J. Org. Chem, Vol. 72, No. 15, 2007 5753

Yang and Coward
TABLE 2. SN2 Alkylation of p-Aminophenyl Phosphinic Acid Esters: Optimization Studies
a Yield based on HPLC analysis, product could be purified by semi-prep HPLC or column chromatography (entries 1 and 4). b Used di-tert-butyl ester.
c Microwave irradiation. d Byproducts detected by analytical HPLC. e Subsequent heating at 55 °C for ca. 20 h led to additional products (HPLC analysis)
and a decreased yield (40%) of the desired product.
is well-known that the pKa of the ammonium conjugate acid of
p-aminobenzoic acid is unusually low (pKa ) 2.30).43 Similarly,
the nucleophilicity of the amine moiety of 7, the phosphinic
acid ester analogue of dimethyl p-aminobenzoylglutamate, is
apparently decreased significantly in comparison to a diester
of p-aminobenzoylglutamic acid (Table 2, entry 1). The attenu-
ated nucleophilicity of 7b is further supported by the slow rate
of its alkylation by benzyl bromide (Table 2, entry 4). Elevation
of the reaction temperature by either conventional heating or
microwave irradiation resulted in the formation of several
unidentified byproducts (Table 2, entries 2 and 3). The formation
of these byproducts was not observed in the previously reported
synthesis of prodrug esters of pseudopeptide FPGS inhibitors
under similar conditions (60 °C, 30 h).27 These observations
provide further evidence for attenuation of the nucleophilicity
of both the phosphorus PIII species derived from 4 and the aryl
amine in 7, due to the para disposition of the phosphorus
substituent and the amino group. Due to the decreased solubility
of 8a, alkylation of 7a or 7b by 8a is slower than the alkylation
of 7a or 7b by 8b (Table 2, entries 5 and 6). In all cases, the
reaction rates were significantly improved by using 1.5 equiv
of the electrophile on a larger scale (ca. 50 mg), presumably
due to the increased concentration of all reactants. By using
these optimized conditions, reaction of 7a and 7b with 8a or
8b afforded 10a,c,e,f, fully protected aryl phosphinic acid
analogues of folic acid (Scheme 5). In addition to the desired
phosphinic acid methyl esters (P-OMe), 10a,c,e,f, the free
phosphinic acids (P-OH), 10a′,c′,e′,f′, were also isolated from
SCHEME 5
reaction mixture. Demethylation of these methyl esters is
attributed to the presence of bromide ion formed in the alkylation
reaction, as previously noted in our syntheses of pseudopeptide
(43) Jencks, W. P. In Handbook of biochemistry: selected data for
molecular biology; Sober, H. A., Ed.; Chemical Rubber Co: Cleveland,
OH, 1968; pp J144-J149.
5754 J. Org. Chem., Vol. 72, No. 15, 2007

p-Aminophenyl H-Phosphinic Acids and Esters
alkylation is the use of N2-pivaloylated or acetylated heterocycles
in the former reaction, which allows for purification of the
product by column chromatography. Reductive amination via
a two-step pathway, involving imine formation followed by
reduction of intermediate imine with BH3‚NHMe2 as reductant,
was investigated. As previously reported by Taylor’s group for
the synthesis of N2-acetyl 5-deazafolate, the imine intermediate
formed by reaction of 9c and p-AB-Glu(OMe)2 is insoluble in
acetic acid, and precipitates from reaction solution within 10
min, thus driving the reaction to completion in 6 h.29 However,
under the same reaction conditions, no precipitate was observed
in the reaction of 9c and 7b, and additional 9c (1.1 equiv) was
required to drive the reaction to completion after a total reaction
time of 27 h as determined by HPLC and MS. In contrast, a
rapid reductive amination was observed in the reaction of 9a
with 7b (1.1 equiv) or 7c (1.1 equiv), in agreement with reports
in the literature for the preparation of folic acid.28 Imine
formation was complete within 15 min, and subsequent reduc-
tion required only an additional 10 min. Extended exposure to
BH3‚NHMe2 led to a tetrahydropteroyl derivative as detected
by LC-MS.
SCHEME 6
Alternatively, reductive amination of 9a or 9b can be effected
via a one-pot procedure, which involves imine formation and
in situ reduction of intermediate imine by using NaBH3CN as
reductant. In the case of reductive amination of pterin 9a by
either 7b or 7c, reduction of the pyrazine ring led to the 5,6,7,8-
tetrahydro derivative of 10b1 (77%) or 10b2 (57%). However,
reductive amination of the 5-deaza compound 9b (1.5 equiv)
by 7b mediated by NaBH3CN provided compound 10d1 in
modest yield. The advantage of the reductive amination method
(Scheme 6) over SN2 alkylation (Scheme 5) is that the fully
protected products 10b1,2 and 10d1,2 can be purified by column
chromatography and the reductive amination reaction is less
affected by the attenuation of amine nucleophilicity in 7. In
addition, this method is more amenable to scale up due to the
rapid reaction and ease of product purification.
a Method A: Two-step reaction, imine formation, reduction by
BH3‚NHMe2. bMethod B: One-step reaction, imine formation, in situ
reduction by NaBH3CN. cHydrolyzed from 10b1. dHydrolyzed from 10b2.
f
eHydrolyzed from 10d1. Hydrolyzed from 10d2.
prodrug esters.27 In the alkylation of 7b by 8b, a small amount
(8%) of an N-10-dialklylated compound 13 was also isolated.
