(R)- and (S)-N-(Benzyloxycarbonyl)-3,4-epoxybutylamine. New 4-amino-2- hydroxybutyl synthons for the synthesis of hypusine reagents and (R)-6- and (S)-7-hydroxyspermidine

Article abstract of DOI:10.1016/S0957-4166(99)00474-7

The synthesis and application of the chiral reagents (R)- and (S)-N- (benzyloxycarbonyl)-3,4-epoxybutylamine is described for the first time. These 4-amino-2-hydroxybutyl synthons are successfully employed in the assembly of two hydroxylated triamines, (R)-6- and (S)-7-hydroxyspermidine, and a previously described hypusine reagent, (2S,9R)-11- [(benzyloxycarbonyl)amino]-7-(benzyloxycarbonyl)-2-[(9- fluorenylmethoxycarbonyl)amino]-9-(tetrahydropyran-2-yloxy)-7-azaundecanoic acid, useful for solution- and solid-phase peptide synthesis. (C) 1999 Elsevier Science Ltd.

Full text of DOI:10.1016/S0957-4166(99)00474-7

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 10 (1999) 4285–4294
(R)- and (S)-N-(Benzyloxycarbonyl)-3,4-epoxybutylamine.
New 4-amino-2-hydroxybutyl synthons for the synthesis of
hypusine reagents and (R)-6- and (S)-7-hydroxyspermidine
Raymond J. Bergeron,∗ Jörg Bussenius,† Ralf Müller, Bruce H. McCosar and
James S. McManis
Department of Medicinal Chemistry, University of Florida, Gainesville, FL 32610-0485, USA
Received 2 August 1999; accepted 10 October 1999
Abstract
The synthesis and application of the chiral reagents (R)- and (S)-N-(benzyloxycarbonyl)-3,4-epoxybutylamine
is described for the first time. These 4-amino-2-hydroxybutyl synthons are successfully employed in the assembly
of two hydroxylated triamines, (R)-6- and (S)-7-hydroxyspermidine, and a previously described hypusine reagent,
(2S,9R)-11-[(benzyloxycarbonyl)amino]-7-(benzyloxycarbonyl)-2-[(9-fluorenylmethoxycarbonyl)amino]-9-(tet-
rahydropyran-2-yloxy)-7-azaundecanoic acid, useful for solution- and solid-phase peptide synthesis. © 1999
Elsevier Science Ltd. All rights reserved.
1. Introduction
The polyamines putrescine, spermidine, and spermine are found in all eukaryotes and are essential
to cell growth. Although polyamine fragments have been identified in a host of siderophores,1 spi-
der toxins,2 squalamine,3 etc., very few natural products contain polyamines in which the methylene
backbone itself is modified. The notable exceptions are hydroxyputrescine,4 the triamines (S)-6- and
7-hydroxyspermidine, and hypusine. The latter three systems are the subject of this paper. Whereas 7-
hydroxyspermidine occurs free in Pseudomonas species,5,6 (S)-6-hydroxyspermidine is a component of
several alkaloids isolated from marine microorganisms.7,8 Hypusine, (2S,9R)-2,11-diamino-9-hydroxy-
7-azaundecanoic acid (1),9 is an unusual amino acid which has attracted great interest in recent years
because of its role in the replication of HIV.10 It is formed by the post-translational modification of
∗
Corresponding author. Department of Medicinal Chemistry, J. Hillis Miller Health Science Center, PO Box 100485,
Gainesville, FL 32610-0485, USA. Tel: (352) 846-1956; fax: (352) 392-8406; e-mail: bergeron@mc.cop.ufl.edu
†
Present address: Albany Molecular Research, Inc., 21 Corporate Circle, Albany, NY 12203, USA.
0957-4166/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00474-7
tetasy 3110

4286
R. J. Bergeron et al. / Tetrahedron: Asymmetry 10 (1999) 4285–4294
eukaryotic initiation factor 5A (eIF-5A).11 Lysine 50 of this protein is coupled with an aminobutyl frag-
ment derived from spermidine followed by hydroxylation at C-9.12 Thus, the hydroxylated polyamines
are interesting from a synthetic standpoint because of the role they play in potential targets for various
therapeutic strategies.
Although there are currently two synthetic approaches to the hydroxyspermidines, one leads to a
mixture of 6- and 7-isomers;6 the other yields racemic 6-hydroxyspermidine.13 The synthons describ-
ed below, (R)- and (S)-N-(benzyloxycarbonyl)-3,4-epoxybutylamine, allow access to enantiomerically
pure 6- or 7-hydroxyspermidine. The same fragments also provide for an alternate, more direct, as-
sembly of the hypusine reagent, (2S,9R)-11-[(benzyloxycarbonyl)amino]-7-(benzyloxycarbonyl)-2-[(9-
fluorenylmethoxycarbonyl)amino]-9-(tetrahydropyran-2-yloxy)-7-azaundecanoic acid.14,15 This reagent
is a very powerful tool in synthesizing hypusine-containing peptides, which are key to expanding our
understanding of how eIF-5A functions.
