Efficient scaled-up synthesis of Nα-Fmoc-4-phosphono(difluoromethyl)- L-phenylalanine and its incorporation into peptides

Article abstract of DOI:10.1055/s-0030-1260206

A rapid, robust and efficient scaled-up synthetic strategy for N - Fmoc-4-phosphono(difluoromethyl)-l-phenylalanine and its direct incorporation into peptides is presented herein. Georg Thieme Verlag Stuttgart · New York.

Full text of DOI:10.1055/s-0030-1260206

PAPER
3255
Efficient Scaled-Up Synthesis of N-a-Fmoc-4-Phosphono(difluoromethyl)-L-
phenylalanine and Its Incorporation into Peptides
Christoph Meyer, Maja Köhn*
European Molecular Biology Laboratory, Meyerhofstr. 1, 69117 Heidelberg, Germany
Fax +49(6221)387518; E-mail: koehn@embl.de
Received 30 May 2011; revised 29 July 2011
fluorinating agent diethylaminosulfur trifluoride (DAST)
Abstract: A rapid, robust and efficient scaled-up synthetic strategy
or diazomethane are used, which are difficult to scale-
for N-a-Fmoc-4-phosphono(difluoromethyl)-L-phenylalanine and
up.9–13 In addition, these routes have only been described
its direct incorporation into peptides is presented herein.
on small scale (mg quantities in cases where the scale was
reported). Therefore, we sought to optimize and scale-up
the synthesis of N-Fmoc-F2Pmp.
Key words: phosphorylation, inhibitors, solid-phase synthesis,
peptides, bioorganic chemistry
OH
P
OH
P
Reversible phosphorylation of proteins is an essential
principle in cell signaling.1 The generation of phosphoty-
rosyl residues (pTyr) by protein tyrosine kinases (PTK)
creates recognition sites for the binding of modular do-
mains of other signaling proteins, such as Src homology 2
(SH2) and phosphotyrosine binding (PTB) domains.2 At
the same time, these sites become high affinity substrates
for their respective protein tyrosine phosphatases (PTP),
which cleave the phosphate from tyrosine residues.3 Per-
turbation of the concerted signaling events leads to the de-
velopment of various diseases including cancer and
diabetes.1,3 Therefore, the development of tools to study
and inhibit pTyr-mediated interactions is highly desirable.
Hydrolytically stable pTyr mimetics have proven to be
useful building blocks for the rational design of inhibitors
of these interactions.4,5 Among other pTyr analogues,
O
OH
O
OH
F
O
F
OH
OH
H2N
H2N
O
O
pTyr (1)
F2Pmp (2)
Figure 1 Comparison of phosphotyrosine (pTyr; 1) and its non-
hydrolyzable analogue phosphonodifluoromethyl-L-phenylalanine
(F2Pmp; 2)4–6
Due to its high coupling efficiency, we chose to follow the
approach of Qabar and co-workers,9 adapted from Burton
and co-workers,14 as starting point. This approach is based
on CuCl-promoted coupling of phosphono-difluorometh-
yl-CdBr (7) to commercially available N-Fmoc-L-4-io-
dophenylalanine (3; Scheme 1).
phosphonodifluoromethyl-L-phenylalanine
(F2Pmp),
which was developed by Burke and co-workers, has been
shown to be especially valuable for the development of
highly potent peptide-based inhibitors of PTPs
(Figure 1).4,5,6
In the first step of the synthesis, the carboxylic acid moi-
ety of 3 (4.8 g) was protected as an ester of 2-trimethylsi-
lylethanol (TMSE). This protecting group was chosen
because the ester bond can be removed under mild acidic
conditions. Thus, after coupling of the phosphonogroup,
the subsequent deprotection was expected to be easier to
control with respect to the integrity of the Fmoc group
than the previously described basic removal of a methyl
ester with LiOH.9 The esterification showed complete
conversion (TLC analysis) of the starting material after
four hours. Subsequent aqueous work-up and drying un-
der high vacuum gave a brown oil, which was used in the
next step without further purification.
For Fmoc solid-phase peptide synthesis, N-Fmoc-F2Pmp
(5) or its respective protected diethyl-phosphonate can be
used (Scheme 1).7,8 Furthermore, an on-resin direct con-
version of iodo-phenylalanine into F2Pmp(OEt2) has been
reported.9 However, the preferred method of F2Pmp-
incorporation into peptides uses the free phosphonate
building block due to side reactions caused by the harsh
conditions needed to remove the ethyl protecting
groups.8,9 Recently, N-Fmoc-F2Pmp (5) has become com-
mercially available, however, it is relatively expensive.
Several different synthetic routes to N-Fmoc-L-F2Pmp
have been described so far.4,7–13 Albeit very useful, these
routes have individual drawbacks in that they either require
several steps that need purification,8,10,11 or provide rela-
tively low overall yields.7,10–13 Often, the highly explosive
In the next reaction, the cadmium-containing compound
7, obtained by treatment of diethyl bromodifluorometh-
ylphosphonate (6) with Cd, was coupled to TMSE-pro-
tected 3 in the presence of a copper halide.9 Like Qabar
and co-workers,9 we also observed that an excess of cop-
per (2–3 equiv) was needed for the reaction. Additionally,
we found that the performance of CuBr was superior to
that of CuCl and that a second addition of the cadmium re-
SYNTHESIS 2011, No. 20, pp 3255–3260
x
x.
x
x
.