Conditions for hydrolysis of the fully protected precursors,
10, obtained by reductive amination of the N2-protected
heterocyclic aldehydes, 9, to the target aryl phosphinic acids,
1, varied depending on the nature of the protecting groups.
Precursor 10b1 was treated with aqueous base (1 N NaOH, room
temperature, 5-7 days) leading to free acid 1a in good yield.
The closely related precursor 10b2 was first treated with aqueous
base (1 N NaOH, room temperature), which resulted in complete
removal of the pivaloyl protecting group, accompanied by partial
removal of the tert-butyl ester as indicated by LC-MS analysis.
Similarly, 10d1 or 10d2 was treated with aqueous base (1 N
NaOH, room temperature), resulting in complete removal of
the pivaloyl or acetyl protecting group as well as all carboxylic
acid methyl esters. The basic reaction solutions (10b2, 10d1,
or 10d2) were then neutralized (HCl) and the solutions
lyophilized. Treatment of the resulting residue with TFA in CH2-
Cl2 provided the free acids, 1a or 1c, in good yield after
crystallization. The more facile cleavage of methyl â-carboxy-
phophinates under both alkaline and acidic conditions is
attributed to intramolecular catalysis involving the formation
of a five-membered cyclic mixed anhydride intermediate.44
Under acidic conditions, the free â-carboxyphosphinic acid was
obtained rapidly. Under basic condition, a similar mechanistic
pathway proceeded at a much slower rate. On the basis of these
The target compounds 1a,b,d,e were obtained after hydrolysis
of the carboxylic acid esters under basic conditions. Hydrolysis
of the phosphinic acid methyl esters (P-OMe) of 10a,c,e,f
proceeded very slowly and was complete only after several days
at room temperature. In contrast, the corresponding phosphinic
acid 10f′ was hydrolyzed overnight in excellent yield. It should
be noted that the N-10 methyl derivatives 10c and 10e were
hydrolyzed more rapidly than the N-H derivatives 10a and 10f.
Although the SN2 alkylation was successful in forming the C-N
bond between heterocycles and phosphinic acid esters, several
experimental limitations were also discovered in the course of
this investigation, such as low solubility and high polarity of
the mixture of free phosphinic acids and the corresponding
methyl esters. Purification via crystallization or column chro-
matography was not generally applicable to these compounds.
As a result, purification of the desired target compounds, 1, or
its precursors 10a/a′,c/c′,e/e′,f/f′ required the use of preparative
HPLC, with associated loss of product during purification.
As an alternative approach, reductive amination was pursued
(Scheme 6). The advantage of reductive amination over SN2
(44) Georgiadis, D.; Dive, V.; Yiotakis, A. J. Org. Chem. 2001, 66,
6604-6610.
J. Org. Chem, Vol. 72, No. 15, 2007 5755

Yang and Coward
mmol), tetramethyl orthosilicate (0.66 mL, 4.43 mmol), 1,3-bis-
(diphenylphosphino)propane (dppp) (60 mg, 0.145 mmol), and Pd-
(OAc)2 (30 mg, 0.134 mmol) was added CH3CN (20 mL) via a
cannula under an Ar atmosphere. The reaction mixture was stirred
under reflux (85 °C). After an additional 2 h, the reaction mixture
was allowed to cool to room temperature. The reaction was
monitored with 31P NMR, and the integration ratio of the starting
material 3 and product 4d indicated when the reaction was complete.
A solution of KHSO4 (1 M) saturated by NaCl (20 mL) was added
to the reaction mixture, followed by ethyl acetate (50 mL). The
organic layer was separated. The aqueous layer was back-extracted
with ethyl acetate (3 × 50 mL). The combined organic layers were
washed with brine, dried over Na2SO4, filtered, and concentrated
in vacuo. The crude product was purified by silica gel column
chromatography, using CH2Cl2:ethyl acetate ) 1:1 as the eluant,
to afford 4d (320 mg, 1.05 mmol, 71%) as a white solid: mp 108-
110 °C; 1H NMR (CDCl3) δ 7.74-7.65 (2 H, m), 7.64-7.52 (2 H,
m), 7.51 (1 H, d, JHP ) 567 Hz), 7.47-7.30 (6 H, m), 5.21 (2 H,
s), 3.75 (3 H, d, JHP ) 12.0 Hz); 13C NMR (CDCl3) δ 152.9, 142.7
(d, 4JCP ) 2.9 Hz), 135.6, 132.3 (d, 2JCP ) 12.8 Hz), 128.6, 128.5,
128.4, 123.0 (d, 1JCP ) 136.7 Hz), 118.1 (d, 3JCP ) 14.3 Hz), 67.3,
52.0 (d, 2JPOC ) 6.6 Hz); 31P NMR (CDCl3) δ 28.0; MS (ESI) m/z
328.1 ([M + Na]+, 100); ESI-HRMS (m/z) calcd for C15H16NO4-
PNa [M + Na]+ 328.0715, found 328.0711.
