Full text of DOI:10.1021/jo061496r

Enantioselective Synthesis of Protected
L-4-[Sulfonamido(difluoromethyl)]phenylalanine and
L-4-[Sulfonamido(methyl)]phenylalanine and an Examination of
Hexa- and Tripeptide Platforms for Evaluating pTyr Mimics for
PTP1B Inhibition
Bryan Hill, Vanessa Ahmed, Daniel Bates, and Scott D. Taylor*
Department of Chemistry, UniVersity of Waterloo, 200 UniVersity AVenue West, Waterloo,
Ontario, Canada N2L 3G1
s5taylor@sciborg.uwaterloo.ca
ReceiVed July 19, 2006
The first enantioselective syntheses of L-4-(sulfonamidomethyl)phenylalanine and L-[sulfonamido-
(difluoromethyl)]phenylalanine suitably protected for peptide syntheses are described. A key step in the
synthesis of L-(sulfonamidomethyl)phenylalanine was an oxidative chlorination on Ac-L-Phe(4-CH2-
SCOCH3)-OEt to give crude Ac-L-Phe(4-CH2SO2Cl)-OEt, which could be reacted with amines to give
the corresponding sulfonamides. Key to the preparation of L-[sulfonamido(difluoromethyl)]phenylalanine
was a highly enantioselective reaction involving William’s auxiliary and a benzylic bromide intermediate.
These amino acids were incorporated into two peptide sequences, DADE-X-LNH2 and FmocGlu(OBn)-
X-LNH2, which have previously been employed as platforms for assessing pTyr mimics for inhibition of
protein tyrosine phosphatase 1B (PTP1B). Inhibition studies with these and other peptides and PTP1B
revealed that good inhibition could be obtained using the tripeptide platform, although the presence of
a pTyr mimic was not required for good inhibition. These results suggest that the FmocGlu(OBn)-X-
LNH2 tripeptide platform is not suitable for assessing pTyr mimics for PTP1B inhibition.
Introduction
enzyme that is involved in the down regulation of insulin
signaling and a well-validated therapeutic target for the treatment
of diabetes and obesity.2,3
The design and synthesis of hydrolytically stable phospho-
tyrosine (pTyr) mimetics has expanded in recent years to the
point where it has become almost an entire field of study unto
itself.1 This is mainly due to their application as inhibitors and
probes of therapeutically significant enzymes involved in protein
tyrosine phosphorylation or dephosphorylation and proteins that
bind to phosphotyrosine.1 Such proteins include protein tyrosine
kinases (PTKs), protein tyrosine phosphatases (PTPs), and SH2
domain binding proteins. The application of pTyr mimics to
the development of inhibitors of PTPs has been a particularly
active area of research, with a distinct emphasis on PTP1B, an
pTyr mimics for PTP1B inhibition are often initially assessed
by incorporating them into peptide platforms with the D-A-D-
E-X-LNH2 (X ) pTyr mimic) sequence being the most
common. This hexapeptide corresponds to a phosphorylated
sequence (X ) pTyr) in the epidermal growth factor receptor
and is a good substrate for PTP1B. The most effective pTyr
mimic, in the context of this hexapeptide platform and PTP1B
inhibition, is difluorophosphonomethylphenylalanine (F2Pmp,
(2) (a) Taylor, S. D. Curr. Top. Med. Chem. 2002, 3, 759. (b) Taylor, S.
D.; Hill, B. Exp. Opin. InVest. Drugs 2004, 13, 199.
(3) Ramachandran, C.; Kennedy, B. Curr. Top. Med. Chem. 2002, 3,
749.
(1) Burke, T. R.; Lee, K. Acc. Chem. Res. 2003, 36, 426.
10.1021/jo061496r CCC: $33.50 © 2006 American Chemical Society
Published on Web 09/20/2006
8190
J. Org. Chem. 2006, 71, 8190-8197

EValuatiion of pTyr Mimics for PTP1B Inhibition
Indeed, tripeptides bearing monoanionic carboxylic acid-based
mimics exhibited affinities for PTP1B only 2-3-fold greater
than the analogous F2Pmp-bearing tripeptide.8,9 These results
led us to question whether a neutral pTyr mimic could also be
effective in the tripeptide platform. An effective neutral pTyr
mimic would be very useful in increasing the bioavailability of
PTP inhibitors. Neutral pTyr mimics are rare. Researchers at
Sugen have suggested, on the basis of studies with nonpeptidyl
PTP1B inhibitors, that the trifluoromethylsulfonamido group is
an effective neutral phosphate mimic.10 We chose to examine
the sulfonamide 3 as a potential pTyr mimic. The sulfonamide
group has found widespread use as a pharmacophore in
medicinal chemistry and numerous bioactive agents bear this
functionality.11 It has a geometry similar to that of the phosphate
group yet does not bear a negative charge. Recently, Chen et
al. reported a racemic synthesis of amino acid 3 and its
incorporation into the D-A-D-E-X-LNH2 platform.12 Although
these workers found this peptide to be a very poor inhibitor of
PTP1B, this amino acid was never examined as a pTyr mimic
in the FmocGlu(OBn)-X-LNH2 platform where the potential
constraints of the hexapeptide platform mentioned above may
not be as severe. Burke and co-workers have shown that peptides
bearing F2Pmp can be up to 1000-fold more potent PTP1B
inhibitors than their nonfluorinated analogues.4a-c pH studies,4c
computational studies, and X-ray crystallographic analyses13 of
a DFMP-bearing inhibitor complexed to PTP1B strongly suggest
that the effect of the fluorines is due to interaction of the
fluorines with residues in the active site and is not due to pKa
effects. Consequently, we also decided to examine the fluori-
nated analogue of 3, compound 4, as a pTyr mimic. Here, we
report an enantioselective synthesis of amino acids 3 and 4,
suitably protected for peptide synthesis, their incorporation into
the tri- and hexapeptide platforms described above, and an
evaluation of these peptides as inhibitors of PTP1B. These
studies yielded some important results concerning the utility of
the tripeptide scaffold for assessing pTyr mimics.
FIGURE 1. pTyr mimics 1-4.
1) (Figure 1).4 The difluoromethylenephosphonic acid (DFMP)
group has also been shown to be a very effective mimic even
in nonpeptidyl scaffolds.2,5 However, it has been suggested that
the dianionic nature of this mimic limits its cell permeability.1
Consequently, a wide variety of other pTyr mimics have been
synthesized and evaluated in this hexapeptide.6 We recently
demonstrated that difluorosulfonomethylphenylalanine (F2Smp,
2) is the most effective nonhydrolyzable monoanionic pTyr
mimic for PTP1B inhibition when incorporated into this
hexapepide platform.7 Most other monoanionic pTyr mimics
bear a carboxylic acid moiety, though these are not particularly
effective in the hexapeptide platform.6 Gao et al. have provided
a possible explanation as to why the D-A-D-E-X-LNH2 peptides
bearing certain pTyr mimics have a poor affinity for PTP1B.6
Although the peptide provides beneficial binding contacts with
residues outside of the active site, it limits the depth of insertion
and freedom of the pTyr mimic within the catalytic pocket.
