4220
J. Yu, J. B. Spencer / Tetrahedron Letters 42 (2001) 4219–4221
OH
OTs
OBn
OBn
OH
OBn
(a)
(b)
(c)
OBn
OBn
OBn
OBn
H2N
OH
OH
N3
BnO
OBn
OBn
BnO
BnO
HO
8
10
11
2
Scheme 2. Reagents and conditions: (a) tosyl chloride, pyridine (anhydrous), CHCl3, 0°C, 90%; (b) NaN3, DMF, 80°C, 95%; (c)
Pd/C, H2, HCl (conc.)/AcOH/MeOH, 99%.
OTs
H(T)
OH
H(T)
OBn
OH
OBn
OBn
(a)
(b)
(c)
(T)H
BnO
OBn
OBn
H2N
OH
OH
(T)H
BnO
OBn
OBn
N3
BnO
OBn
OBn
HO
12
14
15
16
Scheme 3. Reagents and conditions: (a) tosyl chloride, pyridine (anhydrous), CHCl3, 0°C, 95%; (b) NaN3, DMF, 80°C, 92%; (c)
Pd/C, H2, HCl (conc.)/AcOH/MeOH, 98%.
To establish a sensitive assay for the transformation
of 2-deoxy-scyllo-inosamine to deoxystreptamine the
synthesis was modified to prepare tritium labelled 2.
Compound 9 (0.23 mmol) was reduced with NaBH3T
(100 mCi in 8 ml of EtOH) to give a mixture of
tritium labelled cyclohexanol 12 (axial -OH) and 13
(equatorial -OH) from which pure 12 was isolated in
62% yield. 12 was smoothly tosylated to give 14,
which was then subjected to azidation to afford 15.
The tritium labelled azide 15 was then hydrogenated
to give the final product 16 (45 mCi per mmol)
(Scheme 3).
furanose have been reported. However, both synthesis
involved many steps and were achieved in low
yield.4,5 In our previous work, we have reported a
successful stereoselective deoxygenation reaction using
lithium triethylborohydride (LTBH) for the removal
of a hydroxyl group from carbocyclic diols.6 To take
advantage of this method, we set out to synthesise
both 1 and 2 from myo-inositol.
The two cis-hydroxyls of myo-inositol 3 were pro-
tected in the form of 1,2-O-isopropylidene-myo-inosi-
tol 4. The remaining hydroxyls were reacted with
benzyl bromide to give the fully protected myo-inosi-
tol 5. Treatment with 2 M HCl leads to the deprotec-
tion of the acetal to give cis-diol 6. The equatorial
hydroxyl group of the diol 6 was then selectively
tosylated using dibutyltin oxide, tosyl chloride and
benzyl triethylammonium chloride to give the
monotosylate 7. The monotosylate was subjected to
the stereoselective deoxygenation system using LTBH.
The displacement of the tosyl group via a proposed
1,2-hydride shift and subsequent reduction of the
ketone intermediate by LTBH gave the cyclohexanol
8 in good yield (79%). Compound 8 was oxidised
using Jones’ reagent to afford 9 and subsequently
deprotected to give the required product 1, in a
seven-step synthesis with overall yield of 60% (Scheme
1).
In summary, an efficient route for the synthesis of
two key intermediates on the biosynthetic pathway to
deoxystreptamine is established,9 which has also pro-
vided ready access to the radiolabelled version of 2.
Detailed enzymatic studies utilising these compounds
are under the way in our laboratory.
Acknowledgements
The authors would like to thank the BBSRC, St.
John’s College, Cambridge University for a Junior
Research Fellowship (J.Q.Y.) and the Royal Society
(J.B.S.) for funding.
The ready availability of cyclohexanol 8 from this
synthetic route offers an obvious route to synthesise
compound 2 through displacement of the hydroxyl
with an azide moiety followed by hydrogenation.
One-pot azidation of 8 by a modified Mitsunobu
reaction using either zinc azide/bis-pyridine complex
or diphenylphosphoryl azide was attempted with no
product isolated.7,8 Compound 8 was therefore tosyl-
ated and reacted with sodium azide to give 11 in 95%
yield. Subsequent hydrogenation of 11 gives the final
product 2. This leads to an eight-step synthesis of 2
with overall yield of 52% (Scheme 2).
References
1. Ota, Y.; Tamegai, H.; Kudo, F. J. Antibiot. 2000, 53, 1158.
2. Goda, S. K.; Akhtar, M. J. Chem. Soc., Chem. Commun.
1987, 1, 12.
3. Kudo, F.; Hosomi, Y.; Tamegai, H. J. Antibiot. 1999, 52,
81.
4. Yamauchi, N.; Kakinuma, K. J. Antibiot. 1992, 45, 756.
5. Barer, H. H.; Siemsen, L.; Astles, D. J. Carbohydr. Res.
1986, 156, 247.
6. Yu, J. Q.; Spencer, J. B. J. Org. Chem. 1996, 61, 3234.
Yu, Jinquan