2
0
1
1
Advanced online publication: 06.09.2011
DOI: 10.1055/s-0030-1260206; Art ID: T55011SS
© Georg Thieme Verlag Stuttgart · New York

3256
C. Meyer, M. Köhn
PAPER
OEt
P
termined by analytical HPLC), resulting in an overall
yield of 54% based on the amount of starting material 3.
O
OEt
F
I
a) 2-TMSE, EDC, DMAP,
DMF, CH2Cl2, 0 °C,
4 h, full conv.
b) 7, CuBr, DMF, 19 h, r.t.
Unpurified N-Fmoc-F2Pmp (5) was then directly used in
solid-phase peptide synthesis. The coupling took place ef-
ficiently on rink amide resin using standard coupling re-
agents and only three equivalents of N-Fmoc-F2Pmp-OH
(Scheme 2). After cleavage from the resin, HPLC analysis
showed no fragments due to incomplete coupling, and
model peptides 8 (Scheme 2) and 9 (Figure 2) were ob-
tained in good yield after HPLC purification. Peptide 8 is
a protein tyrosine phosphatase 1B (PTP1B) inhibitor,6,15
and the sequence of 9 was chosen to incorporate non-polar
and polar non-charged amino acids in a longer peptide.
F
OH
FmocHN
60%
O
FmocHN
SiMe3
O
O
3
4
a) 25% TFA,
CH2Cl2,
r.t., overnight,
full conv.
OEt
P
F
F
OEt
89%
O
Br
b) (I) TMSBr,
2x overnight,
(II) 10% H2O–
MeCN, r.t., 2x 2h
O
6
OH
P
Cd, DMF, r.t., 3 h
Fmoc
Rink amide
OH
F
a) piperidine–DMF
b) Fmoc-Leu-OH (2 x), HBTU, HOBt, DIPEA, DMF
c) piperidine–DMF
d) Fmoc-F2Pmp-OH, TBTU, HOBt, DIPEA, DMF
F
OEt
F
F
P
OEt
OH
FmocHN
O
BrCd
OH
O
7
O
P
OH
F
5
F
Scheme 1 Synthesis of N-Fmoc-F2Pmp-OH (5); 2-(trimethylsi-
lyl)ethanol (2-TMSE), N-ethyl-N¢-(3-dimethylaminopropyl)carbodi-
imide (EDC), 4-(N,N-dimethylamino)pyridine (DMAP), N,N¢-
dimethylformamide (DMF), trifluoroacetic acid (TFA), trimethylsilyl
bromide (TMSBr)
O
H
N
Rink amide
FmocHN
N
H
O
agent and CuBr helped to drive the reaction faster to com-
pletion. After 19 hours, conversion into the product was
more than 93% (determined by analytical HPLC). After
aqueous work-up and column chromatography 3.8 g (60%
over two steps) of compound 4 was isolated. Minor
amounts of impurities were still detected after this purifi-
cation, two species in the 31P and one in the 1H NMR spec-
tra of compound 4, which correspond to an impurity in
starting material 6 and presumably to a side product de-
rived from compound 7. These impurities were not re-
moved because they did not interfere with subsequent
reactions of the final product N-Fmoc-F2Pmp, which was
to be incorporated into peptide- or peptide-mimic-based
binders of PTPs or PTBs. This is evident because the im-
purities in the 31P NMR are absent in the spectrum of the
corresponding peptides (see below). If necessary, howev-
er, it is possible to remove these impurities by HPLC pu-
rification.