Hence, it was suggested that studies using this hexapeptide
platform may not allow one to assess the true effectiveness of
certain pTyr mimics.6 More recently, Lee et al. demonstrated
that a wide variety of monoanionic and dianionic pTyr mimics
exhibiting poor affinity for PTP1B in the hexapeptide platform6
were considerably more effective in tripeptide platforms such
as FmocGlu(OBn)-X-LNH2 though none were as potent as the
F2Pmp-bearing hexapeptide.8,9 Similar results were also found
when these tripeptides were examined as inhibitors of Yersinia
PTP (YopH). On the basis of these results, it was suggested
that certain monoanionic COOH-based pTyr mimics, which
were considered to be of limited utility based on the hexapeptide
studies, were indeed good pTyr mimics.8
Results and Discussion
Chen et al. recently reported a racemic synthesis of Fmoc-
protected 3.12 After incorporation into the D-A-D-E-X-LNH2
hexapeptide, the resulting diastereomeric peptides were sepa-
rated by HPLC before inhibition studies. We wished to develop
an enantioselective synthesis of protected 3 to avoid having to
separate diastereomeric peptides. Chen et al.’s synthesis of
Fmoc-protected racemic 3 was achieved by reacting compound
5 with sodium sulfite followed by conversion of the resulting
sulfonate to the sulfonyl chloride using PCl5, which was reacted
with ammonium hydroxide to give sulfonamide 6 in 31% yield
from 5 (Scheme 1). Refluxing the sulfonamide in 6 N HCl
followed by reaction with Fmoc-OSu gave Fmoc-protected
racemic 3. Miranda et al. have shown that crude L-Phe(4-CH2-
SO3Na)-OH can be prepared by reacting the hydrochloride salt
of L-Phe(4-CH2Cl)-OH with sodium sulfite.14 Although this
presented a potential route to protected L-3 starting with
The difference in potency between several of the monoanionic
and dianionic mimics studied by Lee at al, when incorporated
into the tripeptide mentioned above, was surprisingly small.
(4) (a) Burke, T. R.; Kole, H. K.; Roller, P. P. Biochem. Biophys. Res.
Commun. 1994, 204, 129. (b) Kole, H. K.; Akamatsu, M.; Ye, B.; Yan, X.;
Barford, D.; Roller, P. P.; Burke, T. R. Biochem. Biophys. Res. Commun.
1995, 209, 817. (c) Chen, L.; Wu, L.; Otaka, A.; Smyth, M. S.; Roller, P.
R.; Burke, T. R.; den Hertog, J.; Zhang, Z.-Y. Biochem. Biophys. Res.
Commun. 1995, 216, 976. (d) Huyer, G.; Kelly, J.; Moffat, J.; Zamboni,
R.; Jia, Z.; Gresser, M. J.; Ramachandran, C. Anal. Biochem. 1998, 258,
19.
(5) Lau, C. K.; Bayly, C.; Gauthier, J. Y.; Li, C. S.; Therien, M.; Asante-
Appiah, E., Cromlish, W.; Boie, Y.; Forghani, F.; Desmarais, S.; Wang Q,
Skorey K, Waddleton D, Payette P, Ramachandran, C.; Kennedy, B. P.;
Scapin, G. Bioorg. Med. Chem. Lett. 2004, 1043.
(6) Gao, Y.; Wu, L.; Luo, J. H.; Guo, R.; Yang, D.; Zhang, Z.-Y.; Burke,
T. R. Bioorg. Med. Chem. Lett. 2000, 10, 923.
(7) Leung, C.; Lee, J.; Meyer, N., Jia, C.; Grzyb, J.; Liu, S.; Hum G.;
Taylor, S. D. Bioorg. Med. Chem. 2002, 10, 2309-2323.
(8) Lee, K.; Gao, Y.; Yao, Z.-J.; Phan, J.; Wu, L.; Liang, J.; Waugh, D.
S.; Zhang, Z-Y.; Burke, T. R., Bioorg. Med. Chem. Lett. 2003, 13, 2577.
(9) Lee, K.; Boovanahalli, S. K.; Nam, K.-Y.; Kang, S.-U.; Lee, M.;
Phan, J.; Wu, L.; Waugh, D. S.; Zhang, Z.-Y.; No, K. T.; Lee, J. J.; Burke,
T. R. Bioorg. Med. Chem. Lett. 2005, 15, 4037.
(10) Huang, P.; Ramphal, J.; Wei, J.; Liang, C.; Jallal, B.; Mcmahon,
G.; Tang, C. Bioorg. Med. Chem. 2003, 11, 835.
(11) Bowman, W. C.; Rand, M. J. Textbook of Pharmacology, 2nd ed.;
Blackwell: London, 1979; Chapter 34.
(12) Chen, Y. T.; Xie, J.; Seto, C. T. J. Org. Chem. 2003, 68, 4123.
(13) Burke, T. R., Jr.; Ye, B.; Yan, X.; Wang, S.; Jia, Z.; Chen, L.; Zhang,
Z.-Y.; Barford, D. Biochemistry 1996, 35, 15989.
(14) Miranda, M. T. M.; Liddle, R. A.; Rivier, J. E. J. Med. Chem. 1993,
36, 1681.
J. Org. Chem, Vol. 71, No. 21, 2006 8191

Hill et al.
SCHEME 3. Attempted Synthesis of Protected 4 by a
SCHEME 1. Chen et al.’s Synthesis of Fmoc-Protected
Racemic 3
Cross-Coupling Approach
SCHEME 2. Synthesis of Fmoc-Protected 3
of approximately 1.5 pKa units per fluorine.16 The pKa of the
sulfonamide moiety in 3 can be estimated to be approximately
10.5.17 On the basis of Harrington’s studies, the pKa of the
sulfonamide moiety in 4 would be expected to be approximately
7.5. Although the nonfluorinated amino acid 3 was readily
incorporated into the hexapeptide platform mentioned above
with the sulfonamide group unprotected, we were concerned
that the lower pKa of the sulfonamide protons in compound 4
could potentially result in side reactions during peptide synthesis
that might not be encountered with 3.18,19 Therefore, we decided
to incorporate amino acid 4 into peptides with the sulfonamide
protected. To reduce the number of synthetic manipulations, it
was deemed advantageous to use the same sulfonamide protect-
ing group for the synthesis of protected amino acid 4 and its
incorporation into peptides. We recently reported that R,R-
difluorosulfonamides can be prepared by electrophilic fluorina-
tion using N-fluorobenzenesulfonimide (NFSi).20 In this paper,
we introduced the bis(2,4-dimethoxybenzyl) (DMB) group as
a new sulfonamide protecting group and this was particularly
effective for the preparation of R-fluorosulfonamides. Benzylic
sulfonamides protected with this group readily underwent
electrophilic fluorination and were stable to wide variety of
reaction conditions yet the group could be easily removed using
TFA/CH2Cl2. Thus, we anticipated that the DMB group would
be suitable for the construction of protected 4 and for peptide
synthesis using Fmoc chemistry.
We initially examined a cross-coupling route to protected 4
since it represented a relatively direct route to this compound
and has been applied to the synthesis of other pTyr mimics
including compounds 1 and 2 (Scheme 3).21 However, numerous
attempts to couple bis(dimethoxybenzyl)-protected sulfonamide
1220 to zincate 13 with a variety of catalysts and under different
reaction conditions were unsuccessful. Therefore, we examined
an alternative approach using William’s auxiliary, a route that
has also been used in the synthesis of other pTyr mimics
(Scheme 4).6 Compound 12 was treated with n-BuLi followed
by the addition of DMF to give the aldehyde 14 in 77% yield.
Conversion to the benzyl bromide 16 was achieved by reducing
the aldehyde moiety of 14 to alcohol 15 with NaBH4 (90%)
followed by treatment of 15 with CBr4/PPh3 (95%). Reaction
of 16 with the lithium enolate of William’s lactone gave
compound 17 in 80% yield. The desired protected amino acid
18 was readily obtained by hydrogenation of 17 using PdCl2 as
protected L-Phe(4-CH2Cl)-OH and a series of reactions involving
sodium sulfite, PCl5 and finally ammonia, the modest yields of
these reactions obtained by Chen et al. on compound 5 prompted
us to investigate an alternative route, preferably one which did
not require the synthesis and purification of polar sulfonate salts.
