Syntheses of The Enantiomers of 1-Deoxynojirimycin and 1-Deoxyaltronojirimycin Via Chemo-And Diastereoselective Olefinic Oxidation of Unsaturated Amines
Syntheses of The Enantiomers of 1-Deoxynojirimycin and 1-Deoxyaltronojirimycin Via Chemo-And Diastereoselective Olefinic Oxidation of Unsaturated Amines
Syntheses of The Enantiomers of 1-Deoxynojirimycin and 1-Deoxyaltronojirimycin Via Chemo-And Diastereoselective Olefinic Oxidation of Unsaturated Amines
org/joc
steve.davies@chem.ox.ac.uk
Received September 10, 2010
DOI: 10.1021/jo101756g Published on Web 11/02/2010 J. Org. Chem. 2010, 75, 8133–8146 8133
r 2010 American Chemical Society
JOC Article Bagal et al.
a
Reagents and conditions: (i) HBF4 (40% w/w in H2O), m-CPBA (4 equiv), CH2Cl2, BnOH, rt, 24 h; (ii) HBF4 (40% w/w in H2O), m-CPBA
(1.5 equiv), CH2Cl2, BnOH, rt, 24 h; (iii) MsCl, Et3N, CH2Cl2, 0 °C, 1.5 h; (iv) F3CCO2H, F3CCO3H, CH2Cl2, 0 °C to rt, 6 h; (v) Cl3CCO2H, CH2Cl2, rt,
16 h, then NaOH (2 M, aq); (vi) AgOAc, DMF, 65 °C, 24 h; (vii) K2CO3, MeOH, rt, 16 h; (viii) H2, Pd(OH)2/C, MeOH, rt, 72 h, then HCl (aq).
SCHEME 3a SCHEME 4a
a
Reagents and conditions: (i) HBF4 3 OEt2, m-CPBA (4 equiv), CH2Cl2,
BnOH, rt, 24 h; (ii) HBF4 3 OEt2, m-CPBA (1.5 equiv), CH2Cl2, BnOH, rt,
24 h.
SCHEME 5a SCHEME 6a
a
a Reagents and conditions: (i) HBF4 3 OEt2, m-CPBA (4 equiv),
Reagents and conditions: (i) HBF4 3 OEt2, m-CPBA (1.5 equiv),
CH2Cl2, BnOH, rt, 24 h; (ii) chromatography; (iii) preparative chiral
CH2Cl2, BnOH, rt, 24 h; (ii) chromatography; (iii) preparative chiral
HPLC (Chiralpak AD-H column).
HPLC (Chiralpak AD-H column).
Having determined the absolute configurations within
and (-)-(R,R)-5 via chemical correlation to the enantiomers (þ)-5 and (-)-5, the absolute configurations within (þ)-8
of 1-deoxynojirimycin 2. Thus, ring contraction of (þ)-5 was and (-)-8 were established via chemical correlation. Thus,
effected upon treatment with MsCl to give 6 in 75% yield oxidation of (þ)-(S,S)-5 (>99% ee) with m-CPBA in the
and>99:1 dr.12 Chemo- and diastereoselective oxidation of presence of aqueous HBF414 gave (þ)-8 in >99% ee15 {[R]25D
the olefin within 6 was achieved upon treatment with þ54.1 (c 1.0 in CHCl3)}, while oxidation of (-)-(R,R)-5 (>99%
F3CCO3H in the presence of F3CCO2H, giving piperidine ee) gave (-)-8 in >99% ee16 {[R]25D -54.3 (c 1.0 in CHCl3)},
epoxide 7 in 98:2 dr, which was isolated in 62% yield. Upon thus allowing the absolute configurations within (þ)-8 and
treatment with Cl3CCO2H, epoxide ring-opening occurred (-)-8 to be unambiguously assigned as (þ)-(3S,4R,5S,6S)-8
at C(5) with modest levels of regioselectivity13 [the ratio of and (-)-(3R,4S,5R,6R)-8. We have previously demonstrated
C(4):C(5) ring opened products in the crude reaction mixture the elaboration of (()-8 into (()-1-deoxyaltronojirimycin,8 and
was 88:12], with saponification of the crude reaction mixture therefore, elaboration of (þ)-(3S,4R,5S,6S)-8 via an analogous
with aqueous NaOH giving 12 in 51% isolated yield and 99:1 series of reactions should culminate in the preparation of
dr. Subsequent displacement of the chloride functionality with- (þ)-1-deoxyaltronojirimycin, while similar elaboration of
in 12 gave acetate ester 1312 and was followed by treatment (-)-(3R,4S,5R,6R)-8 should permit the synthesis of (-)-1-deoxy-
with K2CO3 in MeOH to effect transesterification to give 14, altronojirimycin (Scheme 8).
with subsequent hydrogenolysis and acidification giving Synthesis of Polyhydroxylated Piperidines Employing Dia-
(þ)-1-deoxynojirimycin hydrochloride (þ)-2 3 HCl in 30% yield stereoselective Olefinic Oxidation of an Enantiopure Substrate:
from 12 [7% overall yield from (þ)-5 in 6 steps] and 99:1 dr. Application to the Synthesis of (þ)-1-Deoxyaltronojirimycin.
Similar elaboration of (-)-5 gave a sample of (-)-1-deoxy- Having developed a route to the antipodes of 1-deoxynojirimycin
nojirimycin hydrochloride (-)-2 3 HCl in 8% overall yield in six 2 (and 1-deoxyaltronojirimycin 3) relying on resolution of func-
steps, and 99:1 dr. The values of the specific rotations of our tionalized tetrahydroazepine 5 and azepane 8, we turned our
samples of the antipodes of 2 {for (þ)-2 3 HCl, [R]25D þ31.0 attention to the development of a de novo asymmetric
(c 0.45 in H2O); lit.3a for sample isolated from natural source synthesis.17 Our strategy in this area centered on employing a
[R]22D þ38.0 (c 1.0 in H2O); lit.9m [R]23D þ36.9 (c 1.1 in H2O); chiral N-protecting group as it was envisaged that this would
for (-)-2 3 HCl, [R]25D -34.0 (c 0.45 in H2O); lit.9i [R]25D -46.0 break the symmetry of the 7-membered ring, potentially allow-
(c 1.3 in H2O); lit.9m [R]24D -38.7 (c 1.0 in H2O)} allowed the ing diastereoselective oxidation. A range of N-substituted di-
absolute configurations within (þ)-5 and (-)-5 to be unam- hydroazepanes 19-22, all bearing a chiral N-protecting group
biguously assigned as (þ)-(S,S)-5 and (-)-(R,R)-5, and the based upon the R-methylbenzyl scaffold, was prepared from
absolute configurations within the synthetic intermediates 6, 7, cis,cis-muconic acid 15. Methylation of 15 with MeI and K2CO3
and 12-14 were therefore also unambiguously established. gave diester 16 in 91% yield. Reduction of 16 with DIBAL-H
Furthermore, given the enantiomeric purities of (þ)-5 and
(-)-5 (>99% ee in both cases), the enantiomeric purities of (14) Aqueous HBF4 was used for these reactions as it was found to give
(þ)-2 3 HCl, (-)-2 3 HCl, 6, 7, and 12-14 can be inferred as cleaner crude reaction mixtures than the analogous procedures using
HBF4 3 OEt2.
>99% ee (Scheme 7). (15) The enantiomeric purity of (þ)-8 prepared in this manner was
inferred from the enantiomeric purity of the starting material (þ)-5
(i.e., >99% ee).
(12) This reaction presumably proceeds via the intermediacy of the (16) The enantiomeric purity of (-)-8 prepared in this manner was
corresponding aziridinium ion; see ref 8. inferred from the enantiomeric purity of the starting material (-)-5
(13) For a discussion concerning the regioselectivity of ring-opening, see (i.e., >99% ee).
ref 8. (17) Brown, J. M.; Davies, S. G. Nature 1989, 342, 631.
a
Reagents and conditions: (i) MsCl, Et3N, CH2Cl2, 0 °C, 1.5 h; (ii) F3CCO2H, F3CCO3H, CH2Cl2, 0 °C to rt, 6 h; (iii) Cl3CCO2H, CH2Cl2, rt, 16 h,
then NaOH (2 M, aq); (iv) AgOAc, DMF, 65 °C, 24 h; (v) K2CO3, MeOH, rt, 16 h; (vi) H2, Pd(OH)2/C, MeOH, rt, 72 h, then HCl (aq).
