A detailed account of the development of a general strategy for synthesis of the C-19 methyl-substituted alkaloids including total synthesis of 19(S),20(R)-dihydroperaksine-17-al (1), 19(S),20(R)-dihydroperaksine (2), and peraksine (6) is presented. Efforts directed toward the total synthesis of macrosalhine chloride (5) are also reported. Important to success is the sequence of chemical reactions which include a critical haloboration reaction, regioselective hydroboration, and controlled oxidation (to provide sensitive enolizable aldehydes at C-20). In addition, the all-important Pd-catalyzed α-vinylation reaction has been extended to a chiral C-19 alkyl-substituted substrate for the first time. Synthesis of the advanced intermediate 64 completes an improved formal total synthesis of talcarpine (26) and provides a starting point for synthesis of macroline-related alkaloids 27-31. Similarly, extension of this synthetic strategy in the ring A oxygenated series should provide easy access to the northern hemisphere 32b of the bisindoles angustricraline, alstocraline, and foliacraline (Figure 4 ).
A detailed account of the development of a general strategy for synthesis of the C-19 methyl-substituted alkaloids including total synthesis of 19(S),20(R)-dihydroperaksine-17-al (1), 19(S),20(R)-dihydroperaksine (2), and peraksine (6) is presented. Efforts directed toward the total synthesis of macrosalhine chloride (5) are also reported. Important to success is the sequence of chemical reactions which include a critical haloboration reaction, regioselective hydroboration, and controlled oxidation (to provide sensitive enolizable aldehydes at C-20). In addition, the all-important Pd-catalyzed α-vinylation reaction has been extended to a chiral C-19 alkyl-substituted substrate for the first time. Synthesis of the advanced intermediate 64 completes an improved formal total synthesis of talcarpine (26) and provides a starting point for synthesis of macroline-related alkaloids 27-31. Similarly, extension of this synthetic strategy in the ring A oxygenated series should provide easy access to the northern hemisphere 32b of the bisindolesangustricraline, alstocraline, and foliacraline (Figure 4 ).
The C-19 methyl-substituted
indole alkaloids represented in Figure 1 are
a relatively rare class of alkaloids belonging
to the sarpagine–ajmaline series[1] with over 17 members isolated
to date.[2] In addition to the asymmetriccenters at C-3(S), C-5(S), and C-15(R) of the sarpagine/ajmaline framework, these alkaloids share a distinctive feature which is
the presence of a β-methyl group at C-19. The sarpagineindole alkaloids (+)-19(S),20(R)-dihydroperaksine-17-al (1), (+)-19(S),20(R)-dihydroperaksine
(2), and (+)-10-hydroxydihydroperaksine (3) were isolated from the hairy root culture of Rauwolfia
serpentina by Stöckigt et al.[3] The root of Rauwolfia serpentina Benth (N.0.Apocyanaciae)
has been in use in India for hundreds of years for a host of unrelated
ailments.[4] The plant has gained universal
acclamation as a useful therapeutic agent in the treatment of high
blood pressure.[5] The presence of high concentrations
of alkaloids and other phytochemicals has provided a basis for the
ethnomedical use of this plant in treating various medical conditions.[6]O-Acetyl preperakine (4) was isolated from the stem bark of Rauwolfia volkensii by Akinloye et al.[7] However, the stereochemistry
of the C-19 methyl and C-20 aldehydic groups was not determined during
isolation. The presence of a hexacyclic hemiacetal ring is a unique
feature of the C-19 methyl-substituted alkaloidsmacrosalhine chloride 5 (isolated from Alstonia macrophylla),[8] peraksine 6 (isolated from Rauwolfia perakensis),[9] verticillatine 7 (isolated from Rauwolfia verticillata),[10] and alstoyunine A (8) and B 9 (isolated from Alstonia yunnanensis).[11] Perakine 10 (isolated from Rauwolfia vomitoria),[12] raucaffrinoline 11 (isolated from Rauwolfiacaffra),[13] 10-methoxyperakine (12) and vincawajine 14 (isolated from Vinca major),[14] 10-methoxyraucaffrinoline 13 (isolated
from Vinca herbacea),[15] rauvotetraphylline D 15 (isolated from Rauwolfia
tetraphylla),[16] as well as alstoyunine
C (16) and D 17 (isolated from Alstonia
yunnanensis)[11] are the eight ajmaline-related alkaloids which also belong to this group.
Figure 1
Sarpagine-ajmaline alkaloids
containing a C-19(S) methyl group.
Sarpagine-ajmaline alkaloidscontaining a C-19(S) methyl group.The sarpagine group[1a] of alkaloids (represented by 18) is the largest class
of natural products related to the macroline(17) and ajmaline(1b) bases, and both series originate from common biogenetic
intermediates. Sarpagine alkaloidscan be converted
to macroline19 by a retro-Michael reaction
or a Hoffmann elimination[18] reaction of
the Nb-methyl intermediate of 18a/b, as demonstrated by LeQuesne et al.[19] and latter confirmed in Milwaukee,[20] whereas
a biogeneticconnection between the sarpagine and
the ajmaline alkaloids was confirmed by conversion
of 16-epivellosimine 20 into vinorine 21 by Stöckigt et al. (Scheme 1).[21]
Scheme 1
Biomimetic and Biosynthetic Relationship
between the Sarpagine, Macroline, and Ajmaline Alkaloids
Based on this, the C-19 methyl-substituted basicsarpagine framework 22 (Figure 2) could
serve as the key template for synthesis of the ajmaline alkaloids 10–17 and the macroline bases 22–31 (Figure 3).[16,17,22] Extension of this synthetic strategy to the ring-A oxygenated series
should provide the macroline intermediate 32b, the northern
hemisphere of the bisindoles(+)-angusticraline, (+)-alstocraline,
and (+)-foliacraline (Figure 4).[22b,23] Herein we provide a detailed
account of the development of a general strategy for synthesis of
the C-19 methyl-substituted alkaloids including total synthesis of
(+)-19(S),20(R)-dihydroperaksine
(1),[24] (+)-19(S),20(R)-dihydroperaksine-17-al (2),[24] (+)-peraksine (6), and formal total
synthesis of (−)-talcarpine (26) as well as efforts
directed toward total synthesis of (+)-macrosalhine chloride (5).
Figure 2
Pentacyclic intermediates 22.
Figure 3
Macroline-related indole alkaloids.[16,17,22]
Figure 4
Alstonia bisindoles and their proposed biogenesis.
Pentacyclic intermediates 22.Macroline-related indole alkaloids.[16,17,22]Alstonia bisindoles and their proposed biogenesis.
Results and Discussion
As illustrated
in Scheme 2, in a retrosynthetic
sense, dihydroperaksine-17-al (1) and dihydroperaksine
(2) could presumably be obtained from 36 by an oxidation/epimerization/reduction sequence. The β-primary
alcohol in 36 could also serve as the precursor for peraksine
(6) and can be expected to originate from the terminal
olefinic moiety in aldehyde35a via a regioselective
hydroboration process. The thermodynamically stable α-aldehydes
at C-16 would arise via homologation of the pentacyclic ketones 22a/b, which should be accessible via the recently modified
α-vinylation process from iodides 34a/b. The chiral
methyl group could be introduced by alkylating the asymmetricNb-H tetracyclic ketones 33a/b with
a suitably substituted optically active unit.
Scheme 2
Retrosynthetic Analysis
As planned, the basic tetracycliccore of the Na-H/Na-Me tetracyclicketones 33a/b, which contained
rings ABCD required for synthesis
of the C-19 methyl-substituted sarpagineindole alkaloids,
was constructed enantiospecifically and stereospecifically via the
asymmetric Pictet–Spengler/Dieckmann protocol on a 400 g scale
(from the commercially available d-(+)-tryptophan) following
the procedure developed in Milwaukee.[25] Inclusion of the C-19 β-methyl group in the sarpagine skeleton required synthesis of a suitably protected R-acetylenic alcohol 40 in high optical purity (Scheme 3). Several methods have been reported for synthesis
of enantiomerically enriched alkynols.[26] However, many of these approaches are not completely satisfactory
because of low or moderate chemical yields and/or enantioselectivity,
especially those observed for the substrate, 1-alkyn-3-ol, which has
been attributed to the small difference between the medium and the
large substituents.[27] Although the enzymatic
resolution of 1-alkyn-3-ols with a long alkyl chain was successful,
this method was difficult to apply to short chain analogs such as
3-butyn-2-ol because of poor enantioselectivities.[26g] This prompted us to employ the 1-trialkylsilyl-1-alkyn-3-ols
instead as the chiral unit for introduction of the C-19 methyl group
into the sarpagine framework. The 1-trialkylsilyl-1-alkyn-3-ols
have been synthesized in higher enantiomeric excess (ee) and high
yield by lipase-mediated kinetic resolutions of the racemicalcohols[28] as well as by asymmetric reduction of the corresponding
ketone.[24,26e,26h,29] Because of the potential loss of material which results
from the low volatility of the TMS-propargylicalcohols, it was decided
to synthesize the more stable TIPS analog. Our initial report on
synthesis of the optically active TIPS propargylicalcohol 40 began with the asymmetric Noyori hydrogenation, which when performed
on a 25 g scale resulted in 40 with an erosion of ee
(80% ee).[24] In this paper, the lipase-mediated
kinetic resolution has been developed as an alternative approach for
synthesis of 40, which can be performed on a large scale
with higher ee (>95%). As illustrated in Scheme 3, synthesis of 40 began with large-scale preparation
of the racemic TIPS-acetylenic alcohol 37 from commercially
available TIPS acetylene following the procedure of Jones et al.[30]
Scheme 3
Lipase-Mediated Kinetic Resolution of Racemic
Alcohol 37
The racemicalcohol 37 on treatment with
vinyl acetate
in distilled hexanes under catalysis by crude lipase AK20 provided
the desired (R)-acetate 38 in 95% ee
and 46% yield. The unreacted (S)-alcohol 39 was also recovered in 94% ee in 44% yield. The reaction was found
to be reproducible on a 104 g scale with no change in the ee’s
of the isolated products. To our knowledge this is the first report
of a lipase-catalyzed resolution of the TIPS analog 37 on a large scale. The ee of the (R)-acetate 38 was determined by Eu(hcf)3-aided analysis of
the 1H NMR spectrum.[28] Saponification
of the acetoxy group in 38 with K2CO3/MeOH provided the (R)-alcohol 40 without
loss of stereochemical integrity. The recovered (S)-alcohol 39 was recycled to the (R)-alcohol 40 by a Mitsunobu inversion (following the
procedure of Crimmins et al.)[31] followed
by saponification of the ester, thereby increasing the efficiency
of the process. In order to perform the SN2 reaction, the
secondary alcohol in 40 was converted to propargyl tosylate 41, which was employed for alkylation without any further
purification.The Nb-alkylation
of the secondary
amines 33a/b (individually) with optically active (R)-tosylate 41 in THF/DMF/ethanol with K2CO3 resulted in very little conversion or complete
consumption of 33a/b accompanied by baseline impurities.