To this end, Ac-L-Phe(4-CH2Cl)-OEt14,15 was reacted with
potassium thioacetate to give thioacetate 7 in 96% yield (Scheme
2). Oxidative chlorination of 7 with Cl2 in dichloromethane-
water provided crude sulfonyl chloride 8. Reaction of crude 8
with aq NH4OH in various solvents or NH3/MeOH gave
sulfonamide 9; however, the yields were inconsistent ranging
from 10 to 50%, and purification of the product was difficult.
However, reaction of crude 8 with 2,4-dimethoxybenzylamine
gave secondary sulfonamide 10 in 77% yield (from 7). Acid
hydrolysis of 10 gave amino acid 3 in 82% yield. Treatment of
3 with Fmoc-OSuc yielded the desired Fmoc-protected amino
acid 11 in 93% yield. HPLC analysis of two diastereomeric
dipeptides, Fmoc-3-L-Leucine-NH2 and Fmoc-3-DL-Leucine-
NH2, revealed that the ee of 11 was >97%.
(16) Trepka, R. D.; Harrington J. K.; Belisle, J. B. J. Org. Chem. 1974,
39, 1094.
(17) Blackburn, G. M.; Turkmen, Org. Biomol. Chem. 2005, 3, 225-
226.
(18) Alexsiev, B.; Stoev, S.; Kostov, M. Pharmazie 1971, 26, 18.
(19) Videnov, G.; Alexsiev, B.; Stoev, M.; Paipanov, T.; Jung, G. Liebigs
Ann. Chem. 1993, 941.
(20) Hill, B.; Liu, Y.; Taylor, S. D. Org. Lett. 2004, 6, 4285.
(21) See: Liu, S.; Dockendorf, C.; Taylor, S. D. Org. Lett. 2001, 3, 1571
and references therein.
Before a synthesis of protected 4 was undertaken, the issue
of the protection of the sulfonamide moiety was addressed.
Harrington and co-workers demonstrated that the presence of
fluorines alpha to a sulfonamide leads to a linear acidity increase
(15) Bo, L.; Jiarong, Z. Synth. Commun. 1999, 29, 2293.
8192 J. Org. Chem., Vol. 71, No. 21, 2006

EValuatiion of pTyr Mimics for PTP1B Inhibition
TABLE 1. Inhibition of PTP1B with Peptides 19-25
SCHEME 4. Synthesis of Fluorosulfonamide 18
peptide
IC50 (µM)
DADE-3-LNH2 (19)
DADE-4-LNH2 (20)
FmocGlu(OBn)-3-LNH2 (21)
FmocGlu(OBn)-4-LNH2 (22)
FmocGlu(OBn)-2-LNH2 (23)
DADE-2-LNH2 (24)
0% inhibition at 1 mM
0% inhibition at 1 mM
3.4 ( 0.4
6.4 ( 1.0
7.4 ( 1
10 ( 1a
FmocGlu(OBn)-Phe-LNH2 (25)
4.1 ( 1.0
a From ref 7.
FIGURE 2. Tripeptide inhibitor (26) of YopH and PTP1B and a
hydrophobic inhibitor (27) of PTP1B.
as well as F2Pmp (1), in the same tripeptide platform.8 The
similarity of the IC50’s of tripeptides 21 and 22 to tripeptide 23
and to those of Lee et al. prompted us to question if the pTyr
mimics were contributing at all to the potency of the tripeptides.
An examination of the data reported by Lee et al.8 revealed
that tripeptide FmocGlu(OBn)-Phe-LNH2 (25), a control in
which no phosphate mimic is present on the phenylalanine side
chain, had not been evaluated as a PTP1B inhibitor. Therefore,
we synthesized peptide 25 and evaluated it as a PTP1B inhibitor.
This peptide exhibited an IC50 of 4.1 µM which is similar to
the IC50’s for 21-23 and many of the tripeptides reported by
Lee et al. with PTP1B. This result raises the possibility that the
phosphate “mimic” portion of the tripeptide inhibitors contrib-
utes little or not at all to their inhibitory potency. It is possible
that the phosphate-mimicking portion of these tripeptides do
not bind or binds very poorly in the active site or elsewhere on
the protein and that the majority of the binding energy is a result
of the interaction of a hydrophobic group on the inhibitor with
a hydrophobic region on the protein. Lee et al. have recently
shown that the presence of a hydrophobic aromatic hydrocarbon
moiety at the N-terminus was important for obtaining good
PTP1B inhibition with their tripeptides.9
catalyst followed by treating the crude product with Fmoc-OSuc
in dioxane-aq Na2CO3 (99%, two steps). The ee of the protected
F2Samp, 18, was determined by incorporating 18 into the two
dipeptides, Fmoc-4-L-leucine and Fmoc-4-DL-leucine. HPLC
analysis of the dipeptides revealed that 18 was obtained in a
>97% ee.22
Amino acids 11 and 18 were incorporated into the hexa- and
tripeptide platform mentioned above using standard solid-phase
peptide synthesis protocols employing Fmoc chemistry. Puri-
fication by preparative RP-HPLC gave the desired peptides
DADE-3-LNH2 (19), DADE-4-LNH2 (20), FmocGlu(OBn)-3-
LNH2 (21), and FmocGlu(OBn)-4-LNH2 (22).
IC50’s with PTP1B were performed in bis-tris buffer at pH
6.5, containing 15 mM NaCl, 2 mM EDTA, 0.001% Triton
X-100 and 7.5% DMSO using 6,8-difluoro-4-methylumbelliferyl
phosphate (DiFMUP) as substrate at Km concentration (20
µM).23 The IC50’s are given in Table 1. Hexapeptides 19 and
20 showed no inhibition even at 1.0 mM. These results are
consistent with Chen et al.’s studies on peptide 19, which
showed that this compound was a very poor inhibitor of
PTP1B.12 However, tripeptides 21 and 22 exhibited IC50’s of
3.4 and 6.4 µM, respectively. Clearly, the fluorines in peptides
20 and 22 had little or no impact on inhibitory potency. We
also prepared tripeptide 23 bearing F2Smp (2) and found that it
exhibited an IC50 similar to 21 and 22 as well as hexapeptide
24.7 Indeed, the IC50 values for tripeptides 21-23 are, in general,
not significantly different from those reported by Lee et al. for
a variety of monoanionic carboxylic acid-based pTyr mimics,
Lee et al. examined in some detail the kinetics of YopH
inhibition with tripeptide 26 (Figure 2) which has an IC50 of
1.8-2.8 µM with YopH and 2.9-4.6 µM with PTP1B.8,9 It did
not exhibit a strictly competitive mode of inhibition. It showed
a steep IC50 curve with a slope factor of 3-5 depending upon
the enzyme concentration suggesting that more than one
molecule of inhibitor binds to YopH. IC50 values increased as
the concentration of YopH increased. They also observed a time
dependency in that preincubation with YopH for 5 min
decreased the IC50 from 3.2 to 1.5 µM. Treatment of YopH
with 200 µM inhibitor 26 for 30 min followed by dilution into
excess substrate resulted in a 40% loss of activity. It is
interesting to note that some of these phenomena were also
encountered with hydrophobic inhibitor 27 (Figure 2) and
PTP1B in that it too displayed time-dependent inhibition and
(22) The ee of 18 could also be determined by 19F NMR. 19F NMR of
the F2Samp-DL-leucine dipeptide exhibited a singlet corresponding to one
diastereomeric dipeptide as well as an ABQ doublet of doublets corre-
sponding to the other diastereomeric dipeptide. 19F NMR of the F2Samp-
L-leucine dipeptide exhibited only a single peak. We could not detect any
of the ABQ doublet of doublets corresponding to the other diastereomer
confirming that F2Samp was prepared in very high (>97%) enantiomeric
purity.
(23) Welte, S.; Baringhaus, K. H.; Schmider, W.; Muller, G.; Petry, S.;
Tennagels, N. Anal. Biochem. 2005, 38, 32-38.