SCHEME 8a SCHEME 9a
a
Reagents and conditions: (i) HBF4 (40% w/w in H2O), m-CPBA
(4 equiv), CH2Cl2, BnOH, rt, 24 h. a
Reagents and conditions: (i) Mel, K2CO3, DMF, rt, 24 h; (ii) DIBAL-
H, CH2Cl2, 0 °C to rt, 16 h; (iii) PBr3, Et2O, 0 °C to rt, 7 h; (iv)
followed by treatment of the resultant diol 17 with PBr3 gave ArCH(Me)NH2, K2CO3, THF, rt, 24 h. bIsolated as a 93:7 mixture of
dibromide 18 in 83% yield (two steps). Addition of a range of 22:26. 2-Nap = 2-naphthyl. 1-Nap = 1-naphthyl. 2-Tol = 2-tolyl.
chiral primary amines18 to 18 furnished, in each case, the
corresponding dihydroazepanes 19-22 as the major products, Cl3CCO2H (5 equiv) over 21 h6a proceeded to full conversion
along with the corresponding dihydropyrroles 23-26 as minor to give a chromatographically inseparable 65:35 mixture of two
products.19 Chromatography allowed isolation of 19-22 in diastereoisomeric diols 27 and 31, which were assigned the
good yield (Scheme 9). relative anti-configurations on the basis of the reaction proceed-
The oxidation of N-R-methylbenzyl protected dihydro- ing via epoxidation followed by stereospecific SN2-type epoxide
azepine 19 with m-CPBA (2 equiv) in the presence of ring-opening.20 Under identical reaction conditions, oxidation
of 2-naphthyl substrate 20 and 2-tolyl substrate 22 gave incom-
plete conversion to chromatographically inseparable 65:35
(18) Enantiopure (R)-R-methylbenzylamine (99% ee), (R)-1-(20 -naphthyl)-
ethylamine, and (R)-1-(10 -naphthyl)ethylamine (98% ee) are commercially avail-
able. A racemic sample of 1-(20 -tolyl)ethylamine was prepared according to the (20) The absolute configurations within the major diastereoisomeric diols
procedure outlined by Li et al. for the preparation of (RS)-1-(p-xylyl)- 27-30 resulting from these oxidation reactions were tentatively assigned as
ethylamine; see: Li, Y.; Selvaratnam, S.; Vittal, J. J.; Leung, P.-H. Inorg. Chem. (3S,4S,10 R) by analogy to the stereochemical outcome observed upon
2003, 42, 3229. oxidation of 21 with HBF4 in the presence of BnOH, which gave the
(19) Peak overlap and the presence of unidentified impurities in the 1H corresponding (3S,4S,10 R)-diastereoisomer 38 as the major product, the
NMR spectra of the crude reaction mixtures precluded the determination of the stereochemistry of which was unambiguously established by single-crystal
dihydroazepine to dihydropyrrole product ratios. Dihydropyrroles 23-26 were X-ray analyses of derivatives and by chemical correlation to (þ)-1-deoxy-
formed as single diastereoisomers of unknown relative configuration. altronojirimycin 3.
a
Reagents and conditions: (i) HBF4 3 OEt2, m-CPBA (4 equiv), BnOH,
rt, 24 h; (ii) MsCl, Et3N, CH2Cl2, 0 °C, 1.5 h. 1-Nap = 1-naphthyl.
a
38 was assigned by subsequent chemical correlation with 35,
Reagents and conditions: (i) Cl3CCO2H, m-CPBA, CH2Cl2, rt, while the absolute configuration within 39 was assigned as the
21 h, then NaOH (2 M, aq). 2-Nap = 2-naphthyl. 1-Nap = 1-naphthyl.
2-Tol = 2-tolyl. alternative stereochemical outcome resulting from an epoxida-
tion step followed by an SN2-type ring-opening process. These
mixtures of the corresponding anti-diols 28 and 32 and 30 and observations are consistent with the N-1-(10 -napthyl)ethyl group
34, respectively. Meanwhile, oxidation of 1-naphthyl substrate within 21 promoting the diastereoselective formation of the
21 gave incomplete conversion to an 80:20 mixture of the intermediate epoxide 37 (in 80:20 dr), which undergoes ring-
diastereoisomeric anti-diols 29 and 33, with chromatographic opening via an SN2-type process upon attack of BnOH exclu-
purification allowing partial separation of the diastereo- sively at C(4), which is both distal to the protonated nitrogen
isomers20,21 (Scheme 10). atom26 and an activated allylic position,27 to give an 80:20
Encouraged by these initial results, the oxidation of 1-naphthyl mixture of the monobenzyl protected diols 38 and 39. Mean-
substrate 21 was further investigated. Using our previously while, oxidation of the 80:20 mixture of 38 and 39 using aqueous
optimized conditions for N-benzyl protected dihydroazepine HBF414 and m-CPBA gave a crude reaction mixture containing
4, oxidation of 21 with 4 equiv of m-CPBA in the presence of 35 and a diastereoisomeric compound (assigned as 40) in the
5 equiv of HBF4 3 OEt2 in BnOH22 gave quantitative conversion ratio of 80:20, along with ∼10% of other unidentifiable pro-
to a mixture of products with the functionalized tetrahydro- ducts. Purification gave an 80:20 mixture of 35:40 in 60%
azepine 35 as the major component.23 Purification gave a sample combined yield. This result suggests that 38 undergoes comple-
of 35 in 25% isolated yield and >99:1 dr. Treatment of 35 with tely diastereoselective epoxidation to give 35 and that 39 under-
MsCl in the presence of Et3N effected ring contraction to give goes completely diastereoselective epoxidation to give 40. Given
piperidine 36 as a single diastereoisomer, which was isolated in that the analogous tetrahydroazepine 5 bearing an N-benzyl
66% yield. Single-crystal X-ray analysis of 36 unambiguously group (which is incapable of promoting a diastereoselective
established its relative configuration, with the absolute (2S,3R, reaction) undergoes highly diastereoselective epoxidation to give
4S,5S,10 R)-configuration being assigned from the known 8 as a single diastereoisomer, these studies imply that the
(R)-stereocenter of the 1-(10 -naphthyl)ethyl group.24 Therefore, configuration of the N-1-(10 -naphthyl)ethyl group has little or
the absolute (3S,4R,5S,6S,10 R)-configuration within 35 could no effect in determining the facial selectivity of the epoxidation
also be confidently assigned. Given the known enantiomeric reaction of 38 and 39 (Scheme 12).
purity of the (R)-(þ)-1-(10 -naphthyl)ethylamine (98% ee)18 used There are four potential sites for oxidation within the ammo-
in the preparation of 21, the enantiomeric purities of 35 and 36 nium ions derived from dihydroazepines 19-22: two olefins,
can be inferred as 98% ee (Scheme 11). each with two faces. Assuming that the chiral N-protecting
In order to investigate the origin of selectivity in the oxidation group acts as a steric block to oxidation at one of these sites, with
of 21, the reaction was repeated using only 1.5 equiv of oxidation occurring with equal probability at the remaining
m-CPBA, which gave incomplete (∼50%) conversion to a three sites, then the maximum diastereoselectivity attainable in
chromatographically inseparable 80:20 mixture of two diastereo- this scenario would be 67:33 (two of the three epoxides becom-
isomeric diols 38 and 39, which were isolated in 32% combined ing equivalent on deprotonation during basic aqueous workup):
yield and 80:20 dr.25 The absolute configuration within essentially the same as observed experimentally for oxidation of
R-methylbenzyl 19, 2-naphthyl 20, and 2-tolyl 22 with m-CPBA
in the presence of Cl3CCO2H. In the case of 1-naphthyl 21,
(21) We also investigated N(1)-(10 -phenyl-20 -hydroxyethyl)-2,7-dihydro-
1H-azepine (derived from the reaction of phenylglycinol with dibromide 18)
however, an enhanced diastereoselectivity of 80:20 is noted for
as a potential substrate for the oxidation reaction, but all conditions both oxidation under m-CPBA/Cl3CCO2H conditions and
investigated returned only starting material or gave rise to a complex mixture m-CPBA/HBF4/BnOH conditions and therefore these results
of products.
(22) Unlike N-benzyl-substituted dihydroazepine 4, N-1-(10 -naphthyl)- cannot be accounted for solely by invoking a sterically driven
ethylamine-substituted dihydroazepine 21 was freely soluble in BnOH and process. Examination of the X-ray crystal structure of
therefore the addition of CH2Cl2 to the reaction mixture to aid solubility was 1-naphthyl 2128 reveals that the 7-membered ring preferentially
not necessary; this modification to the experimental conditions did not affect
the reaction diastereoselectivity or efficiency.