e) piperidine–DMF
f) Fmoc-AA-OH, HBTU, HOBt, DMF
4 x
{Fmoc-Glu(Ot-Bu)-OH (2 x), Fmoc-Asp(Ot-Bu)-OH (2 x),
Fmoc-Ala-OH (2 x), Fmoc-Asp(Ot-Bu)-OH (2 x)}
capping/cleavage:
g) Ac2O–pyridine (1:9)
h) TFA, i-Pr3SiH
OH
P
O
OH
F
i) preparative HPLC
O
O
O
F
O
OH
H
O
OH
H
O
O
H
N
N
N
N
N
N
H
NH2
H
H
O
O
8
HO
O
42%
Scheme 2 Solid-phase synthesis of model peptide 8 containing
F2Pmp;
O-(1H-benzotriazol-1-yl)-N,N,N¢,N¢-tetramethyluronium
Subsequent hydrolysis of the trimethylsilylethyl ester
took place quantitatively overnight in 25% trifluoroacetic
acid (TFA) in dichloromethane. After coevaporation of
the acid with toluene and drying, the N-Fmoc-
F2Pmp(OEt2) building block was obtained and used di-
rectly in the next step without purification. The phospho-
nic acid diethyl esters were hydrolyzed quantitatively,
without the need to add scavengers, by treatment with tri-
methylsilyl bromide (TMSBr) in dichloromethane. A sin-
gle repetition of the treatment led to the highest purity of
the final product 5. Thus, N-Fmoc-F2Pmp (5; 2.6 g) was
isolated as a yellowish glass in more than 90% purity (de-
hexafluorophosphate (HBTU), 1-hydroxybenzotriazole (HOBt), N,N-
diisopropylethylamine (DIPEA)
In order to probe the two peptides for their biological ac-
tivity, we determined the IC50 values of both peptides to-
wards PTP1B, which is a phosphatase involved in
diabetes and obesity,16 with the well-established para-
nitrophenolphosphate (pNPP) competition assay.6 First,
the Michaelis–Menten constant (KM) of PTP1B towards
pNPP was measured (4.6 0.3 mM) to establish an appro-
priate pNPP concentration with which the IC50 values
were then determined. The resulting IC50 value
Synthesis 2011, No. 20, 3255–3260 © Thieme Stuttgart · New York

PAPER
Efficient Scaled-Up Synthesis of N-Fmoc-L-F2Pmp-OH
3257
OH
O
P
OH
F
F
OH
O
O
O
O
O
O
O
H
H
H
H
H
N
N
N
N
N
N
N
N
NH2
H
H
H
H
H
O
O
O
O
OH
9
Figure 2 F2Pmp-containing model peptide 9
(1.7 0.1 mM) of the known peptide 8,6,15 which contains
several acidic residues, was found to be approximately
ten-fold better than our model peptide 9 (15.3 1.9 mM),
which is composed of only neutral, non-charged amino
acids around F2Pmp. These results are in agreement with
the known preference of PTP1B for acidic amino acids in
close vicinity to the pTyr of the substrates.16 In addition,
this confirms that, although F2Pmp is essential for effi-
cient binding of a peptide to PTP1B, the surrounding
amino acids can effectively modulate the strength of the
interaction and cause differences of up to at least one or-
der of magnitude in inhibitory potency.
a pump rate of 1.5 mL/min. For preparative separations, a Macherey
Nagel C18 VP 250/10 NUCLEODUR 110–5 C18 ec column with a
pump rate of 5 mL/min was used. The solvent mixtures were either
H2O–MeCN or H2O–MeOH, if needed 0.05% TFA was added.
Mass spectra were recorded with a MALDI micro MX mass spec-
trometer (Waters, Manchester, UK) equipped with a reflectron ana-
lyzer, used in positive ion mode with delayed extraction activated.
HRMS were recorded with a LTQ Obritrap Velos (ThermoFisher
Scientific, US). TLC analyses were conducted with Merck precoat-
ed silica gel plates (Merck, 60 F254) using UV light (254 nm) or a
solution of ceric ammonium molybdate (prepared by addition of 40
mL conc. H2SO4, 360 mL H2O, 10 g ammonium molybdate and 4 g
ceric ammonium sulfate). Preparative column chromatography was
performed using silica gel from Merck (silica 60; grain size 0.063–
0.200 mm; 70–230 mesh ASTM).
In summary, we present here a fast, robust, and efficient
method to synthesize N-Fmoc-F2Pmp in very good overall
yield. The synthesis, which requires only one purification
step, can be easily applied to obtain the final product in
gram scale and in greater than 90% purity. Furthermore,
the application of the unpurified building block in solid-
phase peptide synthesis to generate two model peptides
was demonstrated, and the biological activity of these
peptides was shown. Rapid and simple access to N-Fmoc-
F2Pmp will enable broad use of this building block to gen-
erate inhibitors and tools for protein tyrosine phosphatase
and pTyr-binding domain research.
(S)-2-(Trimethylsilyl)ethyl 2-[(9H-Fluoren-9-yloxy)carbonyl-
amino]-3-{4-[(diethoxyphosphoryl)difluoromethyl]phenyl}pro-
panoate (4)
In a mixture of anhydrous CH2Cl2 (20 mL) and anhydrous DMF (30
mL), N-Fmoc-L-4-iodophenylalanine (3; 4.80 g, 9.35 mmol) was
dissolved under an argon atmosphere. The solution was cooled in an
ice bath, then TMSE (2.50 mL, 25.6 mmol, 2.7 equiv), DMAP (914
mg, 7.48 mmol, 0.8 equiv), and EDC (3.58 g, 23.1 mmol, 2.5 equiv)
were added. The reaction was monitored by TLC. After four hours,
full conversion of the starting material was observed. CH2Cl2 (200
mL) was added to the reaction and the mixture was extracted with
aq NH4Cl (2 × 250 mL) and once with brine (250 mL). The organic
phase was dried (Na2SO4), the solvent was removed under reduced
pressure, and the resulting brown oil [Rf = 0.35 (cyclohexane–
EtOAc, 6:1)] was used in the next reaction without further purifica-
tion.
All chemicals and anhydrous solvents were obtained from commer-
cial sources (Sigma–Aldrich, VWR) and used without further puri-
fication.
