Synthesis of orthogonally protected S,S-2,6-diaminopimelic acid via olefin cross-metathesis

Article abstract of DOI:10.1080/00397910600616701

A short synthesis of orthogonally protected S,S-diaminopimelic acid, DAP, via a selective cross-metathesis between S-allyl glycine and S-vinyl glycine followed by hydrogenation is reported. By analogy, the S,S-diaminosuberic acid, DAS, is produced by self

Full text of DOI:10.1080/00397910600616701

Synthetic Communicationsw, 36: 1707–1713, 2006
Copyright # Taylor & Francis Group, LLC
ISSN 0039-7911 print/1532-2432 online
DOI: 10.1080/00397910600616701
Synthesis of Orthogonally Protected
S, S-2,6-Diaminopimelic Acid via
Olefin Cross-Metathesis
Ernest G. Nolen, Catherine J. Fedorka, and Brooke Blicher
Department of Chemistry, Colgate University,
Hamilton, New York, USA
Abstract: A short synthesis of orthogonally protected S,S-diaminopimelic acid, DAP,
via a selective cross-metathesis between S-allyl glycine and S-vinyl glycine followed
by hydrogenation is reported. By analogy, the S,S-diaminosuberic acid, DAS, is
produced by self-metathesis of S-allyl glycine, and the self-metathesis of S-vinyl
glycine is described.
Keywords: Cross-metathesis, diaminopimelic acid, Grubbs catalyst
The preparation of diamino dicarboxylic acids or bis(amino acids) has drawn
considerable attention because of their ability to covalently link peptides.
Shown in Fig. 1 are three bis(amino acids) with four, three, and two interven-
ing methylenes. 2,7-Diaminosuberic acid (DAS) 1 represents the oxidatively
stable carbon analogue of cystine and has demonstrated utility as the
linchpin of two cyclic nonapeptides with hematoregulatory and oxytocin
antagonistic activities.[1] S,S-2,6-Diaminopimelic acid (DAP) 2 is an inter-
mediate in the biosynthesis of L-lysine via meso-DAP, which is a
component of the peptidoglycan of most bacteria.[2] The truncated 2,5-diami-
noadipic acid (DAA) 3 is also of interest as a mimic of these biochemically
important diamino dicarboxylic acids.
These compounds have been prepared by nonstereoselective means[3] and
several stereoselective methods. For example, by taking advantage of the
Received in the USA December 12, 2005
Address correspondence to Ernest G. Nolen, Department of Chemistry, Colgate
University, 13 Oak Drive, Hamilton, NY 13346, USA. E-mail: enolen@mail.
colgate.edu
1707

1708
E. G. Nolen, C. J. Fedorka, and B. Blicher
Figure 1. Diamino dicarboxylic acids.
resident chirality of natural amino acids, Kolbe electrolysis of protected glu-
tamates and aspartates produces low yields of DAS and DAA in only one
step,[4] but this procedure is limited to an even number of linking methylenes
and symmetrically protected bis(amino acids). In contrast, de novo formation
of the a-carbon stereochemistry of diamino dicarboxylic acids has been
achieved by asymmetric hydrogenations of the enamide following Wittig ole-
fination[5] or following Suzuki coupling.[6] Also, stereoselective glycinyl
enolate alkylations[7] and epoxide opening of azides[8] have been reported
to yield various bis(amino acid) derivatives.
Ruthenium-catalyzed metathesis reactions have been employed to furnish
diamino dicarboxylic acid products. The first report entailed a ring-closing
metathesis (RCM) of allyl glycines using the first-generation Grubbs
ruthenium catalyst to form DAS-linked cyclic peptides.[9] Other RCM
reactions have been utilized to form DAS peptides[10] and protected DAS
derivatives.[11] A ring-closing alkyne metathesis for symmetrically protected
DAS derivatives has also been described,[12] and recently a RCM reaction
for the preparation of orthogonally protected meso-DAP was published.[13]
To date, there has been only one report of a direct self-metathesis reaction
to produce diamino dicarboxylic acids,[14] and surprisingly no cross-
metathesis reactions have been exploited to form the S,S-diaminopimelic
acid, DAP.