(23) Following preparation of an authentic sample, 40 could also be (26) Parker, R. E.; Isaacs, N. S. Chem. Rev. 1959, 59, 737. Addy, J. K.;
identified as a constituent of the crude reaction mixture; the ratio of 35:40 was Parker, R. E. J. Chem. Soc. 1963, 915.
∼80:20. (27) Smith, M. B.; March, J. March’s Advanced Organic Chemistry -
(24) Crystallographic data (excluding structure factors) have been depos- Reactions, Mechanisms, and Structure, 5th ed.; John Wiley & Sons, Inc.:
ited with the Cambridge Crystallographic Data Centre as supplementary New York, 2001; p 434.
publication number CCDC 788733. (28) Crystallographic data (excluding structure factors) have been depos-
(25) Attempts to drive the reaction to full conversion of starting material ited with the Cambridge Crystallographic Data Centre as supplementary
resulted in overoxidation to 35 and 40. publication number CCDC 788732.
SCHEME 13a
a
Reagents and conditions: (i) Cl3CCO2H, CH2Cl2, rt, 16 h, then
NaOH (2 M, aq); (ii) AgOAc, DMF, 65 °C, 24 h. 1-Nap =1-naphthyl.
a
Reagents and conditions: (i) HBF4 3 OEt2, m-CPBA (1.5 equiv),
BnOH, rt, 24 h; (ii) HBF4 (40% w/w in H2O), m-CPBA (4 equiv), or no part in determining the facial selectivity of the ensuing
BnOH, rt, 24 h. 1-Nap =1-naphthyl. epoxidation reaction. Given the known conformational lability
of 7-membered rings,29a the origin of selectivity in the second
adopts an envelope-type conformation, with a near planar epoxidation reaction of 5 to 8, of 38 to 35, and of 39 to 40 is not
arrangement of the six carbon atoms and the nitrogen atom clear, although it may plausibly be a result of hydrogen-bonded
out of this plane. The chiral N-protecting group adopts a pseudo- delivery of the oxidant, or epoxidation on the sterically most
axial position, with the C(10 )-H bond oriented directly over the accessible face, or a combination of both factors.
7-membered ring. 1H NMR NOE data was supportive of an With a diastereo- and enantiomerically pure sample of
analogous solution phase conformation for ammonium ion 41 piperidine epoxide 36 in hand, its elaboration to (þ)-1-deoxy-
(prepared from 21 and F3CCO2H in CDCl3). When viewed as altronojirimycin 3 was pursued. Epoxide ring-opening upon
a Newman projection along the N-C(10 ) bond, the 1-naphthyl treatment of 36 with Cl3CCO2H proceeded exclusively at C(5)
group shields the lower face of one of the olefins, presumably to give trichloroacetate ester 42 as a single regio- and diastereo-
blocking reaction at this site. The origin of enhanced diastereo- isomer, that could be isolated as a 98:2 mixture of 42:43 upon
selectivity in this case is consistent with selective epoxidation workup using 0.1 M aqueous NaHCO3, followed by recrystal-
occurring on the face of the olefin away from the 1-naphthyl lization of the crude reaction mixture. Alternatively, workup
group within ammonium ion 41, with a rate acceleration for using 2 M aqueous NaOH effected ester hydrolysis to give diol
oxidation at this site promoted by stabilization of the partial 43 directly in 67% yield, >99:1 dr and 98% ee.30 The relative
positive charges that develop within the epoxidation transition configuration within 43 was assigned on the basis of 1H NMR
state on the olefinic carbons by interaction with the π-system of 3
J coupling constant analysis. Subsequent treatment of 43 with
the 1-naphthyl group, resulting in the production of epoxide 37 AgOAc in DMF at 65 °C resulted in conversion to a 90:10
as the major product (Figure 1). Regioselective ring-opening of mixture of piperidine 45 and azepane 46. Chromatography
37 by BnOH gives 38, with further diastereoselective oxidation facilitated separation of this mixture, giving piperidine 45 in
of 38 giving 35. It may be assumed that after the first oxidation, 60% yield, >99:1 dr and 98% ee,30 and azepane 46 in 7% yield,
the conformation of the 7-membered ring switches from >99:1 dr and 98% ee.30 This product distribution is consistent
envelope-like to chairlike,29 with the N-1-(10 -naphthyl)ethyl with the reaction proceeding via the intermediacy of the
group occupying a pseudoequatorial position, being some- aziridinium ion 44, which may undergo ring-opening at either
what remote from the reaction site and therefore playing little the least substituted site (leading to piperidine 45) or the more
substituted site (leading to azepane 46). The relative configuration
(29) Both cycloheptene and cycloheptene oxide have been shown to favor
a chairtype conformation in solution; see: (a) Leong, M. K.; Mastryukov,
V. S.; Boggs, J. E. J. Mol. Struct. 1998, 445, 149. (b) Abraham, R. J.; (30) The enantiomeric purities of 43, 45, and 46 were inferred from the
Castellazzi, I.; Sancassan, F.; Smith, T. A. D. J. Chem. Soc., Perkin Trans. 2 enantiomeric purity of the (R)-(þ)-1-(10 -naphthyl)ethylamine used to pre-
1999, 99. pare the original starting material 21 (i.e., 98% ee).
SCHEME 14a Elix UV-10 system. Organic solvents were used as supplied
(analytical or HPLC grade) without prior purification. Organic
layers were dried over MgSO4. Thin layer chromatography was
performed on aluminum plates coated with 60 F254 silica. Plates
were visualized using UV light (254 nm), iodine, 1% aqueous
KMnO4, or 10% ethanolic phosphomolybdic acid. Flash
column chromatography was performed either on Kieselgel
60 silica on a glass column or on an automated flash column
chromatography platform.
Melting points are uncorrected. IR spectra were recorded as
a
either a thin film on NaCl plates (film) or a KBr disk (KBr), as
Reagents and conditions: (i) K2CO3, MeOH, rt, 16 h; (ii) H2, stated. Selected characteristic peaks are reported in cm-1. NMR
Pd(OH)2/C, MeOH, rt, 72 h, then HCl (aq). 1-Nap =1-naphthyl.
spectra were recorded in the deuterated solvent stated. The field
was locked by external referencing to the relevant deuteron
within 45 was unambiguously established by single crystal resonance. 1H-1H COSY and 1H-13C HMQC analyses were
X-ray analysis, with the absolute (2R,3R,4S,5R,10 R)-config- used to establish atom connectivity.