2-[((4-(N-Benzyloxycarbonyl-N-methylamino)phenyl)(methoxy)-
phosphinoyl)methyl]pentane-1,5-dioic Acid Dimethyl Ester (6a,
Table 1, entry 3). BSA-Mediated Michael Addition Reaction
of ArP(OTMS)2 to R,â-Unsaturated Esters and Subsequent
Methylation of the Resulting Phosphinic Acid: General Pro-
cedure (Table 1, Method A). To an oven-dried round-bottomed
flask containing 4a (83 mg, 0.272 mmol) and dimethyl 2-methyl-
eneglutarate 5a (94 mg, 0.544 mmol) was added anhydrous CH2-
Cl2 (1 mL), followed by N,O-(bistrimethylsilyl)acetamide (BSA)
(0.34 mL, 1.36 mmol) under an Ar atmosphere. The reaction
mixture was stirred at room temperature for 24 h. The starting
material was consumed as indicated by 31P NMR. The reaction was
quenched by the addition of a 1 N solution of HCl (5 mL) and
stirring was continued for an additional 30 min. The layers were
separated by adding ethyl acetate (30 mL) and water (30 mL). The
aqueous layer was back-extracted with ethyl acetate (3 × 20 mL).
The combined organic layers were dried over Na2SO4, filtered, and
concentrated in vacuo to afford the crude product as a precursor of
6a as a light brown oil (31P NMR δ 44.3), which was used for next
step reaction without further purification. To a flask containing of
a solution of crude product (130 mg) in MeOH (8 mL) was added
a large excess of a solution of TMS-diazomethane (2 M) in diethyl
ether until a persistent yellow color was observed at 0 °C. In
addition, gas evolution was observed during the addition. After an
additional 2 h at room temperature, TLC indicated the starting
material was consumed. The reaction mixture was quenched by
adding acetic acid. The solvent and the excess of acetic acid were
removed in vacuo. The crude product was purified by silica gel
chromatography, using ethyl acetate as the eluant, to afford 6a (60
mg, 0.122 mmol, 45% (3 steps)) as a colorless oil: 1H NMR
(CDCl3) δ 7.76-7.71 (2 H, m), 7.44-7.29 (7 H, m), 5.20 (2 H, s),
3.69-3.38 (12 H, m), 2.87-2.80 (1 H, m), 2.50-2.29 (3 H, m),
2.09-1.91 (3 H, m); 13C NMR (CDCl3) δ 174.2 (dd, 3JCP ) 22.9,
7.0 Hz), 172.7 (d, 5JCP ) 5.4 Hz), 154.9, 147.0 (t, 4JCP ) 3.6 Hz),
136.0, 132.4 (dd, 2JCP ) 16.3, 10.8 Hz), 128.5, 128.1, 127.9, 126.3
(dd, 1JCP ) 127.0, 42.8 Hz), 124.8 (dd, 3JCP ) 12.4, 6.7 Hz), 67.7,
51.8 (d, 2JPOC ) 17.4 Hz), 51.6 (d, 7JPOC ) 1.3 Hz), 51.1 (dd, 5JPOC
) 6.4, 2.5 Hz), 38.4, 37.1, 31.7 (dd, 1JCP ) 101.6, 40.8 Hz), 31.1,
28.5 (d, 2JCP ) 11.9 Hz); 31P NMR (CDCl3) δ 41.7, 41.4; MS (ESI)
m/z 514.1 ([M + Na]+, 100).
observations, an improved procedure involved hydrolysis of the
carboxylic acid ester under basic condition (1 N NaOH:MeOH
) 2:1) followed by lyophilization of the neutralized reaction
mixture, and treatment of the resulting residue with TFA in CH2-
Cl2 (TFA:CH2Cl2 ) 1:5). This provides the free phosphinic acid
in high purity following a shorter reaction time.
In summary, several p-aminophenyl H-phosphinic acids and
esters (4) have been synthesized from the corresponding
substituted aryl halides via a Pd-catalyzed cross-coupling
reaction. Subsequent conversion of the aryl H-phosphinic acids
or esters to phosphinic acid ester analogues of p-aminobenzoyl
glutamate (6) was pursued by using three methods (Table 1).
Elaboration to the target compounds, 1, was investigated via
two synthetic routes: (1) alkylation by a bromomethyl hetero-
cycle (Scheme 5) or (2) reductive amination of a heterocyclic
aldehyde (Scheme 6). In the course of this research, it was
observed that the nucleophilicity of both substituents in p-
aminophenylphosphinic acid derivatives is markedly reduced.
Thus, the aniline nitrogen and various PIII species derived from
the aryl phosphinic acid each react at significantly lower rates
than the unsubstituted counterpart.
Preliminary biological evaluation of these compounds indi-
cates that the fully oxidized compounds, 1a-e, are at best only
modest inhibitors of glutamate ligation catalyzed by the DHFS-
FPGS bifunctional protein, folC, from E. coli (A. Bognar,
University of Toronto, personal communication). However, this
is not surprising since the natural substrate for this reaction is
a 7,8-dihydropterin derivative. Synthesis and biochemical evalu-
ation of the more labile 7,8-dihyrdo derivative of 1a is currently
in progress. Results of these studies will be reported in a separate
publication.
Experimental Section
General Procedures. See the Supporting Information. Repre-
sentative procedures are given in this Experimental Section. Details
on the synthesis of additional compounds using the procedures
described below are given in the Supporting Information.