Since it was well known that the nucleophilic strength was dependent
on the solvent employed in the reaction,[32] accordingly acetonitrile proved to be the most suitable solvent
for this process as the reaction went to completion in 12–14
h under refluxing conditions to give 42a and 42b (individually) along with the minor diastereomers 42a′/b′ in a combined yield of 90% and 92%
in Na-H and the Na-Me series, respectively. Both diastereomers were completely
separable in both the Na-H and the Na-Me series by silica gelcolumn chromatography
(Scheme 4).
Scheme 4
Synthesis of the
Acetylenic Ketones 43a/b
Desilylation of 42a/b with tetrabutylammonium fluoride
(TBAF·xH2O) at 0 °C in THF provided
the acetylenic compounds 43a/b, respectively, in 96%
yield (Scheme 4). The “S” configuration at C-12 (C-19 for the sarpagine skeleton) was confirmed by X-ray crystal analysis in the Na-Me series (Scheme 4). With successful introduction of the chiral methyl group into
the developing sarpagine framework, attention now
turned to the synthesis of ring E via the vinyl iodides 34a/b (see Scheme 2).The simplest method for converting terminal alkynes to
vinyl iodides
in a single step is by use of HI,[33] which
is not useful especially for sensitive substrates. Although other
methods are reported[34] there are very few
reports on direct synthesis of α-vinyl iodides from terminal
alkynes.[35] Suzuki et al.[36] reported a single-step conversion of terminal alkynes into
2-halo-1-alkenes in excellent yields and very high regioselectivity
using B-bromo- or B-iodo-9-BBN [BBN = borabicyclo(3.3.1)nonane].Initial attempts at haloboration of 43a/b with B-I-9-BBN
resulted in very little conversion. Attempts to improve this outcome
included heating at reflux, excess reagent (1.2–8 equiv), use
of freshly distilled reagent, and portionwise addition, but none of
them provided any significant improvement on the outcome of the reaction.
At best only 30% yield of the desired vinyl iodides 34a/b could be achieved. It was felt that addition of the first few equivalents
of the borane reagent resulted in complexation of the boron reagent
to the ketone in 43a/b and created a stericcloud, which
prevented another molecule of B-I-9-BBN from approaching the acetylenic
moiety. Due to less than satisfactory outcome of the key step, investigation
of other haloborating agents was investigated. The order of reactivity[37] of the haloborating reagents commonly used is
B-I-9-BBN, BBr3 > BCl3 > B-Br-9-BBN, B-Cl-9-BBN.
Although BBr3 is equivalent in reactivity to B-I-9-BBN,
the propensity of the Nb-nitrogen in 43a/b to form stable boranecomplexes with smaller
boranes as observed earlier in a similar sarpagine framework[38] and possibly altering the
course of the reaction at the terminal alkyne rendered it unsuitable.
The consistently poor yields obtained during this process led us to
two metal-catalyzed approaches to achieve this critical conversion.
The Pd-catalyzed silastannation reaction as well as the Mo-catalyzed
hydrostannation were employed to this effect (Scheme 5).
Scheme 5
Functionalization of the Acetylenic Tetracyclic Ketones 43a/b
The Mo-catalyzed hydrostannation
process reported by Kazmaier et
al. is a useful method for synthesis of α-vinyl stannanes.[39] Secondary terminal propargylic systems are reported
to give higher α-selectivities with good to excellent yields
under these conditions. The acetylene 43a on treatment
with the catalyst Mo(CO)3(CNt-Bu)3 in the presence of Bu3SnH in THF at 60 °C
gave a mixture of both the desired α-stannylated product 45a as well as the β-stannylated product 45a′ in a 3:2 ratio in 60% yield (Scheme 5). Formation of the β-stannylated alkenes 45a′
suggested that the isonitrile ligands were not bulky enough to impart
regioselectivity in the system under study here. No further optimization
was performed to improve the selectivity in the process.The
first examples of the Pd-catalyzed regioselective silastannation
of alkynes were reported by the groups of Mitchell[40] and Chenard.[41] Terminal acetylenes
on treatment with silylstannanes in the presence of catalytic amounts
of tetrakis(triphenylphosphine)palladium are reported to give highly
regio- and stereoselective addition products (cis addition with tin
always added to the internal position). Tanner et al.[42] reported the silastannation reaction of several secondary
terminal propargylicalcohols (protected and unprotected) with palladium
dibenzylidene chloroform adduct [Pd2(dba)3·CHCl3] as the catalyst in combination with 1–2 equiv of
Ph3P per Pd, thereby improving the scope of this reaction.
Both Na-H and Na-methyl acetylenic ketones 43a/b, when subjected to
the conditions of silastannation,[42] underwent
complete conversion into the desired silastannanes 44a/b in 3.5 h (Scheme 5). The silastannane reagent,
trimethylsilyl tri-n-butylstannane (Bu3SnSiMe3), was prepared on large scale, according to the
procedure of Rajanbabu et al.[43] Syn addition
of the silastannane reagent to the alkyne was confirmed by X-ray analysis
of 44b (Scheme 5). Although efforts
were continued to improve the haloboration approach, the poor selectivity
obtained in the Mo-catalyzed hydrostannation process necessitated
use of the lengthier silastannation method for synthesis of vinyl
iodide 34a/b.Conversion of the silastannanes 44a/b to the respective vinyl iodides required
two additional steps. The
first step of desilylation proved to be a challenging task. Vinyl
silanes are known to undergo desilylation under conditions of nucleophiliccatalysis specifically wherein good nucleophiles for silicon (i.e.,
F– or alkoxide) are employed. Conditions tend to
be harsh or prolonged for efficient desilylation unless other factors
come into play. Additionally, due to the acid-sensitive nature of
the substrate, protodesilylation was not a feasible option. Desilylation
of 44a/b with 1–1.5 equiv of TBAF in THF at room
temperature (rt) resulted in yields consistently below 10% due to
very little conversion. Due to the absence of stabilizing groups in
the silastannane substrates 44a/b, it was felt the protodesilylation
would require harsher conditions. A large excess of TBAF·xH2O (up to 25 equivalents at 70 °C for
almost 20 h, 1.0 M solution in THF) was required for complete consumption
of 44a/b (Scheme 6). At best, only 65% conversion was observed in both Na-H and Na-Me substrates.
It was then decided to increase the nucleophilicity of the fluoride
ion by employing DMF as the co-solvent. A mixture of THF:DMF (3:2)
improved the rate and yield of the reaction drastically. Upon further
optimization it was found that performing the reaction in anhydrous
DMF at 65 °C, with 13 (Na-H series)
and 4 (Na-Me series) equiv of TBAF·xH2O (solid), resulted in a much cleaner reaction
with the desired vinyl stannanes 45a/b obtained
in 84% and 88% yield, respectively.
Scheme 6
Synthesis of the
Vinyl Iodides 34a/b via Halodestannation
In order to obtain the iodo-olefins 34a/b, the next step was to subject the vinyl stannanes 45a/b to the conditions of halodestannation. Under the standard
conditions
for halodestannations, very low yields of the desired iodo-olefins 34a/b were obtained. Addition of excess iodine
or performing the reaction at reflux (in CHCl3 with I2/N-iodosuccinimide) were detrimental to the yield of the process.
Although complete conversion was achieved after performing the reaction
at 40 °C with 3.5 (for 45a) to 2 (for 45b) equivalents of iodine, the isolated yield of the process was still
low at 35% (for 34a) and 50% (for 34b),
Scheme 6.Although the desired vinyl iodides 34a/b were synthesized via the silastannation approach, efforts
were made
to make this approach more efficient. The haloboranesB-I-9-BBN and
dicyclohexyliodoborane [I-B(Cy)2] have been employed for
synthesis of (Z)-enolborinates exclusively from
various ethyl ketones.[44] I-B(Cy)2 is a highly stereoselective reagent for executing enolboration of
esters[45] and tertiary amides.[46] If all other parameters are kept the same, the
inability of B-I-9-BBN to drive the haloboration reaction to completion
was felt to be due to the rigid skeleton of the bicyclo ring of B-I-9-BBN
as a major factor. In contrast, the two cyclohexyl rings of I-B(Cy)2 are conformationally more flexible and should be able to
adjust to the neighboring environment. This steric difference prompted
substitution of the I-B(Cy)2 for the B-I-9-BBN in the
haloboration sequence. Moreover, the regioselectivity of the process
should remain unchanged on making this switch. After numerous trials
it was found that addition of 2.5 equiv of I-B(Cy)2 (1.0
M solution in hexanes) and subsequent protonolysis with acetic acid
provided the vinyl iodides 34a/b in 74−79% yield
with complete regioselectivity (Scheme 7).
This was a huge improvement over the previous haloboration method
and also provided a shorter and more efficient route to the core pentacyclic
framework. To our knowledge, this is also the first example of haloboration
on terminal acetylenes using I-B(Cy)2 as the haloborating
reagent. The silastannation, protodesilylation, and the halodestannation
steps could now be avoided, and the acetylenic compounds 43a/b (individually) could now be converted directly into the desired
iodo-olefins 34a/b in a single step in good
yields.
Scheme 7
Haloboration of the Acetylenic Tetracyclic
Ketones 43a/b
In keeping with the retrosynthetic analysis, the iodo-olefins 34a/b were subjected to the conditions of the recently developed
palladium-catalyzed intramolecular cross-coupling reaction.[47] The iodo-olefins 34a/b were stirred
(individually) with Pd2(dba)3, DPEPhos, and
NaOt-Bu in refluxing THF to provide the desired pentacyclicketones 22a/b in 60–68% yield (Scheme 8). The key Na-H pentacyclicketone 22a could now be used for synthesis of the C-19
methyl-substituted sarpagine and ajmaline alkaloids. The Na-Me intermediate 22b, on the other hand, could now be employed for synthesis
of macroline andsarpagine alkaloids
as discussed earlier (see Figure 2). The following
sections detail the further functionalization of the ketones 22a/b toward achieving synthesis of the target
alkaloids.
Scheme 8
Synthesis of the Key Pentacyclic Ketones 22a/b
With the C-19 methyl pentacyclicketones 22a/b now
available in gram quantities, 22a was subjected to a
one carbon atom homologation process to provide the C-16 aldehyde
in 35a. Execution of a Wittig reaction followed by acid
hydrolysis of the corresponding two stereoisomeric enol methyl ethers 46 afforded the thermodynamically stable α-aldehyde.
