The use of peptiCLEC-TR in the preparation of dipeptides

Article abstract of DOI:10.1021/op0000430

Dipeptides are important intermediates in many pharmaceutical products. To support a research programme on matrix metalloproteinases (MMP) we needed to prepare the dipeptide, Leu-Phe-NHMe (3). Chemical methods provided the material but not to the quality

Full text of DOI:10.1021/op0000430

Organic Process Research & Development 2000, 4, 563−566
Technical Notes
The Use of PeptiCLEC-TR in the Preparation of Dipeptides
Susan J. Faulconbridge,*,† Karen E. Holt,‡ Christopher J. Lock,† Stephen J. C. Taylor,‡ and Martin Woods†,§
Celltech-Chiroscience Ltd., U.K., and Chirotech Technology Ltd, Cambridge Science Park, Milton Road,
Cambridge, CB4 0WG, U.K.
Scheme 1
Abstract:
Dipeptides are important intermediates in many pharmaceutical
products. To support a research programme on matrix me-
talloproteinases (MMP) we needed to prepare the dipeptide,
Leu-Phe-NHMe (3). Chemical methods provided the material
but not to the quality required, and thus an alternative
preparation was sought. The use of enzymes in chemical
synthesis was a tool that we had used before but not in the
preparation of dipeptides. This contribution will show the utility
of PeptiCLEC-TR in the preparation of dipeptide (3) and its
application to other dipeptides. It will highlight the amenability
of PeptiCLEC-TR to scale-up and demonstrate the recycling
of this enzyme, which will make this a cost-effective route.
Initially dipeptide (3) was prepared by chemical coupling
of CBZ-leucine and amino amide (2) in 10 vol of dichlo-
romethane at 10-20 °C using 1.1 equiv of 1-[3-dimethyl-
amino)propyl]-3-ethylcarbodiimine hydrochloride (EDAC).
The dipeptide product was filtered off directly from the
reaction mixture and a second crop isolated by stripping the
liquors. In the chemically mediated coupling there is little
discrimination of amino acid impurities present in the
feedstock other than that obtained during isolation by
crystallisation. A particular issue in the synthesis of Z-Leu-
Phe-NHMe was the impurity Z-Ile-Phe-NHMe which results
from the presence of isoleucine in the leucine feedstock. This
gives rise to a significant impurity which is impossible to
remove from the final product by crystallisation. In addition
a small but significant amount of racemisation of the leucine
occurs during the coupling giving rise to ∼0.1% of the RS,SR
diastereoisomers after recrystallisation. In view of this,
alternative methods were required.
Although the use of enzymes in organic synthesis is a
well-established area,6 application of lipases as amidation
catalysts has received much less attention.7 At the outset of
this work the use of enzymes in the preparation of dipeptides
was not widespread, particularly on a large scale. There were
some notable exceptions to this, which included the industrial
production of aspartame.8 More recently the use of cross-
linked enzyme crystals (CLECs) of thermolysin9 and sub-
Introduction
The matrix metalloproteinases (MMPs) are a large and
expanding family of zinc-dependent endopeptidases that have
attracted much attention as targets for drug discovery over
the past decade.1 The discovery that the MMP enzymes are
overexpressed in many pathological conditions has led to
the belief that inhibitors of the MMPs could be useful for
the treatment of a range of inflammatory disorders for which
there is an unmet therapeutic need. From the literature it can
be seen that the amino-carboxylate families of inhibitors all
contain dipeptides of the form (1) as a common motif2
(Scheme 1).
To support a research program dipeptides of the form (1)
were prepared on a gram to kilogram scale3-5 (Scheme 2).
* Author for correspondence. Sue Faulconbridge, Celltech-Chiroscience Ltd.,
Granta Park, Great Abington, Cambridge, CB1 6GS, U.K. Telephone: 01223
896300. Fax: 01223 896400. E-mail: suefaulconbridge@chiroscience.com.
† Celltech-Chiroscience Ltd.
‡ Chirotech Technology Ltd.
§ Current address. Glaxo-Wellcome Research & Development, Glaxo Wellcome
Medicines Research centre, Gunnels Wood Road, Stevenage, Hertfordshire, SG1
2NY, U.K.
(6) Margolin, A. L. Enzyme Microb. Technol. 1993, 15, 266.
(7) Zaks, A.; Klibanov, A. M. Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 3192.
Quiros, M.; Sanchez, V. M.; Brieva, R.; Rebolledo, F.; Gotor, V.
Tetrahedron: Asymmetry 1993, 4, 1105. Garcia, M. J.; Rebolledo, F.; Gotor,
V. Tetrahedron: Asymmetry, 1993, 4, 2199. Gotor, V.; Menendez, E.;
Mouloungui, Z.; Gaset, A. J. Chem. Soc., Perkin Trans. 1 1993, 2453.
Adamczyk, M.; Grote, J. Tetrahedron Lett. 1996, 37, 7913. Sanchez, V.
M.; Francisca, F.; Gotor, V. Tetrahedron: Asymmetry 1997, 8, 37.
Adamczyk, M.; Grote, J. Tetrahedron: Asymmetry 1997, 8, 2099.
(8) Oyama, K. The Industrial Production of Aspartame. In Chirality in Industry;
Collins, A. N., Sheldrake, G. N., Crosby, J., Eds., Wiley: Chichester, 1992;
p 237.
(1) Davidson, A. H.; Drummond, A. H.; Galloway, W. A.; Whittaker, M. Chem.
Ind. (London), 1997, 258.
(2) Xue, C.; He, X.; Roderick, J.; DeGrado, W. F.; Decicco, C.; Copeland, R.
A. Bioorg. Med. Chem. Lett. 1996, 6, 379.
(3) Baxter, A. D.; Bhogal, R.; Bird, J.; Buckley, G. M.; Gregory, D. S.; Hedger,
P. C.; Manallack, D. T.; Massil, T.; Minton, K. J.; Montana, J.; Neidle, S.;
Owen, D. A.; Watson, R. J. Bioorg. Med. Chem. Lett. 1997, 7, 2765.
(4) Baxter, A. D.; Bird, J.; Bhogal, R.; Massil, T.; Minton, K. J.; Montana, J.;
Owen, D. A. Bioorg. Med. Chem. Lett. 1997, 7, 897.
(5) Hunter, D. J.; Bird, J.; Cassidy, F.; De Mello, R. C.; Harper, G. P.; Karran,
E. H.; Markwell, R. E.; Miles-Williams, A. J.; Ward, R. W. Bioorg. Med.
Chem. Lett. 1994, 4, 2833.
(9) Persichetti, R. A.; St. Clair, N. L.; Griffith, J. P.; Navia, M. A.; Margolin,
A. L. J. Am. Chem. Soc. 1995, 117, 2732.
10.1021/op0000430 CCC: $19.00 © 2000 American Chemical Society and The Royal Society of Chemistry
Published on Web 08/10/2000
Vol. 4, No. 6, 2000 / Organic Process Research & Development
•
563