(4-(N-Benzyloxycarbonyl-N-methylamino)phenyl)phosphin-
ic Acid (4a). Palladium-Catalyzed Cross-Coupling Reaction with
Pd(PPh3)4: General Procedure. To an oven-dried round-bottom
flask containing 2a (482 mg, 1.51 mmol), anilinium hypophosphite
(3) (642 mg, 4.03 mmol), and Pd(PPh3)4 (60 mg, 52 µmol) were
added Et3N (1 mL) and CH3CN (12 mL) via syringes under a N2
atmosphere. The reaction mixture was heated under reflux for 24
h. The reaction was monitored by 31P NMR, and the integration
ratio of the starting material of 3 and product 4a indicated when
the reaction was complete. The reaction mixture was then cooled
to room temperature. Water (5 mL) and a solution of KHSO4 (1
M) saturated with NaCl (5 mL) were added to the reaction mixture,
followed by addition of ethyl acetate (30 mL). The organic layer
was separated. The aqueous layer was back-extracted with ethyl
acetate (3 × 25 mL). The combined organic layers were dried over
Na2SO4, filtered, and concentrated in vacuo. The concentrated
solution was passed through a Celite filter to remove the remaining
catalyst. The solvent was removed in vacuo to afford 4a as a brown
oil with sufficient purity (450 mg, 98% (31P NMR yield)), which
was used without further purification: 1H NMR (CDCl3) δ 12.39
(1 H, br s), 7.76-7.64 (2 H, m), 7.59 (1 H, d, JHP ) 561 Hz),
7.57-7.35 (7 H, m), 5.20 (2 H, s), 3.34 (3 H, s); 31P NMR (CDCl3)
δ 19.3; MS (ESI) m/z 304.0 ([M - H]-, 100).
(4-(N-Benzyloxycarbonylamino)phenyl)phosphinic Methyl Es-
ter (4d). Palladium-Catalyzed Cross-Coupling Reaction with
Pd(OAc)2 and dppp: General Procedure. To an oven-dried
round-bottomed flask containing 2b (520 mg, 1.47 mmol), anilinium
hypophosphite (3) (703 mg, 4.42 mmol), DABCO (496 mg, 4.42
2-[((4-(N-Benzyloxycarbonylamino)phenyl)(methoxy)phos-
phinoyl)methyl]pentane-1,5-dioic Acid Dimethyl Ester (6b,
Table 1, entry 6). BSA-Mediated Michael Addition Reaction
of ArP(OMe/OEt)OTMS to R,â-Unsaturated Esters: General
Procedure (Table 1, Method B). To an oven-dried round-bottomed
5756 J. Org. Chem., Vol. 72, No. 15, 2007

p-Aminophenyl H-Phosphinic Acids and Esters
2
flask containing 4d (224 mg, 0.734 mmol) and dimethyl 2-meth-
yleneglutarate (5a) (253 mg, 1.47 mmol) was added anhydrous CH2-
Cl2 (5 mL), followed by BSA (1.82 mL, 7.34 mmol) under an Ar
atmosphere. The reaction mixture was heated under reflux for 7
days. The reaction was quenched by the addition of water (10 mL)
and stirred for 30 min. The layers were separated by adding ethyl
acetate (50 mL) and brine (10 mL). The aqueous layer was back-
extracted with ethyl acetate (3 × 40 mL). The combined organic
layers were dried over Na2SO4, filtered, and concentrated in vacuo.
The crude product was purified by silica gel chromatography, using
ethyl acetate as the eluant, to afford 6b (250 mg, 0.524 mmol, 71%)
as a colorless oil: 1H NMR (CDCl3) δ 7.75 (1 H, br s), 7.69-7.63
(2 H, m), 7.62-7.57 (2 H, m), 7.38-7.31 (5 H, m), 5.20 (2 H, s),
3.65-3.48 (9 H, m), 2.84-2.76 (1 H, m), 2.46-2.25 (3 H, m),
2.01-1.90 (3 H, m); 13C NMR (CDCl3) δ 174.3 (dd, 3JCP ) 18.9,
) 12.8, 8.6 Hz), 52.4, 51.9 (d, JPOC ) 10.6 Hz), 51.6, 51.1 (d,
5JPOC ) 6.6 Hz), 38.5, 31.7 (dd, 1JCP ) 101.8, 32.9 Hz), 31.2, 28.5
(d, JCP ) 11.5 Hz); 31P NMR (CDCl3) δ 43.8, 43.7, 43.4,43.3;
2
MS (ESI) m/z 402.1 ([M + Na]+, 100).
2-[((4-Aminophenyl)(methoxy)phosphinoyl)methyl]pentane-
1,5-dioic Acid (12). A solution of 6e (135 mg, 0.336 mmol) in 5
mL of 6 N HCl was heated under reflux for 10 h. The solution
was allowed to cool to room temperature. Solvent was removed in
vacuo to afford 12 as a colorless oil: 1H NMR (CD3OD) δ 7.73-
7.66 (2 H, m), 7.31-7.12 (2 H, m), 2.42-2.38 (1 H, m), 2.13-
1.99 (3 H, m), 1.80-1.57 (3 H, m); 31P NMR (D2O) δ 36.8. Note:
Compound 12 obtained from 6c has identical spectral data as that
from 6e.