2. Results and discussion
The enantiomeric epoxides 6a and 6b were utilized in the construction of the triamines (R)-6- and
(S)-7-hydroxyspermidine (8 and 13, respectively) and the hypusine reagent (18). These oxirane synthons
were synthesized from tosylates 5a and 5b, which were in turn made from D- or L-malic acid, respectively
(Scheme 1). Compounds 5a,b were also prepared by Shiba et al. from the requisite malic acid by a differ-
ent route and were converted using LiBr in acetone to (R)- or (S)-4-(benzyloxycarbonyl)amino-1-bromo-
2-butanol. Next, each bromohydrin was condensed with the benzyl ester of Nα-(benzyloxycarbonyl)-L-
lysine, tosylate salt; catalytic removal of the benzyl-derived protecting groups gave hypusine itself or its
epimer in 20% optimized yield from tosylate 5a or 5b, respectively.16
Scheme 1.
2.1. Epoxides 6a and 6b
The synthetic sequences began with the conversion of benzyl esters 2a,b, which are readily
available from D- or L- malic acid,17 respectively, to their N-hydroxysuccinimide esters and
subsequent reaction with ammonia to provide amide esters 3a,b in 65% yield. Reduction of

R. J. Bergeron et al. / Tetrahedron: Asymmetry 10 (1999) 4285–4294
4287
3a,b using borane–tetrahydrofuran, followed by selective protection of the amino group with N-
(benzyloxycarbonyloxy)succinimide (CBZ-NHS), yielded the diols 4a,b (61% and 63%). Selective
O-tosylation of 4a,b in dichloromethane and pyridine afforded the mono-tosylates 5a,b.16 Finally,
ring closure of 5a,b with potassium carbonate (1.1 equiv., 1 h) in methanol gave the desired epoxides
6a,b in 31% and 34% yield from 3a and 3b, respectively. Although oxiranes 6a,b may have formed
as intermediates in Shiba’s base-promoted construction of the hypusine framework,16 the first directed
synthesis of these chiral epoxide synthons has been accomplished.
2.2. (R)-6-Hydroxyspermidine 8
The synthesis of (R)-6-hydroxyspermidine (8, Scheme 2) was achieved by coupling oxirane 6a
with tert-butyl N-(3-aminopropyl)carbamate to provide the protected triamine 7 (65%). The CBZ
and BOC protective groups were then removed in one step using HBr in acetic acid to give (R)-6-
hydroxyspermidine (8) in 82% yield after conversion to its trihydrochloride salt, thus accomplishing
the first synthesis of this polyamine in enantiomerically pure form.
Scheme 2.
2.3. (S)-7-Hydroxyspermidine 13
Similar methodology was also applied to the first synthesis of (S)-7-hydroxyspermidine (13, Scheme
3). Benzyl (S)-N-(3,4-epoxybutyl)carbamate (6b), derived from L-malic acid (Scheme 1), was treated
with dibenzylamine (2 equiv.) in refluxing ethanol for 1 day, giving amino alcohol 9 in 53% yield.
The benzyloxycarbonyl (CBZ) protecting group of 9 was cleaved by treatment with 30% HBr in
HOAc/TFA (phenol/pentamethylbenzene/triisopropylsilane)15 at rt for 1 h to furnish (S)-N1,N1-dibenzyl-
2-hydroxyputrescine (10) in 56% yield. Exposure of 10 to an equivalent of acrylonitrile in methanol for
1 d at rt led to mono-cyanoethylated18 diamine 11 in 54% yield, thus completing the polyamine skeleton.
An attempt to catalytically cleave the benzyls and saturate the nitrile of 11 (Pd black/formic acid) failed
to give the final polyamine because of incomplete reaction; thus, a stepwise approach was employed.
The cyano moiety of 11 was selectively reduced using borane–dimethyl sulfide complex in THF (15 min,
reflux),19 generating (S)-N8,N8-dibenzyl-7-hydroxyspermidine (12) in quantitative yield.
The gem-dibenzyl blocking groups of 12 were not cleanly removed by dissolving metal reduction
(Na/NH3/THF/–78°C) or prolonged hydrogenolysis over 10% Pd–C in methanolic HCl. Successful
debenzylation required exposure to two different catalysts. First, hydrogenation of 12 over Pearl-
man’s catalyst [20% Pd(OH)2–C/MeOH/3 days],20,21 followed by reduction over palladium black (1N
HCl/MeOH), provided the target 13 as its trihydrochloride salt (32%).