Scheme 2
tilisin10 has been recommended for the formation of di- and
polypeptides and peptide alkylamides. CLECs offer the
advantages of increased product purity and potential for
enzyme recycling reducing the cost of the process and
minimising waste.
Enzyme Preparation of Z-Leu-Phe-NHMe (3). Thus,
having reviewed the literature, we investigated the use of
the thermolysin CLEC (PeptiCLEC-TR) for the synthesis of
Z-Leu-Phe-NHMe (3). The initial conditions focused on the
use of 50% aqueous ethanol as solvent with dry PeptiCLEC-
TR, which furnished a 68% yield of isolated product 3 after
24 h. Disappointingly, recycle of the enzyme gave only a
34% yield of 3. The use of the enzyme in the commercially
available slurry form gave much better results; 87% yield
both on first run and first recycle after a total of 24 h.
Changing the solvent to 40% aqueous ethanol was even more
encouraging and the results are summarised in the table.
Although some deactivation of the enzyme was apparent a
total of 255 g of 3 was produced using 3 g of PeptiCLEC-
TR over five runs (Table 1).
The influence of solvent was then examined especially
the use of ethyl acetate, the extraction solvent used in the
preparation of 2. Gratifyingly, it was found that there was
no significant loss in the reaction rate. This had the benefit
that the enzyme charge could be maintained at 1% w/w,
could still achieve 95% conversion in 7.5 h, and could use
(2) without the need to isolate the solid. On a larger scale
(10) Wang, Y.-F.; Yakovlevsky, K.; Margolin, A. L. Tetrahedron Lett. 1996,
37, 5317; Wang, Y.-F.; Yakovlevesky, K.; Zhang, B. and Margolin, A. L.
J. Org. Chem. 1997, 62, 3488.
564
•
Vol. 4, No. 6, 2000 / Organic Process Research & Development