J. Org. Chem, Vol. 71, No. 21, 2006 8193

Hill et al.
nonclassical inhibition patterns.24,25 Using 23 as a model
inhibitor, we have found that some of the above phenomena
encountered with 26 and 27 also occur with PTP1B. A
Lineweaver-Burk plot did not exhibit classical inhibition
patterns and we were unable to determine a Ki from the plot.
The IC50’s depended upon PTP1B concentration. At 1.5 nM
enzyme, the IC50 of 23 was 4.3 µM while at 12 nM enzyme the
IC50 increased to 12 µM, however, the slope factors were 1.2-
1.4. We did not observe a significant time dependence upon
inhibition. We also examined the reversibility of inhibition by
incubating 150 nM PTP1B with 50 µM of 23 for either 15 s or
60 min and then diluting the mixture 50-fold into a solution
containing 1.0 mM DIFMUP and assaying enzyme activity. In
each case, only 5-10% of the activity was recovered. Moreover,
even after extensive dialysis, no activity was recovered. As
pointed out by Lee et al. and others,8,24 these results are
consistent with the properties associated with nonspecific
promiscuous inhibitors.26-28 However, these workers also point
out that some of the observed effects are not as pronounced as
previously reported for promiscuous inhibitors.8
In summary, we have described the first enantioselective
synthesis of protected L-4-[sulfonamido(methyl)]phenylalanine
(3) and L-4-[sulfonamido(difluoromethyl)]phenylalanine (4) and
incorporated these species into a hexapeptide that has been
widely used for assessing pTyr mimics as well as the tripeptide,
FmocGlu(OBn)-X-LNH2, that has recently been used by others
for evaluating pTry mimics.8 Although the tripeptides bearing
these mimics were relatively good inhibitors, our results suggest
that the phosphate mimicking portion of the tripeptides may
not be contributing significantly to their potency. Moreover,
studies with inhibitor 23 reveals that this class of tripeptide
inhibitor exhibits nonclassical inhibition patterns with PTP1B.
As mentioned above, it has been suggested that the hexapeptide
platform may limit the depth of insertion and freedom of the
pTyr mimic within the catalytic pocket and this may result in
artificially low potencies for certain pTyr mimics.6,8 Our results
do not suggest that this hypothesis is incorrect though they do
raise serious questions as to the value of the FmocGlu(OBn)-
X-LNH2 peptide as a platform for assessing pTyr mimics for
PTP1B inhibition. Indeed, our results emphasize an important
point about assessing pTyr mimics in any platform in that it is
prudent to compare the platform bearing the pTyr mimic to the
analogous structure lacking the phosphate mimicking portion.
Although amino acids 3 and 4 were not good pTyr mimics
for PTP1B, they may also prove to be useful in the development
of inhibitors and probes of other enzymes. Chen et al. have
shown that peptide 19 was a modest inhibitor of Yersinia PTP.12
The presence of fluorines R to the sulfonamide moiety might
enhance its potency with Yersinia PTP. It is also worthy of note
that tyrosine sulfation has been recognized as an important post-
translational modification.29 The growing list of proteins that
bind to sulfotyrosine include viral proteins that recognize
sulfated receptor proteins on cell surfaces.30,31 Hence, a need
has also arisen for hydrolytically stable sulfotyrosine mimics.
We have recently demonstrated that the difluoromethylene-
sulfonamide group is a relatively good nonhydrolyzable replace-
ment of the sulfate group in estrone sulfate in that this species
is a relatively good reversible inhibitor of steroid sulfatase.32
Thus, amino acid 4 may prove to be an effective sTyr mimic
and useful in the development of inhibitors of protein-protein
interactions.
Experimental Section
Syntheses. L-2-Acetylamino-3-(4-acetylsulfanylmethylphenyl)-
propionic Acid Ethyl Ester (7). To a stirred 0 °C solution of L-(4-
chloromethyl)phenylalanine14,15 (48 g, 12.3 mmol), TBAI (452 mg,
1.23 mmol), and DMF (49 mL) was added potassium thiol acetate
(1.54 g, 13.5 mmol). The reaction was gradually warmed to rt and
stirred overnight. The reaction was quenched with water, extracted
with EtOAc (×3), washed with brine, dried over MgSO4, and
concentrated. The crude product was purified by flash chromatog-
raphy (60:40:4 hexanes/EtOAc/MeOH) yielded 3.81 g (96%) of 7
as a white solid: mp 47-48 °C; 1H NMR (300 MHz) δ 7.20 (2H,
d, J ) 7.9 Hz), 7.03 (2H, d, J ) 7.9 Hz), 5.90 (1H, bd, J ) 7.5
Hz), 4.84 (1H, dd, J ) 7.6, 5.8 Hz), 4.17 (2H, q, J ) 7.1 Hz), 4.08
(2H, s), 3.09 (2H, d, J ) 5.6 Hz), 2.35 (3H, s), 1.99 (3H, s), 1.24
(3H, t, J ) 7.1 Hz); 13C NMR (75 MHz) δ 194.8 (CdO), 171.5
(CdO), 169.6 (CdO), 136.2 (Carom), 134.9 (Carom), 129.3 (CHarom),
128.7 (CHarom), 61.2 (CH2), 53.0 (CH), 37.3 (CH2), 32.8 (CH2),
30.1 (CH3), 22.8 (CH3), 13.9 (CH3); LR-EIMS m/z (relative
intensity) 323 (M+, 15), 264 (M+ - NH2Ac, 100); HR-EIMS m/z
calcd for C16H21NO4S 323.1191 found 323.1194.
L-2-Acetylamino-3-{4-[(2,4-dimethoxybenzylsulfamoyl)methyl]-
phenyl}propionic Acid Ethyl Ester (10). Chlorine gas was slowly
bubbled through a cooled 0 °C solution of thiol acetate 7 (1.98 g,
6.12 mmol) and CH2Cl2/H2O (34 mL: 7 mL). Once a yellow color
appeared, the Cl2 was stopped and argon was bubbled through to
remove the excess Cl2. The mixture was poured into EtOAc, washed
with H2O and brine, and then dried over MgSO4. Removal of the
solvent by rotary evaporation yielded crude sulfonyl chloride that
was used in the next step without further purification. A solution
of dimethoxybenzylamine (1.53 g, 9.18 mmol) and NEt3 (929 mg,
9.18 mmol) in THF (40 mL + 10 mL) was added via cannula to
a 0 °C solution of sulfonyl chloride and THF (50 mL). The reaction
was allowed to warm to rt and stir overnight. The reaction was
quenched with NH4Cl, extracted with EtOAc (×3), washed with
NaHCO3 and brine, and dried over MgSO4. The solvent was
removed in vacuo and purified by flash chromatography (50:50 to
80:20 EtOAc/hexanes) yielded 1.36 g (77%) of 10 as a off-white
solid: mp 91-93 °C; 1H NMR (300 MHz) δ 7.16 (1H, d, J ) 8.0
Hz), 7.06 (4H, s), 6.47-6.44 (2H, m), 6.30 (1H, d, J ) 7.8 Hz),
5.09 (1H, t, J ) 6.1 Hz), 4.80 (1H, dd, J ) 13.9, 6.2 Hz), 4.18-
4.11 (4H, m), 4.02 (2H, s), 3.80 (3H, s), 3.76 (3H, s), 3.09 (1H,
dd, J ) 13.9, 6.0 Hz), 3.00 (1H, dd, J ) 13.9, 6.4 Hz), 1.93 (3H,
s), 1.23 (3H, t, J ) 7.1 Hz); 13C NMR (75 MHz) δ 171.4 (CdO),
169.7 (CdO), 160.9 (Carom), 158.4 (Carom), 136.4 (Carom), 130.6
(Carom), 130.3 (CHarom), 129.3 (CHarom), 127.9 (CHarom), 117.6
(Carom), 103.8 (CHarom), 98.5 (CHarom), 61.3 (CH2), 58.5 (CH2), 55.2
(24) Johnson, T. O.; Ermolieff, J.; Jirousek, M. R. Nature ReV. 2002, 1,
696.