uration being assigned from the known (R)-stereocenter of the (þ)-(S,S)- and (-)-(R,R)-N(1)-Benzyl-3-hydroxy-4-benzyloxy-
1-(10 -naphthyl)ethyl group.31 This analysis also affirms the 2,3,4,7-tetrahydro-1H-azepine 5. HBF4 3 Et2O (4.07 mL, 29.7 mmol)
relative configurations within 42 and 43 (and thereby also was added to a stirred solution of 4 (1.1 g, 5.94 mmol) in BnOH/
confirms the regioselectivity of the ring-opening reaction). CH2Cl2 (v/v 2:1, 33 mL), and the resultant solution was stirred for 5
min at rt. m-CPBA (75%, 2.05 g, 8.91 mmol) was then added, and
The absolute (3R,4S,5S,6R,10 R)-configuration within azepane
the resultant solution was stirred for 24 h before the addition of solid
46 was assigned on the basis of an SN2-type ring-opening of Na2SO3 (∼4 g) until starch-iodide paper indicated that no oxidant
aziridinium 44 at the more substituted site (Scheme 13). remained. The mixture was diluted with CH2Cl2 (50 mL), and the
Finally, treatment of piperidine 45 with K2CO3 in MeOH organic layer was washed with 2 M aq NaOH (2 100 mL). The
was followed by hydrogenolysis of the resultant triol 47 to combined aqueous washings were extracted with CHCl3/iPrOH (v/v
give (þ)-1-deoxyaltronojirimycin (3), which was isolated as 3:1, 2 100 mL), and the combined organic extracts were dried and
its hydrochloride salt (þ)-3 3 HCl in 71% yield (two steps) concentrated in vacuo. The residue was then passed through a
and >99:1 dr (Scheme 14). The spectroscopic data of our Biotage SCX-2 scavenger column, eluting first with CH2Cl2/MeOH
sample of (þ)-3 3 HCl were in excellent agreement with those (v/v 1:1), then 2 M NH3 in MeOH. The ammonia-containing eluent
previously reported in the literature {[R]25D þ31.1 (c 0.5 in was concentrated in vacuo to give a 75:25 mixture of 5:8. Purification
via flash column chromatography (gradient elution, 5f50% EtOAc
MeOH); lit.9i [R]23D þ31.0 (c 2.0 in MeOH); lit.10f [R]25D
in 30-40 °C petroleum ether) gave (()-5 as a colorless oil (580 mg,
þ33.2 (c 0.5 in MeOH)}. Given the enantiomeric purity of 32%, >99:1 dr):33 IR νmax (film) 3450 (O-H), 3028, 2918 (C-H);
the (R)-(þ)-1-(10 -naphthyl)ethylamine (98% ee) used in the NMR δH (400 MHz, CDCl3) 2.88 (1H, dd, J = 13.4, 7.6 Hz,
preparation of 21, the enantiomeric purity of (þ)-3 3 HCl C(7)HA), 3.04-3.29 (4H, m, C(2)H2, C(7)HB, OH), 3.69 (2H, AB
prepared in this manner can be inferred as 98% ee. system, NCH2Ph), 3.85 (1H, td, J = 7.0, 3.7 Hz, C(3)H), 4.30-4.35
(1H, m, C(4)H), 4.55 (1H, d, J = 11.6 Hz, OCHAHBPh), 4.73 (1H,
Conclusion d, J = 11.6 Hz, OCHAHBPh), 5.69-5.84 (2H, m, C(5)H, C(6)H),
7.25-7.42 (10H, m, Ph); NMR δC (100 MHz, CDCl3) 55.0 (C(7)),
In conclusion, oxidation of enantiomerically pure (R)-N(1)- 59.3 (C(2)), 62.5 (NCH2Ph), 69.6 (C(3)), 71.6 (OCH2Ph), 80.2
10 -(100 -naphthyl)ethyl-2,7-dihydro-1H-azepine with m-CPBA in (C(4)), 127.2, 127.8, 127.9, 128.4, 128.5, 128.9 (o,m,p-Ph), 130.0,
the presence of HBF4 and BnOH gave (3S,4R,5S,6S,10 R)-N(1)- 130.1 (C(5), C(6)), 138.1, 138.6 (i-Ph); MS m/z (ESIþ) 332 ([M þ
10 -(100 -naphthyl)ethyl-3-hydroxy-4-benzyloxy-5,6-epoxyaze- Na]þ, 40%), 310 ([M þ H]þ, 100); HRMS (ESIþ) C20H24NO2þ
pane as the major product and as a single diastereoisomer ([M þ H]þ) requires 310.1802, found 310.1798. Preparative chiral
after chromatography. Elaboration of this highly functiona- HPLC (Chiralpak AD-H [250 21.2 mm (i.d.)], mobile phase:
lized intermediate via ring contraction to (2S,3R,4S,5S,10 R)- MeOH/EtOH [v/v 1:1]) gave (-)-(R,R)-5 as a colorless oil (240 mg,
13% from 4, >99:1 dr, >99% ee): [R]25D -81.8 (c 1.0 in CHCl3).
N(1)-benzyl-2-chloromethyl-3-benzyloxy-4,5-epoxypiperidine
Anal. Calcd for C20H23NO2: C, 77.6; H, 7.5; N, 4.5. Found: C, 77.8;
followed by regioselective epoxide ring-opening, functional H, 7.7; N, 4.6. Further elution gave (þ)-(S,S)-5 as a colorless oil
group manipulation, and deprotection gave (þ)-1-deoxyaltro- (280 mg, 15% from 4, >99:1 dr, >99% ee): [R]25D þ82.8 (c 1.0 in
nojirimycin. Alternatively, resolution of (RS,RS)-N(1)-benzyl- CHCl3).
3-hydroxy-4-benzyloxy-2,3,4,7-tetrahydro-1H-azepine or (þ)-(S,S)-N(1)-Benzyl-2-chloromethyl-3-benzyloxy-1,2,3,6-tetra-
(3RS,4SR,5RS,6RS)-N(1)-benzyl-3-hydroxy-4-benzyloxy- hydropyridine 6. Et3N (0.35 mL, 2.49 mmol) and MsCl (0.14 mL,
5,6-epoxyazepane by preparative chiral HPLC and subsequent 1.87 mmol) in CH2Cl2 (12 mL) were added sequentially to a stirred
elaboration allows access to the enantiomers of 1-deoxynojirimy- solution of (þ)-(S,S)-5 (385 mg, 1.24 mmol) in CH2Cl2 (12 mL) at
cin and 1-deoxyaltronojirimycin, respectively. 0 °C, and the resultant solution was stirred for 1.5 h at 0 °C before
being concentrated in vacuo. Purification via flash column chroma-
tography (gradient elution, 2f20% EtOAc in 30-40 °C petroleum
Experimental Section
ether) gave (þ)-(S,S)-6 as a yellow oil (304 mg, 75% >99:1 dr,
General Experimental Details. m-CPBA was supplied as a >99% ee):33 [R]25D þ49.6 (c 1.0 in CHCl3); IR νmax (film) 3063,
70-77% slurry in water and titrated according to the procedure 3031, 2868 (C-H); NMR δH (400 MHz, CDCl3) 3.09-3.13 (2H, m,
of Swern32 immediately before use. Water was purified by an C(6)H2), 3.15-3.21 (1H, m, C(2)H), 3.55 (1H, dd, J 11.3,
8.0, CHAHBCl), 3.71 (1H, d, J = 13.1 Hz, NCHAHBPh), 3.88
(1H, dd, J= 11.3, 3.5 Hz, CHAHBCl), 4.01 (1H, d, J=13.1 Hz,
(31) Crystallographic data (excluding structure factors) have been depos-
ited with the Cambridge Crystallographic Data Centre as supplementary
publication no. CCDC 788734. (33) The synthesis of racemic 5-8 and 12 has previously been reported by
(32) Swern, D. Org. React. 1953, VII, 392. us; see ref 8.
and the combined organic extracts were dried and concentrated ether) gave (þ)-(2R,3S,4S,5R)-12 as a colorless oil (65 mg, 50%,
in vacuo. The residue was then passed through a Biotage SCX-2 99:1 dr):12 [R]25D þ26.9 (c 1.0 in CHCl3).
scavenger column, eluting first with CH2Cl2/MeOH (v/v 1:1) (þ)-(2R,3R,4R,5S)-2-Hydroxymethyl-3,4,5-trihydroxypiperidine
and then 2 M NH3 in MeOH. The ammonia-containing eluent Hydrochloride [(þ)-1-Deoxynojirimycin hydrochloride] 2 3 HCl. Step
was concentrated in vacuo. Purification via flash column chro- 1. AgOAc (21 mg, 0.12 mmol) was added to a stirred solution of
matography (eluent 30-40 °C petroleum ether/EtOAc, 2:1) gave (-)-(2S,3R,4R,5S)-12 (30 mg, 0.08 mmol) in DMF (1 mL), and the
(-)-(3R,4S,5R,6R)-8 as a beige solid (35 mg, 61%, >99:1 dr, resultant suspension was stirred for 24 h at 65 °C. The reaction mixture
>99% ee): [R]25D -54.3 (c 1.0 in CHCl3). was then allowed to cool to rt before the addition of H2O
(RS,RS)-1-Benzyl-2,3,4,7-tetrahydro-1H-azepine-3,4-diol 11. (4 mL). The mixture was extracted with Et2O (3 2 mL), and the
Cl3CCO2H (1.54 g, 9.45 mmol) was added to a stirred solution combined organic extracts were washed with H2O (3 6 mL) before
of 4 (350 mg, 1.89 mmol) in CH2Cl2 (3.5 mL), and the resultant being dried and concentrated in vacuo to give (2R,3R,4R,5S)-13 as a
solution was stirred for 5 min at rt. m-CPBA (75%, 870 mg, 3.78 colorless oil (20 mg) that was used without purification.
mmol) was then added, and the resultant solution was stirred for Step 2. K2CO3 (143 mg, 1.04 mmol) was added to a stirred
21 h before the addition of solid Na2SO3 (∼500 mg) until starch- solution of (2R,3R,4R,5S)-13 (20 mg, 0.05 mmol) in MeOH
iodide paper indicated that no oxidant remained. The mixture (1 mL), and the resultant suspension was stirred for 16 h at rt
was diluted with CH2Cl2 (50 mL), and the organic layer was before being concentrated in vacuo. The residue was dissolved in
washed with 2 M aq NaOH (2 50 mL). The combined aqueous H2O (2 mL) and extracted with CHCl3/iPrOH (v/v 3:1, 2
washings were extracted with CH2Cl2 (2 50 mL), and the 2 mL). The combined organic extracts were dried and concen-
combined organic extracts were dried and concentrated in vacuo. trated in vacuo to give (2R,3R,4R,5S)-14 as a colorless oil
Purification via flash column chromatography (eluent 40-60 °C (18 mg) that was used without purification.