2.4. Extension to the hypusine reagent 18
The application of the epoxide 6a was extended to the synthesis of 18, an adaptable hypusine reagent
(Scheme 4).14,15 In this reagent all three nitrogens and the 9-hydroxyl are protected while the carboxyl

4288
R. J. Bergeron et al. / Tetrahedron: Asymmetry 10 (1999) 4285–4294
Scheme 3.
group remains free for peptide coupling. The α-amino nitrogen is protected as the 9-fluorenylmethyl
carbamate (FMOC); the two remaining amines are masked as carbobenzyloxy (CBZ) moieties. The 9-
hydroxyl is protected as a tetrahydropyranyl ether (THP). For further peptide coupling, the α-amino
group can be cleanly unmasked in the presence of the other protecting groups. Thus, 18 allows for
the assembly of a variety of hypusine-containing peptides. Two syntheses of this reagent have been
published from our laboratories.14,15 In both, the chiral 4-amino-2-hydroxybutyl group was attached at
the ε-nitrogen of a lysine derivative employing epichlorohydrin, followed by displacement of the halide
with cyanide. This part of the synthesis does not lend itself to scaleup. In the first route,14 the internal
nitrogen of the reagent was protected by a benzyl group. Cyanide replaced the halide of the intermediate
3-amino chlorohydrin in the following step; racemization can occur if the reaction conditions are not
carefully controlled. This was thought to occur by formation of an azetidinium intermediate, ring opening
of which would lead to epimerization. In the second route,15 the secondary nitrogen was protected
by a CBZ, vitiating the nucleophilicity of the nitrogen’s non-bonding pair and preventing azetidinium
formation.
Scheme 4.
In our new synthetic approach, the tert-butyl ester of Nα-BOC-L-lysine, hydrochloride salt (14),15 was
treated with sodium bicarbonate solution (Scheme 4) and then heated for 1 day with the (R)-epoxide 6a
(0.5 equiv.) in ethanol producing amino alcohol 15 (56% yield), which possesses the hypusine skeleton.
The assembly of orthogonally-protected hypusine 15 from tosylate 5a was accomplished in 52% yield
for the two steps, which is more than twice as efficient as Shiba’s production of hypusine via its benzyl

R. J. Bergeron et al. / Tetrahedron: Asymmetry 10 (1999) 4285–4294
4289
group-protected derivative, also from intermediate 5a.16 Protection of the secondary amine of 15 was
effected with benzyl chloroformate (NEt3, CHCl3) to furnish 16a in 79% yield.14,15 The final three steps,
i.e., removal of the tert-butyl protective groups of 16a with TFA/triethylsilane and protection of the 9-
hydroxyl and the α-amino functions as the THP ether and FMOC derivative, respectively, followed our
previously described procedures14,15 and provided the (9R)-hypusine reagent 18 in good yield (44%,
Scheme 4). The (9S)-diastereomer of 18 could be readily prepared by using (S)-epoxide reagent 6b in
Scheme 4.
To confirm the stereochemical integrity at the C-9 position of 18, alcohol 16a (Scheme 4) was
converted to its Mosher ester22 derivative 17a (88%). We also synthesized Mosher ester 17b from 16b
(70%), which was produced by a different method.15 Resonances at −71.23 and −71.37 ppm were
observed in the 19F NMR spectrum of 17b at 25°C. However, these peaks had coalesced at 45°C to a
sharp signal at −71.41 ppm, most likely indicating the presence of rotamers. The peak at −71.41 ppm
was distinct from the corresponding resonance of 17a at −71.74 ppm. Since we were not able to observe
the −71.41 ppm resonance of 17b in the spectrum of 17a nor the −71.74 ppm signal of 17a in the
NMR of 17b, it is concluded that the synthesis of reagent 6a, its coupling with the protected lysine 14
to produce 15, and the subsequent steps in Scheme 4 had not been accompanied by racemization of the
secondary alcohol at C-9.
3. Conclusions
In summary, new enantiomeric epoxide reagents lead efficiently to chain functionalized polyamines
in which a four-carbon unit has a chiral secondary alcohol. The epoxide synthons also provide stereo-
chemically controlled access to a versatile hypusine reagent, which is in turn a highly useful synthon in
accessing the eIF-5A pentapeptide sequence and eIF-5A mimics.
4. Experimental
Reagents were purchased from Aldrich (Milwaukee, WI), Fluka (Milwaukee, WI), or Sigma (St. Louis,
MO) Chemical Co. and were used without purification. L- and D-Malic acids were obtained from Aldrich
in 99% and 98% e.e., respectively. Fisher Optima grade solvents were routinely used. THF was distilled
from sodium metal and benzophenone, and CH3OH and EtOH were also distilled. Organic extracts were
dried over Na2SO4, and silica gel 32–60 (40 µm ‘flash’) from Selecto, Inc. (Kennesaw, GA) was used for
flash column chromatography unless otherwise indicated. Melting points are uncorrected. Proton NMR
spectra were run at 300 MHz in deuterated organic solvents or in D2O with chemical shifts in parts
per million downfield from tetramethylsilane or 3-(trimethylsilyl)propionic-2,2,3,3-d4 acid, sodium salt,
respectively. Coupling constants (J) are in hertz. High resolution mass spectra were run in a 3-nitrobenzyl
alcohol matrix. Elemental analyses were performed by Atlantic Microlabs, Norcross, GA.