During our synthetic studies of carbon-linked glycosyl amino acids by a
cross-metathesis methodology,[15] we encountered the ready production of the
self-metathesis by-products from the starting S-allyl glycine 4 and S-vinyl
glycine 5 precursors (Fig. 2). Here, we report the use of the allyl glycine
and vinyl glycine self-metathesis reaction to yield DAS and DAA derivatives,
respectively, and the successful cross-metathesis between allyl and vinyl
glycines for DAP analogues.
Because allyl glycine 4 is considered a type 1, highly reactive, metathesis
partner according to criteria suggested by Grubbs,[16] it was expected that its
self-metathesis product would be isolated upon refluxing in dichloromethane
in the presence of 15 mol% of the N-hetereocyclic Grubbs second-generation
catalyst. We obtained a 71% yield of the symmetrical product 6.
Although not as reactive to the metathesis conditions as allyl glycine,
the bis(trimethylsilylethyl) S-vinyl glycine 5 in the presence of the second-
generation ruthenium catalyst affords self-metathesis product 7 in 62%

S,S-2,6-Diaminopimelic Acid
1709
Figure 2. Self-metatheses and cross-metathesis reactions.
yield. This is in sharp contrast to results using (Cy3P)2Cl2Ru ¼ CHPh as
catalyst, in which no self-metathesis was observed, and a recent report by
Schmittmann et al.[14b]
In view of this reactivity pattern, it seemed likely that a selective cross-
metathesis may be achieved between the more reactive allyl glycine and the
less reactive vinyl glycine. To this end, a 2 : 1 ratio of vinyl to allyl glycine
was subjected to the cross-metathesis conditions, and a 69% yield of 8,
based on 4, was collected after chromatography. The E-isomer was dis-
tinguishable by the 15.4-Hz coupling constant between the vinyl protons at
postitions 3 and 4. This reaction also produced the readily separable
(DRf . 0.1), less polar, vinyl glycine self-metathesis product 7 in 39%
yield based on 5.
Hydrogenation of each of the metathesis adducts 6, 7, and 8 should lead to
DAS, DAA, and DAP, respectively. As shown in Table 1, hydrogenation was
successful for the production of the symmetrically protected S,S-DAS 9 and
for the orthogonally protected S,S-DAP 10. Production of 10 required the
use of platinum on alumina to prevent hydrogenolysis of the Cbz-protecting
group.
Attempts to hydrogenate the alkene of compound 7 led to mixtures. The
rate of hydrogenation is impeded by the proximity of the protecting groups,
and alkene isomerization to the a,b-unsaturated ester, presumably due to
metal coordination, is observed; therefore, any hydrogenated product
affords an uncontrolled mixture of diastereomers. Diimide reduction of 7
using tosylhydrazide and sodium acetate at 708C also gave a thermal isomer-
ization, whereas attempted diimide reductions at room temperature failed to
reduce the alkene in our hands.[17]

1710
E. G. Nolen, C. J. Fedorka, and B. Blicher
Table 1. Hydrogenation of olefin metathesis adducts
Catalyst/
Yield
(%)
Alkene
reagent
Product
6
Pd/C
96
8
Pt/Al2O3
92
In summary, the self-metathesis reactions of S-allyl and S-vinyl glycines
produced alkenes in good yield. Taking advantage of the differential reactivity
of these glycine derivatives led to a moderately selective cross-metathesis
reaction using the second-generation Grubbs ruthenium catalyst. The cross-
metathesis adduct 8 can be hydrogenated to yield an orthogonally protected
S,S-diaminopimelic acid 10. In principle, the meso-diaminopimelic acid
can be produced through hydrogenation of the cross-metathesis product
from R-allyl glycine[18] and S-vinyl glycine.