petroleum ether/EtOAc, 1:1) gave 11 as a colorless oil (124 mg, Step 3. Pd(OH)2/C (18 mg) was added to a stirred solution of
30%, >99:1 dr): IR νmax (film) 3385 (O-H), 3079, 2985, 2971 (2R,3R,4R,5S)-14 (18 mg, 0.05 mmol) in degassed MeOH
(C-H); NMR δH (400 MHz, CDCl3) 2.68 (1H, dd, J = 13.4, 7.3 (0.5 mL), and the resultant suspension was stirred for 72 h under
Hz, C(7)HA), 3.09-3.19 (3H, m, C(7)HB, C(2)H2), 3.64 (2H, A2 an atmosphere of H2 (1 atm). Concentrated aq HCl (10 μL) was
system, NCH2Ph), 3.70-4.00 (3H, m, C(3)H, 2 OH), 4.36 (1H, then added, and the resultant suspension was stirred for a further
ddd, J = 7.3, 3.7, 1.4 Hz, C(4)H), 5.60-5.69 (1H, m, C(6)H), 5 min before being filtered through Celite (eluent MeOH). The
5.73-5.80 (1H, m, C(5)H), 7.23-7.36 (5H, m, Ph); δC (100 MHz, filtrate was concentrated in vacuo, and the residue was purified via
CDCl3) 55.4 (C(7)), 60.1 (C(2)), 62.1 (NCH2Ph), 71.4 (C(3)), 73.1 flash column chromatography (eluent CHCl3/MeOH, 4:1) to give
(C(4)), 127.4, 128.5, 128.8, 129.0 (o,m,p-Ph, C(6)), 132.1 (C(5)), (þ)-(2R,3R,4R,5S)-2 3 HCl as a white semisolid (5 mg, 30% from
138.2 (i-Ph); MS m/z (ESIþ) 220 ([M þ H]þ, 100); HRMS (ESIþ) (-)-(2S,3R,4R,5S)-12, 99:1 dr, >99% ee):2 [R]25D þ31.0 (c 0.45 in
C13H18NO2þ [M þ H] þ requires 220.1332, found 220.1332. H2O) [lit.3a for natural sample [R]22D þ38.0 (c 1.0 in H2O); lit.9m
(-)-(2S,3R,4R,5S)-N(1)-Benzyl-2-chloromethyl-3-benzyloxy- [R]23D þ36.9 (c 1.1 in H2O)]; NMR δH (500 MHz, D2O) 3.17 (1H,
4,5-dihydroxypiperidine 12. Cl3CCO2H (1.05 g, 6.40 mmol) dd, J = 13.6, 2.8 Hz, CHAHBOH), 3.27-3.35 (2H, m, C(4)H,
was added to a stirred solution of (þ)-(2S,3R,4R,5R)-7 (110 CHAHBOH), 3.77 (1H, dd, J = 12.6, 6.9 Hz, C(6)HA), 3.91 (1H, dd,
mg, 0.32 mmol) in CH2Cl2 (1.1 mL), and the resultant solution was J = 12.6, 3.2 Hz, C(6)HB), 3.94-3.99 (2H, m, C(3)H, C(5)H), 4.08-
stirred for 16 h at rt. The reaction mixture was diluted with CH2Cl2 4.11 (1H, br m, C(2)H); NMR δC (125 MHz, D2O) 43.9 (CH2OH),
(10 mL) and then washed with 2 M aq NaOH (2 50 mL). The 55.8 (C(4)), 58.2 (C(6)), 63.6 (C(5)), 66.2 (C(2)), 68.5 (C(3)).
combined aqueous layers were extracted with CHCl3/iPrOH (v/v (-)-(2S,3S,4S,5R)-2-Hydroxymethyl-3,4,5-trihydroxypiperi-
3:1, 2 20 mL). The combined organic extracts were dried and dine Hydrochloride [(-)-1-Deoxynojirimycin Hydrochloride]
concentrated in vacuo. Purification via flash column chromatogra- (2 3 HCl). Step 1. AgOAc (21 mg, 0.12 mmol) was added to a
phy (gradient elution, 10f100% EtOAc in 30-40 °C petroleum stirred solution of (þ)-(2R,3S,4S,5R)-12 (30 mg, 0.08 mmol) in
ether) gave (-)-(2S,3R,4R,5S)-12 as a colorless oil (59 mg, 51%, DMF (1 mL), and the resultant suspension was stirred for 24 h at
99:1 dr, >99% ee):33 [R]25D -26.7 (c 1.0 in CHCl3); IR νmax (film) 65 °C. The reaction mixture was then allowed to cool to rt before
3387 (O-H) 3030, 2915 (C-H); NMR δH (400 MHz, CDCl3) 1.98 the addition of H2O (4 mL). The mixture was extracted with
(1H, app t, J = 10.5 Hz, C(6)HA), 2.45 (1H, app d, J = 8.8 Hz, Et2O (3 2 mL), and the combined organic extracts were
C(2)H) 2.78 (1H, br s, OH), 2.97-3.05 (2H, m, C(6)HB, OH), 3.16 washed with H2O (3 6 mL) before being dried and concen-
(1H, d, J = 12.9 Hz, NCHAHBPh), 3.42-3.66 (3H, m, C(3)H, trated in vacuo to give (2S,3S,4S,5R)-13 as a colorless oil
C(4)H, C(5)H), 4.00 (1H, dd, J = 12.6, 3.3 Hz, CHAHBCl), 4.09 (20 mg) that was used without purification.
(1H, dd, J = 12.6, 1.8 Hz, CHAHBCl), 4.14 (1H, d, J = 12.9 Hz, Step 2. K2CO3 (143 mg, 1.04 mmol) was added to a stirred
NCHAHBPh), 4.82 (1H, d, J = 11.4 Hz, OCHAHBPh), 4.90 (1H, d, solution of (2S,3S,4S,5R)-13 (20 mg, 0.05 mmol) in MeOH
J = 11.4 Hz, OCHAHBPh), 7.25-7.43 (10H, m, Ph); NMR δC (100 (1 mL), and the resultant suspension was stirred for 16 h at rt
MHz, CDCl3) 41.8 (CH2Cl), 54.7 (C(6)), 56.0 (NCH2Ph), 64.6 before being concentrated in vacuo. The residue was dissolved
(C(2)), 69.7 (C(5)), 75.3 (OCH2Ph), 78.7 (C(4)), 79.3 (C(3)), in H2O (2 mL) and extracted with CHCl3/iPrOH (v/v 3:1, 2
127.4, 128.0, 128.1, 128.4, 128.7, 129.2 (o,m,p-Ph), 137.7, 138.2 2 mL). The combined organic extracts were dried and concen-
(i-Ph); MS m/z (ESIþ) 386 ([M þ Na]þ, 37Cl, 34), 384 ([M þ Na]þ, trated in vacuo to give (2S,3S,4S,5R)-14 as a colorless oil
35
Cl, 100), 364 ([MþH]þ, 37Cl, 24), 362 ([M þ H]þ, 35Cl, 64); (18 mg) that was used without purification.
HRMS (ESIþ) C20H2535ClNO3þ ([M þ H]þ) requires 362.1517, Step 3. Pd(OH)2/C (18 mg) was added to a stirred solution of
found 362.1516. (2S,3S,4S,5R)-14 (18 mg, 0.05 mmol) in degassed MeOH
(þ)-(2R,3S,4S,5R)-N(1)-Benzyl-2-chloromethyl-3-benzyloxy- (0.5 mL), and the resultant suspension was stirred for 72 h under
4,5-dihydroxypiperidine 12. Cl3CCO2H (1.19 g, 7.27 mmol) an atmosphere of H2 (1 atm). Concentrated aq HCl (10 μL) was
was added to a stirred solution of (-)-(2R,3S,4S,5S)-7 (125 then added, and the resultant suspension was stirred for a
mg, 0.36 mmol) in CH2Cl2 (1.3 mL), and the resultant solution was further 5 min before being filtered through Celite (eluent
stirred for 16 h at rt. The reaction mixture was diluted with CH2Cl2 MeOH). The filtrate was concentrated in vacuo, and the residue
(10 mL) and then washed with 2 M aq NaOH (2 50 mL). The was purified via flash column chromatography (eluent CHCl3/
combined aqueous layers were extracted with CHCl3/iPrOH (3:1, MeOH, 4:1) to give (-)-(2S,3S,4S,5R)-2 3 HCl as a white semi-
2 20 mL). The combined organic extracts were dried and solid (6 mg, 36% from (þ)-(2R,3S,4S,5R)-12, 99:1 dr, >99%
concentrated in vacuo. Purification via flash column chromatogra- ee):2 [R]25D -34.0 (c 0.45 in H2O) [lit.9i [R]25D -46.0 (c 1.3 in
phy (gradient elution, 10f100% EtOAc in 30-40 °C petroleum H2O); lit.9m [R]24D -38.7 (c 1.0 in H2O)].