4.1. General procedure for generation of malic acid derivatives 3a,b
4.1.1. Benzyl (R)-4-amino-2-hydroxy-4-oxobutanoate 3a
N-Hydroxysuccinimide (287 mg, 2.49 mmol) and DCC (514 mg, 2.49 mmol) were added to a solution
of 2a17 (508 mg, 2.27 mmol) in CH2Cl2 (25 mL) under N2. After 90 min, the mixture was filtered, the
DCC was washed with CH2Cl2, and the solvent was removed in vacuo. The residue was dissolved in

4290
R. J. Bergeron et al. / Tetrahedron: Asymmetry 10 (1999) 4285–4294
dry THF (70 mL), cooled to 0°C, and ammonia gas was introduced into the solution over 20 min. The
reaction mixture was stirred for 20 min at 0°C and then concentrated to dryness. Flash chromatography
of the residue (12:1 CHCl3:CH3OH) afforded 324 mg (64%) of 3a as a colorless solid, mp 77–79°C.
1H NMR (CDCl3): 2.66 (dd, 1H, J=15.6, 7.0), 2.77 (dd, 1H, J=15.6, 3.9), 3.80 (br s, 1H), 4.53 (dd, 1H,
J=7.0, 3.9), 5.23 (s, 2H), 5.80 (br s, 1H), 6.00 (br s, 1H), 7.36 (m, 5H). [α]23 +18.6 (c 2.0, CH3OH) [lit.
D
ent-3a:23 [α]25 –17.1 (c 2, CH3OH)]. Anal. calcd: C, 59.19; H, 5.87; N, 6.27. Found: C, 59.20; H, 5.80;
D
N, 6.28.
4.1.2. Benzyl (S)-4-amino-2-hydroxy-4-oxobutanoate 3b
Compound 2b17 (15.2 g, 67.7 mmol) was reacted by the method of 3a to give 9.95 g (66%) of 3b
as a colorless solid. 1H NMR (CDCl3): 2.66 (dd, 1H, J=15.6, 7.3), 2.77 (dd, 1H, J=15.6, 3.7), 3.70 (d,
1H, J=4.8), 4.53 (m, 1H), 5.23 (s, 2H), 5.55 (br m, 1H), 5.91 (br m, 1H), 7.36 (m, 5H). A sample was
recrystallized from cyclohexane/EtOAc, mp 79.2–80.5°C (lit. 3b:23 78–80°C). Anal. calcd: C, 59.19; H,
5.87; N, 6.27. Found: C, 58.98; H, 5.82; N, 6.41.
4.2. General procedure for reduction to alcohols 4a,b
4.2.1. (R)-4-(Benzyloxycarbonyl)amino-1,2-butanediol 4a
A 1 M solution of BH3 in THF (100 mL) was added to a stirred solution of 3a (4.35 g, 19.5
mmol) in THF (100 mL) at rt under N2. After the mixture was stirred for 17 h, 4N aqueous HCl was
slowly added until pH 4 was achieved. The reaction mixture was stirred at rt for 4 h. The bulk of the
THF was removed by rotary evaporation; the residue was neutralized with 1N NaOH and concentrated
to dryness under high vacuum.24 Water (60 mL) and Et2O (100 mL) were added to the residue. N-
(Benzyloxycarbonyloxy)succinimide (5.34 g, 21.4 mmol) and NaHCO3 (1.64 g, 19.5 mmol) were added
at 0°C, and the mixture was stirred at rt for 18 h. The layers were separated, and the aqueous layer was
further extracted with EtOAc (3×100 mL). The organic layers were washed with NaHCO3 (50 mL), the
NaHCO3 layer was re-extracted with EtOAc (2×50 mL), and the solvent was removed in vacuo. Flash
chromatography (12:1 CHCl3:CH3OH) gave 2.82 g (61%) of 4a as a colorless solid, mp 64–65°C (lit.16
67–68°C). 1H NMR (CD3OD): 1.52 (m, 1H), 1.71 (m, 1H), 3.25 (m, 2H), 3.45 (m, 2H), 3.63 (m, 1H),
22
5.06 (s, 2H), 7.31 (m, 5H). [α] +6.3 (c 1.97, CHCl3) [lit.16 [α]22 +5.8 (c 1.9, CHCl3)].
D
D
4.2.2. (S)-4-(Benzyloxycarbonyl)amino-1,2-butanediol 4b
Compound 3b (10.60 g, 47.50 mmol) was treated as was 3a to give 7.18 g (63%) of 4b as a colorless
solid. [α] –6.0 (c 1.9, CHCl3) [lit.16 [α]25 –6.0 (c 2.2, CHCl3)].
21
D
D
4.3. General procedure for epoxide reagents 6a,b
4.3.1. (R)-N-(Benzyloxycarbonyl)-3,4-epoxybutylamine 6a
Potassium carbonate (151 mg, 1.09 mmol) was added to a solution of 5a16 (391 mg, 0.99 mmol)
in CH3OH (30 mL). The mixture was stirred at rt under N2 for 50 min, then poured into a mixture
of H2O (50 mL) and CH2Cl2 (40 mL). The layers were separated, the aqueous layer was extracted with
CH2Cl2 (4×30 mL), and the solvent was removed in vacuo. Flash chromatography (80:1 CHCl3:CH3OH)
afforded 182 mg (83%) of 6a as a colorless oil. 1H NMR (CD3OD): 1.63 (m, 1H), 1.76 (m, 1H), 2.47 (m,
1H), 2.71 (m, 1H), 2.95 (m, 1H), 3.27 (m, 2H), 5.07 (s, 2H), 7.33 (m, 5H). [α]22 +21.4 (c 1.00, CHCl3).