Table 1. Showing yield and % purity of product when
recycling 3 g PeptiCLEC TR over five runs
(Scheme 2). The HPLC conversion was 93%, and the modest
isolated yield was attributed to the workup in which some
of the water-soluble dipeptide was lost during the aqueous
washes.
Z-Leu-tert-Leu-NHMe (11). To test the reaction out with
respect to steric constraints the formation of the dipeptide
(10) from tert-leucine N-methylamide (Scheme 2) was
examined. Although product was formed only 10-20%
conversion was reached after 3 days!
run
yield of (3)/%
L.C. purity/%
reaction time/h
1
2
3
4
5
83
88
88
91
91
94
94
92
88
94
6.5
14
19
20
20
a Conditions: 1wt %/wt PeptiCLEC-TR, 20 g/L Z-leu, 36 g/L amine, 40 °C.
Conclusions
with increased substrate loading a 91% yield of 96% pure 3
was obtained and demonstrated three recycles without
significant loss of activity. The economic benefits of this
are clearly apparent. The increased substrate loading was
achieved by way of multiple additions of Z-leucine and
N-methylamide to prevent the mixture becoming too thick
to stir.
An added advantage of the enzyme-coupling procedure
was that the product was essentially free of any isoleucine
contaminant. It is presumed that the low levels result from
the fact that isoleucine is more sterically encumbered than
leucine and hence undergoes coupling at a slower rate than
leucine.
A comparison of Boc-leucine with Z-leucine in the
coupling reaction (Scheme 2) was examined. The former
reagent had a number of potential advantages over the latter.
First, Boc-leucine is a crystalline solid and may be purified
by crystallisation, whereas the Z-leucine is a viscous oil.
Second the resultant Boc dipeptide (5) can be converted to
the desired dipeptide (1) by simple acidolysis. In the event
a 77% yield of isolated 5 was obtained after 70 h using
PeptiCLEC-TR. Thus, the reaction of Boc-leucine appears
to be about 2-3 times slower when compared to that of
Z-leucine.
In summary, the utility of PeptiCLEC-TR in the prepara-
tion of diastereomerically enriched dipeptides has been
shown. This method has limitations in that the more sterically
hindered the components, the slower the reaction, but
diastereomeric integrity is maintained. When compared with
traditional chemical methods this is a simple means of
ensuring chemical and diasteriomeric purity. This method
also has the advantage of recycling of the enzyme, thus
reducing costs and minimising waste.
Experimental Section
General Procedure for Dipeptide Formation. Protected
leucine (1 mol equiv) is dissolved in 3:2 0.2 M sodium
acetate, 50 mM calcium acetate solution, pH 6.5:ethyl alcohol
at a concentration of 20 g/L. Amine (2 mol equiv) is added
and the mixture warmed to 40 °C. Enzyme (3.5 mg/mL) is
added and the reaction shaken vigorously at 40 °C for the
required length of time with regular analysis by HPLC.
HPLC analysis method: C18 250 × 4.6 mm i.d.
Hypersil Shandon HPLC column, 1 mL/min, 210 nm. 50:
50 methanol:10 mM phosphate solution pH 3.1 hold for 5
min, gradient to 100% methanol at 20 min, then back to
original conditions over 2 min.
A further advantage of using an enzyme for this coupling
is the fact that the procedure is racemisation free without
the need for further purification as seen for the chemical
method. Indeed the highly stereoselective nature of the
reaction would allow one to use racemic 4 in the coupling
and only produce the desired (S,S)-dipeptide (5). In this way,
particularly during early-stage research, where material
supply to a specified quality standard may override cost
issues, the need for developing chiral assays could be avoided
and instead a de assay (HPLC or NMR) utilised to measure
the quality of the dipeptide (5).
Preparation of Boc-Leu-Phe-NHMe (5): as general
procedure, 77% yield. H NMR (CDCl3, 200 MHz) δ 0.9
1
(6H, 2xCH3), 1.4 (9H, m, tBu), 1.7 (3H, m, CH2, CH), 2.75
(3H, d, NHMe), 3.1 (2H, dd, CH2Ph), 4.0 (1H, m, CH), 4.7
(2H, m, NH, CH), 6.45 (1H, m, NH), 6.6 (1H, m, NH), 7.3
(5H, s, Ph).
Preparation of Z-Leu-2-Pyridylalanine-NHMe (7): as
general procedure, new peak at 14.8 min using HPLC method
above.
Preparation of Z-Leu-Taz-NHMe (9): as general pro-
cedure, 66% yield. 1H NMR (CDCl3, 200 MHz) δ 0.8 (6H,
d, 2xCH3), 1.5 (3H, m, CH2, CH), 2.5 (3H, d, NHMe), 2.8
(1H, dq, CH2), 4.1(1H, m, CH), 4.8 (1H, m, CH), 5.0 (2H,
dd, CH2Ph), 5.8 (2H, dd, CH2Ph), 6.9 (1H, d, NH), 7.05
(1H, s, CH), 7.2 (5H, s, Ph).
Preparation of Z-Leu-tert-Leu-NHMe (11): as general
procedure, new peak at 16.4 min using HPLC method above.
Preparation of Z-leucine-phenylalanine-N-methyl-
amide (3). Enzyme preparation. PeptiCLEC TR (3.75 mL)
in storage solution was filtered until no surface water could
be seen and then re-filtered with 2 × 3.75 mL absolute
ethanol until a paste.
To investigate the generality of PeptiCLEC-TR for the
synthesis of leucine dipeptides, a range of amino acids were
tested in the coupling.
Z-Leu-2-Pyridylalanine-NHMe (7). A possible replace-
ment for phenylalanine is 2-pyridylalanine. CLEC coupling
of both racemic 6 and 82% ee 6 gave rise to a single new
peak in the HPLC assay (Scheme 2). Furthermore, with the
racemic material 6 an ee assay of the recovered starting
material confirmed that it was being resolved (50% ee after
4 h). This example highlights the specificity of the enzyme
to only couple in one enantiomer.
Z-Leu-Taz-NHMe (9). The TAZ-amide (8) was coupled
to Z-leucine to give the dipeptide (9) in 66% isolated yield
A mixture of Z-leucine (37.5 g, 0.14 mol) and PheNHMe
amide (27.4 g, 0.15 mol) in ethyl acetate (500 mL) and 10
Vol. 4, No. 6, 2000 / Organic Process Research & Development
•
565

