Organic Letters
Letter
phosphate, and the desired THCA 11 was isolated in excellent
yield (92%).
Scheme 3. Selective Cyclizations of Aldehyde 4
Careful monitoring of these reactions via both thin-layer
chromatography (TLC) and liquid chromatography−mass
spectrometry (LC−MS) indicated that the cyclization did
indeed occur first, followed by external nucleophilic attack, in a
domino Friedel−Crafts type of process.7,8 Furthermore, alcohol
5 could be converted to 11 upon reaction with indole and
diphenyl phosphate in CH2Cl2 in 80% yield (Scheme 7), thus
validating the overall strategy outlined in Scheme 1.
Scheme 7. Conversion of a Type 1 Product into a Type 2
Product
Scheme 4. Validation of the Structure of 6 and Crystal
Structure of 8
We then addressed the question of enantioselectivity by
using chiral phosphoric acids9,10 to catalyze the type 2
cyclization. After preliminary screening experiments, reaction
conditions were optimized for the use of (R)-3,3′-bis(2,4,6-
triisopropylphenyl)-1,1′-binaphthyl-2,2′-diyl hydrogen phos-
phate ((R)-TRIP) in CH2Cl2 at −50 °C.11 When aldehyde 4
was exposed to a range of external nucleophiles, the desired
products were obtained only with electron-rich heterocycles
and thiols (Scheme 8). Attempted reactions with poorer carbon
nucleophiles such as furan and trimethoxybenzene, as well as
with alcohols, gave no conversion to the desired products. As
shown in Scheme 8 (13a−j), the substitution pattern and the
steric bulk of the external indole nucleophile had considerable
impact on both the yields (35−74%) and the enantioselectivity
(11−94% ee). For example, the highest enantioselectivity (94%
ee) was obtained using 2-tert-butylindole, which can be
compared with the 2-methyl analogue (25% ee), the isolated
yields being comparable (Scheme 8, 13b−c). The substituent
pattern of the aldehyde (9b, 12a−d) also proved highly
important, as the products 13k−o were isolated in moderate to
excellent yields (35−94%, generally >78%) and with low to
moderately high ee (6−78%, Scheme 8).
tional reaction, involving the indole nitrogen, giving compound
7 in 80% yield. The structure of 7 was confirmed after acylation
with 3,5-dinitrobenzoyl chloride to give 8 as a single
diastereomer. The crystal structure of 8 is shown in Scheme 4.
Not unexpectedly, the ease of the type 1 cyclization was very
dependent on the electronic properties of the indole (Scheme
5). The electron-rich indoles 9a and 9b gave the desired
Scheme 5. Type 1 Cyclization to THCAs
Returning to reactions of aldehyde 4, reactions of some other
nucleophilic heterocycles and thiols gave good to excellent
yields (up to 95%) but generally low enantioselectivity (46% ee
at best).
Encouraged by the excellent enantioselectivity provided by 2-
tert-butylindole, we explored the possibility of using the
removable trimethylsilyl (TMS) group in the 2-position of
the indole nucleophile (Table 1). When aldehyde 4 was
exposed to a range of readily available 2-TMS-indoles (14a−f),
followed by removal of the TMS group, the desired products
(15a−f) were generally isolated in excellent enantiopurity (95
to >99% ee) and moderate to good yields (33−70%, Table 1,
entries 1−4).
products (10a and 10b) in superb yields (91−95%), while for
the less nucleophilic 9c no conversion was observed, even at
elevated temperatures. Notably, when the aldehyde function-
ality was changed to a ketone, no conversion was observed for
any of the substrates.
The formation of compound 6 indicated that a type 2
reaction (cf. Figure 1) should also be possible (Scheme 6).
Aldehyde 4 was therefore reacted with indole as external
nucleophile in the present of catalytic amounts of diphenyl
Finally, we turned our attention to the long-sought exocyclic
Pictet−Spengler reaction, for which a new optimization study
proved to be necessary.10 As shown in Table 2, a variety of
functional groups are tolerated, and yields are generally
acceptable. Unfortunately, reactions catalyzed by (R)-TRIP
gave only low enantioselectivity (<10% ee), and a synthetically
useful asymmetric version of the exocyclic Pictet−Spengler
reaction remains elusive.
Scheme 6. Type 2 Cyclization to THCAs
B
Org. Lett. XXXX, XXX, XXX−XXX