The
chiral
starting
material
N-Fmoc-L-4-
For the next step, cadmium metal (6.00 g, 53.4 mmol, purum p.a. for
filling reductors, Sigma–Aldrich) was washed by stirring under ar-
gon with 1 N HCl (5 mL, 15 min), H2O (5 mL, 3 × 1 min) and ace-
tone (5 mL, 3 × 1 min). The metal was dried overnight under high
vacuum while stirring until a metallic shine was observed. While
warming the reaction vessel by hand, anhydrous DMF (10 mL) and
diethyl bromodifluoromethylphosphonate (12.5 mL, 31.2 mmol)
were added dropwise over 15 min to the metal under argon. The
slightly exothermic reaction proceeded and was allowed to stir at r.t.
for 3 h.
iodophenylalanine was purchased from Amatek Chemical Co., Ltd.,
Zhangjiagang, China. Diethyl bromodifluoromethylphosphonate
was obtained from Matrix Scientific, Columbia, USA. Fmoc-pro-
tected amino acids and Rink amide resin (200–400 mesh; 0.62 or
0.57 mmol/g) were purchased from Novabiochem, Darmstadt, Ger-
many. Peptide synthesis was performed with an automatic peptide
1
synthesizer Syro I from Multisyntech, Witten, Germany. H, 13C,
and 31P NMR spectra were recorded with a 400 MHz Bruker
Avance DPX spectrometer. Chemical shifts (d) are reported in ppm
1
and coupling constants (J) are given in Hz. H and 13C chemical
shifts were referenced to the solvent peaks (d = 7.26 and 77.0 ppm
for CDCl3). Splitting patterns are designated as follows: s = singlet,
d = doublet, t = triplet, q = quartet, m = multiplet, dd = doublet of
doublet. 13C and 31P NMR spectra were broadband hydrogen decou-
pled. For 1H assignment, COSY and HMQC spectra were recorded.
HPLC analyses and purifications were carried out with a Shimadzu
HPLC-MS LCMS-2010EV instrument fitted with a UV/Vis Photo-
diode array detector (SPD-M20A Prominence). For analytical and
preparative HPLC, the solvent delivery module LC-20AD was
used. RP-HPLC analytical runs were carried out with a Macherey
Nagel C18 EC 250/4.0 NUCLEODUR 100–5 C18 ec column with
TMSE-protected 3 was dissolved in anhydrous DMF (10 mL), then
CuBr (2.70 g, 18.8 mmol, 2 equiv) and half of the supernatant solu-
tion of the Cd reagent (ca. 1.7 equiv) was added in a dropwise man-
ner. After 3 h, further CuBr (1.40 g, 9.41 mmol, 1 equiv) and the
other half of the Cd reagent solution were added. The reaction was
allowed to stir at r.t. for 19 h in total. The progress of the reaction
was checked by HPLC (Nucleodur C18; MeOH–H2O, 85%; 15
min), the AUC of the peak of the starting material (Rt = 13.8 min)
was 6.5% compared to the area of the product peak (Rt = 7.5 min).
The reaction mixture was diluted with EtOAc (250 mL), filtered
through Celite, and extracted with aq NH4Cl (2 × 250 mL) and brine
Synthesis 2011, No. 20, 3255–3260 © Thieme Stuttgart · New York

3258
C. Meyer, M. Köhn
PAPER
(250 mL). After drying (Na2SO4), the solvents were removed in
vacuo and product 4 was isolated by flash chromatography (cyclo-
hexane–EtOAc, 6:1→2:1).
tion was stirred for 2 h at r.t., and the solvents were removed by
evaporation (2 ×). The product 5 was obtained in >90% purity (es-
timated by HPLC) and used without further purification in solid-
phase peptide synthesis.
Yield: 3.80 g (5.65 mmol, 60% yield over two steps); transparent
yellowish oil; Rf = 0.62 (cyclohexane–EtOAc, 2:1).
Yield: 2.60 g (5.03 mmol, 89% yield over two steps); yellowish
glass; tR = 7.6 min (RP-HPLC, Nucleodur C18, MeOH–H2O, 30 →
70% in 15 min).