EXPERIMENTAL
(S)-Allyl glycine 4 and (S)-vinyl glycine 5 were prepared according to the lit-
erature procedures.[14] All NMR spectral assignments were determined on a
1
Bruker Avance spectrometer by H (400 MHz) and 13C (100 MHz) attached
proton tests, Correlation Spectroscopy (COSY), and Heteronuclear Multiple
Quantum Correlation Spectroscopy (HMQC) two-dimensional techniques in
CDCl3. Mass spectra were recorded on a Bruker Daltonics Autoflex
MALDI-ToF using a-cyano-4-hydroxycinnamic acid as the matrix. Spectral
data for compound 6 (4b and 13) and hydrogenation product 9 (8) are in
agreement with literature values.
Dimethyl (2S, 7S)-Bis-(tert-butoxycarbonyl)amino-4-octenedioate 6. A
solution of S-allyl glycine 4 (105 mg, 0.458 mmol) in dichloromethane (6 mL)
was bubbled with Ar for 5 min, and then cannulated into a flask containing
15 mol% of the Grubbs II catalyst (29 mg, 0.034 mmol). The dark solution
was refluxed overnight (16 h). The reaction mixture was concentrated onto
SiO2 followed by flash column chromatography using 20% EtOAc/hexanes
(Rf ¼ 0.47 in 30% EtOAc/hex) to yield a light brown oil, 90mg (71% yield).
1H NMR: d 5.42 (bs, 2H, H-4 and H-5), 5.13 (bd, J ¼ 7.7 Hz, 2H, 2NH), 4.35
(m, 2H, H-2 and H-7), 3.75 (s, 6H), 2.47 (m, 4H, H-3 and H-6), 1.45 (s, 18H).

S,S-2,6-Diaminopimelic Acid
1711
Bis(2-trimethylsilylethyl) (2S, 5S)-2,5-bis[(phenylmethoxycarbonyl)
Expeamino]-3-hexenedioate 7. A solution of S-vinyl glycine 5 (150 mg,
0.427 mmol) in dichloromethane (3 mL) was bubbled with Ar for 5 min, and
then cannulated into a flask containing the Grubbs II catalyst (36 mg,
0.043 mmol). The dark solution was refluxed overnight (18 h). The reaction
mixture was concentrated onto SiO2 followed by flash column chromato-
graphy using 10–20% EtOAc/hexanes (Rf ¼ 0.30 in 20% EtOAc/hex) to
yield a light brown oil, 93 mg (62% yield). 1H NMR: d 7.4 (m, 10H,
2C6H5), 5.87 (bs, 2H, H-3 and H-4), 5.42 (bs, 2H, 2NH), 5.14 (s, 4H,
2PhCH2), 4.95 (bd, J ¼ 7.2 Hz, 2H, H-2 and H-5), 4.25 (bdt, J ¼ 8.8 and
12.1 Hz, 4H), 1.03 (bt, J ¼ 8.8 Hz, 4H), 0.07 (s, 18H). 13C NMR d 170.1,
155.4, 136.1, 128.5, 128.2, 128.1, 127.8, 67.2, 64.6, 55.1 (C-2 and C-5),
17.3, 21.5 ppm. MALDI-ToF MS: m/z ¼ 665.3 (Mþ þ Na), 681.2
(Mþ þ K). HRMS calcd. for C32H46N2O8Si2Na 665.2690; found 665.2698.
1-Trimethylsilylethyl
amino-6-(tert-butoxycarbonyl)amino-3-heptenedioate 8.
7-methyl (2S,6S)-E-2-(phenylmethoxycarbonyl)
solution of
A
S-allyl glycine 4 (100 mg, 0.436 mmol) and S-vinyl glycine 5 (292 mg,
0.872mmol) in dichloromethane (6 mL) was bubbled with Ar for 5 min, and
then cannulated into a flask containing the Grubbs II catalyst (74mg,
0.087mmol). The dark solution was refluxed overnight (18h). The reaction
mixture was concentrated onto SiO2 followed by flash column chromatography
using 10–20% EtOAc/hexanes to yield a light brown oil, 161 mg (69% yield).