The aldehyde at C-16 was located in the more stable configuration
even with the presence of the C-19 methyl group in the β-position.
The aldehydic group in 35a was protected as the cyclicacetal 47 with 1,3-dioxolane in the presence of para
toluenesulfonic acid monohydrate (pTSA·H2O) in refluxing
benzene (Scheme 9). The hydroboration–oxidation
sequence to generate the desired primary alcohol 36 was
achieved under kineticcontrol;[38] 9 equiv
of boranedimethylsulfidecomplex was added at rt in one portion to
a solution of the acetal 47 in THF to minimize the Markovnikov
alcohol 48 and facilitate formation of the anti-Markovnikov
(kinetic) monol 36 (Scheme 9).
The primary and tertiary alcohols existed as Nb-BH3complexes (in a ratio of 25:1) because of
the highly basic (highly exposed) nature of the Nb-nitrogen atom. The presence of the Nb-boranecomplex was felt to be the reason for formation
of a small amount of tertiary alcohol 48. Inverse addition
of the acetal 47 to a solution of boranedimethylsulfidecomplex in THF did not eliminate formation of 48. Decomplexation
of the Nb-boranecomplexes of 36 and 48 was achieved by stirring the mixture with 5
equiv of Na2CO3 in refluxing methanol for 5
h, after which the desired primary alcohol 36 along with
tertiary alcohol 48 were obtained with the free nitrogen
function in 76% yield. Clearly hydroboration of the C20–C21
terminal olefin bond had taken place from the α-face and supported
its hindered nature from the β-face.
Scheme 9
Synthesis of the
C-20 Primary Alcohol 48
With the all important primary alcohol 36 in hand,
the next step was oxidation[48] to obtain
the β-aldehyde followed by epimerization. The basicity of the Nb-nitrogen function in the sarpagine framework
and the electron-rich character of the indole nucleus (especially
with the indoleNa-H function) contributed
in making oxidation of the primary alcohol in 36 more
problematic. Swern oxidation gave a complex mixture of products, whereas
Dess–Martinperiodinane (DMP) oxidation led to aldehydeNb-oxide formation and some overoxidized byproduct
as identified by mass spectrometry. Because of the susceptibility
of the Nb-nitrogen atom in the monol 36 to overoxidation, attention focused on the much milder
Corey–Kim oxidation. During initial attempts, addition of
the alcohol 36 in CH2Cl2 at rt
to 5 equiv of the Corey–Kim salt at −78 °C generated
a small amount of the aldehydeNb-oxide
(not shown). Reduction of the equivalents of the Corey–Kim
reagent from 5 to 3.5 accompanied by addition of the cooled solution
(−78 °C) of 36 in CH2Cl2 to the reagent at −78 °C resulted in a much cleaner
reaction and completely eliminated formation of any overoxidized byproduct.
The reaction mixture was then treated with 16 equiv of triethylamine
and allowed to stir at −78 °C for 1 h and rt for 1 h.
Since an epimeric mixture of aldehydes was obtained (1H
NMR spectrum), it was decided to carry out complete epimerization
in the same reaction vessel by allowing the reaction mixture to stir
at rt for an additional 2 h. Chromatographic purification of the crude
material furnished the α-aldehyde 49 as a colorless
oil (Scheme 10).
Scheme 10
Completion of the
Total Synthesis of 1 and 2
Reduction of the stable α-aldehyde 49 with NaBH4 in ethanol for 3 h at rt furnished
the α-alcohol 50 in 94% yield (Scheme 10). This was
followed by hydrolysis of the acetal group in 50 under
acidicconditions (1.38 N aqueous HCl) in refluxing acetone to provide
19(S),20(R)-dihydroperaksine-17-al 1 in 96% yield. A modified workup which took advantage of
the basicity of the Nb-nitrogen function
was sufficient to remove all the hydrocarbon impurities, and the product 1 thus obtained required no further purification. Signals
in the 1H and 13C NMR spectra of 1 were in excellent agreement with the literature values.[3] The 19(S),20(R)-dihydroperaksine-17-al (1) was then further subjected
to reduction with NaBH4 to complete the total synthesis
of 19(S),20(R)-dihydroperaksine 2 (Scheme 10). The synthetic 1 and 2 were also identical on silica gel thin
layer chromatography (TLC) including a mixed TLC sample to the sample
of natural 19(S),20(R)-dihydroperaksine-17-al
(1) and natural 19(S),20(R)-dihydroperaksine (2) kindly supplied by Professor
Joachim Stöckigt.[49]To effect
formation of the hemiacetal ring in peraksine (6), the
monol 36 was heated to reflux for 24
h in THF in the presence of 10 equiv of 1 N aqHCl. Examination of
the 1H NMR spectrum of the crude material after 24 h indicated
formation of the hemiacetal ring, but the cyclic acetal ring opened
only half way to give an ether linkage (see 51) as a
single diastereomer (Scheme 11). The same result
was obtained upon heating the material in THF:H2O (1:1),
at reflux for 24 h. Eventually, continued heating of ether 51 with an additional 10 equiv of 1 N aqueous HCl for 4 days resulted
in formation of trace amount of 6.
Scheme 11
Completion of the
Total Synthesis of Peraksine (6)
Because of the resistance of the ethereal linkage in 51 to hydrolysis due to reversible formation of the 1,3-dioxolane
ring,
the cyclic acetal was converted into the dimethoxy acetal. The dimethoxyacetal group was not as stable as the 1,3-dioxolane group to the hydroboration
reaction conditions and resulted in only 35% yield of the desired
monol 52. The monol 52 when subjected to
reflux in acidicconditions effected the hemicacetal ring formation
and thus provided peraksine 6 as an epimeric mixture
at C-17 in 52% yield (Scheme 11) analogous
to Arthur et al.[9c]
Studies Directed toward
Total Synthesis of Macrosalhine Chloride
(5) as Well as Partial Total Synthesis of Talcarpine
(26)
With the successful synthesis of the C-19
methyl-functionalized sarpagine alkaloids 1, 2 and 6 in the Na-H series, synthesis of the hemiacetal alkaloid (in the Na-methyl series) macrosalhine chloride 5 was next attempted. Macrosalhine 5 was isolated
from the stem bark of Alstonia macrophylla WALL.
by Schmid et al. in small amounts as the chloride or the thiocyanate
salt. The structure of this alkaloid was determined on the basis of
NMR and by X-ray analysis of its bromide salt.[8] With the key pentacyclic template 22b in hand, attention
was focused on synthesis of the hemiacetal ring formation in 5. It was decided to first reduce the aldehyde at C-16 into
an alcohol, protect it with a suitable acid-labile group, and then
functionalize the terminal olefin to the least stable β-aldehyde
by a hydroboration–oxidation sequence. Cyclization of this
β-aldehyde with the protected primary alcohol under acidicconditions
would then enable hemiacetal ring formation in 5.The pentacyclic ketone 22b was thus converted to the
aldehyde 35b by a Wittig–hydrolysis sequence.
The α-position of the aldehyde at C-16 was confirmed by X-ray
analysis (Scheme 12). Reduction of 35b with NaBH4 afforded the alcohol (95%), which was then
protected as its silyl ether to give 53. Hydroboration
of the terminal olefin in 53 provided the desired monol 54, accompanied by the expected tertiary alcohol 55 and the unusual protodeboronation byproduct 56 (Scheme 12).
Scheme 12
Synthesis of C-20 Primary Alcohol 54
Due to the susceptibility
of the more basicNb-nitrogen to form N-oxides, as observed in
the Na-H series, it was decided to employ
the milder Corey–Kim oxidation to carry out the oxidation,
although it was felt some of the more stable α-aldehyde 57b might form in this process. The oxidation was carried
out under identical conditions developed in the Na-H series and resulted in a mixture of aldehydes 57a/b (reaction mixture was not allowed to warm to rt after
addition of Et3N). The ratio of the two aldehydes as determined
by 1H NMR was 4:1 in favor of the β-aldehyde 57a. The mixture of aldehydes 57a/b was then
heated at reflux under mild acidicconditions to attempt formation
of the cyclic hemiacetal. However, upon stirring in acidic media,
complete epimerization to the α-aldehyde accompanied by desilylation
of the TIPS ether was observed (Scheme 13).
Reduction of the equivalents of Et3N (from 16 to 3) failed
to improve the outcome of the process. Thus, the inability to prevent
formation of the α-aldehyde in the Corey–Kim oxidation
was detrimental to hemiacetal ring formation.
Scheme 13
Corey–Kim
Oxidation
As performed in the Na-H series, the
Dess–Martinperiodinane (DMP) oxidation was next attempted.
As expected, the N-oxide of the desired β-aldehyde 59 was obtained as the only stereoisomer (Scheme 14), in spite of efforts to prevent overoxidation
by either using fewer equivalents of DMP or performing the reaction
at lower temperature. While attempts at using trifluoroacetic acid
(TFA) [equivalents varied from 1.5 to 1.1 equiv] in the above case
(to protonate the Nb-nitrogen) failed
to prevent N-oxide formation, oxidation with IBX/DMSO
resulted in no reaction. Since macrosalhine chloride 5 is a quaternary salt, the Nbnitrogen
in amine 54 was quaternized with excess methyl iodide
to give salt 61. However, oxidation of alcohol 61 with DMP resulted in a complex mixture with no evidence
of aldehyde 62 formation. In all the above processes
strict anhydrous conditions were maintained.
Scheme 14
DMP Oxidation and
Acid-Mediated Hemiacetal Ring Formation
Since the β-aldehyde 59 was obtained
as the
sole product during the above oxidation process, it was decided to
first attempt hemiacetal ring formation and then perform a reduction
of the N-oxide to complete synthesis of 5. Thus, refluxing the aldehyde 59 in 1 N aqueous HCl
for 24 h furnished the cyclic hemiacetal 60. At this
point with only limited material left in hand, efforts to reduce the N-oxide with Zn/AcOH failed to produce the free Nb-nitrogen. Further work toward attempting this
key step is necessary to complete the total synthesis of 5.Although the Corey–Kim oxidation could not be utilized
for
synthesis of 5, it provided a much milder method for
synthesis of the epimericaldehydes 57a/b (Scheme 13) and then to the thermodynamically stable α-aldehyde 57b (Scheme 15) in a one-pot process.
Further quaternization of the Nb-nitrogen
atom in 57b with excess MeI provided salt 63, which when subjected to a retro-Michael reaction provided the macroline framework in olefin 64 as a single
isomer (configuration not determined). Desilylation of the TIPS group
in olefin 64 to the free alcohol would generate the all-important
macroline equivalent 32a (Figure 4b), which was the intermediate involved in partial synthesis of
talcarpine 26 by Sakai et al.[50] This resulted in the formal total synthesis of 26 here.