1H NMR (400 MHz, CDCl3): d = 7.77 (d, 3J = 7.3 Hz, 2 H, 2 × Ar-
H-Fmoc), 7.63–7.51 (m, 4 H, 2 × Ar-H-Fmoc, 2 × Ar-H-Phe), 7.40
(t, 3J = 7.5 Hz, 2 H, 2 × Ar-H-Fmoc), 7.32 (t, 3J = 7.4 Hz, 2 H, 2 ×
1H NMR (400 MHz, CDCl3, drops of CD3OD): d = 7.76 (d,
3
Ar-H-Fmoc), 7.20 (d, J = 7.7 Hz, 2 H, 2 × Ar-H-Phe), 5.26 (d,
3J = 7.5 Hz, 2 H, 2 × Ar-H-Fmoc), 7.57 (t, 3J = 7.1 Hz, 2 H, 2 × Ar-
3J = 8.0 Hz, 1 H, NH-Phe), 4.70–4.60 (m, 1 H, a-H-Phe), 4.50–4.34
(m, 2 H, CH2-Fmoc), 4.34–4.07 (m, 7 H, CH-Fmoc, 2 × CH2-Ethyl,
H-Fmoc), 7.52 (d, J = 7.8 Hz, 2 H, 2 × Ar-H-Phe), 7.40 (t, J =
3
3
7.4 Hz, 2 H, 2 × Ar-H-Fmoc), 7.30 (t, 3J = 6.9 Hz, 2 H, 2 × Ar-H-
3
3
2
CH2-TMSE), 3.27–3.08 (m, 2 H, CH2-Phe), 1.30 (t, J = 7.1 Hz,
Fmoc), 7.13 (d, J = 7.8 Hz, 2 H, 2 × Ar-H-Phe), 4.59 (dd, J =
5.8 Hz, 3J = 8.4 Hz, 1 H, a-H-Phe), 4.41 (dd, 2J = 6.7 Hz, 3J = 10.5,
1 H, CHa-Fmoc), 4.32 (dd, 2J = 6.7 Hz, 3J = 10.6, 1 H, CHb-Fmoc),
4.18 (dd, 2J = 6.3 Hz, 3J = 13.8 Hz, 1 H, CH-Fmoc), 3.18–3.04 (m,
2 H, CH2-Phe).
6 H, 2 × CH3-Ethyl), 0.97 (t, 3J = 8.3 Hz, 2 H, CH2-TMSE), 0.04 (s,
9 H, 3 × CH3-TMSE).
13C NMR (100 MHz, CDCl3): d = 171.2 (O=C-TMSE ester), 155.5
(O=C-Fmoc), 143.8 (C-F2P), 143.7, 141.3, 139.0, 131.4 (6 × Ar-C),
129.5, 127.7, 127.1, 126.5, 125.1, 125.0, 120.0 (12 × Ar-CH), 66.9
(CH2-Fmoc), 64.7 (CH2-TMSE), 64.1 (2 × CH2-Ethyl), 54.7 (a-CH-
Phe), 47.2 (CH-Fmoc), 38.0 (CH2-Phe), 17.4 (CH2-TMSE), 16.3 (2
× CH3-Ethyl), –1.6 (3 × CH3 TMSE).
31P (126 MHz, CDCl3): d = 6.4 (t, J = 115 Hz).
HRMS (ESI): m/z [M + Na]+ calcd for C34H42F2NO7PSiNa:
13C NMR (100 MHz, CDCl3, drops of CD3OD): d = 173.3 (O=C-
carboxy), 155.8 (O=C-Fmoc), 143.7 (C-F2P), 143.5, 141.2, 138.5,
131.9 (6 × Ar-C), 129.3, 127.7, 127.0, 126.3, 124.9, 124.8, 120.0
(12 × Ar-C), 66.8 (CH2-Fmoc), 54.6 (a-CH-Phe), 47.0 (CH-Fmoc),
37.8 (CH2-Phe).
31P (126 MHz, CDCl3, drops of CD3OD): d = 6.1 (t, J = 116 Hz).
HRMS (ESI): m/z [M + H]+ calcd for C25H23F2NO7P: 518.11747;
696.23284; found: 696.23318.
found: 518.11778.
(S)-2-{[(9H-Fluoren-9-yl)methoxy]carbonylamino}-3-{4-[(di-
ethoxyphosphoryl)difluoromethyl]phenyl}propanoic Acid [N-
Fmoc-F2Pmp(OEt2)]
Ac-Asp-Ala-Asp-Glu-F2Pmp-Leu-NH2 (8)
As solid support for the synthesis of hexapeptide 8, Fmoc protected
Rink amide resin (40 mg, 200–400 mesh, loading: 0.62 mmol/g)
was chosen. The synthesis was performed with a Syro I automated
peptide synthesizer, Fmoc deprotection was achieved by treatment
with 40% piperidine in DMF (3 min) and 20% piperidine in DMF
(12 min). Peptide couplings were carried out by reaction with
Fmoc-protected amino acids (5 equiv), HBTU (5 equiv), HOBt (5
equiv), and DIPEA (10 equiv) in DMF for 40 min. After cleavage
of the first Fmoc group, Fmoc-Leu-OH was coupled to the resin in
a double coupling, and subsequently Fmoc deprotected. Next, unpu-
rified Fmoc-F2Pmp-OH (39 mg, 75 mol, 3 equiv) was coupled in
DMF (1 mL) by manual addition of TBTU (23.8 mg, 75 mmol, 3
equiv), HOBt (10.25 mg, 75 mmol, 3 equiv), and DIPEA (43.6 mL,
150 mmol, 6 equiv) for 3 h. Fmoc deprotection was performed auto-
matically and the next four amino acids Fmoc-Glu(O-t-Bu)-OH,
Fmoc-Asp(O-t-Bu)-OH, Fmoc-Ala-OH, and Fmoc-Asp(O-t-Bu)-
OH, were double coupled and Fmoc deprotected according to the
standard protocol. The terminal amino group was acetylated by ad-
dition of acetic anhydride in pyridine (1:9, 800 mL). After thorough
washing with DMF (3 × 1 min), CH2Cl2 (3 × 1 min) and drying in
high vacuum, the peptide was cleaved and side chain deprotected by
treatment with TFA (950 mL) and triisopropylsilane (50 mL) for
2.5 h. The supernatant solution was filtered into ice-cold Et2O (20
mL) and the colorless precipitate was centrifuged and washed with
cold Et2O (2 × 20 mL). The crude product was dissolved in MeCN–
H2O (10%) and purified by preparative RP-HPLC (Nucleodur C18;
MeCN–H2O, 10 → 75% + 0.05% TFA in 25 min). The completely
deprotected peptide 8 was obtained as a colorless lyophilizate.