(25) Wrobel, J.; Sredy, J.; Moxham, C.; Dietrich, A.; Li, Z.; Sawicki,
D. R.; Seestaller, L.; Wu, L.; Katz, A.; Sullivan, D.; Tio, C.; Zhang, Z.-Y.
J. Med. Chem. 1999, 42, 3199-3202.
(26) Recent studies have shown that enzyme inhibition can occur due to
a process of colloid formation (see refs 27 and 28). Dynamic light scattering
measurements by Lee at al (ref 8) on one of their Fmoc-Glu(OBn)-X-LNH2
tripeptides bearing a monoanionic pTyr mimic suggested that colloid
formation was not occurring. Although inhibition by colloid formation has
never been demonstrated to occur with PTP1B inhibitors of any kind, such
a phenomenon cannot be ruled out with the tripeptides examined here or
others examined by Lee et al.
(29) Kehoe, J. W.; Bertozzi, C. R. Chem. Biol. 2000, 7, 57-61.
(30) Farzan, M.; Mirzabekov, T.; Kolchinsky, P.; Wyatt, R.; Cayabyab,
M.; Gerard, N. P.; Gerard, C.; Sodroski, J.; Choe, H. Cell 1999, 96, 667.
(31) Farzan, M.; Vasilieva, N.; Schnitzler, C. E., Chung, S.; Robinson,
J.; Gerard, N. P.; Gerard, C.; Choe, H.; Sodroski, J. J. Biol. Chem. 2000,
275, 33516.
(27) McGovern, S. L.; Caselli, E.; Grigorieff, N.; Shoichet, B. K. J. Med.
Chem. 2002, 45, 1712.
(28) McGovern, S. L.; Shoichet, B. K. J. Med. Chem. 2003, 47, 1478.
(32) Liu, Y.; Ahmed, V.; Hill, B.; Taylor, S. D. Org. Biomol. Chem.
2005, 3, 3329-3335.
8194 J. Org. Chem., Vol. 71, No. 21, 2006

EValuatiion of pTyr Mimics for PTP1B Inhibition
(CH3), 55.1 (CH3), 53.0 (CH), 43.2 (CH2), 37.3 (CH2), 22.8 (CH3),
intensity) 560 (M + Na+, 100); HR-ESIMS calcud for C26H29F2-
NO7SNa (M + Na+) 560.1525, found 560.1518.
13.9 (CH3); LR-EIMS m/z (relative intensity) 326 (M+ - C9H12O2,
+
3), 249 (M+ - C9H12SO4N, 15); LR-CIMS (NH3) 496 (M + NH4
,
Bis(2,4-dimethoxybenzyl)difluoro(4-benzyl bromide)methane-
sulfonamide (16). To a cooled 0 °C solution of benzyl alcohol 15
(1.69 g, 3.14 mmol), PPh3 (1.23 g, 4.71 mmol), and CH2Cl2 (175
mL) was added CBr4 (1.56 g, 4.71 mmol) portionwise. The ice
bath was removed and the solution stirred overnight. Silica gel was
added to the mixture and the solvent removed in vacuo. Flash
chromatography (25:75 to 33:67 EtOAc/hexanes) afforded 1.80 g
of benzyl bromide 16 as a white solid (95%): mp 113-114 °C;
1H NMR (300 MHz) δ 7.64 (2H, d, J ) 8.2 Hz), 7.49 (2H, d, J )
8.2 Hz), 7.17 (2H, d, J ) 8.4 Hz), 6.38 (2H, dd, J ) 8.4, 2.3 Hz),
6.28 (2H, d, J ) 2.3 Hz), 4.50 (2H, s), 4.49 (4H, s), 3.77 (6H, s),
3.65 (6H, s); 13C NMR (75 MHz) δ 160.3 (Carom), 158.1 (Carom),
141.3 (Carom), 130.3 (CHarom), 129.4 (t, J ) 23.0 Hz, Carom), 129.0
(CHarom), 127.6 (t, J ) 5.9 Hz, CHarom), 122.0 (t, J ) 282.7 Hz,
CF2), 116.5 (Carom), 103.8 (CHarom), 97.6 (CHarom), 55.3 (CH3), 54.9
(CH3), 46.7 (CH2), 32.0 (CH2); 19F NMR (282 MHz) δ -100.9;
LR-ESIMS m/z (relative intensity) 624 (M + Na+, 44), 622 (M +
Na+, 44), 602 (M + H+, 29), 600 (M + H+, 29), 301 (100); HR-
ESIMS calcd for C26H29BrF2NO7S (M + H+) 600.0872, found
600.0869.
30); positive-ion HR-ESIMS calcd for C23H31N2O7S (M + H+)
479.1852, found 479.1842.
L-4-[Sulfonamido(methyl)]phenylalanine (3). A solution of
amino acid 10 (1.00 g, 2.09 mmol) in 6 N HCl (17.4 mL) was
refluxed overnight. The solution was cooled and the solvent
removed in vacuo. The resulting solid was dissolved in EtOH (85
mL), and propylene oxide (2 mL) was added. The reaction was
stirred overnight, which resulted in the precipitation of a white solid.
The solid was filtered to yield pure amino acid 540.6 mg (82%) of
amino acid 3 as a white solid. Spectral data matched literature data:
12 1H NMR (300 MHz, D2O) δ 3.00 (1H, dd, J ) 14.7, 7.7 Hz),
3.13 (1H, dd, J ) 14.7, 5.4 Hz), 3.87 (1H, overlapping dd, J ) 6.8
Hz), 4.34 (2H, s), 7.19 (2H, d, J ) 8.0 Hz), 7.29 (2H, d, J ) 8.0
Hz).
L-Fmoc-4-[sulfonamido(methyl)]phenylalanine (11). To a stirred
0 °C mixture of amino acid 3 (392 mg, 1.52 mmol), NaHCO3 (687
mg, 8.49 mmol), and H2O (15 mL) were added Fmoc-OSu (624
mg, 1.85 mmol) and dioxane (15 mL). The reaction was gradually
warmed to room temperature and stirred overnight. The mixture
was acidified with 1.2 N HCl, extracted with EtOAc (×4), washed
(brine), dried (MgSO4), filtered, and concentrated. Recrystallization
of the residue in EtOAc yielded 677 mg (93%) of compound 11.
Spectral data matched literature data:12 1H NMR (300 MHz, DMSO-
d6) δ 2.87 (1H, dd, J ) 14.1, 10.4 Hz), 3.08 (1H, dd, J ) 13.7, 4.5
Hz), 4.19 (6H, m), 6.82 (2H, s), 7.31 (6H, m), 7.39 (2H, m), 7.65
(2H, dd, J ) 7.2, 4.2 Hz), 7.70 (2H, d, J ) 8.4 Hz), 7.87 (2H, m,
J ) 7.5 Hz).