2-[((4-(N-Methylamino)phenyl)(methoxy)phosphinoyl)methyl]-
pentane-1,5-dioic Acid Dimethyl Ester (7a). Preparation of
p-Aminophenylphosphinic Acid Methyl Esters via Hydrogena-
tion To Remove the Cbz Protecting Group: General Procedure.
To a suspension of 6a (270 mg, 0.549 mmol) and 5% Pd/C (200
mg) in THF (20 mL) was bubbled H2 under liquid for 10 min,
then the mixture was kept under H2 atmosphere for 24 h. Pd/C
was removed by filtration through a Celite pad. The filtrate was
concentrated in vacuo. The crude product was purified by a silica
gel chromatography, using ethyl acetate as the eluant, to afford 7a
(160 mg, 0.448 mmol, 82%) as a colorless oil: 1H NMR (CDCl3)
δ 7.55-7.50 (2 H, m), 6.64-6.61 (2 H, m), 4.35 (1 H, br s), 3.72-
3.51 (9 H, m), 2.88-2.81 (3.5 H, m), 2.80-2.75 (0.5 H, s), 2.44-
2.28 (3 H, m), 2.03-1.86 (3 H, m); 13C NMR (CDCl3) δ 174.6
5
7.2 Hz), 172.8 (d, JCP ) 5.2 Hz), 153.1, 142.3, 135.7, 133.0 (dd,
2JCP ) 13.7, 10.9 Hz), 128.6, 128.4, 128.3, 123.1 (dd, 1JCP ) 130.3,
3
2
41.7 Hz), 118.2 (dd, JCP ) 12.8, 8.7 Hz), 67.2, 51.9 (d, JPOC
)
5
4
10.6 Hz), 51.6, 51.1 (d, JPOC ) 5.4 Hz), 38.5 (d, JCP ) 4.8 Hz),
1
2
31.7 (dd, JCP ) 101.9, 32.9 Hz), 31.2, 28.5 (d, JCP ) 11.5 Hz);
31P NMR (CDCl3) δ 44.4, 44.0; MS (ESI) m/z 500.1 ([M + Na]+,
100); ESI-HRMS (m/z) calcd for C23H28NO8PNa [M + Na]+
500.1450, found 500.1457.
2-[((4-(N-Benzyloxycarbonylamino)phenyl)(methoxy)phos-
phinoyl)methyl]pentane-1,5-dioic Acid Dimethyl Ester (6b,
Table 1, entry 7). BSA-Mediated Michael Addition Reaction
of ArP(OMe/OEt)OTMS to R,â-Unsaturated Ester under
Microwave Conditions: General Procedure (Table 1, Method
B, Microwave). To an oven-dried microwave reaction vessel
containing 4d (32 mg, 0.105 mmol) and dimethyl 2-methylene-
glutarate (5a) (36 mg, 0.209 mmol) was added anhydrous CH2Cl2
(1 mL), followed by BSA (0.26 mL, 1.05 mmol) under an Ar
atmosphere. The reaction mixture was heated by microwave reactor
at 80 °C for 4 h (250 W). The reaction was quenched by the addition
of HCl (0.1 N, 2 mL) and water (5 mL) and stirred for 10 min.
The layers were separated by adding ethyl acetate (20 mL) and
brine (10 mL). The aqueous layer was back-extracted with ethyl
acetate (3 × 20 mL). The combined organic layers were dried over
Na2SO4, filtered, and concentrated in vacuo. The crude product was
purified by silica gel chromatography, using ethyl acetate as the
eluant, to afford 6b (25 mg, 52 µmol, 50%) as a colorless oil.
Spectra are identical with those obtained for 6b by Method B.
2-[((4-(N-Benzyloxycarbonylamino)phenyl)(methoxy)phos-
phinoyl)methyl]pentane-1,5-dioic Acid Dimethyl Ester (6b,
Table 1, entry 10). Michael Addition Reaction of ArP(O)-
(OMe)-H to R,â-Unsaturated Esters Promoted by KOt-Bu
(Table 1, Method C: General Procedure). To an oven-dried
round-bottomed flask containing 4d (110 mg, 0.360 mmol) and
KOt-Bu (49 mg, 0.437 mmol) was added anhydrous THF(5 mL)
at 0 °C. The reaction mixture was stirred for an additional 20 min
at 0 °C before a solution of dimethyl 2-methyleneglutarate 5a (105
mg, 0.610 mmol) in THF (5 mL) was added. After an additional 2
h at room temperature, the starting material was consumed as
indicated by 31P NMR. The reaction was quenched by the addition
of water (5 mL) and stirred for 10 min. The layers were separated
by adding ethyl acetate (20 mL) and brine (10 mL). The aqueous
layer was back-extracted with ethyl acetate (3 × 20 mL). The
combined organic layers were dried over Na2SO4, filtered, and
concentrated in vacuo. The crude product was purified by silica
gel chromatography, using ethyl acetate as the eluant, to afford 6b
(87 mg, 0.182 mmol, 51%) as a colorless oil. Spectra are identical
with those obtained for 6b by Method B.