D
Anal. calcd: C, 65.14; H, 6.83; N, 6.33. Found: C, 65.15; H, 6.80; N, 6.33.

R. J. Bergeron et al. / Tetrahedron: Asymmetry 10 (1999) 4285–4294
4291
4.3.2. (S)-N-(Benzyloxycarbonyl)-3,4-epoxybutylamine 6b
In the same manner as for 6a, 5b16 (8.74 g, 22.2 mmol) was converted to 3.72 g (76%) of 6b as a
22
D
colorless oil. [α] –21.8 (c 1.07, CHCl3). Anal. calcd: C, 65.14; H, 6.83; N, 6.33. Found: C, 65.07; H,
6.89; N, 6.27.
4.4. (R)-N1-(tert-Butoxycarbonyl)-N8-(benzyloxycarbonyl)-6-hydroxyspermidine 7
A solution of 6a (152 mg, 0.69 mmol) and tert-butyl N-(3-aminopropyl)carbamate (239 mg, 1.37
mmol) in EtOH (20 mL) was heated to reflux for 18 h under Ar. The reaction mixture was concentrated
in vacuo and purified by flash chromatography (10%, then 50% CH3OH/CHCl3) to give 175 mg (65%)
of 7 as a white solid, mp 90–92°C. 1H NMR (CD3OD): 1.43 (s, 9H), 1.50–1.77 (m, 4H), 2.50–2.78 (m,
23
D
4H), 3.09 (t, 2H, J=6.6), 3.25 (t, 2H, J=7.0), 3.74 (m, 1H), 5.06 (s, 2H), 7.23–7.45 (m, 5H). [α] +2.8
(c 1.05, CH3OH). HRMS m/z calcd for C20H34N3O5: 396.2498; found: 396.2523.
4.5. (R)-6-Hydroxyspermidine trihydrochloride 8
HBr in acetic acid (30%, 400 µL) was added to a solution of 7 (155 mg, 0.39 mmol), phenol (500
mg), pentamethylbenzene (500 mg), and triisopropylsilane (200 µL) in TFA (5 mL) at 0°C under Ar.
The ice bath was removed after 10 min, and the reaction mixture was stirred at rt for an additional 50
min and concentrated in vacuo. The residue was dissolved in 10% acetic acid (20 mL) and extracted
with tert-butyl methyl ether (3×25 mL). The aqueous layer was concentrated and dried in vacuo. The
residue was dissolved in H2O (5 mL) and the pH was adjusted to 7 with saturated NaHCO3. The solution
was concentrated in vacuo and purified by flash chromatography (Kieselgel 60, 70–230 mesh, Merck),
eluting with freshly prepared 1:2:1 CH2Cl2:CH3OH:conc. NH4OH. The oil was dissolved in CH3OH
(4 mL) and filtered through a syringe filter (Whatman PVDF 0.2 µm). The filtrate was concentrated in
vacuo, dissolved in H2O (2 mL), acidified with 1N HCl (4 mL), and concentrated. The resulting solid was
dissolved in CH3OH (5 mL) and precipitated with EtOAc (9 mL) to give 87 mg (82%) of 8 as a white,
hygroscopic powder, mp 192–193°C. 1H NMR (D2O): 1.76–2.04 (m, 2H), 2.04–2.24 (m, 2H), 3.04–3.38
(m, 8H), 4.08 (tt, 1H, J=9.5, 3.2). 13C NMR (75 MHz, D2O, dioxane=67.19 ppm): 24.20, 31.96, 37.03,
37.18, 45.28, 52.88, 65.32. [α]24 –7.1 (c 1.02, H2O). HRMS m/z calcd for C7H20N3O: 162.1606; found:
D
162.1604. Anal. calcd: C, 31.07; H, 8.19; Cl, 39.30; N, 15.53. Found: C, 31.01; H, 8.07; Cl, 39.14; N,
15.32.
4.6. (S)-4-[(Benzyloxycarbonyl)amino]-1-(dibenzylamino)-2-butanol 9
Dibenzylamine (1.02 g, 5.17 mmol) and 6b (513 mg, 2.32 mmol) were heated at reflux in EtOH (20
mL) for 20 h under N2. The solution was concentrated to dryness, and the residue was purified twice by
1
flash chromatography (2% CH3OH/CH2Cl2) affording 520 mg (53%) of 9 as a colorless oil. H NMR
(CD3OD): 1.31 (m, 1H), 1.76 (m, 1H), 2.43 (m, 2H), 3.14 (m, 2H), 3.57 (s, 4H), 3.72 (m, 1H), 5.05 (s,
2H), 7.17–7.37 (m, 15H). 13C NMR (75 MHz, CD3OD, ref. solvent 49.0 ppm): 36.0, 38.3, 38.7, 60.0,
60.8, 67.3, 67.6, 128.1, 128.7, 128.9, 129.3, 129.4, 130.2, 138.4, 140.5, 158.8. Anal. calcd: C, 74.61; H,
7.22; N, 6.69. Found: C, 74.32; H, 7.15; N, 6.65.