Bis(2,4-dimethoxybenzyl)difluoro(4-benzaldehyde)methane-
sulfonamide (14). To a stirred -78 °C solution of iodo compound
12 (3.49 g, 5.50 mmol) in THF (110 mL) was added 1.98 M n-BuLi
(3.05 mL, 6.05 mmol). After 5 min at -78 °C, DMF (1.28 mL,
16.5 mmol) was added. The reaction was stirred for 15 min and
then quenched with saturated NH4Cl. The aqueous layer was
extracted with EtOAc (×3), and the combined organics were dried
(MgSO4), filtered, and concentrated. Flash chromatography (15:
85 to 25:75 to 40:60 EtOAc/hexanes) yielded 2.28 g of aldehyde
14 as a white solid (77%): mp 141-142 °C; 1H NMR (300 MHz)
δ 10.10 (1H, s), 7.98 (2H, d, J ) 8.2 Hz), 7.84 (2H, d, J ) 8.2
Hz), 7.17 (2H, d, J ) 8.4 Hz), 6.38 (2H, dd, J ) 8.4, 2.4 Hz), 6.29
(2H, d, J ) 2.3 Hz), 4.51 (4H, s), 3.77 (6H, s), 3.66 (6H, s); 13C
NMR (75 MHz) δ 191.3 (CdO), 160.3 (Carom), 158.1 (Carom), 138.2
(Carom), 135.0 (t, J ) 22.8 Hz, Carom), 130.2 (CHarom), 129.4
(CHarom), 127.9 (t, J ) 5.9 Hz, CHarom), 121.6 (t, J ) 283.2 Hz,
CF2), 116.2 (Carom), 103.8 (CHarom), 97.6 (CHarom), 55.2 (CH3), 54.8
(CH3), 46.7 (CH2): 19F NMR (282 MHz) δ -101.5; LR-EIMS m/z
(relative intensity) 535 (M+, 11), 151 (100), 178 (69); HR-ESIMS
calculated for C26H27F2NO7S (M + Na+) 558.1368, found 558.1389.
Bis(2,4-dimethoxybenzyl)difluoro(4-benzyl alcohol)methane-
sulfonamide (15). To a stirred 0 °C solution of aldehyde 14 (1.58
g, 2.95 mmol) in THF (15 mL) and methanol (15 mL) was added
NaBH4 (111 mg, 2.95 mmol). After 1 h, the reaction was quenched
with saturated NH4Cl. The mixture was partitioned between CH2-
Cl2 and water. The aqueous layer was extracted with CH2Cl2 (×3),
and the combined organics were dried (MgSO4), filtered, and
concentrated. Flash chromatography of the residue (25:75 to 33:
67 to 40:60 to 50:50 EtOAc/hexanes) gave 1.42 g of alcohol 15 as
a white solid (90%): mp 123 °C; 1H NMR (300 MHz) δ 7.66 (2H,
d, J ) 8.1 Hz), 7.47 (2H, d, J ) 8.0 Hz), 7.18 (2H, d, J ) 8.4 Hz),
6.37 (2H, dd, J ) 8.4, 2.2 Hz), 6.27 (2H, d, J ) 2.2 Hz), 4.77 (2H,
s), 4.49 (4H, s), 3.77 (6H, s), 3.65 (6H, s); 13C NMR (75 MHz) δ
160.2 (Carom), 158.1 (Carom), 144.8 (Carom), 130.3 (CHarom), 128.3
(t, J ) 22.9 Hz, Carom), 127.3 (t, J ) 6.0 Hz, CHarom), 126.5
(CHarom), 122.3 (t, J ) 282.6 Hz, CF2), 116.6 (Carom), 103.8
(CHarom), 97.6 (CHarom), 64.4 (CH2), 55.3 (CH3), 54.9 (CH3), 46.7
(CH2); 19F NMR (282 MHz) δ -100.6; LR-ESIMS m/z (relative
3-(4-{[Bis(2,4-dimethoxybenzyl)sulfamoyl]difluoromethyl}-
benzyl)-2-oxo-5,6-diphenylmorpholine-4-carboxylic Acid Benzyl
Ester (17). To a stirred -78 °C solution of benzyl bromide 16
(1.80 g, 3.00 mmol) and William’s lactone (969 mg, 2.50 mmol)
in THF/HMPA (75 mL:7.5 mL) was added 0.96 M LiHMDS (2.7
mL, 2.62 mmol) dropwise. After 3 h at -78 °C, the reaction was
quenched with EtOAc. The layers were separated, and the EtOAc
layer was washed with water and brine. After drying (MgSO4),
filtration, and concentration, the crude lactone was purified by flash
chromatography (25:75 to 33:67 EtOAc/hexanes) to provide 1.82
g of lactone 17 as a foamy white solid (80%), mp 91-94 °C. Two
conformers were observed at room temperature in a 1:2 ratio: 1H
NMR (300 MHz) δ major comformer 7.68 (2H, d, J ) 8.0 Hz),
7.39-7.34 (2H, m), 7.21-7.03 (11H, m), 6.78 (2H, d, J ) 7.0
Hz), 6.59 (2H, d, J ) 7.3 Hz), 6.48 (2H, d, J ) 7.6 Hz), 6.37 (2H,
dd, J ) 8.4, 2.0 Hz), 6.27 (2H, d, J ) 1.9 Hz), 5.34-5.32 (1H,
m), 5.07 (1H, d, J ) 12.3 Hz), 4.96 (1H, d, J ) 12.3 Hz), 4.85
(1H, d, J ) 2.6 Hz), 4.51 (4H, s), 4.45 (1H, d, J ) 2.6 Hz) 3.74
1
(6H, s), 3.62 (6H, s), 3.48-3.41 (2H, m); H NMR (300 MHz,
DMSO-d6, 373K) δ 7.62-7.59 (2H, m), 7.49 (2H, bm), 7.35 (2H,
s), 7.25-7.14 (7H, m), 7.10-7.04 (4H, m), 6.98-6.96 (2H, m),
6.60-6.57 (2H, m), 6.42-6.41 (4H, m), 5.89 (1H, s), 5.27-5.26
(1H, bs), 5.16-5.12 (1H, m), 5.02 (2H, s), 4.36 (4H, s), 3.73 (6H,
s), 3.66 (6H, s), 3.59-3.52 (2H, m); 13C NMR (75 MHz) δ 168.9
(maj, CdO), 168.6 (min, CdO), 160.2 (Carom), 158.0 (Carom), 154.3
(maj, CdO), 153.8 (min, CdO), 140.1 (maj, Carom), 139.5 (min,
Carom), 135.6 (maj, Carom), 135.4 (Carom), 135.0 (min, Carom), 133.5
(min, Carom), 133.4 (maj, Carom), 130.2 (CHarom), 129.9 (CHarom),
129.7 (CHarom), 129.0 (t, J ) 23.1 Hz, Carom), 128.7 (CHarom), 128.4
(CHarom), 128.2 (CHarom), 127.8 (CHarom), 127.8 (CHarom), 127.4
(CHarom), 127.4 (CHarom), 126.6 (CHarom), 126.4 (CHarom), 122.1 (t,
J ) 282.5 Hz, CF2), 116.4 (CHarom), 103.7 (CHarom), 97.6 (CHarom),
79.0 (min, CH), 78.7 (maj, CH), 68.3 (min, CH), 67.7 (maj, CH),
60.1 (CH2), 58.8 (CH), 55.2 (CH3), 54.8 (CH3), 46.6 (CH2), 40.1
(min, CH2), 38.6 (maj, CH2); 19F NMR (282 MHz) δ -100.5 (ABQ,
J ) 260.4 Hz), -100.6 (s); LR-EIMS m/z (relative intensity) 906
(M+, 6), 151 (100), 91 (72), 178 (60); HR-ESIMS calcd for
C50H48F2N2O10SNa (M + Na+) 929.2889, found 929.2936.