3
5
4
(dd, JCP ) 23.7, 7.4 Hz), 172.9 (d, JCP ) 5.3 Hz), 152.5 (t, JCP
2
1
) 2.2 Hz), 133.4 (dd, JCP ) 18.5, 11.3 Hz), 115.0 (dd, JCP
)
)
136.4, 53.7 Hz), 111.6 (dd, 3JCP ) 13.5, 7.9 Hz), 51.8 (d, 2JPOC
7
5
9.3 Hz), 51.6 (d, JPOC ) 1.9 Hz), 50.7 (d, JPOC ) 6.3 Hz), 38.6
4
1
(dd, JCP ) 5.1, 2.7 Hz), 32.0 (dd, JCP ) 102.3, 34.4 Hz), 31.3,
30.0, 28.6 (dd, JCP ) 10.3, 1.4 Hz); 31P NMR (CDCl3) δ 44.8,
2
44.4; MS (ESI) m/z 358.1 ([M + H]+, 100); ESI-HRMS (m/z) calcd
for C16H25NO6P [M + H]+ 358.1420, found 358.1412.
[((4-(N-((2-Amino-3,4-dihydro-4-oxo-6-pteridinyl)methyl)ami-
no)phenyl)(methoxy)phosphinoyl)methyl]pentane-1,5-dioic Acid
Dimethyl Ester (10a, Table 2, entry 6)). Preparation of Het-
erocyclic Phosphinic Acid Methyl Esters by SN2 Alkylation:
General Procedure. A mixture of 7b (45 mg, 0.131 mmol) and
8a (66 mg, 0.196 mmol) in DMA (2 mL) was stirred at room
temperature. The reaction mixture was monitored by analytical
HPLC, and consumption of the limiting reagent, 7b, was complete
after 14 d. The crude product was purified by semipreparative HPLC
(see the Supporting Information for details) and solvent was
removed via lyophilization to afford 10a (P-OMe) (9 mg, 17 µmol,
13%) as a light brown solid and 10a′ (P-OH) (26 mg, 52 µmol,
39%) as a greenish yellow solid. 10a: mp >300 °C (210 °C dec);
1H NMR (DMSO) δ 8.64 (1 H, s), 7.38-7.33 (2 H, m), 7.16-
7.12 (1 H, m), 7.05 (1 H, br s), 6.76-6.71 (2 H, m), 4.48 (2 H, d,
J ) 5.5 Hz), 3.56-3.35 (9 H, m), 2.57-2.51 (1 H, m), 2.26-1.92
(4 H, m), 1.85-1.68 (2 H, m); 31P NMR (DMSO) δ 43.8, 43.6;
MS (ESI) m/z 541.1 ([M + Na]+, 100); ESI-HRMS (m/z) calcd
for C22H27N6O7PNa [M + Na]+ 541.1577, found 541.1589; UV
λmax (0.1 N NaOH) 235, 267, 329, 369 nm; (0.1 N HCl) 268, 320
nm; analytical HPLC tR ) 19.3 min. 10a′: mp >300 °C (210 °C
dec); 1H NMR (DMSO) δ 8.62 (1 H, s), 7.37-7.33 (2 H, m), 7.28
(2 H, br s), 6.73 (1 H, br s), 6.59-6.58 (2 H, m), 4.44 (2 H, s),
3.60-3.38 (6 H, m), 2.55-2.45 (2 H, m), 2.17 (2 H, t, J ) 7.5
Hz), 1.89-1.78 (2 H, m), 1.73-1.63 (1 H, m), 1.63-1.53 (1 H,
m); 13C NMR (DMSO) δ 175.5, 173.1, 161.9, 156.6, 154.8, 149.6
4
2
(d, JCP ) 114.4 Hz), 149.6, 148.1, 132.8 (dd, JCP ) 168.3, 18.2
1
3
If NaOMe was used as base, in addition to the desired product,
6b, a side-product compound 6e was isolated (33%) as a colorless
oil: 1H NMR (CDCl3) δ 8.01 (1 H, br s), 7.70-7.61 (4 H, m),
3.76-3.50 (12 H, m), 2.88-2.72 (1 H, m), 2.46-2.25 (3 H, m),
2.00-1.90 (3 H, m); 13C NMR (CDCl3) δ 174.3 (dd, 3JCP ) 17.9,
Hz), 128.3, 124.9 (d, JCP ) 125.8 Hz), 111.7 (d, JCP ) 156.6
Hz), 53.4, 52.3, 51.1, 49.9, 47.5, 46.5, 45.4, 39.0, 38.8, 38.6, 38.5,
32.4, 31.4, 30.4, 29.6, 28.6, 27.6; 31P NMR (DMSO) δ 27.1; MS
(ESI) m/z 527.1 ([M + Na]+, 100); ESI-HRMS (m/z) calcd for
C21H25N6O7PNa [M + Na]+ 527.1420, found 527.1425; UV λmax
(0.1 N NaOH) 259, 366 nm; (0.1 N HCl) 263, 322 nm; analytical
HPLC tR ) 14.4 min.
