Table 1 Base-mediated fragmentation of 4
Entrya
Base
Solvent
Rxn. time/h
Yieldb (%)
1
2
3
4
LitBuO
LitBuO
LiOH
Et3N
THF
DMF
EtOH
THF
1.5
2
72
24
84
71
—
o2
Scheme 2 Preparation and fragmentation of 7a and 7b.
a
b
All reactions were carried out at 50 1C. Isolated yields.
chromatographic analysis of this reaction proved to be futile due
to the use of volatile THF as a solvent. However, a careful GC
analysis of the reaction of 4 with LitBuO in DMF for 2 h at 50 1C
revealed the release of 64 mol% of N2O as a reaction product
along with the isolation of 5 in 71% yield following an aqueous
workup. 1H NMR analysis of an aliquot of this reaction mixture
also revealed the formation of methanol as a product. It is also to
be noted that the potential fragmentation of 4 in the presence of
weak bases was examined and was found to be extremely
sluggish (Table 1, entries 3 and 4) in contrast to that of reactions
with a strong base, such as LitBuO (Table 1, entries 1 and 2).
These results strongly indicate that the treatment of 4 with a
strong base, such as LitBuO, triggers in situ b-elimination of
the protected diazeniumdiolate leading to the formation of the
corresponding imine, N2O and methanol in a near stoichiometric
ratio. We postulate that the acid–base properties of the hydroxyl
group in 4 play a pivotal role in this chemical transformation.
To confirm this, we studied the effect of base on several
other protected diazeniumdiolates with structures analogous
to that of 4. Sodium salts of N-diazeniumdiolated N-benzyl-
ethylamine (6a) and N,N-dibenzylamine (6b) were prepared
following the reported procedure9 and the corresponding
O2-alkylated derivatives were synthesized. Reaction of 7a with
LitBuO in THF was found to be extremely sluggish even after
reflux for 15 h. 1H NMR analysis of the crude reaction mixture
revealed that the majority of the starting material was
unreacted. However, formation of the corresponding imine
8a in only 6% estimated yield was determined by comparing
the 1H NMR spectrum of the crude reaction mixture with that
of the authentic imine obtained commercially. Also, gas
chromatographic examination of the reaction of 7a with
LitBuO in DMF revealed the release of only 5% N2O which
is consistent with the estimated yield of 8a as determined by
1H NMR. A similar outcome was also observed for the NMR
and GC analyses of the base-mediated reaction of 7b, indicating the
formation of imine 8b and N2O in about 2% yield each. Analogs
7a and 7b, being devoid of any free N-hydroxyethyl functionality,
proved to be much less susceptible towards base-induced
fragmentation under identical conditions (Scheme 2).
Scheme 3 Preparation and reaction of 9 in base.
the reactivities of 4 and its above-mentioned analogs 7a, 7b,
and 9 points towards a key role of the N-hydroxyethyl group
in initiating a presumably concerted elimination by abstracting
one of the benzylic protons in 4 via in situ formation of the
potentially six-membered cyclic transition state (11) and leading
to the facile formation of imine 5 (Fig. 2). Based on this tentative
mechanism for the proposed b-elimination reaction, we presume
that the base-mediated reactivity of the diazeniumdiolated
benzylamines with longer N-hydroxyalkyl chains will be
substantially sluggish due to the formation of potentially
unfavorable cyclic transition states.
In summary, we have reported for the first time the preparation
of a unique secondary benzylamine diazeniumdiolate, the
O2-protected prodrugs of which can fragment directly to N2O
via a novel base-mediated b-elimination reaction triggered by
the potential acidity of the benzylic protons. Even though these
caged diazeniumdiolates have not been developed as potential
N2O donors in biological systems, this unexpected finding
should be borne in mind when working with such prodrugs,
especially diazeniumdiolated benzylamine derivatives and other
diazeniumdiolates bearing hydroxyl groups with similarly
situated acidic protons a to the amino nitrogen.
This project has been funded with Federal funds from the
National Cancer Institute, National Institutes of Health,
under contract HHSN261200800001E, and by the Intramural
Accordingly, 9, prepared by the treatment of 3 with iodomethane
in the presence of 15-crown-5 ether, when reacted with LitBuO
in THF at 50 1C, afforded only imine 10, albeit in trace quantity
(Scheme 3). Identification of the product was achieved by
comparing the NMR and HR-MS data of the crude reaction
mixture with an authentic sample of 10.10 The anomaly between
Fig. 2 Mechanism for the formation of imine 5.
c
5932 Chem. Commun., 2012, 48, 5931–5933
This journal is The Royal Society of Chemistry 2012