4292
R. J. Bergeron et al. / Tetrahedron: Asymmetry 10 (1999) 4285–4294
4.7. (S)-4-Amino-1-(dibenzylamino)-2-butanol 10
HBr in HOAc (30%, 0.20 mL) was added to a solution of 9 (520 mg, 1.24 mmol), phenol (250 mg),
pentamethylbenzene (250 mg), and triisopropylsilane (0.10 mL) in TFA (5.0 mL) at 0°C under N2. The
solution was stirred for 5 min at 0°C, then for 30 min at rt. Solvent removal gave a crude oil, to which
was added 90% aqueous HOAc (20 mL) at 0°C followed by extraction with tert-butyl methyl ether
(3×50 mL). Concentration of the aqueous layer in vacuo and purification by flash chromatography (3%
concentrated NH4OH/CH3OH) gave 197 mg (56%) of 10 as a colorless oil. 1H NMR (CD3OD): 1.37 (m,
1H), 1.71 (m, 1H), 2.44 (d, 2H, J=6.9), 2.69 (m, 2H), 3.60 (s, 4H), 3.78 (m, 1H), 7.19–7.38 (m, 10H).
HRMS m/z calcd for C18H25N2O: 285.1967; found: 285.2024.
4.8. (S)-8-(Dibenzylamino)-7-hydroxy-4-azaoctanenitrile 11
Acrylonitrile (37.6 mg, 0.709 mmol) was slowly added to 10 (197 mg, 0.693 mmol) in CH3OH (1.0
mL) under N2; the reaction solution was gradually warmed to rt and stirred for a total of 21 h. The solvent
was evaporated, and the residue was purified by flash chromatography (4% CH3OH/CH2Cl2) to give 126
mg (54%) of 11 as a colorless oil. 1H NMR (CD3OD): 1.31–1.46 (m, 1H), 1.68–1.81 (m, 1H), 2.45 (d,
2H, J=6.5), 2.51–2.62 (m, 4H), 2.78 (t, 2H, J=6.8), 3.59 (dd, 4H, J=17.7, 13.3), 3.79 (m, 1H), 7.19–7.38
(m, 10H). HRMS m/z calcd for C21H28N3O: 338.2232; found: 338.2195.
4.9. (S)-N8,N8-Dibenzyl-7-hydroxyspermidine 12
Borane–methyl sulfide complex in THF (2 M, 1.8 mL, 3.6 mmol) was added by syringe to 11 (125.5
mg, 0.373 mmol) under N2. The reaction was stirred at rt for 15 min, then refluxed for an additional 15
min. After cooling, 6 M HCl (2.5 mL) was added and the solution was stirred for 40 min. The solvent
was evaporated and the crude product was redissolved in 1N NaOH (15 mL), then extracted with CHCl3
(3×25 mL). The organic layer was washed with H2O (20 mL), then concentrated to give 127 mg (100%)
of 12 as a colorless oil. 1H NMR (CD3OD): 1.40 (m, 1H), 1.63 (m, 2H), 1.67–1.81 (m, 1H), 2.45 (d, 2H,
J=6.4), 2.48–2.62 (m, 4H), 2.66 (t, 2H, J=7.1), 3.59 (dd, 4H, J=15.4, 13.5), 3.77 (m, 1H), 7.19–7.38 (m,
10H). HRMS m/z calcd for C21H32N3O: 342.2545; found: 342.2545.
4.10. (S)-7-Hydroxyspermidine trihydrochloride 13
Compound 12 (127.3 mg, 0.373 mmol) was stirred with 20% Pd(OH)2–C (62 mg) in CH3OH (7.5
mL) under hydrogen for 3 days at 1 atm. The reaction mixture was filtered through Celite, evaporated
to dryness, and hydrogenated again at ambient pressure and temperature for 24 h over Pd black (32 mg)
in CH3OH (12 mL) and 1N HCl (1.6 mL). After filtration through Celite, the solvent was evaporated.
The crude product was treated with 1N HCl (10 mL) and evaporated to give a white crystalline powder.
Recrystallization from 25% Et2O/CH3OH (15 mL) gave 32 mg (32%) of 13 as fine white crystals, mp
203–206°C. 1H NMR (D2O): 1.78–2.17 (m, 4H), 2.97 (dd, 1H, J=13.3, 9.5), 3.07–3.36 (m, 7H), 4.00 (tt,
1H, J=9.5, 3.3). 13C NMR APT (75.4 MHz, D2O, dioxane=67.19 ppm, APT inverse peaks underlined):
24
D
24.4, 30.8, 37.2, 44.9, 45.26, 45.31, 66.0. [α] –0.5 (c 1.00, 1N HCl), +1.1 (c 1.00, CH3OH). Anal.
calcd: C, 31.07; H, 8.19; Cl, 39.30; N, 15.53. Found: C, 31.34; H, 8.36; Cl, 39.55; N, 15.24.