7.2 Hz), 172.8 (d, 5JCP ) 5.5 Hz), 153.8, 142.6, 132.9 (dd, 2JCP
)
14.1, 11.1 Hz), 122.9 (dd, 1JCP ) 130.7, 41.9 Hz), 118.1 (dd, 3JCP
J. Org. Chem, Vol. 72, No. 15, 2007 5757

Yang and Coward
[((4-((2-Pivaloylamino-4-hydroxy)pteroylamino)phenyl)-
(methoxy)phosphinoyl)methyl]pentane-1,5-dioic Acid Dimethyl
Ester (10b1). Preparation of Heterocyclic Phosphinic Acid
Esters by Reductive Amination via a Two-Step Sequence:
General Procedure (Method A). The solution of phosphinic
methyl ester 7b (55 mg, 0.160 mmol) and 9a (40 mg, 0.145 mmol)
in AcOH (2 mL) was stirred at room temperature for 15 min. A
solution of BH3‚NHMe2 (17 mg, 0.289 mmol) in AcOH (0.5 mL)
was added to reduce the imine intermediate. After an additional
10 min, the solvent was removed in vacuo to afford the crude
product, which was purified by silica gel chromatography, using
ethyl acetate as eluant, to remove excess 7b, followed by a mixture
of 5% of MeOH and 95% of ethyl acetate to afford 10b1 (54 mg,
90 µmol, 61%) as a yellow solid: mp 80-82 °C; 1H NMR (CDCl3)
δ 12.43 (1 H, br s), 8.89 (1 H, s), 8.52 (1 H, br s), 7.57-7.53 (2
H, m), 6.75-6.73 (2 H, m), 5.30 (1 H, br s), 4.73 (2 H, d, J ) 5.6
Hz), 3.64-3.49 (9 H, m), 2.86-2.76 (1 H, m), 2.44-2.28 (3 H,
m), 2.05-1.86 (3 H, m), 1.36-1.34 (9 H, m); 13C NMR (CDCl3)
δ 180.9, 174.5 (dd, 3JCP ) 21.8, 7.1 Hz), 172.9 (d, 5JCP ) 4.5 Hz),
solid was precipitated out by addition of a very small amount of
ice-cold water to the residue, collected by centrifuge, washed with
ice-cold water twice, and dried in vacuo to afford the first batch
1a (16.3 mg). After the supernatant was dried by lyophilization,
the resulting residue was dissolved in DMA. The mixture was
filtered through a nylon filter before it was purified by semiprepara-
tive HPLC (see the Supporting Information for details) and solvent
was removed via lyophilization to afford the second batch 1a (2
mg) (18.3 mg (total), 38 µmol, 65%) as a yellow solid: mp >300
°C dec; 1H NMR (D2O) δ 8.63 (1 H, s), 7.37 (2 H, t, J ) 8.7 Hz),
6.72 (2 H, d, J ) 8.0 Hz), 4.53 (2 H, s), 2.41-2.32 (1 H, m), 2.10
(2 H, t, J ) 7.4 Hz), 2.02-1.93 (1 H, m), 1.65-1.58 (3 H, m); 31
P
NMR (D2O) δ 32.5, 32.3; MS (ESI) m/z 477.1 ([M + H]+, 100);
ESI-HRMS (m/z) calcd for C19H22N6O7P [M + H]+ 477.1288,
found 477.1293. UV λmax (0.1 N NaOH) 258, 364 nm; (0.1 N HCl)
263, 321 nm; analytical HPLC tR ) 9.6 min.
Method B: Preparation of Heterocyclic Aryl Phosphinic
Acids from the Corresponding Protected 2-Amino Heterocycles
and Glutarate Di-tert-butyl Esters via Hydrolysis of the Pivaloyl
Group under Basic Conditions Followed by TFA-Mediated
2
159.5, 154.5, 151.9, 150.5, 149.4, 149.2, 133.6 (dd, JCP ) 16.8,
11.4 Hz), 130.6, 116.6 (dd, 1JCP ) 135.6, 48.4 Hz), 112.5 (dd, 3JCP
Hydrolysis of the tert-Butyl Ester: General Procedure.45
A
2
5
) 13.4, 7.1 Hz), 51.8 (d, JPOC ) 8.5 Hz), 51.6, 50.9 (d, JPOC
)
)
degassed solution of 10b2 obtained by reductive amination (Scheme
6) (50 mg, 73 µmol) and 1 N NaOH (3.0 mL) in CH3OH (1.5 mL)
was stirred at room temperature protected from the light. The
reaction mixture was monitored by analytical HPLC and LC-MS
frequently. Complete removal of the N-pivaloyl group and partial
removal of tert-butyl esters was observed after 1 day. The crude
reaction mixture was concentrated in vacuo to remove CH3OH,
followed by acidification with 1 N HCl to pH 2-3 at 0 °C. After
the solvent was removed via lyophilization, a suspension of the
resulting solid in CH2Cl2 (2.5 mL) was treated overnight with TFA
(0.5 mL), after which solvent was removed in vacuo. The desired
product was precipitated by addition of a very small amount of
ice-cold water, collected by centrifuge (SpeedVac SPD111V, 4
min), washed with ice-cold water twice, and dried in vacuo to afford
1a (30 mg, 63 µmol, 86%) as an orange-yellow solid. Spectra are
identical with those obtained for 1a by Method A.
6.3 Hz), 46.6, 40.5, 38.6 (dd, 4JCP ) 7.9, 2.8 Hz), 31.8 (dd, 1JCP
102.3, 37.4 Hz), 31.2, 28.5 (d, JCP ) 11.4 Hz), 26.8; 31P NMR
(CDCl3) δ 44.6, 44.3; MS (ESI) m/z 625.2 ([M + Na]+, 100); ESI-
HRMS (m/z) calcd for C27H35N6O8PNa [M + Na]+ 625.2152, found
625.2150; UV λmax (0.1 N NaOH) 267, 356 nm; (0.1 N HCl) 270,
326 nm; analytical HPLC tR ) 26.8 min.