114.1 (CHdCH2), 123.8, 124.8, 125.2, 125.5, 125.6, 127.0, 127.1, was washed with 2 M aq NaOH (2 50 mL). The combined
128.8 (C(3), CHdCH2, Ar), 131.2, 134.0, 141.3, 141.9 (C(4), Ar); MS aqueous washings were extracted with CH2Cl2 (2 50 mL), and
m/z (ESIþ) 250 ([M þ H]þ, 100); HRMS (ESIþ) C18H20Nþ ([M þ the combined organic extracts were dried and concentrated in
H]þ) requires 250.1590, found 250.1594. Further elution gave 21 as a vacuo. The residue was then passed through a Biotage SCX-2
white solid (4.67 g, 75%, 98% ee): mp 55-57 °C; [R]25D þ76.4 (c 1.0 scavenger column, eluting first with CH2Cl2/MeOH (v/v 1:1) and
in CHCl3); IR νmax (film) 3008, 2974, 2934, 2874, 2791 (C-H); then 2 M NH3 in MeOH. The ammonia-containing eluent was
NMR δH (400 MHz, CDCl3) 1.72 (3H, d, J = 6.8 Hz, C(10 )Me), concentrated in vacuo to give a 65:35 mixture of 27:31. Purification
3.54 (2H, dd, J = 16.7, 3.5 Hz, C(2)HA, C(7)HA), 3.74 (2H, dd, J = via flash column chromatography (eluent 40-60 °C petroleum
16.7, 3.5 Hz, C(2)HB, C(7)HB), 4.72 (1H, m, C(10 )H), 5.70 (2H, dt, ether/EtOAc, 2:1) gave a 65:35 mixture of 27:31 as a yellow oil (122
J = 8.8, 3.5 Hz, C(3)H, C(6)H), 5.97-6.06 (2H, m, C(4)H, C(5)H), mg, 35%): IR νmax (film) 3381 (O-H), 3084, 3061, 3027, 2981,
7.42-7.55 (3H, m, Ar), 7.66-7.73 (1H, m, Ar), 7.74-7.81 (1H, m, 2971 (C-H); NMR δC (100 MHz, CDCl3) 17.8, 18.1 (2 C(R)-
Ar), 7.84-7.94 (1H, m, Ar), 8.51 (1H, d, J = 4.3 Hz, Ar); NMR δC Me), 53.4, 54.1 (2 C(7)), 56.3, 56.6 (2 C(2)), 62.8, 63.0 (2
(100 MHz, CDCl3) 21.0 (C(10 )Me), 53.6 (C(2), C(7)), 55.2 (C(10 )), C(R)), 71.9, 72.1 (2 C(3)), 72.8, 73.0 (2 C(4)), 127.3, 127.5,
124.2, 125.2, 125.4, 125.6, 126.5, 127.2, 128.7 (Ar, C(4), C(5)), 131.7 128.5, 129.8, 130.0, 130.8 (2 C(5), 2 C(6), 2 o-,m-,p-Ph,),
(C(3), C(6)), 133.2, 134.0, 141.4 (Ar); MS m/z (ESIþ) 250 ([M þ 142.8, 143.1 (2 i-Ph); MS m/z (ESIþ) 234 ([M þ H]þ, 100);
H]þ, 100); HRMS (ESIþ) C18H20Nþ ([M þ H]þ) requires 250.1590, HRMS (ESIþ) C14H19NNaO2þ ([M þ Na]þ) requires 256.1308,
found 250.1593. Anla. Calcd for C, 86.7; H, 7.7; N, 5.6. Found: C, found 256.1304. Data for 27: NMR δH (400 MHz, CDCl3) 1.40
86.8; H, 7.6; N, 5.5. (3H, d, J = 6.6 Hz, C(10 )Me), 2.69 (1H, dd, J = 13.8, 6.7 Hz,
X-ray Crystal Structure Determination for 21. Data were collected C(2)HA), 3.08 (1H, m, C(2)HB), 3.19-3.24 (2H, m, C(7)H2),
using an Enraf-Nonius κ-CCD diffractometer with graphite-mono- 3.73-3.88 (2H, m, C(3)H, C(10 )H), 4.34-4.39 (1H, m, C(4)H),
chromated Mo KR radiation using standard procedures at 150 K. 5.61-5.71 (1H, m, C(6)H), 5.74-5.85 (1H, m, C(5)H), 7.24-7.40
The structure was solved by direct methods (SIR92); all non-hydro- (5H, m, Ph). Data for 31: δH (400 MHz, CDCl3) (selected peaks)
gen atoms were refined with anisotropic thermal parameters. Hydro- 2.63 (1H, dd, J = 13.6, 7.1 Hz, C(2)HA), 3.11-3.17 (1H, m,
gen atoms were added at idealized positions. The structure was C(2)HB), 3.25-3.28 (1H, m, C(7)HA), 4.26-4.34 (1H, m, C(7)HB).
refined using CRYSTALS.35 (3S,4S,rR)- and (R,R,R)-N(1)-10 -(100 -Naphthyl)ethyl-3,4-dihy-
X-ray crystal structure data for 21 [C18H19N]: M = 249.36, droxy-2,3,4,7-tetrahydro-1H-azepine (3S,4S,rR)-29 and (R,R,R)-33.
orthorhombic, space group P212121, a=7.1891(2) Å, b=7.2120(2) Cl3CCO2H (1.64 g, 10.0 mmol) was added to a stirred solution
Å, c = 27.0900(9) Å, V = 1404.56(7) Å3, Z = 4, μ = 0.068 mm-1, of 21 (500 mg, 2.01 mmol, 98% ee) in CH2Cl2 (5.0 mL), and the
colorless plate, crystal dimensions =0.12 0.14 0.19 mm3. A total resultant solution was stirred for 5 min at rt. m-CPBA (75%, 923 mg,
of 1860 unique reflections were measured for 5 < θ < 27 and 1860 4.02 mmol) was then added, and the resultant solution was stirred for
reflections were used in the refinement. The final parameters were 21 h before the addition of solid Na2SO3 (∼500 mg) until
wR2 = 0.118 and R1 = 0.084 [I > -3.0σ(I)]. starch-iodide paper indicated that no oxidant remained. The mix-
Crystallographic data (excluding structure factors) has been ture was diluted with CH2Cl2 (100 mL), and the organic layer was
deposited with the Cambridge Crystallographic Data Centre as washed with 2 M aq NaOH (2 100 mL). The combined aqueous
supplementary publication no. CCDC 788732. Copies of the washings were extracted with CHCl3/iPrOH (v/v 3:1, 2 100 mL),
data can be obtained, free of charge, on application to CCDC, and the combined organic extracts were dried and concentrated in
12 Union Road, Cambridge CB2 1EZ, UK (fax: þ44(0)-1223- vacuo. The residue was then passed through a Biotage SCX-2
336033 or e-mail: deposit@ccdc.cam.ac.uk). scavenger column, eluting first with CH2Cl2/MeOH (v/v 1:1), then
(RS)-N(1)-10 -(200 -Tolyl)ethyl-2,7-dihydro-1H-azepine 22. (RS)- 2 M NH3 in MeOH. The ammonia-containing eluent was concen-
1-(20 -Tolyl)ethylamine (600 mg, 4.44 mmol)18 and K2CO3 (1.23 g, trated in vacuo to give an 80:20 mixture of 29:33 (50% conversion
8.87 mmol) were added sequentially to a stirred solution of 18 from 21). Purification via flash column chromatography (eluent
(532 mg, 2.22 mmol) in THF (30 mL) at rt. After 24 h, the reaction 40-60 °C petroleum ether/EtOAc, 2:1) gave 29 as a yellow oil
mixture was concentrated in vacuo. The residue was dissolved in (51 mg, 9%, >99:1 dr, 98% ee): [R]25D þ27.4 (c 1.0 in CHCl3); IR
CH2Cl2 (30 mL) and filtered through Celite (eluent CH2Cl2), and νmax (film) 3385 (O-H), 3048, 2972 (C-H), 1658 (CdC); NMR δH
the filtrate was concentrated in vacuo. Purification via flash (400 MHz, CDCl3) 1.52 (3H, d, J 6.6, C(10 )Me), 2.77 (1H, dd, J =