3-(4-{[Bis-(2,4-dimethoxybenzyl)sulfamoyl]-difluoromethyl}-
phenyl)-2-(9H-fluoren-9-ylmethoxycarbonylamino)propionic Acid
Methyl Ester (18). To a stirred solution of lactone 17 (1.62 g,
1.79 mmol) and ethanol (14 mL) in dioxane (14 mL) was added
PdCl2 (158 mg, 0.893 mmol). This suspension was stirred under a
H2 atmosphere (balloon) for 6.5 h before the suspension was spun
down via centrifuge to remove the catalyst. The solvent was
removed and the crude amino acid titurated with Et2O (×3) to yield
a gummy white solid. The crude amino acid was dissolved in
J. Org. Chem, Vol. 71, No. 21, 2006 8195

Hill et al.
dioxane (12 mL), and a solution of Na2CO3 (568 mg, 5.36 mmol)
in water (14 mL) was added. The mixture was cooled to 0 °C, and
a solution of Fmoc-OSuc (632 mg, 1.88 mmol) and dioxane (2
mL) was added dropwise. The ice bath was removed and the
mixture stirred overnight. The mixture was acidified with 1 N HCl
and the layer separated. The product was extracted with EtOAc
(×5), and the combined organics were washed with brine, dried
(MgSO4), filtered, and concentrated. Flash chromatography (0 to
1 to 2 to 5 to 10% methanol in chloroform) yielded 1.4 g of Fmoc-
protected amino acid 18 as a white foam (99%): mp 79-80 °C;
1H NMR (300 MHz) δ 7.75-7.72 (2H, m), 7.58-7.53 (4H, m),
7.39-7.35 (2H, m), 7.30-7.21 (4H, m), 7.15 (2H, d, J ) 8.4 Hz),
6.34 (2H, d, J ) 8.4 Hz), 6.25 (2H, s), 5.35 (1H, d, J ) 7.3 Hz),
4.69 (1H, m), 4.41 (4H, s) overlapping, 4.36 (1H, m), 4.17 (1H,
m), 3.73 (6H, s), 3.62 (6H, s), 3.17 (2H, m); 13C NMR (75 MHz)
δ 174.9 (CdO), 160.4 (Carom), 158.1 (Carom), 155.9 (CdO), 143.8
(Carom), 141.4 (Carom), 139.9 (Carom), 130.5 (CHarom), 129.6 (CHarom),
128.7 (CHarom), 127.9 (t, J ) 20.9 Hz, Carom), 127.2 (CHarom), 125.2
(CHarom), 122.2 (t, J ) 282.7 Hz, CF2), 120.1 (CHarom), 116.7
(Carom), 104.0 (CHarom), 97.9 (CHarom), 67.2 (CH2), 55.4 (CH3), 55.0
(CH3), 54.5 (CH), 47.2 (CH), 46.8 (CH2), 37.7 (CH2); 19F NMR
(282 MHz) δ -100.6; LR-ESIMS m/z (relative intensity) 839 (M
+ Na+, 12), 834 (M + NH4+, 9), 855 (M + K+, 6); HR-ESIMS
calcd for C43H42N2O10F2SNa (M + Na+) 839.2426, found 839.2422.
General Protocol for Hexapeptide and Tripeptide Synthesis.
Solid-phase peptide synthesis was carried out manually using Fmoc
strategy employing Rink amide AMS resin (0.71 mmol/g). The resin
was preswollen in CH2Cl2 for 2 h (minimum), the Fmoc group
was removed using 20% piperidine/DMF (3 mL, 2 × 15 min), and
then the resin was washed with DMF (3 mL, 4 × 3 min), 1:1 DMF:
CH2Cl2 (3 mL, 1 × 3 min), and CH2Cl2 (3 mL, 1 × 3 min). All
Fmoc deprotections were carried out in this manner. Coupling of
the first amino acid was carried out with the pentafluorophenyl
ester of Fmoc-L-leucine (4.0 equiv) and HOBt (4.0 equiv) in DMF
(stirred 10 min at room temperature before being added to the resin)
and then shaken with resin (2 × 30 to 90 min). The resin was then
rinsed with DMF (6 × 3 min) and CH2Cl2 (2 × 3 min). All
remaining coupling reactions were carried out as follows. A solution
of Fmoc-amino acid (4.0 equiv), HATU or HBTU (4.0 equiv),
HABt or HOBt (4.0 equiv), DIEA (4.0 equiv), and DMF (3 mL)
was stirred at room temperature for 10 min. This solution was then
added to the resin and shaken for 30-90 min. Each coupling was
carried out twice, except for unnatural Fmoc-amino acids 9 and 16
which were coupled once overnight (2.0-4.0 equiv). The resin was
washed after the second coupling with DMF (3 mL, 6 × 3 min)
and CH2Cl2 (3 mL, 2 × 3 min). The side chains of both the Asp
and Glu were protected as tert-butyl esters for peptides 19 and 20
or benzyl ester for the Glu in the case of tripeptides 21-23 and
25. Once the peptide was assembled, the final Fmoc group was
removed and the resin dried overnight. The resin was placed in a
small round-bottom flask, and Reagent K (82.5% TFA, 2.5%
ethanedithiol, 5% water, 5% thioanisole, and 5% phenol) was added
(1.5 mL per 100 mg of resin) and stirred for 2.5 h at room
temperature. The mixture was filtered and the residue evaporated.
Hexapeptide Workup. To this mixture was added cold methyl
tert-butyl ether (0 °C) and the suspension transferred to a 35 mL
centrifuge tube. The mixture was centrifuged at 3500 rpm at 0 °C
for 10 min, and then the solvent was decanted off. This process
was repeated two more times affording a white crude peptide. The
peptide was dried overnight in the fumehood, then placed in water
and cooled and lyophilized yielding a fluffy white powder.
Tripeptide Workup. The peptides were precipitated with the
addition of diethyl ether or chloroform and filtered to give fluffy
white powders. Purification was accomplished by RP-HPLC or
silica gel column chromatography.
powder. Analytical HPLC (Vydac 218TP54 C18 column) using a
linear gradient of 90/10 H2O (0.025% TFA)-CH3CN to 70/30 H2O
(0.025% TFA)-CH3CN over 36 min showed a single peak
(retention time ) 14.0 min): positive-ion HRESMS m/z calcd for
C32H49N8O14S (M + H)+ 801.3089, found 801.3017.
DADE-4-LNH2 (20). Preparative HPLC of the precipitated
peptide (Vydac 218TP1022-C18 reversed-phase column) using a
linear gradient of 90/10 H2O (0.025% TFA)-CH3CN to 70/30 H2O
(0.025% TFA)-CH3CN over 36 min gave pure 20 as white powder.
Analytical HPLC (Vydac 218TP54 C18 column) using a linear
gradient of 90/10 H2O (0.025% TFA)-CH3CN to 70/30 H2O
(0.025% TFA)-CH3CN over 36 min showed a single peak
(retention time ) 21.3 min): negative-ion HRESMS m/z calcd for
C32H45F2N8O14S (M - H)- 835.2744, found 835.2701.
FmocGlu(OBn)-3-LNH2 (21). Purification by column chroma-
tography (0 to 1 to 2.5 to 5% MeOH/CHCl3) gave pure 21 as a
white powder. Analytical HPLC (Phenomenex Jupiter Proteo
reversed-phase column) using H2O (0.1% TFA)-CH3CN as eluent
(linear gradients of 60/40 to 25/75 from 0 to 30 min, 25/75 to 1/99
from 30 to 40 min, 1/99 from 40 to 55 min, retention time ) 18.5
min) indicated greater than 98.6% purity: positive-ion HRESMS
m/z calcd for C43H50N5O9S (M + H)+ 812.3298, found 812.3329.
FmocGlu(OBn)-4-LNH2 (22). Purified by column chromatog-
raphy (0 to 1 to 2.5 to 5% MeOH/CHCl3) gave pure 22 as a white
powder. Analytical HPLC (Phenomenex Jupiter Proteo reversed-
phase column) using H2O (0.1% TFA)-CH3CN as eluent (linear
gradients of 60/40 to 25/75 from 0 to 30 min, 25/75 to 1/99 from
30 to 40 min, 1/99 from 40 to 55 min, retention time ) 25.7 min)
indicated greater than 98.7% purity: positive-ion HRESMS m/z
calcd for C43H48F2N5O9S (M + H)+ 848.3126, found 848.3141.
