Winterfeldt oxidation did not provide any pyrroloquinolone
products from benzoyl- or Boc-protected â-carbolines 7b and
7c (entries 2 and 3). Initially, we tested the KO2 method on
these two substrates, in which R1 is a phenyl group bearing
an electron-donating group. To our delight, we could isolate
the pyrroloquinolone product in moderate yields. In the case
where R1 was a 3,4-methylene-dioxyphenyl group, the
product yield was dependent upon the functional group R2
on the amine (entries 2-4, R2 ) Bz, Boc, and Bn and the
yields were 24, 37 and 53% for substrates 7b, 7c, and 7d
respectively). When R1 was an alkyl group (entries 9 and
10), similar functional groups on the amine (Boc or Bz) led
to slightly higher yields (75 and 48% for substrates 7i and
7j, respectively). When R1 was a phenyl group, â-carboline
7e (entry 5) proceeded to give pyrroloquinolone 8e in good
yield. However, when R1 was a p-NO2 phenyl group,
â-carboline 7f (entry 6) was dehydrogenated upon treatment
with KO2, even at -60 °C. Interestingly, when R1 was a
p-Cl phenyl group, â-carboline 7g (entry 7) proceeded well
to provide pyrroloquinolone 8g. For optically pure starting
material 7h (entry 8), this method provided product 8h
smoothly without any epimerization.
The following cases distinguished the KO2 method more
from the Winterfeldt conditions. For entry 12, the KO2
method provided the desired product 8l in 52% yield without
epimerization, while the KO-t-Bu/O2 method only led to
dehydrogenated or hydrolyzed starting material. Moreover,
the KO-t-Bu/O2 method failed to provide any desired
pyrroloquinolones in entries 13 and 14 for substrates 7m and
7n, due to the fact that KO-t-Bu isomerized the allyl group
(entry 13) and cleaved the N-SO2Ph bond (entry 14). In
entry 14, Winterfeldt oxidation actually generated fully
aromatized â-carboline as the major product. Yet, because
of the mild basicity of KO2, 7m and 7n proceeded well to
pyrroloquinolones 8m and 8n.
provide the desired quinolone probably due to the ring
opening of the epoxide. Judged by HPLC-MS, the product
mixture contained several diol derivatives. Moreover, sub-
strates bearing radical-sensitive functionalities such as aryl
bromide12a or 1,2-diphenol12b should not be used in this
reaction due to the possible side reactions caused by indolyl
radical intermediate generated during the reaction process.
Since Winterfeldt oxidation might go through an anionic
intermediate, in this sense, KO2 oxidation method is comple-
mentary to the Winterfeldt condition.
In conclusion, we have developed a mild and efficient
method for the synthesis of pyrroloquinolones by KO2.
â-Carbolines bearing a variety of functional groups can be
effectively oxidized using inexpensive and widely available
reagents. This is a superior method to Winterfeldt oxidation
for base-sensitive substrates.
Acknowledgment. The authors thank Ms. Mary Evan-
gelisto and Dr. Naresh Jain for technical support and Dr.
Raymond Ng for helpful discussions.
Supporting Information Available: Experimental details
and characterization for all new compounds (1H NMR and
mass spectral data). This material is available free of charge
OL0271279
(12) (a) Yamaguchi, Y.; Van der Plas, H. C. Rec. TraV. 1977, 96, 89.
(b) Lee-Ruff, E.; Lever, A. B. P. Can. J. Chem. 1976, 54, 1837.
(13) Typical procedure of KO2 oxidation. Preparation of 1,2,3,4-
tetrahydro-2-benzyl-3-(2,3-dihydrobenzofuran-5-yl)-9H-pyrrolo-[3,4-b]-
quinolin-9-one (8a). To a solution of 7a (entry 1 in Table 2) (60 mg, 0.16
mmol) and 18-crown-6 (42 mg, 0.16 mmol) in DMF (1 mL) was added
KO2 (45 mg, 0.63 mmol) in one portion at 25 °C. The reaction mixture
turned red and was stirred for 16 h. Several drops of water were added to
consume the extra KO2. The mixture was then partitioned between ethyl
acetate and water. The aqueous phase was extracted three times with ethyl
acetate. The combined organic layer was washed with brine, dried over
anhydrous Na2SO4, filtered, and concentrated to give the crude product,
which was purified by silica gel column chromatography using 1:1 EtOAc-
hexane as eluent to give 8a as a white solid (32.5 mg, 52%): 1H NMR 300
MHz (CDCl3) δ 3.21 (t, J ) 8.7 Hz, 2H), 3.51∼3.72 (m, J ) 11.8 Hz,
2H), 4.02 (d, J ) 12.2 Hz, 1H), 4.41 (d, J ) 11.8 Hz, 1H), 4.61 (t, J ) 8.7
Hz, 2H), 4.95 (s, 1H), 6.81 (d, J ) 8.1 Hz, 1H), 7.21-7.41 (m, 8H), 7.55
(t, J ) 8.3 Hz, 1H), 7.91 (s, 1H), 8.41 (d, 1H, J ) 7.56 Hz); MS (m/z) 395
(MH+); HRMS calcd MH+ for C26H22N2O2, 395.1759; found, 395.1743.
Anal. Calcd for C26H22N2O2‚0.6 H2O: C, 77.05; H, 5.77; N, 6.91. Found
C, 77.17; H, 5.50; N, 6.87.
This method also has its limitations. For 3-chloro-1-(6-
methoxy-1,3,4,9-tetrahydro-â-carbolin-2-yl)-propan-1-one 7o,
even the mild basicity of KO2 led to the â-elimination of
HCl and consequently generated the R,â-unsaturated amide,
2-acryloyl-7-methoxy-1,2,3,4-tetrahydro-pyrrolo[3,4-b]quino-
lin-9-one in only 15% yield. KO2 oxidation of 2-oxiranyl-
methyl-2,3,4,9-tetrahydro-1H-â-carboline 7p also failed to
46
Org. Lett., Vol. 5, No. 1, 2003