column chromatography (gradient elution, 0f10% EtOAc in 13.9, 6.0 Hz, C(2)HA), 3.20 (1H, dd, J = 13.9, 5.0 Hz, C(2)HB),
30-40 °C petroleum ether) gave a 93:7 mixture of 22:26 as a pale 3.32-3.35 (2H, m, C(7)H2), 3.63 (1H, app q, J = 6.0 Hz, C(3)H),
yellow oil (405 mg, 86%): IR νmax (film) 2970 (C-H), 1605 4.31-4.36 (1H, m, C(4)H), 4.59 (1H, q, J = 6.6 Hz, C(10 )H),
(CdC); MS m/z (ESIþ) 214 ([M þ H]þ, 100); HRMS (ESIþ) 5.61-5.71 (2H, m, C(5)H, C(6)H), 7.43-7.56 (4H, m, Ar),
C15H20Nþ ([M þ Hþ]) requires 214.1590, found 214.1589. Data 7.77-7.78 (1H, m, Ar), 7.85-7.89 (1H, m, Ar), 8.32 (1H, d, J =
for 22: NMR δH (400 MHz, CDCl3) 1.31 (3H, d, J = 6.6 Hz, 8.5 Hz, Ar); NMR δC (100 MHz, CDCl3) 15.0 (C(10 )Me), 53.1
C(10 )Me), 2.33 (3H, s, ArMe), 3.48 (2H, dd, J = 17.2, 3.1 Hz, (C(7)), 56.8 (C(2)), 59.4 (C(10 )), 72.6, 72.7 (C(3), C(4)), 123.9, 124.4,
C(2)HA, C(7)HA), 3.64 (2H, dd, J = 17.2, 3.1 Hz, C(2)HB, 125.1, 125.6, 126.1, 128.1, 128.9 (Ar), 128.7, 130.5 (C(5), C(6)), 131.7,
C(7)HB), 4.19 (3H, q, J = 6.6 Hz, C(10 )H), 5.68 (2H, app dt, 134.1, 139.0 (Ar); MS m/z (ESIþ) 306 ([M þ Na]þ, 100), 284 ([M þ
J 12.4, 3.1, C(3)H, C(6)H), 5.90-6.00 (2H, m, C(4)H, C(5)H), H]þ, 85); HRMS (ESIþ) C18H22NO2þ ([MþH]þ) requires 284.1645,
7.08-7.15 (2H, m, Ar), 7.20 (1H, app t, J = 7.1 Hz, Ar), 7.56 (1H, found 284.1648. Further elution gave a 70:30 mixture of 29:33 as a
d, J = 7.6 Hz, Ar); NMR δC (100 MHz, CDCl3) 19.5 (ArMe), 20.8 yellow oil (116 mg, 29%). Further elution gave a 24:76 mixture of
(C(10 )Me), 53.4 (C(10 ), C(2), C(7)), 126.2 (C(4), C(5)), 126.1, 29:33 as a yellow oil (15 mg, 3%). Data for 33: NMR δH (400 MHz,
126.2, 126.5, 126.9, 130.3, 133.1 (C(3), C(6), Ar), 135.6, 143.6 (Ar). CDCl3) 1.52 (3H, d, J = 6.6 Hz, C(10 )Me), 2.80 (1H, dd, J = 13.9,
(3S,4S,rR)- and (R,R,R)-N(1)-r-Methylbenzyl-3,4-dihydroxy- 5.7 Hz, C(2)HA), 3.11 (1H, dd, J = 13.9, 4.7 Hz, C(2)HB), 3.38-3.42
2,3,4,7-tetrahydro-1H-azepine (3S,4S,rR)-27 and (R,R,R)-31. (2H, m, C(7)H2), 3.63-3.68 (1H, m, C(3)H), 4.26-4.32 (1H, m,
Cl3CCO2H (1.23 g, 7.5 mmol) was added to a stirred solution C(4)H), 4.61 (1H, q, J = 6.6 Hz, C(10 )H), 5.58-5.75 (2H, m, C(5)H,
of 19 (300 mg, 1.5 mmol, >99% ee) in CH2Cl2 (3.0 mL), and C(6)H), 7.40-7.60 (4H, m, Ar), 7.79 (1H, d, J = 7.9 Hz, Ar), 7.87
the resultant solution was stirred for 5 min at rt. m-CPBA (75%, (1H, d, J = 7.9 Hz, Ar), 8.26 (1H, d, J = 8.2 Hz, Ar); NMR δC (100
693 mg, 3.01 mmol) was then added, and the resultant solution was MHz, CDCl3) (selected peaks) 14.6 (C(10 )Me), 53.1 (C(10 )), 54.5
stirred for 21 h before the addition of solid Na2SO3 (∼500 mg) (C(2)), 54.7 (C(7)), 72.9 (C(4)), 73.2 (C(3)).
until starch-iodide paper indicated that no oxidant remained. The (3S,4R,5S,6S,10 R)-N(1)-10 -(100 -Naphthyl)ethyl-3-hydroxy-4-ben-
mixture was diluted with CH2Cl2 (50 mL), and the organic layer zyloxy-5,6-epoxyazepane 35. From 21. HBF4 3 Et2O (1.93 mL,
0.07 mmol) in CH2Cl2 (0.3 mL), and the resultant solution was (1H, m, Ar), 7.84-7.89 (1H, m, Ar), 8.67-8.74 (1H, m, Ar); NMR
stirred for 16 h at rt. The reaction mixture was diluted with CH2Cl2 δC (100 MHz, CDCl3) 11.4 (C(10 )Me), 21.1 (COMe), 45.1 (C(6)),
(5 mL) and then washed with 2 M aq NaOH (2 10 mL). The 52.4 (C(10 )), 57.4 (C(2)), 59.8 (CH2OAc) 67.0 (C(5)) 67.5 (C(4)),
combined aqueous layers were extracted with CHCl3/iPrOH (v/v 71.4 (OCH2Ph), 73.7 (C(3)), 124.0, 125.0, 125.3, 125.5, 125.9, 128.0,
3:1, 3 5 mL). The combined organic extracts were dried and 128.2, 128.4, 128.5, 128.7, 129.1, 131.9, 134.3, 137.2 (Ar, Ph), 170.7
concentrated in vacuo. Purification via flash column chromatog- (COMe); MS m/z (ESIþ) 472 ([M þ Na]þ, 100), 450 ([M þ H]þ,
raphy (gradient elution, 10f100% EtOAc in 30-40 °C petroleum 98); HRMS (ESIþ) C27H32NO5þ ([M þ H]þ) requires 450.2275,
ether) gave 43 as a colorless oil (20 mg, 67%, >99:1 dr, 98% ee): found 450.2273.
[R]25D -12.8 (c 1.0 in CHCl3); IR νmax (film) 3443 (O-H), 3089, X-ray Crystal Structure Determination for 45. Data were
3051, 3034, 2924, 2973 (C-H); NMR δH (400 MHz, CDCl3) 1.52 collected using an Enraf-Nonius κ-CCD diffractometer with
(3H, d, J = 6.8 Hz, C(10 )Me), 2.17 (1H, br s, OH), 2.23 (1H, dd, graphite-monochromated Mo KR radiation using standard
J = 12.2, 4.3 Hz, C(6)HA), 2.42 (1H, br, s, OH), 2.79 (1H, app d, procedures at 150 K. The structure was solved by direct methods
J = 12.2 Hz, C(6)HB), 3.21 (1H, app d, J = 7.8 Hz, C(2)H), (SIR92); all non-hydrogen atoms were refined with anisotropic
3.50-3.60 (1H, m, C(5)H), 3.92-4.00 (1H, m, C(4)H), 4.11-4.19 thermal parameters. Hydrogen atoms were added at idealized
(2H, m, C(3)H, CHAHBCl), 4.22-4.29 (1H, m, CHAHBCl), positions. The structure was refined using CRYSTALS.35
4.62-4.72 (2H, m, OCH2Ph), 5.04 (1H, q, J = 6.8 Hz, C(10 )H), X-ray crystal structure data for 45 [C27H31NO5]: M = 449.55,
7.31-7.58 (9H, m, Ar, Ph), 7.80 (1H, d, J = 7.6 Hz, Ar), 7.86 monoclinic, space group P21, a = 9.3063(2) Å, b = 9.0446(2) Å,
(1H, d, J = 8.1 Hz, Ar), 8.91 (1H, d, J = 8.3 Hz, Ar); NMR δC (100 c = 14.3162(3) Å, β = 103.4925(10)°, V = 1171.76(4) Å3, Z = 2,
MHz, CDCl3) 12.0 (C(10 )Me), 41.8 (CH2Cl), 45.3 (C(6)), 53.5 μ = 0.087 mm-1, colorless plate, crystal dimensions = 0.08
(C(10 )), 58.9 (C(2)), 67.1 (C(5)), 72.3 (C(4)), 75.3 (OCH2Ph), 0.13 0.17 mm3. A total of 2818 unique reflections were measured
77.3 (C(3)), 124.7, 124.9, 125.3, 125.7, 126.1, 128.1, 128.2, 128.5, for 5 < θ < 27 and 2818 reflections were used in the refinement.