Compound 4 (3.80 g, 5.65 mmol) was dissolved in anhydrous
CH2Cl2 (22.5 mL), then TFA (7.5 mL, 25%) was added and the mix-
ture was stirred overnight (TLC showed complete conversion with-
out the formation of side products). The solvents were removed in
vacuo, and the remaining oil was dissolved in toluene (40 mL),
again evaporated (2 ×) and dried under high vacuum. The resulting
transparent brownish oil was used in the next reaction without fur-
ther purification.
Rf = 0.85 (cyclohexane–EtOAc, 1:1).
1H NMR (400 MHz, CDCl3): d = 7.76 (d, 3J = 7.6 Hz, 2 H, 2 × Ar-
H-Fmoc), 7.56 (d, 3J = 7.7 Hz, 2 H, 2 × Ar-H-Fmoc), 7.51 (d, 3J =
3
7.7 Hz, 2 H, 2 × Ar-H-Phe), 7.40 (t, J = 7.4 Hz, 2 H, 2 × Ar-H-
Fmoc), 7.30 (t, 3J = 7.5 Hz, 2 H, 2 × Ar-H-Fmoc), 7.22 (d,
3
3J = 7.8 Hz, 2 H, 2 × Ar-H-Phe), 5.49 (d, J = 7.8 Hz, 1 H, NH-
Phe), 4.74–4.65 (m, 1 H, a-H-Phe), 4.50–4.43 (m, 1 H, CH2-Fmoc),
4.40–4.34 (m, 1 H, CH2-Fmoc), 4.33–4.27 (m, 1 H, CH-Fmoc),
4.23–4.09 (m, 4 H, 2 × CH2-Ethyl), 3.30–3.09 (m, 2 H, CH2-Phe),
1.33–1.23 (m, 6 H, 2 × CH3-Ethyl).
13C NMR (100 MHz, CDCl3): d = 173.2 (O=C-carboxy), 155.9
(O=C-Fmoc), 143.9 (C-F2P), 143.7, 141.4, 139.3, 131.0 (6 × Ar-C),
129.8, 127.9, 127.2, 126.5, 125.2, 125.1, 120.1 (12 × Ar-CH), 67.1
(CH2-Fmoc), 65.4 (2 × CH2-Ethyl), 54.5 (a-CH-Phe), 47.3 (CH-
Fmoc), 37.7 (CH2-Phe), 16.3 (2 × CH3-Ethyl).
31P (126 MHz, CDCl3): d = 6.1 (t, J = 122 Hz).
HRMS (ESI): m/z [M + H]+ calcd for C29H31F2NO7P: 574.18007;
found: 574.18034.
Yield: 9.30 mg (42%, 10.6 nmol); RP-HPLC: tR = 6.4 min (Nucle-
odur C18; 215 nm; MeCN–H2O, 10 → 75% + 0.05% TFA in
15 min).
(S)-2-{[(9H-Fluoren-9-yl)methoxy]carbonylamino}-3-{4-[di-
fluoro(phosphono)methyl]phenyl}propanoic Acid (N-Fmoc-
F2Pmp; 5)
The material obtained in the previous step was dissolved in anhy-
drous CH2Cl2 (30 mL) and cooled in an ice bath. Under argon, tri-
methylsilyl bromide (12.4 mL, 94.0 mmol, 16.6 equiv) was added
dropwise. The solution was stirred and allowed to warm to r.t. over-
night. The solvents were removed in vacuo and the procedure was
repeated once more. Then, MeCN–H2O (10%) was added, the solu-
1H NMR (400 MHz, CD3OD): d = 8.55 (m, NH), 8.17 (d, J =
3
7.0 Hz, NH), 8.04 (d, 3J = 7.0 Hz, NH), 7.91 (d, 3J = 7.0 Hz, 2NH),
7.81 (d, 3J = 8.3 Hz, NH), 7.52 (d, 3J = 7.5 Hz, 2 H), 7.38 (d, J =
3
7.4 Hz, 2 H), 4.76–4.69 (m, 1 H), 4.65–4.51 (m, 2 H), 4.37–4.29
(m, 1 H), 4.26–4.17 (m, 2 H), 3.24–3.05 (m, 2 H), 2.97–2.72 (m,
4 H), 2.37–2.24 (m, 2 H), 2.10–1.87 (m, 5 H), 1.72–1.48 (m, 3 H),
1.41 (d, 3J = 7.4 Hz, 3 H), 0.93 (d, 3J = 6.7 Hz, 3 H), 0.87 (d, 3J =
5.7 Hz, 3 H).
Synthesis 2011, No. 20, 3255–3260 © Thieme Stuttgart · New York

PAPER
Efficient Scaled-Up Synthesis of N-Fmoc-L-F2Pmp-OH
3259
13C NMR (100 MHz, CD3OD): d = 179.4, 176.1, 175.1, 174.3,
173.7, 173.1, 172.7, 172.4, 172.1, 171.6, 151.2, 139.4, 133, 128.9,
126.3, 55.3, 53.8, 51.4, 51.0, 50.0, 40.2, 36.6, 35.3, 34.5, 29.8, 26.0,
24.3, 22.2, 21.1, 20.3, 15.8.