R. J. Bergeron et al. / Tetrahedron: Asymmetry 10 (1999) 4285–4294
4293
4.11. tert-Butyl (2S,9R)-11-[(benzyloxycarbonyl)amino]-2-[(tert-butoxycarbonyl)amino]-9-hydroxy-
7-azaundecanoate 15
A solution of 1415 (171 mg, 0.51 mmol) in CHCl3 (20 mL) was extracted with saturated NaHCO3
(2×20 mL). The combined aqueous layers were extracted with CHCl3 (2×20 mL). The combined organic
layers were washed with saturated NaCl (20 mL) and concentrated to give the primary amine (136 mg,
89%) as a colorless oil, a portion of which (52 mg, 172 mmol) was heated at reflux for 23 h under Ar with
6a (19 mg, 86 mmol) in EtOH (10 mL). The solvent was removed in vacuo and the residue was purified
by flash chromatography (2%, then 20% CH3OH/CHCl3) to give 29 mg (63%) of 15 as a colorless oil. 1H
NMR (CD3OD): 1.35–1.82 (m, 8H), 1.44 (s, 9H), 1.46 (s, 9H), 2.48–2.74 (m, 4H), 3.25 (t, 2H, J=6.8),
22
D
3.74 (m, 1H), 3.94 (dd, 1H, J=8.6, 5.1), 5.06 (s, 2H), 7.24–7.40 (m, 5H). [α] –12.3 (c 1.00, CH3OH).
HRMS m/z calcd for C27H46N3O7: 524.3336; found: 524.3261.
4.12. tert-Butyl (2S,9R)-11-[(benzyloxycarbonyl)amino]-2-[(tert-butoxycarbonyl)amino]-9-hydroxy-
7-(benzyloxycarbonyl)-7-azaundecanoate 16a
Benzyl chloroformate (10 µL, 70 µmol) and triethylamine (15 µL, 110 µmol) were added to a solution
of 15 (27 mg, 52 µmol) in CHCl3 (2 mL) at 0°C under Ar. The reaction mixture was stirred at rt for 1.5 h,
diluted with CHCl3 (20 mL), and successively washed with 1N HCl (10 mL), H2O (10 mL), and saturated
NaCl (10 mL). The organic layer was dried over MgSO4 and concentrated in vacuo. Purification by flash
chromatography (50% hexane/EtOAc) gave 27 mg (79%) of 16a as a colorless oil. 1H NMR (CD3OD):
1.25–1.80 (m, 8H), 1.43 (s, 9H), 1.44 (s br, 9H), 3.08–3.48 (m, 6H), 3.82 (m, 1H), 3.92 (dd, 1H, J=8.0,
5.3), 5.06 (s, 2H), 5.11 (s, 2H), 7.24–7.39 (m, 10H). [α]22 –12.5 (c 1.00, CH3OH).
D
4.13. General procedure for derivatization of 16a,b to Mosher esters 17a,b
4.13.1. tert-Butyl (2S,9R)-11-[(benzyloxycarbonyl)amino]-2-[tert-(butoxycarbonyl)amino]-7-(benzyl-
oxycarbonyl)-9-[(S)-α-methoxy-α-(trifluoromethyl)phenylacetoxy]-7-azaundecanoate 17a
The reaction was carried out in an oven-dried 5×175 mm NMR tube, fitted with a rubber septum,
under Ar. The reagents were injected via syringe in the following order: anhydrous pyridine (300 µL),
(R)-(−)-Mosher’s acid chloride (12 µL, 57 µmol), anhydrous CCl4 (200 µL), and a solution of 16a (22
mg, 33 µmol) in anhydrous CCl4 (500 µL). The reaction mixture was shaken and allowed to stand at rt
for 18 h. The reaction mixture was then taken up in CHCl3 (20 mL) and washed with saturated NaHCO3
and saturated NaCl. The organic layer was concentrated in vacuo. The residue was purified by flash
chromatography (25% EtOAc/hexane) to give 26 mg (88%) of 17a as a colorless oil. 1H NMR (CDCl3,
45°C): 1.10–1.88 (m, 8H), 1.44 (s, 9H), 1.45 (s, 9H), 2.80 (m, 1H), 2.95–3.60 (m, 4H), 3.28 (dd, 1H,
J=14.6, 7.5), 3.50 (s, 3H), 4.09 (m, 1H), 4.80–5.38 (m, 7H), 7.25–7.55 (m, 15H). 19F NMR (282 MHz,
21
D
CDCl3, CFCl3 as internal standard, 45°C): −71.74. [α] −12.6 (c 1.00, CHCl3). HRMS m/z calcd for
C45H59F3N3O11: 874.4102; found: 874.4100.