The macroline intermediate 64 would also serve as the
precursor for potential synthesis of the macroline-related alkaloids 27−31. Additionally, 32b the 10-methoxy equivalent of 32a and the
northern hemisphere of the Alstonia bisindoles(+)-angusticraline,
(+)-alstocraline, and (+)-foliacraline (Figure 4A) could be synthesized from 5-methoxy-d-tryptophan ethyl
ester[51] via the route developed here.
Scheme 15
Formal Total Synthesis of Talcarpine (26)
Conclusion
The
first enantiospecific, stereospecific total synthesis of 19(S),20(R)-dihydroperaksine (1), 19(S),20(R)-dihydroperaksine-17-al
(2), and peraksine (6) has been accomplished.
A stereospecific approach toward synthesis of macrosalhine chloride
(5) was also developed. Commercially available d-(+)-tryptophan has served both as the chiral auxiliary and as the
starting material. Moreover, this is the first synthesis which sets
the stereochemistry of the methyl group at C-19 in a stereospecific
fashion. The acetylenic moiety 43a/b was modified to
the key vinyl iodide 34a/b by silyl-stannation and haloboration
approaches. Initial attempts at haloboration were accompanied by inconsistent
reproducibility, purification problems, and very low yield, which
were circumvented by modification of the reaction conditions and
use of a superior haloborating agent [I-B(Cy)2]. It is
important to point out that the palladium-catalyzed α-vinylation
has been extended to the C-19 chiral methyl series which make it a
process of more general applicability. The key Na-H pentacyclic ketone 22a could now be used for
synthesis of the C-19 methyl-substituted ajmaline alkaloids 10, 11, and 15–17 (Figure 1), and a route amenable
to the synthesis of macroline–related alkaloids 27–31 (Figure 3) is now possible due to the efficient synthesis of the advanced
intermediate 64 via a general strategy for synthesis
of C-19 methyl-substituted indole alkaloids. This process provides
a general entry into a whole series of biosynthetically important
monoterpene C-19-substituted indole alkaloids.
Experimental
Section
Experimental details and spectral data for synthesis
of alkaloids 1 and 2 and compounds 22a, 34a (from 43a), 36, 41, 42a, 43a, 47, 49, and 50 are contained in the Supporting
Information
of ref (24a). For general
experimental considerations see the Supporting
Information.
Lipase-Catalyzed Kinetic Resolution of 4-Triisopropylsilyl-3-butyn-2-ol
(37)
A 5 L round-bottom flask, which had been
flame dried, was equipped with an overhead mechanical stirrer and
charged with 69.0 g of ground activated 4 Å molecular sieves
and 50 g (0.5 mass equiv) of the lipase (crude) Amano AK20. To this
were added 4 L of distilled hexanes, 158.2 g (1.83 mmol) of vinyl
acetate (dried over MgSO4), and the racemicalcohol 37 (104 g, 0.46 mol). This suspension was stirred at 25 °C
for 5 days, after which analysis by NMR spectroscopy indicated 50%
conversion to the R-acetate 38. The
solution was filtered and concentrated under reduced pressure, and
the crude product was purified by column chromatography (silica gel)
with 2–10% EtOAc/hexanes as the eluant to afford the R-acetate 38 (63.0 g, 46%) and the S-alcohol 39 (46.0 g, 44%).
R-Acetate 38
R 0.59 (silica gel, EtOAc/hexanes,
1:7). 1H NMR (400 MHz, CDCl3): δ 5.49
(q, 1H, J = 6.8 Hz), 2.10 (s, 3H), 1.51 (d, 3H, J = 6.8 Hz), 1.09 (s, 21H). The % ee of the R-acetate 38 was determined as 95% by Eu(hcf)3-aided 1H NMR analysis. The acetate was not subjected
to any further characterization and used in the next step as is.
S-Alcohol 39
R 0.33 (silica gel, EtOAc/hexanes,
1:7). 1H NMR (300 MHz, CDCl3): δ 4.54
(q, 1H, J = 6.6 Hz), 1.91 (s, 1H), 1.47 (d, 3H, J = 6.6 Hz), 1.07 (s, 21H) [1H NMR was identical
to that reported for R-alcohol 40 herein].
HRMS (APCI-TOF) m/z: [M + H]+ calcd for C13H27OSi 227.1826, found
227.1828.This material was employed directly in the next step
without any further characterization.
Hydrolysis of the R-Acetate 38 to the R-Alcohol 40
The R-acetate 38 (10 g, 37.2 mmol) was dissolved
in a saturated solution (100 mL) of K2CO3 in
MeOH:H2O (15:1). The solution, which resulted, was stirred
at rt for 2 h until disappearance of the starting material 38 as monitored by TLC (silica gel). The solvent was then
removed under reduced pressure, and the mixture, which resulted, was
extracted with ether. The organic layer was separated and dried over
anhydrous Na2SO4 and concentrated to give the R-alcohol 40 as a clear oil (8.1 g, 96%). R 0.33 (silica gel, EtOAc/hexanes,
1:7). [α]20D + 23.38° (c 2.01 CHCl3). 1H NMR (400 MHz, CDCl3): δ 4.56 (q, 1H, J = 6.4 Hz), 2.06 (br s,
1H), 1.49 (d, 3H, J = 6.4 Hz), 1.09 (s, 21H). 13C NMR (75 MHz, CDCl3): δ 109.8, 84.4, 58.8,
24.6, 18.6, 11.1. HRMS (APCI-TOF) m/z: [M + H]+ calcd for C13H27OSi 227.1826,
found 227.1826.
Mitsunobu Inversion of the S-Alcohol 39 to the R-Alcohol 40
To a stirred solution of S-alcohol 39 (28.0 g, 0.124 mol) in dry THF (2.1 L) at rt was added
triphenylphosphine
(64.85 g, 0.247 mol) and benzoic acid (30.2 g, 0.247 mol). The mixture,
which resulted, was then cooled to 0 °C, and to it was added
diethyl azodicarboxylate (41 g, 0.235 mol). The reaction mixture was
allowed to warm to rt and stirred for an additional 2.5 h. The solution
was then concentrated, and the residue was dissolved in ethyl acetate
and filtered through a pad of Celite topped with silica gel. The filtrate
was concentrated, and the residue was dissolved in a solution of MeOH
(200 mL), THF (70 mL), H2O (70 mL), and 15 g of NaOH. This
mixture was stirred for 2 h, and solvents were removed under reduced
pressure. The aq residue was extracted with diethyl ether (3 ×
100 mL). The combined organic extracts were dried over Na2SO4, filtered, and concentrated to give the R-alcohol 40 (25.2 g, 90%). Spectral data of 40 were identical in all respects to that reported above for 40 (obtained from 38). HRMS (APCI-TOF) m/z: [M + H]+ calcd for C13H27OSi 227.1826, found 227.1826.
(6S,10S)-12-((S)-4-(Triisopropylsilyl)but-3-yn-2-yl)-7,8,10,11-terahydro-5H-6,10-epiminocycloocta[b]indol-9(6H)-one (42a) and (6S,10S)-12-((R)-4-(Triisopropylsilyl)but-3-yn-2-yl)-7,8,10,11-terahydro-5H-6,10-epiminocycloocta[b]indol-9(6H)-one (42a′)
An oven-dried
500 mL flask cooled under argon was charged with optically active Na-H, Nb-H tetracyclicketone 33a (5.0 g, 22.1 mmol), freshly distilled acetonitrile
(150 mL), (R)-4-triisopropylsilyl-3-butyn-2-ol tosylate 41 (13.5 g, 35.4 mmol) in dry acetonitrile (50 mL), and anhydrous
potassium carbonate (6.1 g, 44.2 mmol). The mixture which resulted
was allowed to stir at 75 °C (outside oil bath temperature) for
12 h under argon. Analysis by TLC (silica gel, CHCl3/EtOH,
9:1) indicated the absence of tetracyclicketone 33a after
12 h. The reaction mixture was cooled to rt, and the K2CO3 was filtered off by passing the solution through a
bed of Celite using EtOAc as the eluent. After removal of the solvent
under reduced pressure, the crude product was purified by flash chromatography
(silica gel, EtOAc/hexanes) to provide the major (S)-Na-H, TIPS acetylenic tetracyclic ketone 42a as a light yellow-colored solid (8.4 g) in 88% yield and
a small amount of the (R)-Na-H, TIPS acetylenic tetracyclic ketone 42a′
as a buff-colored solid (0.2 g, 2%). A small amount of 42a′ was obtained because of the 95% ee of the starting tosylate 41.
Major [C(12)-S] Diastereomer (42a)
For spectral data see the Supporting Information of ref (24a).
(6S,10S)-5-Methyl-12-((S)-4-(triisopropylsilyl)but-3-yn-2-yl)-7,8,10,11-tetrahydro-5H-6,10-epiminocycloocta[b]indol-9(6H)-one (42b) and (6S,10S)-5-Methyl-12-((R)-4-(triisopropylsilyl)but-3-yn-2-yl)-7,8,10,11-tetrahydro-5H-6,10-epiminocycloocta[b]indol-9(6H)-one (42b′)
An oven-dried 1 L flask cooled under argon was
charged with optically active Na-Me, Nb-H tetracyclic ketone 33b (15.0
g, 0.062 mol). The solid 33b was dissolved in freshly
distilled acetonitrile (1 L), after which a solution of (R)-4-triisopropylsilyl-3-butyn-2-ol tosylate 41 (47.57
g, 0.116 mol) in dry acetonitrile (50 mL) was added. Anhydrous potassium
carbonate (17.27 g, 0.125 mol) was added, and the mixture which resulted
was allowed to stir at 75 °C (outside oil bath temperature) for
12 h under argon. Analysis by TLC (silica gel, CHCl3/EtOH,
9:1) indicated the absence of tetracyclicketone 33b after
12 h. The reaction mixture was cooled to rt, and the K2CO3 was filtered off by passing the solution through a
bed of Celite using EtOAc as eluent. After removal of the solvent
under reduced pressure, the crude product was purified by flash chromatography
(silca gel, EtOAc/hexanes) to provide the (S)-Na-Me, TIPS-protected acetylenic tetracyclicketone 42b as a light yellow-colored solid (25.3 g, 90%)
and a small amount of the (R)-Na-Me, TIPS-protected acetylenic tetracyclic ketone 42b′ as a buff-colored solid (0.51 g, 2%) in a combined yield
of 92%. A small amount of 42b′ was obtained because
of the 95% ee of starting tosylate 41.