Michaelis–Menten Constant Measurement of PTP1B for pNPP
In a 96 well plate, pNPP was added in eight different concentra-
tions. After addition of PTP1B (40 nM) or H2O (blank) to reach a
final volume of 100 mL, absorbance at 405 nm was measured every
60 s over 1 h on a Tecan safire2 (Tecan, Salzburg, Austria) plate
reader at 37 °C. The measurements were performed in duplicate.
The buffer conditions were pH 7.2, 4-(2-hydroxyethyl)-1-pipera-
zine-ethanesulfonic acid (HEPES; 25 mM), EDTA (2.5 mM), DTT
(2.0 mM), NaCl (124.4 mM). After subtracting the blank, the initial
reactions rates were obtained by linear regression of the data points
up to 10 min, and the reaction rates were plotted over the substrate
concentration. The Michaelis–Menten constant was determined by
fitting the data points to the Michaelis–Menten equation using
Sigma plot (Systat Software Inc., San Jose, USA).
31P (126 MHz, CD3OD): d = 5.1 (t, J = 112 Hz).
MS (MALDI-TOF; CHCA): m/z [M
+
Na]+ calcd for
C34H48F2N7NaO16P: 902.3; found: 902.5; m/z [M + K]+ calcd for
C34H48F2KN7O16P: 918.4; found: 918.4.
Ac-Ala-Ser-Gly-Ala-F2Pmp-Ala-Gly-Gly-Ser-Ala-NH2 (9)
The decapeptide 9 was synthesized on Fmoc protected Rink amide
resin (40 mg, 200–400 mesh, loading: 0.57 mmol/g). The synthesis
was performed with a Syro I automated peptide synthesizer, Fmoc
deprotection was achieved by the standard protocol. After Fmoc
deprotection of the solid support, Fmoc-Ala-OH, Fmoc-Ser(O-t-
Bu)-OH, Fmoc-Gly-OH (2 ×) and Fmoc-Ala-OH were double-cou-
pled and Fmoc deprotected under standard coupling conditions. Un-
purified Fmoc-F2Pmp-OH (39 mg, 75 mmol, 3.3 equiv) was coupled
in DMF (1 mL) by manual addition using TBTU (23.8 mg, 75 mmol,
3.3 equiv), HOBt (10.25 mg, 75 mmol, 3.3 equiv) and DIPEA (43.6
mL, 150 mmol, 6.6 equiv) for 3 h. Fmoc deprotection and coupling
of the subsequent amino acids Fmoc-Ala-OH, Fmoc-Gly-OH,
Fmoc-Ser(O-t-Bu)-OH and Fmoc-Ala-OH were performed auto-
mated in the described manner. The terminal amino group was
acetylated by addition of acetic anhydride in pyridine (1:9, 800 mL).
After thorough washing with DMF (3 × 1 min), CH2Cl2 (3 × 1 min)
and drying in high vacuum, the peptide was cleaved and side-chain
deprotected by treatment with TFA (950 mL) and triisopropylsilane
(50 mL) for 2.5 h. The supernatant solution was filtered into ice-cold
Et2O (20 mL). The colorless precipitate was centrifuged and
washed with cold Et2O (2 × 20 mL). The crude product was dis-
solved in MeCN–H2O (5%) and purified by preparative RP-HPLC
(Nucleodur C18; MeCN–H2O, 5 → 10% + 0.05% TFA in 25 min).
The completely deprotected peptide 9 was obtained as a colorless
lyophilizate.
PTP1B Inhibition Assay
In a final reaction volume of 100 mL, PTP1B (40 nM) was incubat-
ed with decreasing concentrations of the inhibitory peptides for
30 min at 37 °C. The buffer conditions were pH 7.2, 4-(2-hydroxy-
ethyl)-1-piperazine-ethanesulfonic acid (HEPES; 25 mM), EDTA
(2.5 mM), DTT (2.0 mM), NaCl (124.4 mM). After the incubation
time, pNPP (20 mL) was added to reach a final concentration of
10 mM. Using a Tecan safire2 (Tecan, Salzburg, Austria) plate read-
er, the absorption at 405 nm was measured every 30 s over 15 min.
All measurements were performed in triplicate. The slopes of the
initial phases were obtained by linear regression of the data points
up to 5 min. The slopes were plotted versus the log of the inhibitor
concentrations, and the IC50 values were obtained by fitting the
curves using the one site competition model of Sigma plot (Systat
Software Inc., San Jose, USA).
Supporting Information for this article is available online at
http://www.thieme-connect.com/ejournals/toc/synthesis.