Rf ¼ 0.19 (20% EtOAc/hex). 1H NMR: d 7.3 (m, 5H, C6H5), 5.64 (dt, J ¼ 8.1
and 15.4Hz, 1H, H-4), 5.54 (dd, J ¼ 5.7 and 15.4Hz, 1H, H-3), 5.44 (bd,
J ¼ 7.3Hz, 1H, NH-2), 5.06 (s, 2H, PhCH2), 5.05 (overlapped, 1H, NH-6),
4.77 (bt, J ¼ 5.6Hz, 1H, H-2), 4.32 (m, 1H, H-6), 4.19 (bdt, J ¼ 8.8 and
10.8 Hz, 2H), 3.66(s, 3H, OCH3), 2.47 (m, 2H, H-5), 1.39 (s,9H, Boc), 1.04(bt,
J ¼ 8.8 Hz, 2H),0.00 (s, 9H). 13C NMR d 172.1, 170.5, 155.4, 155.1, 136.1,
128.6, 128.5, 128.2, 128.1, 128.1, 79.9, 67.0, 64.3, 55.6 (C-6), 52.8 (C-2), 52.2,
35.2 (C-5), 28.2, 17.2, 21.6 ppm. MALDI-ToF MS: m/z ¼ 559.2 (Mþ þ Na),
575.2 (Mþ þ K). HRMS calcd. for C26H41N2O8Si 537.2632; found 537.2627.
Dimethyl (2S,7S)-bis-(tert-butoxycarbonyl)aminooctanedioate 9. EtOAc
(4 mL) was added to the alkene 6 (151 mg, 0.35 mmol) and 20% Pd/C
(75 mg, 0.0 mmol), and the solution was degassed by evacuation and flushing
with Ar (2ꢀ). An atmosphere of H2 was introduced in a like manner. The
reaction mixture was left under a balloon of hydrogen and filtered through a
Celite pad after 12 h. Evaporation of solvent provided 145 mg (96% yield) of
an oil. 1H NMR: d 5.05 (bd, J ¼ 8.0 Hz, 2H, 2NH), 4.27 (bdt, J ¼ 7.7,
5.5 Hz, 2H, H-2 and H-7), 3.73 (s, 6H), 1.76 (m, 2H, H-3a and H-6a), 1.61
(m, 2H, H-3b and H6b), 1.43 (s, 18H), 1.36 (m, 4H, H-4 and H-5).
1-Trimethylsilylethyl 7-methyl (2S, 6S)-2-(phenylmethoxycarbonyl)
amino-6-(tert-butoxycarbonyl)aminoheptanedioate 10. EtOAc (2 mL) was
added to the alkene 8 (95 mg, 0.18 mmol) and 5% Pt/Al2O3 (69 mg,

1712
E. G. Nolen, C. J. Fedorka, and B. Blicher
0.018 mmol), and the solution was degassed by evacuation and flushing with
Ar (2ꢀ). An atmosphere of H2 was introduced in a like manner. The reaction
mixture was left under a balloon of hydrogen and filtered through a Celite pad
after 16 h. Evaporation of solvent provided 88 mg (92% yield) of a dark oil. 1H
NMR: d 7.3 (m, 5H, C6H5), 5.41 (bd, J ¼ 7.3 Hz, 1H, NH), 5.13 (s overlapping
bd, 3H, NH and PhCH2), 4.25 (bm, 4H, H-2, H-6 and OCH2), 3.74 (s, 3H,
OCH3) 1.8–1.6 (2 ꢀ bm, 4H, H-3 and H-5), 1.44 (bs, 11H, H-4 and Boc)
1.02 (bt, J ¼ 8.7 Hz, 2H), 0.06 (s, 9H). 13C NMR d 173.1, 172.4, 156.0,
155.6, 136.2, 128.5, 128.1, 128.1, 79.9, 67.0, 63.9, 53.2 and 52.9 (C-2 and
C-6), 52.3, 32.1 and 32.0 (C-3 and C-5), 28.3, 21.2 (C-4), 17.3, 2 1.6 ppm.