TBAF·xH2O (84 mL, 0.084 mol, 1.0 M solution in THF) was added to a
solution of the Na-Me, TIPS-protected
acetylenic tetracyclic ketone 42b (25 g, 0.056 mol) in
THF (420 mL) at 0 °C. The solution which resulted was allowed
to stir at 0 °C for 0.5 h, after which the ice bath was removed
and the mixture was stirred at rt for 3 h until analysis by TLC indicated
disappearance of the starting material 42b. The reaction
solution was then quenched with H2O (150 mL) at rt, followed
by dilution with EtOAc (500 mL). The two layers were separated. The
organic layer was washed with water, brine, and dried (Na2SO4). The EtOAc was removed under reduced pressure, and
the residue was passed through a small pad of silica gel to give the Na-Me acetylenic tetracyclic ketone 43b as an off-white-colored solid (15.7 g, 96% yield). Part of the solid
was crystallized using EtOAc to give 43b as white crystals
for X-ray analysis. IR (KBr): 3266, 2983, 2099, 1708, 1469, 756, 662
cm–1. 1H NMR (300 MHz, CDCl3): δ 7.48 (d, 1H, J = 7.3 Hz), 7.34 (d, 1H, J = 8.1 Hz), 7.25 (td, 1H, J = 7.0, 1.1
Hz), 7.13 (ddd, 1H, J = 7.8, 6.9, 1.1 Hz), 4.78–4.77
(m, 1H), 3.97 (d, 1H, J = 6.6 Hz), 3.72 (s, 3H),
3.64 (qd, 1H, J = 6.7, 2.2 Hz), 3.16 (dd, 1H, J = 16.8, 6.7 Hz), 2.72 (d, 1H, J = 16.8
Hz), 2.67–2.58 (m, 1H), 2.54–2.47 (m, 1H), 2.31 (d,
1H, J = 2.2 Hz), 2.20–2.02 (m, 2H), 1.48 (d,
3H, J = 6.7 Hz). 13C NMR (75 MHz, CDCl3) δ 210.1 (C), 137.1 (C), 133.5 (C), 126.2 (C), 121.5
(CH), 119.2 (CH), 118.1 (CH), 108.9 (CH), 106.1 (C), 84.7 (C), 72.4
(CH), 61.3 (CH), 49.2 (CH), 47.1 (CH), 34.3 (CH2), 29.2
(CH3), 28.9 (CH2), 21.4 (CH2), 20.4
(CH3). EIMS (m/e, relative
intensity) 292 (M+˙, 81), 235 (100), 196 (24), 183
(46), 168 (26). HRMS (EI-trisector) m/z: calcd for C19H20N2O 292.1576,
found 292.1570.
To a solution of the
silylstannane 44a (4.54 g, 7.1 mmol) in DMF (123 mL)
was added solid TBAF·xH2O (14.80
g, 57 mmol) at rt. The mixture which resulted was heated to 65 °C
(oil bath temperature). After heating for 3 h, an additional 9.25
g (5 equiv) of solid TBAF·xH2O was
added at 65 °C in three portions of 3.7 (2 equiv), 3.7 (2 equiv),
and 1.85 g (1 equiv) in intervals of 2, 2, and 1 h. After heating
for 10 h, analysis by TLC indicated the disappearance of the silylstannane 44a. The reaction mixture was brought to rt and diluted with
water (123 mL). The mixture was stirred for 5 min, and this was followed
by addition of EtOAc (500 mL). The two layers were separated, and
the aq layer was extracted with EtOAc (3 × 50 mL). The combined
organic layers were washed with water (5 × 50 mL) and brine (5
× 40 mL), dried (Na2SO4), and concentrated
under reduced pressure. The residue was purified by chromatography
on basic alumina to yield the vinylstannane 45a as a
white-colored solid (4.0 g, 84%). Part of the oil was crystallized
using EtOAc to give 45a as white crystals for X-ray
analysis. Mp 99–101 °C, [α]20D −70.36 (c 1.12 CHCl3). R 0.4 (hexanes/EtOAc, 6:2).
IR (KBr): 1692 cm–1. 1H NMR (300 MHz,
CDCl3): δ 7.78 (s, 1H), 7.50 (d, 1H, J = 7.6 Hz), 7.35 (d, 1H, J = 7.8 Hz), 7.23–7.11
(m, 2H), 5.64 (d, 1H, J = 2.3 Hz), 5.17 (d, 1H, J = 2.3 Hz), 4.27 (s, 1H), 4.02 (d, 1H, J = 6.4 Hz), 3.36 (q, 1H, J = 9.5, 3.2 Hz), 3.14
(dd, 1H, J = 16.8, 6.6 Hz), 2.67 (d, 1H, J = 16.8 Hz), 2.49–2.39 (m, 1H), 2.37–2.29
(m, 1H), 2.20–2.10 (m, 1H), 2.03–1.97 (m, 1H), 1.58–1.46
(m, 6H), 1.35 (sex, 6H, J = 7.4 Hz), 1.19 (d, 3H, J = 6.4 Hz), 1.00–0.90 (m, 15H). 13C NMR
(75 MHz, CDCl3): δ 210.6 (C), 161.0 (C), 135.7 (C),
132.3 (C), 126.9 (C), 125.6 (CH2), 121.8 (CH), 119.6 (CH),
118.0 (CH), 110.8 (CH), 107.6 (C), 64.2 (CH), 61.2 (CH), 48.7 (CH),
34.5 (CH2), 30.1 (CH2), 29.0 (3 × CH2), 27.3 (3 × CH2), 20.8 (CH2),
19.8 (CH3), 13.6 (3 × CH3), 10.6 (3 ×
CH2). CIMS (m/e, relative
intensity) 571 (M+ + 1, 36), 513 (100), 281 (37), 263 (17),
169 (11). HRMS (ESI-TOF) m/z: (M
+ H)+ calcd for C30H47N2O120Sn 571.2711, found 571.2705.
Molybdenum-Catalyzed Functionalization
of Acetylene 43a To Provide 45a and (6S,10S)-12-((S,E)-4-(Tributylstannyl)but-3-en-2-yl)-5,6,7,8,10,11-hexahydro-9H-6,10-epiminocycloocta[b]indol-9-one (45a′)
In an oven-dried flask were dissolved
the Na-H, acetylenic
tetracyclicketone 43a (451 mg, 1.6 mmol), hydroquinone
(10 mg, 0.1 mmol), and Mo(CO)3(CNt-Bu)3 (230 mg, 0.28 mmol) under argon in THF (5 mL). Then Bu3SnH (1.4 g, 4.9 mmol) was added slowly, and the mixture was
warmed to 55 °C and held at the same temperature for 9 h until
all the starting material 43a was consumed on analysis
by TLC (silica gel). After cooling to rt, the reaction mixture was
concentrated under reduced pressure and subjected to flash chromatography
on silica gel. The excess Bu3SnH was removed using hexanes
as eluent. The stannylated products 45a (360 mg) and 45a′ (190 mg) were obtained using hexanes/ethyl acetatecontaining 1% triethylamine as the eluent in a yield of 60%.
Major Diastereomer
(45a)
The spectral
and physical properties of vinyl stannane 45a were identical
to those described in the above experiment.
An oven-dried single
neck flask was cooled under argon and charged
with the vinylstannane 45a (1.0 g, 1.76 mmol) dissolved
in CH2Cl2 (25 mL). Iodine (0.67 g, 2.63 mmol)
was dissolved in CH2Cl2 (20 mL) and this solution
was added in one portion to the vinylstannane 45a, and
the mixture was placed and stirred in a preheated oil bath (40–45
°C) for 1 h. After 1 h of heating, another 1.5 equiv of iodine
solution (0.67 g dissolved in 20 mL of CH2Cl2) was added to the reaction mixture after which it was allowed to
stir at 40–45 °C for 2 h. The reaction mixture was then
quenched with solutions of 5% aq sodium bisulfite (100 mL) and 5%
KF in methanol (100 mL), and the mixture which resulted was stirred
vigorously for 10 min. The reaction mixture was diluted with additional
CH2Cl2 (50 mL), and the two layers were separated.
The aq layer was extracted with CH2Cl2 (3 ×
20 mL), and the combined organic layers were washed with brine (2
× 50 mL) and dried (Na2SO4). The solvent
was removed under reduced pressure, and the residue was purified by
silica gelchromatography to give the vinyl iodide 34a as a white-colored solid (285 mg, 35%).Spectral data for 34a was identical to that reported in the communication.[24a]
Procedure with iodine:
The reaction
was performed following the same procedure as for 34a except the reaction was performed on a relatively smaller scale.
Vinyl stannane 45b (407 mg, 0.697 mmol) was dissolved
in CHCl3 (25 mL). Iodine (354 mg, 1.39 mmol) was dissolved
in CHCl3 (46 mL) and added to the solution of 45b. Vinyl iodide 34b was obtained as a white-colored solid
(150 mg, 50%). Mp 63 °C. Procedure with I-B(Cy)2:
An oven-dried flask fitted with an addition funnel was cooled under
argon and charged with Na-Me acetylenic
tetracyclicketone 43b (2.10 g, 7.55 mmol) dissolved
in freshly distilled CH2Cl2 (52.5 mL) and hexanes
(7.0 mL). The flask was cooled to 0 °C with ice, and I-B(Cy)2 (30.2 mL, 15.1 mmol, 0.5 M solution in hexanes) was added
dropwise every 0.5 h in three portions, over a total period of 1.5
h. After the last addition the reaction mixture was allowed to stir
at 0 °C for another 0.5 h, after which the ice bath was removed
and the mixture was stirred at rt for 2 h. After stirring at rt for
2 h, another 0.5 equiv of I-B(Cy)2 (7.6 mL, 3.78 mmol)
was added dropwise at rt and the mixture was allowed to stir for another
2 h. The mixture was then treated with glacial acetic acid (4.8 mL,
83.1 mmol) at 0 °C and stirred at rt for 1.15 h. At this point
the flask was again cooled to 0 °C, a solution of cold aq 3
M NaOH (40.3 mL, 121 mmol) and 30% H2O2 (2.6
mL, 23 mmol) was added, and the stirring was maintained for 1 h at
rt. The biphasic solution which resulted was transferred to a bigger
flask, diluted with CH2Cl2 (400 mL) and water
(50 mL), after which the two layers were separated. The original reaction
flask still had some residual solid attached to the bottom of the
flask. The solid was dissolved in acetone (50 mL). The acetone was
evaporated under reduced pressure to 3/4th the volume, and the mixture
was diluted with CH2Cl2 (50 mL). Again, the
two layers were separated, and the combined CH2Cl2 layers were treated with solutions of 5% KF in methanol (160 mL)
and 5% aq sodium bisulfite (160 mL) under vigorous stirring for 5
min. The aq layer was separated, extracted with CH2Cl2 (2 × 80 mL), after which the combined organic layers
were washed with brine, dried (Na2SO4), filtered,
and concentrated under reduced pressure. Purification by chromatography
on silica gel (EtOAc/hexanes, 1:4) afforded vinyl iodide 34b as a white solid (79%, 2.3 g). Mp 63–65 °C. 1H NMR (300 MHz, CDCl3): δ 7.50 (d, 1H, J = 7.7 Hz), 7.35–7.22 (m, 2H), 7.14 (ddd, 1H, J = 7.9, 7.8, 1.1 Hz), 6.22 (d, 1H, J = 0.8 Hz),
5.87 (d, 1H, J = 1.3 Hz), 4.26–4.25 (m, 1H),
4.02 (d, 1H, J = 6.5 Hz), 3.65 (s, 3H), 3.14 (dd,
1H, J = 16.9, 6.5 Hz), 2.78 (d, 1H, J = 16.3 Hz), 2.62 (q, 1H, J = 6.2 Hz), 2.59–2.51
(m, 2H), 2.20–1.95 (m, 2H), 1.95 (d, 3H, J = 6.2 Hz). 13C NMR (75 MHz, CDCl3): δ
210.1 (C), 137.1 (C), 132.9 (C), 126.3 (C), 125.7 (CH2),
122.5 (C), 121.5 (CH), 119.2 (CH), 118.1 (CH), 108.9 (CH), 106.5 (C),
62.3 (CH), 60.7 (CH), 47.6 (CH), 34.3 (CH2), 29.6 (CH2), 29.2 (CH3), 21.1 (CH2), 19.7 (CH3). EIMS (m/e, relative intensity):
420 (M+˙, 78), 363 (100), 293 (46), 265 (34), 239
(32), 211 (15), 196 (27), 183 (78), 168 (42), 154 (17), 128 (14).