Acknowledgment
Yield: 8.20 mg (8.49 nmol, 37%); RP-HPLC: tR = 10.0 min (Nucle-
odur C18; 215 nm; MeCN–H2O, 5 → 10% + 0.05% TFA in 15
min).
M.K. acknowledges financial support from the German Science
Foundation (DFG) within the Emmy-Noether program. The authors
are grateful to Joanna Kirkpatrick and Jeroen Krijgsveld of the
EMBL Proteomics Core Facility for their support in recording the
HRMS.
1H NMR (400 MHz, D2O): d = 7.45 (d, 3J = 7.6 Hz, 2 H), 7.25 (d,
3
3J = 7.6 Hz, 2 H), 4.54 (t, J = 7.0 Hz, 1 H), 4.37–4.31 (m, 2 H),
4.23–4.16 (m, 3 H), 4.16–4.10 (m, 1 H), 3.90 (s, 2 H), 3.84 (s, 4 H),
3.81–3.73 (m, 4 H), 3.17–3.08 (m, 1 H), 3.02–2.94 (m, 1 H), 1.92
(s, 3 H), 1.31–1.23 (m, 9 H), 1.41 (d, 3J = 7.1 Hz, 3 H).
13C NMR (100 MHz, D2O): d = 180.4, 176.8, 175.6, 175.2, 174.8,
174.3, 173.3, 172.9, 171.7, 171.6, 171.3, 156.8, 138.5, 129.2, 128.5,
126.1, 61.0, 55.6, 54.4, 49.8, 49.6, 43.1, 42.5, 40.0, 21.6, 16.6, 16.5,
16.3.
References
(1) Blume-Jensen, P.; Hunter, T. Nature 2001, 411, 355.
(2) Yaffe, M. B. Nat. Rev. Mol. Cell Biol. 2002, 3, 177.
(3) Tonks, N. K. Nat. Rev. Mol. Cell Biol. 2006, 7, 833.
(4) Burke, T. R. Jr. Curr. Top. Med. Chem. 2006, 6, 1465.
(5) Bialy, L.; Waldmann, H. Angew. Chem. Int. Ed. 2005, 44,
3814.
(6) Akamatsu, M.; Roller, P. P.; Chen, L.; Zhang, Z.-Y.; Ye, B.;
Burke, T. R. Jr. Bioorg. Med. Chem. 1997, 5, 157.
(7) Shen, K.; Keng, Y.-F.; Wu, L.; Guo, X.-L.; Lawrence, D. S.;
Zhang, Z.-Y. J. Biol. Chem. 2001, 276, 47311.
(8) Gordeev, M. F.; Patel, D. V.; Barker, P. L.; Gordon, E. M.
Tetrahedron Lett. 1994, 35, 7585.
31P (126 MHz, D2O): d = 4.2 (t, J = 105 Hz).
MS (MALDI-TOF; CHCA): m/z [M
+
H]+ calcd for
C36H55F2N11O16P: 966.4; found 966.4; m/z [M + Na]+ calcd for
C36H54F2N11NaO16P: 988.3; found 988.4; m/z [M + K]+ calcd for
C36H54F2KN11O16P: 1004.4; found 1004.3.
Expression and Purification of Recombinant PTP1B
The T7–7 vector containing the DNA sequence encoding for His-
tagged PTP1B (1–321) was transformed into Escherichia coli strain
BL21 DE3 using standard methods. The cells expressing the recom-
binant His-PTP1B were lysed by sonication in lysis buffer A
[20 mM Tris-HCl, pH 8.0 containing 150 mM NaCl, 10 mM imida-
zole, 1 mM dithiothreitol (DTT) and 0.5 mM phenylmethylsulfonyl
fluoride (PMSF)]. The protein was purified using a FPLC Histrap
HP 1 mL column using an elution gradient of 10–500 mM imida-
zole gradient in buffer A. The integrity of the protein was confirmed
by ESI-MS analysis.
(9) Qabar, M. N.; Urban, J.; Kahn, M. Tetrahedron 1997, 53,
11171.
(10) Wrobel, J.; Dietrich, A. Tetrahedron Lett. 1993, 34, 3543.
(11) Burke, T. R. Jr.; Smyth, M. S.; Otaka, A.; Roller, P. P.
Tetrahedron Lett. 1993, 34, 4125.
(12) Solas, D.; Hale, R. L.; Patel, D. V. J. Org. Chem. 1996, 61,
1537.
(13) Smyth, M. S.; Burke, T. R. Jr. Tetrahedron Lett. 1994, 35,
551.
Synthesis 2011, No. 20, 3255–3260 © Thieme Stuttgart · New York

3260
C. Meyer, M. Köhn
PAPER
(14) Qiu, W.; Burton, D. J. Tetrahedron Lett. 1996, 37, 2745.
(15) Burke, T. R. Jr.; Kole, H. K.; Roller, P. P. Biochem. Biophys.
Res. Commun. 1994, 204, 129.
(16) Salmeen, A.; Andersen, J. N.; Myers, M. P.; Tonks, N. K.;
Barford, D. Mol. Cell 2000, 6, 1401.
Synthesis 2011, No. 20, 3255–3260 © Thieme Stuttgart · New York