MALDI-ToF MS: m/z ¼ 561.3 (Mþ þ Na), 577.2 (Mþ þ K). HRMS calcd.
for C26H43N2O8Si 539.2770; found 539.2797.
ACKNOWLEDGMENTS
Partial support of this work was provided by the National Science Foun-
dation’s Course, Curriculum, and Laboratory Improvement (CCLI) Program
(Department of Undergraduate Education-0088227). C. J. F. thanks Pfizer
for a summer undergraduate research fellowship.
REFERENCES
1. (a) Bhatnagar, P. K.; Agner, E. K.; Alberts, D.; Arbo, B. E.; Callahan, J. F.;
Cuthbertson, A. S.; Engelsen, S. J.; Fjerdingstad, H.; Hartmann, M.;
Heerding, D.; Hiebl, J.; Huffiman, W. F.; Hysben, M.; King, A. G.;
Kremminger, P.; Kwon, C.; LoCastro, S.; Løvhaug, D.; Pelus, L. M.;
Petteway, S.; Takata, J. S. Structure–activity relationships of novel hematoregula-
tory peptides. J. Med. Chem. 1996, 39, 3814–3819; (b) Stymiest, J. L.;
Mitchell, B. F.; Wong, S.; Vederas, J. C. Synthesis of oxytocin analogues with
replacement of sulfur by carbon gives potent antagonists with increased
stability. J. Org. Chem. 2005, 70, 7799–7809.
2. Asada, Y.; Tanizawa, K.; Sawada, S.; Suzuki, T.; Misono, H.; Soda, K.
Stereochemistry of meso-a,1-diaminopimelate decarboxylase reaction. The first
evidence for pyridoxal 50-phosphate dependent decarboxylation with inversion
of configuration. Biochemistry 1981, 20, 6881–6886.
3. (a) Girodeau, J.-M.; Agouridas, C.; Masson, M.; Pineau, R.; Le Goffic, F. The
lysine pathway as a target for a new genera of synthetic antibacterial antibiotics?
´
´
J. Med. Chem. 1986, 29, 1023–1030; (b) Mazon, A.; Najera, C.; Ezquerra, J.;
Pedregal, C. Synthesis of bis(a-amino acids) by palladium-catalyzed allylic
double substitution. Tetrahedron Lett. 1995, 36, 7697–7700.
4. (a) Nutt, R. F.; Strachan, R. G.; Veber, D. F.; Holly, F. W. Useful intermediates for
synthesis of dicarba analogues of cystine peptides: Selectively protected a-amino-
suberic acid and a, a0-diaminosuberic acid of defined stereochemistry. J. Org.
Chem. 1980, 45, 3078–3080; (b) Hiebl, J.; Blanka, M.; Guttman, A.;
Kollmann, H.; Leitner, K.; Mayrhofer, G.; Rovenszky, F.; Winkler, K. A
detailed investigation of the preparation of 2,7-diaminosuberic acid and

S,S-2,6-Diaminopimelic Acid
1713
2,5-diaminoadipic acid derivatives using Kolbe electrolysis. Tetrahedron 1998,
54, 2059–2074.
5. Hiebl, J.; Kollmann, H.; Rovenszky, F.; Winkler, K. Enantioselective synthesis of
diamino dicarboxylic acids. J. Org. Chem. 1999, 64, 1947–1952.
6. Collier, P. N.; Campbell, A. D.; Patel, I.; Raynham, T. M.; Taylor, R. J. K.
Enantiomerically pure a-amino acid synthesis via hydroboration–Suzuki cross-
coupling. J. Org. Chem. 2002, 67, 1802–1815.
7. (a) Bold, G.; Allmendinger, T.; Herold, P.; Moesch, L.; Schaer, H. P.;
Duthaler, R. O. Stereoselective synthesis of (2S,6S)-2,6-diaminoheptanedioic
acid and of unsymmetrical derivatives of meso-2,6-diaminoheptanedioic acid.