Anal. Calcd for C19H21IN2O: C, 54.30;
H, 5.04; N, 6.67. Found: C, 54.09; H, 5.00; N, 6.26.
A mixture of anhydrous
potassium tert-butoxide (1.82 g, 16.2 mmol) and methoxy-methyltriphenylphosphonium
chloride (5.13 g, 15.0 mmol) in dry benzene (82.1 mL) was allowed
to stir at rt for 1 h. The pentacyclic ketone 22a (570
mg, 2.05 mmol) in THF (23 mL) was then added to the above red-colored
solution dropwise at 0 °C. The mixture which resulted was stirred
at rt for 12 h. After 12 h at rt, analysis of the mixture by TLC (silica
gel, CH2Cl2:MeOH, 4.7:0.3, R 0.58) indicated the absence of starting
material 22a. The mixture was then diluted with EtOAc
(100 mL), and the reaction solution was quenched with water (50 mL).
The aq layer was extracted with EtOAc (2 × 15 mL), and the combined
organic layers were washed with brine (2 × 30 mL) and dried (Na2SO4). The solvent was removed under reduced pressure
to afford a mixture of enol ethers 46 as an oil. The
baseline materials (silica gel, TLC) were removed by percolation through
a wash column. This material was employed directly in the next step
without any further characterization and purification.The crude
compound 46 was dissolved in a solution of THF/H2O (1:1, 10 mL). To the above solution was added 12 N aqconcHCl (1.65 mL) and the mixture which resulted was stirred at 55 °C
(oil bath temperature) for 6 h. The reaction mixture was then cooled
to 0 °C and extracted with ethyl ether (4 × 15 mL) to remove
the phosphorus byproducts, after which the aq layer was then brought
to pH 8 with an ice-cold solution of 14% aq NH4OH. The
aq layer was extracted with EtOAc (3 × 15 mL), and the combined
organic layers were washed with brine (2 × 15 mL) and dried (Na2SO4). The solvent was removed under reduced pressure
to afford the α-aldehyde35a as an oil, which was
subjected to the next step without any further purification and characterization.The crude alkenic aldehyde35a (2.0 g, 6.84 mmol)
was dissolved in dry benzene (233 mL), and this was followed by addition
of dry ethylene glycol (4.67 g, 75 mmol) and p-toluenensulfonic
acid monohydrate (1.43 g, 7.52 mmol). The mixture which resulted was
heated to reflux for 6 h followed by removal of water via a DST. Analysis
of the mixture by TLC (silica gel, CH2Cl2:MeOH)
indicated the absence of starting material 35a. The mixture
was allowed to cool to rt, diluted with EtOAc, and at 0 °C brought
to pH 8–9 with 14% aq NH4OH. The aq layer was separated
and then extracted with EtOAc (3 × 20 mL). The combined organic
layers were washed with brine (2 × 20 mL) and dried (Na2SO4). The solvent was removed under reduced pressure and
chromatographed [silica gel, CH2Cl2/MeOH, (10:0.3)
to provide the ethylene acetal 47 (620 mg, 90% yield
over 3 steps from 22a). 1H NMR (300 MHz, CDCl3): δ 7.90 (s, 1H), 7.51 (d, 1H, J =
7.4 Hz), 7.32 (d, 1H, J = 7.3 Hz), 7.15 (td, 1H, J = 7.1, 1.3 Hz), 7.10 (td, 1H, J = 7.2,
1.1 Hz), 4.98 (d, 1H, J = 2.4 Hz), 4.91 (d, 1H, J = 1.5 Hz), 4.82, (d, 1H, J = 8.0 Hz),
4.21 (d, 1H, J = 8.9 Hz), 3.97–3.80 (m, 4H),
3.64–3.57 (m, 1H), 3.36 (t, 1H, J = 6.0 Hz),
3.02 (dd, 1H, J = 15.6, 5.1 Hz), 2.85 (dd, 1H, J = 14.7, 1.1 Hz), 2.54 (t, 1H, J = 1.8
Hz), 2.10 (ddd, 1H, J = 12.3, 10.2, 2.0 Hz), 1.77–1.66
(m, 2H), 1.44 (d, 3H, J = 6.8 Hz). 13C
NMR (75 MHz, CDCl3): δ 151.8 (C), 137.7 (C), 136.3
(C), 127.8 (C), 121.3 (CH), 119.2 (CH), 118.2 (CH), 110.7 (CH), 107.7
(CH2), 106.1 (CH), 105.2 (C), 64.7 (CH2), 64.4
(CH2), 57.9 (CH), 51.8 (CH), 47.0 (CH), 45.6 (CH), 35.6
(CH), 33.5 (CH2), 27.0 (CH2), 16.8 (CH3). EIMS (m/e, relative intensity):
336.5 (M+˙, 100), 293.5 (14), 263.5 (98), 207.4 (12),
169.4 (71), 115.3 (18), 91.3 (12). HRMS (EI) calcd for C21H24N2O2 336.1838, found 336.1831.
A mixture of Na-Me vinyl iodo tetracyclicketone 34b (1.78 g, 4.23 mmol), DPEPhos (88.4 mg, 0.164
mmol), and t-BuONa (610 mg, 6.35 mmol) in a solution
of freshly distilled THF (48 mL) was degassed under reduced pressure
at rt and backfilled with argon (3 times). Pd2(dba)3 (77.5 mg, 0.08 mmol) along with dry THF (3 mL) was introduced
into the reaction mixture, and the system was again degassed under
reduced pressure at rt and backfilled with argon (4 times). The mixture
was then heated to 70 °C (oil bath temperature) under argon for
3.5 h. The mixture was then cooled to rt and quenched with ice–water.
The THF volume was reduced to one-half under reduced pressure, and
the mixture was diluted with EtOAc (70 mL). The aq layer was extracted
with EtOAc (2 × 15 mL), and the combined organic layers were
washed with brine (2 × 30 mL) and dried (Na2SO4). The EtOAc was then removed under reduced pressure, and
the residue was flash chromatographed with CH2Cl2 on basic alumina to provide the cross-coupled pentacyclic ketone 22b as a light brown-colored solid (842 mg, 68%). 1H NMR (300 MHz, CDCl3): δ 7.50 (d, 1H, J = 7.7 Hz), 7.26 (d, 1H, J = 7.2 Hz), 7.20 (ddd,
1H, J = 7.5, 6.9, 1.1 Hz), 7.10 (t, 1H, J = 7.3 Hz), 5.14 (d, 1H, J = 2.7 Hz), 5.03 (d, 1H, J = 2.3 Hz), 4.43 (dd, 1H, J = 9.4, 2.1
Hz), 3.96–3.89 (m, 1H), 3.74 (d, 1H, J = 5.5
Hz), 3.62 (s, 3H), 3.36 (dd, 1H, J = 15.6, 1.4 Hz),
3.07 (dd, 1H, J = 3.7, 2.0 Hz), 2.93 (dd, 1H, J = 15.5, 6.1 Hz), 2.62 (ddd, 1H, J = 12.3,
9.9, 1.9 Hz), 2.15 (ddd, 1H, J = 12.7, 3.8, 2.7 Hz),
1.51 (d, 3H, J = 6.8 Hz). EIMS (m/e, relative intensity): 292 (M+˙,
34), 263 (100), 249 (15), 183 (91), 168 (31). The pentacylic ketone 22b was not subjected to any further characterization. It
was used directly in the next experiment.A mixture of anhydrous
potassium tert-butoxide (4.85 g, 43.2 mmol) and methoxy-methyltriphenylphosphonium
chloride (13.69 g, 39.9 mmol) in dry benzene (82.1 mL) was allowed
to stir at rt for 1 h. The pentacyclic ketone 22b (1.7
g, 5.47 mmol) in THF (20 mL) was then added to the above red-colored
solution dropwise at 0 °C. The mixture which resulted was stirred
at rt for 12 h. After 12 h at rt analysis of the mixture by TLC (silica
gel, CH2Cl2:MeOH, 4.7:0.3, R = 0.58) indicated the absence of starting
material 22b. The mixture was then diluted with EtOAc
(100 mL), and the reaction was quenched with water (50 mL). The aq
layer was extracted with EtOAc (2 × 15 mL), and the combined
organic layers were washed with brine (2 × 30 mL) and dried (Na2SO4). The solvent was removed under reduced pressure
to afford the mixture of enol ethers as a brownish red oil. The baseline
materials (silica gel, TLC) were removed by percolation through a
wash column (silica gel). The solvent was removed under reduced pressure,
and the residue was dissolved (without further purification) in a
solution of THF/H2O (1:1, 28 mL). To the above mixture
aq 12 N concHCl (4.7 mL) was added, and the mixture which resulted
was stirred at 55 °C (oil bath temperature) for 6 h. The reaction
mixture was then cooled to 0 °C and extracted with ethyl ether
(4 × 15 mL) to remove the phosphorus byproducts, after which
the aq layer was then brought to pH 8 with an ice-cold solution of
14% aq NH4OH. The aq layer was extracted with EtOAc (3
× 15 mL), and the combined organic layers were washed with brine
(2 × 15 mL) and dried (Na2SO4). The solvent
was removed under reduced pressure to afford 35b as a
solid (1.5 g, 90%). Part of the solid was crystallized using CH2Cl2:MeOH to give white crystals of 35b for X-ray analysis. 1H NMR (300 MHz, CDCl3): δ 9.64 (s, 1H), 7.49 (d, 1H, J = 7.7 Hz),
7.30 (d, 1H, J = 12.8 Hz), 7.21 (td, 1H, J = 7.0 Hz), 7.11 (ddd, 1H, J = 7.8, 6.9,
1.0 Hz), 4.94 (d, 1H, J = 2.7 Hz), 4.90 (d, 1H, J = 2.3 Hz), 4.34 (d, 1H, J = 8.8 Hz),
3.83 (t, 1H, J = 6.3 Hz), 3.67 (s, 4H), 3.10 (dd,
1H, J = 15.6, 5.1 Hz), 2.86 (t, 1H, J = 1.8 Hz), 2.64 (dd, 1H, J = 15.7, 0.96 Hz), 2.43
(d, 1H, J = 7.6 Hz), 2.24 (ddd, 1H, J = 12.3, 10.0, 2.0 Hz), 1.77 (dt, 1H, J = 12.4,
3.2 Hz), 1.44 (d, 3H, J = 6.8 Hz). 13C
NMR (75 MHz, CDCl3): δ 202.8 (CH), 150.3 (C), 138.8
(C), 137.3 (C), 127.1 (C), 121.0 (CH), 118.9 (CH), 118.1 (CH), 108.7
(CH), 108.4 (CH2), 103.6 (C), 57.9 (CH), 54.9 (CH), 50.8
(CH), 44.4 (CH), 34.9 (CH), 32.4 (CH2), 29.3 (CH3), 27.3 (CH2), 16.6 (CH3). HRMS (ESI-TOF) m/z: [M + H]+ calcd for C20H22N2O 306.1732, found 306.1729.