2
[((4-((2-Pivaloylamino-4-hydroxy)-5-deazapteroylamino)phe-
nyl)(methoxy)phosphinoyl)methyl]pentane-1,5-dioic Acid Dim-
ethyl Ester 10d1. Preparation of Heterocyclic Phosphinic Methyl
Esters by Reductive Amination via a One-Step Sequence:
General Procedure (Method B). The mixture of phosphinic methyl
ester 7b (25 mg, 73 µmol), 9b (30 mg, 0.109 mmol), and NaBH3-
CN (7 mg, 0.111 mmol) in AcOH (1.5 mL) was stirred at room
temperature for 48 h. The solvent was removed in vacuo to afford
the crude product, which was purified by silica gel chromatography,
using a mixture of 5% of MeOH and 95% of EtOAc as eluant, to
afford 10d1 (27 mg, 45 µmol, 61%) as a light yellow solid: mp
Acknowledgment. We thank Drs. James Bristol and Ellen
Dobrusin, Parke-Davis Pharmaceutical Research Division,
Warner-Lambert Co. for their encouragement and support of
this research that initially involved the screening of a selected
series of compounds from the Parke-Davis library of antifolates
for antimalarial activity. In addition, this research was supported
by a grant from the National Cancer Institute (CA28097). We
gratefully acknowledge Prof. Jean-Luc Montchamp for several
stimulating and helpful discussions. We thank Prof. Andrew
Bognar for providing preliminary data on the inhibition of DHFS
activity by compounds 1a-e. We thank Prof. Melanie Sanford
for the use of a microwave reactor. We thank Jason Rush and
David Bartley for initial research on the synthesis of aryl
phosphinic acids, specifically 2a, 4a, and 6a.
1
80-82 °C; H NMR (CDCl3) δ 12.13 (1 H, br s), 8.88 (1 H, s),
8.52 (1 H, br s), 8.49 (1 H, s), 7.56-7.51 (2 H, m), 6.68-6.66 (2
H, m), 4.90 (1 H, br s), 4.54 (2 H, d, J ) 5.7 Hz), 3.69-3.50 (9 H,
m), 2.88-2.77 (1 H, m), 2.45-2.29 (3 H, m), 2.06-1.87 (3 H,
3
m), 1.30 (9 H, s); 13C NMR (CDCl3) δ 180.7, 174.6 (dd, JCP
)
23.3, 7.0 Hz), 173.0 (d, 5JCP ) 4.6 Hz), 161.0, 158.2, 155.8, 150.6,
2
149.1, 134.8, 133.7 (dd, JCP ) 18.6, 11.3 Hz), 131.7, 116.9 (dd,
1JCP ) 134.9, 52.5 Hz), 115.2, 112.5 (dd, JCP ) 13.3, 8.4 Hz),
3
2
5
51.9 (d, JPOC ) 8.3 Hz), 51.7, 50.9 (d, JPOC ) 6.3 Hz), 44.7,
4
1
40.5, 38.7 (dd, JCP ) 5.1, 2.8 Hz), 32.0 (dd, JCP ) 102.4, 39.9
Hz), 31.3, 28.7 (d, JCP ) 10.7 Hz), 26.9; 31P NMR (CDCl3) δ
2
44.4, 44.0; MS (ESI) m/z 624.2 ([M + Na]+, 100); UV λmax (0.1 N
NaOH) 246, 274, 324 nm; (0.1 N HCl) 276, 339 nm; analytical
HPLC tR ) 29.0 min.
[((4-(N-((2-Amino-3,4-dihydro-4-oxo-6-pteridinyl)methyl)ami-
no)phenyl)(hydroxy)phosphinoyl)methyl]pentane-1,5-dioic Acid
(1a). Method A: Preparation of Heterocyclic Aryl Phosphinic
Acids from the Corresponding Unprotected 2-Amino Hetero-
cycles and Glutarate Methyl Esters via Hydrolysis under Basic
Conditions: General Procedure. A degassed solution of 10a
obtained by SN2 alkylation (Scheme 5) (30 mg, 59 µmol) and 1 N
NaOH (3.0 mL) in CH3OH (3.0 mL) was stirred at room
temperature protected from the light. The reaction mixture was
monitored by analytical HPLC, and consumption of 10a was >95%
after 5 days. The crude reaction mixture was concentrated in vacuo
to remove CH3OH, followed by acidification with 1 N HCl to pH
2-3 at 0 °C. After the solvent was removed via lyophilization, a
Supporting Information Available: Experimental procedures
and characterization of compounds 2a,b, 4b-e, compounds noted
in Table 1 (entries 1, 2, 6, 8, 9, and 11), 7b,c, 10b2, 10c/c′, 10d2,
10e/e′, 10f/f′, 11a,b (R4 ) Et), and 1a-e, Table S1 (optimization
1
experiments, Table 1, method B), and H NMR, 13C NMR, and
31P NMR spectra for all new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
JO0707840
(45) This method is also applicable to glutarate dimethyl esters by
completely removing the dimethyl ester under basic condition, followed
by removing the P-OMe by TFA.
5758 J. Org. Chem., Vol. 72, No. 15, 2007