Helv. Chim. Acta 1992, 75, 865–882; (b) Williams, R. M.; Yuan, C. Asymmetric
synthesis of 2,6-diaminopimellic acids. J. Org. Chem. 1992, 57, 6519–6527;
(c) Kremminger, P.; Undheim, K. Asymmetric synthesis of unsaturated and
bis-hydroxylated (S,S)-2,7-diaminosuberic acid derivatives. Tetrahedron 1997,
53, 6925–6936.
8. Hernandez, N.; Martin, V. S. General stereoselective synthesis of chemically
differentiated a-diamino acids: Synthesis of 2,6-diaminopimelic acid and 2,7-
diaminosuberic acids. J. Org. Chem. 2001, 66, 4934–4938.
9. Miller, S. J.; Blackwell, H. E.; Grubbs, R. H. Application of ring-closing metath-
esis to the synthesis of rigidified amino acids and peptides. J. Am. Chem. Soc. 1996,
118, 9606–9614.
10. (a) Miles, S. M.; Leatherbarrow, R. J.; Marsden, S. P.; Coates, W. J. Synthesis and
bio-assay of RCM-derived Bowman–Birk inhibitor analogues. Org. Biomol.
Chem. 2004, 2, 281–283; (b) Pernerstorfer, J.; Schuster, M.; Blechert, S.
Cyclisation/cleavage of macrocycles by ring-closing metathesis on solid
support-conformational studies. Chem. Commun. 1997, 1949–1950.
11. (a) O’Leary, D. J.; Miller, S. J.; Grubbs, R. H. Template-promoted dimerization of
C-allylglycine: A convenient synthesis of (S,S)-2,7-diamnosuberic acid. Tetrahe-
dron Lett. 1998, 39, 1689–1690; (b) Williams, R. M.; Liu, J. Asymmetric
synthesis of differentially protected 2,7-diaminosuberic acid, a ring-closure
metathesis approach. J. Org. Chem. 1998, 63, 2130–2132.
12. Aguilera, B.; Wolf, L. B.; Nieczypor, P.; Rutjes, F. P. J. T.; Overkleeft, H. S.; van
Hest, J. C. M.; Schoemaker, H. E.; Wang, B.; Mol, J. C.; Fu¨rstner, A.;
Overhand, M.; van der Marel, G. A.; van Boom, J. H. Synthesis of diaminosuberic
acid derivatives via ring-closing alkyne metathesis. J. Org. Chem. 2001, 66,
3584–3589.
13. Del Valle, J. R.; Goodman, M. An efficient RCM-based synthesis of orthogonally
protected meso-DAP and FK565. J. Org. Chem. 2004, 69, 8946–8948.
14. (a) Biagini, S. C. G.; Gibson, S. E.; Keen, S. P. Cross-metathesis of unsaturated
a-amino acid derivatives. J. Chem. Soc., Perkin Trans. 1 1998, 2485–2499;
(b) Schmidtmann, F. W.; Benedum, T. E.; McGarvey, G. J. The preparation of
C-glycosyl amino acids—an examination of olefin cross-metathesis. Tetrahedron
Lett. 2005, 46, 4677–4681.
15. Nolen, E. G.; Kurish, A. J.; Wong, K. A.; Orlando, M. D. Short, stereoselective
synthesis of C-glycosyl asparagines via an olefin cross-metathesis. Tetrahedron
Lett. 2003, 44, 2449–2453.
16. Chatterjee, A. K.; Choi, T.-L.; Sanders, D. P.; Grubbs, R. H. A general model for
selectivity in olefin cross metathesis. J. Am. Chem. Soc. 2003, 125, 11360–11370.
17. Pasto, D. J.; Taylor, R. T. Reduction with diimide. Org. React. 1991, 40, 91–155.
18. Williams, R. M. Asymmetric synthesis of a-amino acids. Aldrichimica Acta 1992,
25, 11–25.