((6S,8S,9S,11R,11aS)-11-(1,3-Dioxolan-2-yl)-8-methyl-5,6,8,9,10,11,11a,12-octahydro-6,10-methanoindolo[3,2-b]quinolizin-9-yl)methanol (36) and (6S,8S,9R,11R,11aS)-11-(1,3-Dioxolan-2-yl)-8,9-dimethyl-5,6,8,9,10,11,11a,12-octahydro-6,10-methanoindolo[3,2-b]quinolizin-9-ol (48)
Procedure is
reported in the Supporting Information of ref (24a).
Primary Alcohol (36)
For spectral data
see the Supporting Information of ref (24a).
A
mixture of 35a (100 mg, 0.34 mmol), pTSA·H2O (75 mg, 0.39 mmol), methanol (132 mg, 4.04
mmol), and trimethyl orthoformate (109 mg, 1.03 mmol) was refluxed
in CHCl3 (8 mL) for 6 h. The reaction mixture was brought
to pH 7 with 10% aq NH4OH. The layers were separated. The
CHCl3 layer was washed with brine, dried (Na2SO4), and removed under reduced pressure to provide dimethoxyacetal (114 mg) as an oil in 93% yield. EIMS (m/e, relative intensity): 338 (M+˙, 100),
323 (63), 307 (36), 291 (13), 277 (16), 263 (78), 216 (64), 201 (89),
169 (39). This material was employed directly in the next step without
any further characterization.To a solution of the dimethoxyacetal (114 mg, 0.38 mmol) in dry THF (4 mL) was added BH3·DMS (2.0 M solution in THF, 1.5 mL, 3.0 mmol) at rt. The mixture
which resulted was stirred at rt for 2 h. The reaction mixture was
then quenched by careful addition of ice cold water (4 mL) at 0 °C
(initial addition of water results in a large amount of effervescence).
At this point NaBO3·4H2O (1.2 g, 7.76 mmol)
was added to the mixture in one portion at 0 °C. The mixture
which resulted was allowed to stir at rt for 2 h, after which EtOAc
(100 mL) and H2O (15 mL) were added. The organic layer
was separated, washed with water (2 × 20 mL), brine (2 ×
20 mL), and dried (Na2SO4). The EtOAc was then
removed under reduced pressure to provide the Nb-BH3complex as a mixture of isomers at [C(20)],
the majority of which was the primary alcohol. This material was used
in the next step without any further purification. The above mixture
of isomers was dissolved in freshly distilled MeOH (10 mL), and Na2CO3 (178 mg, 1.7 mmol) was added. The mixture was
then warmed to 60 °C (oil bath) for 5 h under vigorous stirring.
The reaction mixture which resulted was cooled to rt, followed by
filtration through a bed of Celite. The filtrate was concentrated
under reduced pressure to provide a turbid oil, which was redissolved
in CH2Cl2. The CH2Cl2 layer
was washed with H2O (1 × 10 mL), brine (4 × 10
mL), and dried (Na2SO4). The solvent was removed
under reduced pressure to afford a crude solid, which was purified
by flash chromatography [silica gel, CH2Cl2/MeOH
(v/v 10:1)] to furnish the primary alcohol 52 (43 mg,
35%). The other impurities were not characterized.1H NMR (300 MHz, CDCl3): δ 8.41 (br,
s, 1H), 7.50 (d, 1H, J = 7.5 Hz), 7.36 (d, 1H, J = 7.8 Hz), 7.18–7.08 (m, 2H), 4.65 (d, 1H, J = 8.9 Hz), 4.19 (d, 1H, J = 8.4 Hz),
3.79–3.67 (m, 2H), 3.61 (s, 3H), 3.33–3.20 (m, 2H),
3.14 (s, 3H), 2.94–2.93 (m, 2H), 2.08 (br, s, 1H), 1.96 (t,
2H, J = 11.0 Hz), 1.87 (t, 1H, J = 8.6 Hz), 1.67 (dt, 1H, J = 12.2, 3.0 Hz), 1.29
(d, 3H, J = 7.3 Hz). 13C NMR (75 MHz,
CDCl3): δ 137.1 (C), 136.3 (C), 127.6 (C), 121.3
(CH), 119.1 (CH), 118.2 (CH), 111.0 (CH), 105.8 (CH), 104.8 (C), 62.2
(CH2), 54.6 (CH3), 54.0 (CH), 52.3 (CH), 49.2
(CH3), 48.7 (CH), 39.8 (CH), 39.1 (CH), 36.5 (CH2), 26.8 (CH2), 24.7 (CH), 12.8 (CH3). EIMS
(m/e, relative intensity): 356 (M+˙, 34), 341 (66), 323 (100), 293 (13), 281 (26), 207
(28), 169 (17), 129 (170). HRMS (EI-trisector) m/z: calcd for C21H28N2O3 356.2100, found 356.2110.
Peraksine (6)
To a solution of 52 (8 mg, 0.03 mmol) in
THF (2 mL) was added 1 N aqHCl (0.8 mL, 0.6
mmol). The mixture which resulted was heated to reflux for 4 days,
after which it was cooled to rt and brought to pH = 7 with 10% aqNH4OH. The mixture was extracted with CH2Cl2. The CH2Cl2 layer was washed with brine,
dried (Na2SO4), and removed under reduced pressure
to afford an oil. Peraksine (1 mg) was isolated by preparative TLC
(silica gel) as an oil. 1H NMR (300 MHz, CDCl3): δ 7.85 (br, s, 1H), 7.47 (t, 1H, J = 6.8
Hz), 7.35–7.32 (m, 1H), 7.18–7.07 (m, 2H), 5.10 (d,
1/2H, J = 2.0 Hz), 4.71 (d, 1/2H, J = 1.2 Hz), 4.21 (d, 1H, J = 10.9 Hz), 4.04 (dd,
1H, J = 11.8, 2.4 Hz), 3.82 (t, 1H, J = 4.6 Hz), 3.75 (d, 1H, J = 11.1 Hz), 3.56 (t,
1H, J = 5.3 Hz), 3.50 (dd, 1H, J = 12.2, 2.5 Hz), 3.28–3.02 (m, 2H), 2.70 (d, 1H, J = 16.0 Hz), 2.61 (d, 1H, J = 16.0 Hz),
2.30 (br, s, 1H), 2.06–1.94 (m, 2H), 1.47 (d, 3H, J = 7.2 Hz) (The compound is a epimeric mixture as reported).[9c] EIMS (m/e,
relative intensity): 310 (M+˙, 100), 309 (69), 293
(11), 279 (13), 263 (16), 223 (16), 209 (59), 195 (22), 182 (41),
168 (47), 156 (36), 115 (24). HRMS (EI-trisector) m/z: calcd for C19H22N2O2 310.1681, found 310.1664. No further characterization
was carried out.
The aldehyde 35b (530 mg, 1.72 mmol) was dissolved in EtOH (10 mL). NaBH4 (98 mg, 2.59 mmol) was added to the above solution in one
portion at 0 °C and allowed to stir at 0 °C for 8 h. The
reaction mixture was diluted with CH2Cl2 (300
mL) and poured into cold water (50 mL). The aq layer was extracted
with additional CH2Cl2 (3 × 80 mL), and
the combined organic layer was washed with brine (80 mL) and dried
(K2CO3). The solvent was removed under reduced
pressure to afford the crude product, which was purified by chromatography
on silica gel (CH2Cl2/CH3OH = 10:1)
to provide the primary alcohol (480 mg, 90%) as a yellow oil (see
S1 in Supporting Information). 1H NMR (500 MHz, CDCl3): δ 7.44 (d, 1H, J = 7.8 Hz), 7.34 (d, 1H, J = 8.2 Hz), 7.23 (t, 1H, J = 7.6 Hz), 7.12 (t, 1H, J = 7.6 Hz),
4.95 (d, 2H, J = 3.0 Hz), 4.27 (d, 1H, J = 9.8 Hz), 3.68 (s, 4H), 3.57–3.49 (m, 2H), 3.05 (t, 1H, J = 6.2 Hz), 2.99 (dd, 1H, J = 15.3, 5.1
Hz), 2.66 (d, 1H, J = 15.3 Hz), 2.39 (br, s, 1H),
2.17–2.11 (m, 2H), 1.69 (q, 1H, J = 6.8 Hz),
1.60–1.58 (m, 1H), 1.44 (d, 3H, J = 6.8 Hz). 13C NMR (125 MHz, CDCl3): δ 153.0, 139.8,
137.8, 127.8, 121.3, 119.2, 118.6, 109.1, 108.3, 104.4, 65.3, 58.6,
51.4, 48.6, 45.0, 36.6, 33.6, 29.8, 27.5, 17.3. EIMS (m/e, relative intensity): 308 (M+˙,
100), 293 (12), 277 (41), 196 (12), 183 (69), 168 (24): HRMS (ESI-TOF) m/z: (M + H)+ calcd for C20H25N2O, 309.1967, found 309.1976.A solution of the primary alcohol (520 mg, 1.68 mmol) in dry CH2Cl2 (40 mL) was cooled to 0 °C, after which
2,6-lutidine (0.587 mL, 5.05 mmol) was added, and this was followed
by addition of TIPSOTf (774 mg, 2.52 mmol) to the stirred solution.