mM calcium acetate solution (15 mL) was warmed to 50
°C; an insoluble salt formed, but this dissolved after about
45 min. The enzyme prepared as above was added in a small
amount of reaction mixture. After 1 h a pre-warmed mixture
of Z-leucine (20 g, 0.07 mol) and PheNHMe (14.77 g, 0.08
mol) in ethyl acetate (4 mL) was added to bring the
concentration to 100 g/L. Again, the insoluble salt was
present, but this had dissolved after 2 h, and thus a second
addition of the above quantities of Z-leucine and PheNHMe
was performed. The mixture was then stirred overnight at
50 °C.
At completion methanol (410 mL) was added and the
mixture warmed to 50 °C, the enzyme filtered off (was stored
in 10mM calcium acetate solution for future use). The
reaction mixture was cooled to 15 °C, and the product was
filtered off and then re-dissolved in methanol (1.25 L). The
enzyme was then filtered off and discarded before cooling
to recrystallise the product for isolation, giving a 63% yield
first crop and a further 15% yield as second crop on
concentration of the mother liquor. H NMR (DMSO, 400
MHz) δ 0.8 (6H, dd, 2 × CH3), 1.35 (2H, dq, CH2), 1.5
(1H, m, CH), 2.5 (3H, d, NHMe), 2.9 (2H, dq, CH2Ph), 4.0
(1H, m, CH), 4.4 (1H, m, CH), 5.0 (2H, dd, CH2Ph), 7.25
(10H, m, 2 × Ph), 7.4 (1H, d, NH), 7.9 (2H, m, 2 × NH).
1
Acknowledgment
We thank Dr. Peter Tiffin for assistance with the writing
of this paper, Mr. Phil Gilbert for obtaining NMR spectra
and Mrs. Christel Kronig for the development of the
diastereomeric and chiral assays used.
Received for review April 17, 2000.
OP0000430
566
•
Vol. 4, No. 6, 2000 / Organic Process Research & Development