FmocGlu(OBn)-2-LNH2 (23). Rink Resin (284 mg, 0.202
mmol) and Fmoc-L-4-[neopentylsulfono(difluoromethyl)phenyl-
alanine2 (237 mg, 0.403 mmol) were used. Purification by column
chromatography (0 to 1 to 2.5 to 5% MeOH/CHCl3) yielded 59.4
mg (17%) of tripeptide 23 with the sulfonate moiety protected with
a neopentyl group (white solid): positive-ion LR-ESIMS m/z
(relative intensity) 919 (M + H+, 100); positive-ion HR-ESIMS
calcd for C48H57F2N4O10S (M + H+) 919.3763, found 919.3765.
The purified neopentyl tripeptide (48.9 mg, 0.0532 mmol) was
dissolved in methyl ethyl ketone (4 mL) and LiBr (5 mg, 0.0585
mmol) and refluxed for 2 days. The solvent was removed in vacuo
and the residue dissolved in H2O. This solution was washed with
ether (×3), and the aqueous layer was then lyophilized to yield
55.2 mg of crude tripeptide 23. Purification by semipreparative
reversed-phase HPLC (Phenomenex Jupiter Proteo reversed-phase
column) using H2O (0.1% TFA)-CH3CN as eluent (linear gradients
of 50/50 to 30/70 from 0 to 65 min,30/70 to 1/99 from 65 to 75
min, retention time ) 42.7 min) gave 15 mg of pure 23. Analytical
HPLC (Phenomenex Jupiter Proteo reverse phase column) using
H2O (0.1% TFA)-CH3CN as eluent (linear gradients of 50/50 to
20/80 from 0 to 30 min, 20/80 to 1/99 from 30 to 40 min, 1/99
from 40 to 55 min, retention time ) 22.5 min) indicated greater
than98.8%purity: positiveionHRESMScalculatedforC43H47F2N4O10S
(M + H)+ 849.2971, found 849.2981.
FmocGlu(OBn)-Phe-LNH2 (25). Purification by silica gel
column chromatography (0 to 1 to 2.5 to 5% MeOH-CHCl3) gave
tripeptide 25 as a white powder. Analytical HPLC (Phenomenex
Jupiter Proteo reverse phase column) using H2O (0.1% TFA)-
CH3CN as eluent (linear gradients of 60/40 to 25/75 from 0 to 30
min, 25/75 to 1/99 from 30 to 40 min, 1/99 from 40 to 55 min,
retention time ) 28.2 min) indicated greater than 98.5% purity:
positive-ion HRESIMS calcd for C42H47N4O7 (M + H)+ 719.3406,
found 719.3445.
IC50 Determinations. Ten microliters of a solution of a stock
solution of inhibitor in DMSO-water was added to the wells of a
microtiter plate containing 70 µL of 0.1 M bis-tris, pH 6.3, 2 mM
EDTA, 5 mM DTT, 0.001% Triton. A control was prepared by
adding 10 µL of DMSO-water instead of inhibitor. To this, 10
µL of 200 µM diFMUP (Invitrogen) stock in 2% DMSO was added
DADE-3-LNH2 (19). Preparative HPLC of the precipitated
peptide (Vydac 218TP1022-C18 reversed-phase column) using a
linear gradient of 90/10 H2O (0.025% TFA)-CH3CN to 70/30 H2O
(0.025% TFA)-CH3CN over 36 min gave pure 19 as a white
8196 J. Org. Chem., Vol. 71, No. 21, 2006

EValuatiion of pTyr Mimics for PTP1B Inhibition
to bring the volume up to 90 µL. The assay was initiated by the
addition of 10 µL of PTP1B stored in 20 mM tris, pH 7.5, 150
mM NaCl, 5 mM DTT, 0.1 mM EDTA, and 20% glycerol. A
control was performed by adding 10 µL of enzyme storage buffer.
The final concentration of inhibitor in the assay wells was 100,
80, 60, 40, 20, 15, 10, 5, 2.5, 1, 0.5, and 0 µM. The final
concentration of DMSO was 7.5%. The final concentration of
diFMUP substrate in the assay was 20 µM, the previously
determined Km value under these conditions. The final concentration
of PTP1B ranged from 1.5 to 3.0 to 6.0 nM. The production of
fluorescent product diFMU was monitored for 10 min at 30 °C
using a spectrofluorometer platereader with excitation and emission
at 360 and 460 nm, respectively. The initial rates of enzyme activity
in relative fluorescence units per second (RFU/s) were used to
determine the IC50. The ratio of the initial rate in the presence of
inhibitor (Vi) to that in the absence of inhibitor (Vo) was calculated
and plotted as a semilog curve in Grafit, from which the IC50 value
was calculated based on the following equation: Vi ) Vo/[1 + ([I]/
IC50)S] + B, where Vi is the initial rate of reaction at an inhibitor
concentration of [I], Vo is the velocity in the absence of inhibitor,
B is background, and s is the slope factor equal to Vo - B.
Assay for Time-Dependent Inhibition with Inhibitor 23. The
IC50 of 23 was determined using the procedure described above
except the inhibitor solutions were incubated with PTP1B (3 nM)
for 30 min at 30 °C before initiation of the reaction with DiFMUP
(20 µM final concentration).
min of incubation, a 2 µL aliquot was withdrawn at t ) 15 s and
t ) 60 min and added to a 96-well microtiter plate containing 98
µL of 1.0 mM diFMUP (approximately 50 × Km) in 0.1 M bis-
tris, pH 6.3, 2 mM EDTA, 5 mM DTT, 0.001% Triton for a 50-
fold dilution of the incubation mixture. The final concentration of
the inhibitor in the assay was 1 mM, and the final concentration of
enzyme was 150 nM. The production of diFMU was followed for
10 min as described above. The percent of activity remaining in
the presence of inhibitor compared to that of the control was
determined. After the aliquot was withdrawn from the preincubation
mixture after 60 min, the remainder was transferred to a dialysis
membrane (10 kDa cutoff, SpectraPor) and dialyzed against 500
mL of 0.1 M bis-tris, pH 6.3, 2 mM EDTA, 5 mM DTT, 0.001%
Triton, changed twice over 24 h at 4 °C. At 24 h the amount of
activity remaining in the enzyme-inhibitor mixture and its control
was measured by withdrawing 2 µL into 98 µL of 1 mM diFMUP
(approximately 50 × Km) in 0.1 M bis-tris, pH 6.3, 2 mM EDTA,
5 mM DTT, 0.01% Triton as described above.
Acknowledgment. This work was supported by a Discovery
Grant and a Collaborative Research and Development Grant
from the Natural Sciences and Engineering Research Council
of Canada (NSERC) and Merck Frosst Canada and Co.
Supporting Information Available: 1H, 13C, and 19F NMR
(when applicable) for new compounds and HPLC chromatograms
for peptides 19-23 and 25. Experimental procedures for the
synthesis of dipeptides prepared for determining the enantiomeric
excess of compounds 11 and 18 and HPLC chromatograms and
19F NMR (when applicable) of the resulting dipeptides. This mate-
rial is available free of charge via the Internet at http://pubs.acs.org.
Assay for Irreversible Inhibition with Inhibitor 23. Ten
microliters of a 500 µM stock solution of 23 in DMSO-water was
added to 80 µL of 0.1 M bis-tris, pH 6.3, 2 mM EDTA, 5 mM
DTT, 0.001% Triton, and 10 µL of 1.5 µM PTP1B. A control was
similarly made with 10 µL of DMSO-water instead of inhibitor.
Upon initiation of the reaction at t ) 0 and after 20, 40, and 60
JO061496R
J. Org. Chem, Vol. 71, No. 21, 2006 8197