4.13.2. tert-Butyl (2S,9S)-11-[(benzyloxycarbonyl)amino]-2-[tert-(butoxycarbonyl)amino]-7-(benzyl-
oxycarbonyl)-9-[(S)-α-methoxy-α-(trifluoromethyl)phenylacetoxy]-7-azaundecanoate 17b
According to the method described for the preparation of 17a, 16b15 (14 mg, 21 µmol) was reacted
with (R)-(–)-Mosher’s acid chloride (10 µL, 54 µmol) to give 13 mg (70%) of 17b as a colorless oil. 1H
NMR (CDCl3, 45°C): 1.15–1.90 (m, 8H), 1.44 (s, 9H), 1.44 (s, 9H), 2.92–3.60 (m, 6H), 3.44 (s, 3H),
4.10 (m, 1H), 4.80–5.40 (m, 7H), 7.22–7.60 (m, 15H). 19F NMR (282 MHz, CDCl3, CFCl3 as internal

4294
R. J. Bergeron et al. / Tetrahedron: Asymmetry 10 (1999) 4285–4294
22
D
standard, 45°C): −71.41. [α] −9.7 (c 0.98, CHCl3). HRMS m/z calcd for C45H59F3N3O11: 874.4102;
found: 874.4173.
Acknowledgements
This work was supported by SunPharm Corporation, Ponte Vedra Beach, FL. The work of the
Spectroscopy Services Group, Department of Chemistry, University of Florida, for HRMS analyses is
appreciated. We also thank Dr. E. Eiler-Hughes for her editorial comments.
References
1. Bergeron, R. J.; McManis, J. S. Synthesis of catecholamide and hydroxamate siderophores. In CRC Handbook of Microbial
Iron Chelates; Winkelmann, G., Ed.; CRC: Boca Raton, 1991; pp. 271–307.
2. Moya, E.; Blagbrough, I. S. Tetrahedron Lett. 1994, 35, 2061–2062.
3. Moore, K. S.; Wehrli, S.; Roder, H.; Rogers, M.; Forrest, J. N., Jr.; McCrimmon, D.; Zasloff, M. Proc. Natl. Acad. Sci. USA
1993, 90, 1354–1358.
4. Kullnig, R. K.; Rosano, C. L.; Coulter, M. E.; Hurwitz, C. J. Biol. Chem. 1973, 248, 2487–2488.
5. Hamana, K.; Matsuzaki, S. FEMS Microbiol. Lett. 1990, 70, 353–356.
6. Rosano, C. L.; Hurwitz, C.; Bunce, S. C. J. Bacteriol. 1978, 135, 805–808.
7. Palagiano, E.; DeMarino, S.; Minale, L.; Riccio, R.; Zollo, F.; Iorizzi, M.; Carré, J. B.; Debitus, C.; Lucarain, L.; Provost, J.
Tetrahedron 1995, 51, 3675–3682.
8. Jares-Erijman, E. A.; Sakai, R.; Rinehart, K. L. J. Org. Chem. 1991, 56, 5712–5715.
9. Shiba, T.; Mizote, H.; Kaneko, T.; Nakajima, T.; Kakimoto, Y.; Sano, I. Biochim. Biophys. Acta 1971, 244, 523–531.
10. Ruhl, M.; Himmelspach, M.; Bahr, G. M.; Hammerschmid, F.; Jaksche, H.; Wolff, B.; Aschauer, H.; Farrington, G. K.;
Probst, H.; Bevec, D.; Hauber, J. J. Cell Biol. 1993, 123, 1309–1320.
11. Park, M. H.; Cooper, H. L.; Folk, J. E. J. Biol. Chem. 1982, 257, 7217–7222.
12. Park, M. H.; Wolff, E. C.; Folk, J. E. Trends Biochem. Sci. 1993, 18, 475–479.
13. Ponasik, J. A.; Ganem, B. Tetrahedron Lett. 1995, 36, 9109–9112.
14. Bergeron, R. J.; Ludin, C.; Müller, R.; Smith, R. E.; Phanstiel IV, O. J. Org. Chem. 1997, 62, 3285–3290.
15. Bergeron, R. J.; Weimar, W. R.; Müller, R.; Zimmerman, C. O.; McCosar, B. H.; Yao, H.; Smith, R. E. J. Med. Chem. 1998,
41, 3888–3900.
16. Shiba, T.; Akiyama, H.; Umeda, I.; Okada, S.; Wakamiya, T. Bull. Chem. Soc. Jpn. 1982, 55, 899–903.
17. Cammas, S.; Renard, I.; Boutault, K.; Guérin, P. Tetrahedron: Asymmetry 1993, 4, 1925–1930.
18. Bergeron, R. J.; McManis, J. S. J. Org. Chem. 1988, 53, 3108–3111.
19. Brown, H. C.; Choi, Y. M.; Narasimhan, S. Synthesis 1981, 605–606.
20. Pearlman, W. M. Tetrahedron Lett. 1967, 17, 1663–1664.
21. Yoshida, K.; Nakajima, S.; Wakamatsu, T.; Ban, Y.; Shibasaki, M. Heterocycles 1988, 27, 1167–1168.
22. Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512–519.
23. Roy, J.; Gazis, D.; Shakman, R.; Schwartz, I. L. Int. J. Pept. Protein Res. 1982, 20, 35–42.
24. Wang, G.; Hollingsworth, R. I. J. Org. Chem. 1999, 64, 1036–1038.