128.7, 128.8 (o,m,p-Ph, Ar), 131.8, 134.2, 137.6, 137.7 (i-Ph, Ar); The final parameters were wR2 = 0.100 and R1 = 0.048 [I >
MS m/z (ESIþ) 450 ([M þ Na]þ, 37Cl, 6), 448 ([M þ Na]þ, 35Cl, 18), -3.0σ(I)].
428 ([M þ H]þ, 37Cl, 29), 426 ([M þ H]þ, 35Cl, 100); HRMS Crystallographic data (excluding structure factors) has been
(ESIþ) C25H2835ClNNaO3þ ([M þ Na]þ) requires 448.1650, found deposited with the Cambridge Crystallographic Data Centre as
448.1654. supplementary publication no. CCDC 788734. Copies of the
(2R,3R,4S,5R,10 R)-N(1)-10 -(100 -Naphthyl)ethyl-2-acetoxymethyl- data can be obtained, free of charge, on application to CCDC,
3-benzyloxy-4,5-dihydroxypiperidine 45 and (3R,4S,5S,6R,10 R)- 12 Union Road, Cambridge CB2 1EZ, UK (fax: þ44(0)-
N(1)-10 -(100 -Naphthyl)ethyl-3-acetoxy-4-benzyloxy-5,6-dihydroxy- 1223-336033 or e-mail: deposit@ccdc.cam.ac.uk).
azepane 46. AgOAc (89 mg, 0.54 mmol) was added to a stirred (þ)-(2R,3R,4S,5R)-2-Hydroxymethyl-3,4,5-trihydroxypiperi-
solution of 43 (65 mg, 0.15 mmol) in DMF (1 mL), and the dine Hydrochloride [(þ)-1-Deoxyaltronojirimycin Hydrochloride]
resultant suspension was stirred for 24 h at 65 °C. The reaction 3 3 HCl. Step1. K2CO3 (154 mg, 1.10 mmol) was added to a stirred
mixture was then allowed to cool to rt before the addition of H2O solution of 45 (25 mg, 0.06 mmol) in MeOH (1 mL) and the
(8 mL). The mixture was extracted with Et2O (3 10 mL), and the resultant suspension was stirred for 16 h at rt before being con-
combined organic extracts were washed with H2O (3 30 mL) centrated in vacuo. The residue was dissolved in H2O (5 mL)
before being dried and concentrated in vacuo. Purification via flash and extracted with CHCl3/iPrOH (v/v 3:1, 3 10 mL). The
column chromatography (gradient elution, 10f100% EtOAc in combined organic extracts were dried and concentrated in vacuo
30-40 °C petroleum ether) gave 46 as a colorless oil (5 mg, 7%, to give 47 as a colorless oil (17 mg) that was used without
>99:1 dr, 98% ee): [R]25D þ44.8 (c 1.0 in CHCl3); IR νmax (film) purification.
3444 (O-H), 3088, 3050, 3034, 3006, 2917, 2887 (C-H), 1734 Step 2. Pd(OH)2/C (17 mg) was added to a stirred solution of
(CdO); NMR δH (400 MHz, CDCl3) 1.48 (3H, d, J 6.6, C(10 )Me), 47 (17 mg, 0.05 mmol) in degassed MeOH (0.5 mL) and the
2.15 (3H, s, COMe), 2.21 (1H, br s, OH), 2.69 (1H, m, C(7)HA), resultant suspension was stirred for 72 h under an atmosphere of
2.80 (1H, dd, J = 14.0, 2.7 Hz, C(7)HB), 3.17-3.22 (1H, m, H2 (1 atm). Concentrated aq HCl (10 μL) was then added, and
C(2)HA), 3.27 (1H, dd, J = 14.8, 4.1 Hz, C(2)HB), 3.41-3.46 (1H, the resultant suspension was stirred for a further 5 min before
m, C(6)H), 3.87 (1H, app d, J = 6.3 Hz, C(4)H), 3.93-3.97 (1H, m, being filtered through Celite (eluent MeOH). The filtrate was
C(5)H), 4.49 (1H, d, J = 14.5 Hz, OCHAHBPh), 4.55-4.64 (2H, concentrated in vacuo, and the residue was purified via flash
m, C(10 )H, OCHACHBPh), 5.11-5.15 (1H, m, C(3)H), 7.25-7.50 column chromatography (eluent CHCl3/MeOH, 4:1) to give
(9H, m, Ar, Ph), 7.80 (1H, d, J = 7.6 Hz, Ar), 7.88 (1H, d, J = 7.9 (þ)-(2R,3R,4S,5R)-3 3 HCl as a white semisolid (8 mg, 71% from
Hz, Ar), 8.47 (1H, d, J = 8.5 Hz, Ar); NMR δC (100 MHz, CDCl3) 45, >99:1 dr, 98% ee):3 [R]25D þ31.1 (c 0.5 in MeOH) [lit.9i
11.5 (C(10 )Me), 21.4 (COMe), 52.3 (C(7)), 59.0 (C(2)), 60.6 (C(10 )), [R]23D þ31.0 (c 2.0 in MeOH); lit.10f [R]25D þ33.2 (c 0.5 in
70.5 (C(6)), 72.5 (OCH2Ph), 74.1 (C(5)), 74.4 (C(3)), 79.6 (C(4)), MeOH)]; NMR δH (500 MHz, D2O) 3.17 (1H, dd, J 13.6, 2.8,
124.5, 124.7, 125.1, 125.7, 126.0, 127.7, 127.9, 128.4, 128.6, 129.0, CHAHBOH), 3.27-3.35 (2H, m, C(4)H, CHAHBOH), 3.77 (1H,
131.4, 134.2, 138.1, 138.3 (Ar, Ph), 170.3 (COMe); MS m/z (ESIþ) dd, J 12.6, 6.9, C(6)HA), 3.91 (1H, dd, J 12.6, 3.2, C(6)HB),
472 ([M þ Na]þ, 41), 450 ([M þ H]þ, 100); HRMS (ESIþ) 3.94-3.99 (2H, m, C(3)H, C(5)H), 4.08-4.11 (1H, m, C(2)H);
C27H32NO5þ ([M þ H]þ) requires 450.2275, found 450.2273. NMR δC (125 MHz, D2O) 43.9 (CH2OH), 55.8 (C(4)), 58.2
Further elution gave 45 as a white solid (40 mg, 60%, >99:1 dr, (C(6)), 63.6 (C(5)), 66.2 (C(2)), 68.5 (C(3)).
98% ee): mp 129-131 °C; [R]25D þ48.3 (c 1.0 in CHCl3); IR νmax
(film) 3484 (O-H), 3088, 3050, 3035, 2924, 2881, 2850 (C-H), Acknowledgment. We thank Pfizer for a CASE student-
1741 (CdO); NMR δH (400 MHz, CDCl3) 3.01 (3H, d, J = 6.6 Hz, ship (P.M.S.), Laure Hitzel for assistance with preparative
C(10 )Me), 1.60 (1H, br s, OH), 2.19 (3H, s, COMe), 2.27 (1H, dd, Chiral HPLC, and the Oxford Chemical Crystallography
J = 12.3, 4.1 Hz, C(6)HA), 2.45 (1H, br s, OH), 2.81 (1H, dd, J =
Service for the use of their X-ray diffractometers.
12.3, 1.3 Hz, C(6)HB), 3.06-3.15 (1H, m, C(2)H), 3.53-3.61 (1H,
m, C(5)H), 3.90 (1H, dd, J = 9.1, 3.4 Hz, C(3)H), 4.01-4.07 (1H,
m, C(4)H), 4.51 (1H, d, J = 11.3 Hz, OCHAHBPh), 4.56 (1H, dd, Supporting Information Available: Copies of 1H and 13C
J = 12.3, 2.5 Hz, CHAHBOAc), 4.62 (1H, d, J = 11.3 Hz, NMR spectra and CIF (for structures CCDC 788731-788734).
OCHAHBPh), 4.72 (1H, d, J = 12.3, 2.2 hz, CHAHBOAc), 4.92 This material is available free of charge via the Internet at http://
(1H, q, J = 6.6 hz, C(10 )H), 7.32-7.50 (9H, m, Ar, Ph), 7.76-7.82 pubs.acs.org.