The mixture was then allowed to stir for an additional 2 h at 0 °C,
after which cold water (2 mL) was added to quench the reaction. The
reaction mixture was diluted with CH2Cl2 (100
mL) and the layers separated. The aq layer was extracted with additional
CH2Cl2 (2 × 50 mL), and the combined organic
layer was washed with brine (30 mL) and dried (Na2SO4). The solvent was removed under reduced pressure to afford
the crude product, which was dried in vacuo to remove the extra 2,6-lutidine
before the solid was purified by chromatography on silica gel (CH2Cl2/CH3OH = 20:1) to provide the O-TIPS ether as a white-colored solid 53 (744
mg, 95%). R 0.53 (silica
gel, hexanes/EtOAc, 1:4). 1H NMR (300 MHz, CDCl3): δ 7.49 (d, 1H, J = 7.6 Hz), 7.30 (d, 1H, J = 7.7 Hz), 7.19 (td, 1H, J = 7.0, 1.1
Hz), 7.09 (ddd, 1H, J = 14.7, 6.9, 1.1 Hz), 4.93–4.90
(m, 2H), 4.26 (d, 1H, J = 7.9 Hz), 3.66 (s, 4H),
3.62–3.61 (m, 2H), 3.02–2.94 (m, 2H), 2.73 (d, 1H, J = 14.0 Hz), 2.44 (t, 1H, J = 1.8 Hz),
2.16 (ddd, 1H, J = 12.7, 10.1, 2.0 Hz), 1.80 (q,
1H, J = 7.5 Hz), 1.71–1.64 (m, 1H), 1.42 (d,
3H, J = 6.8 Hz), 1.06 (s, 21H). 13C NMR
(75 MHz, CDCl3) δ 152.0 (C), 139.3 (C), 137.2 (C),
127.4 (C), 120.7 (CH), 118.6 (CH), 118.1 (CH), 108.6 (CH), 107.7 (CH2), 104.2 (C), 65.5 (CH2), 58.3 (CH), 51.1 (CH),
48.7 (CH), 44.4 (CH), 35.6 (CH), 33.2 (CH2), 29.2 (CH3), 27.2 (CH2), 18.0 (6 × CH3),
16.8 (CH3), 11.9 (3 × CH). EIMS (m/e, relative intensity): 464 (M+˙,
100), 421 (12), 277 (50), 196 (21), 183 (52), 161 (20). Anal. Calcd
for C29H44N2OSi: C, 74.94; H, 9.54;
N, 6.03. Found: C, 74.74; H, 9.45; N, 6.06.
((6S,8S,9S,11R,11aS)-5,8-Dimethyl-11-(((triisopropylsilyl)oxy)methyl)-5,6,8,9,10,11,11a,12-octahydro-6,10-methanoindolo[3,2-b]quinolizin-9-yl)methanol (54), (6S,8S,9S,11R,11aS)-5,8,9-Trimethyl-11-(((triisopropylsilyl)oxy)methyl)-5,6,8,9,10,11,11a,12-octahydro-6,10-methanoindolo[3,2-b]quinolizin-9-ol (55), and (6S,8S,9S,11R,11aS)-5,8,9-Trimethyl-11-(((triisopropylsilyl)oxy)methyl)-5,6,8,9,10,11,11a,12-octahydro-6,10-methanoindolo[3,2-b]quinolizine (56)
To a solution of
the alkenic TIPS-protected alcohol 53 (208 mg, 0.447
mmol) in dry THF (9.3 mL) was added BH3·DMS (2.0 M
solution in THF, 2.05 mL, 4.11 mmol) at rt. The mixture which resulted
was stirred at rt for 2 h. The reaction mixture was then quenched
by careful addition of ice cold water (8.5 mL) at 0 °C (initial
addition of water results in a large amount of effervescence). At
this point NaBO3·4H2O (1.85 g, 12 mmol)
was added to the mixture in one portion at 0 °C. The mixture
which resulted was allowed to stir at rt for 2 h, after which EtOAc
(200 mL) and H2O (25 mL) were added. The organic layer
was separated, washed with water (2 × 20 mL), brine (2 ×
20 mL), and dried (Na2SO4). The EtOAc was then
removed under reduced pressure to provide the Nb-BH3complex as a mixture of isomers at [C(20)],
the major of which was the primary alcohol. This material was used
in the next step without any further purification. The above mixture
of isomers was dissolved in freshly distilled MeOH (20 mL), and Na2CO3 (237 mg, 2.23 mmol) was added. The mixture
was then warmed to 60 °C (oil bath) for 5 h under vigorous stirring.
The reaction mixture which resulted was cooled to rt, followed by
filtration through a bed of Celite. The filtrate was concentrated
under reduced pressure to provide a turbid oil which was redissolved
in CH2Cl2. The CH2Cl2 layer
was washed with H2O (1 × 10 mL) and brine (4 ×
10 mL) and dried (Na2SO4). The solvent was removed
under reduced pressure to afford a crude solid, which was purified
by flash chromatography [silica gel, CH2Cl2/MeOH
(v/v 10:1)] and furnished the primary alcohol 54 (153
mg, 71%) and the tertiary alcohol 55 (6.48 mg, 3%) along
with the desilylated by product 56 (10 mg, 7%) in a combined
yield of 81%.
To a stirred solution of N-chlorosuccinimide (8.3
mg, 0.062 mmol) in dry CH2Cl2 (1.5 mL) was added
dimethyl sulfide (7.6 μL, 0.103 mmol) at −5 to −15
°C (outside bath temperature) under argon. A white precipitate
appeared immediately after addition of the sulfide, which was stirred
for an additional 0.5 h at the above-mentioned temperature range.
After 0.5 h, the temperature of the reaction mixture was lowered to
−78 °C (EtOAc–dry ice bath). The alcohol 54 (10 mg, 0.02 mmol) in dry CH2Cl2 (1.0
mL) was also cooled at −78 °C and then added via cannula
to the white complex, and stirring was continued for 2 h at −78
°C. A solution of distilled triethylamine (11 μL, 0.08
mmol) was then added to the above mixture dropwise (neat) and the
stirring was continued for an additional 1 h at −78 °C.
Upon completion of the reaction, the reaction mixture was quenched
at −78 °C by addition of excess CH2Cl2 and H2O. The organic layer was separated, washed with
brine, and dried (Na2SO4). The solvent was removed
under reduced pressure to provide the mixture of crude aldehydes 57a/b. Addition of EtOAc from the sides of the flask resulted
in formation of a white insoluble precipitate. The precipitate was
filtered, and the EtOAc layer was concentrated under reduced pressure.
The same process was repeated 4–5 times until one no longer
sees any formation of a precipitate after addition of EtOAc to the
residue to remove the Corey–Kim sulfur impurities. Analysis
of the 1H NMR spectrum of the residue indicated formation
of the epimericaldehydes (57a:57b) in the
ratio of 4:1. The epimeric mixture of aldehydes 57a/b was dissolved in a solution of MeOH (3 mL) and triethylamine (0.17
mL), and the mixture was stirred overnight at rt to effect complete
epimerization. The methanol was then removed under reduced pressure
to give an oil, which was further purified by flash column chromatography
(basic alumina, EtOAc:EtOH, 9:0.1) to give the α-aldehyde 57b as a colorless oil (6.97 mg, 80%). 1H NMR (300
MHz, CDCl3): δ 9.85 (s, 1H), 7.50 (d, 1H, J = 7.7 Hz), 7.30 (d, 1H, J = 8.1 Hz),
7.20 (ddd, 1H, J = 7.5, 7.0, 1.1 Hz), 7.11 (t, 1H, J = 7.3 Hz), 4.20 (d, 1H, J = 8.1 Hz),
3.87–3.78 (m, 2H), 3.62 (s, 3H), 3.50–3.41 (m, 1H),
3.23 (t, 1H, J = 6.1 Hz), 3.00 (dd, 1H, J = 15.4, 5.0 Hz), 2.74 (d, 1H, J = 14.7 Hz), 2.53
(d, 1H, J = 8.5 Hz), 2.46 (d, 1H, J = 2.0 Hz), 1.92 (dd, 1H, J = 11.8, 1.7 Hz), 1.78
(dd, 1H, J = 14.1, 6.7 Hz), 1.46–1.39 (m,
1H), 1.36 (d, 3H, J = 6.8 Hz), 1.10 (s, 21H). 13C NMR (75 MHz, CDCl3): δ 203.5 (CH), 139.1
(C), 137.2 (C), 127.3 (C), 120.8 (CH), 118.7 (CH), 118.0 (CH), 108.6
(CH), 104.3 (C), 64.3 (CH2), 52.2 (CH), 51.3 (CH), 51.0
(CH), 47.6 (CH), 42.4 (CH), 29.5 (CH2), 29.2 (CH3), 26.9 (CH2), 26.6 (CH), 19.3 (CH3), 18.0
(6 × CH3), 11.8 (3 × CH). HRMS (EI-trisector) m/z: calcd for C29H44N2O2Si 480.3172, found 480.3175.
The aldehydes 57a/b were synthesized following the above
procedure for the Corey–Kim oxidation. To a solution of 57a/b (15 mg, 0.03 mmol) in THF (10 mL) was added 1 N aqHCl
(1.6 mL, 0.6 mmol). The mixture which resulted was heated to reflux
for 3 h, after which it was cooled to rt and brought to pH = 7 with
cold 10% aq NH4OH. The mixture was extracted with CH2Cl2. The CH2Cl2 layer was
washed with brine, dried (Na2SO4), and removed
under reduced pressure to afford 58 as a brown oil (8
mg, 80%), which was purified by preparative TLC [silica gel, CH2Cl2/MeOH (v/v 10:1) with 1% NH4OH (14%)]. 1H NMR (300 MHz, CDCl3): δ 9.87 (s, 1H), 7.48
(d, 1H, J = 7.7 Hz), 7.29 (d, 1H, J = 7.8 Hz), 7.20 (m, 1H), 7.09 (m, 1H), 4.18 (d, 1H, J = 12.4 Hz), 3.80–3.75 (m, 2H), 3.62 (s, 3H), 3.51–3.41
(m, 1H), 3.12–3.08 (m, 1H), 3.00 (dd, 1H, J = 15.6, 4.9 Hz), 2.71 (d, 1H, J = 15.0 Hz), 2.47
(br, s, 1H), 2.38 (d, 1H, J = 8.5 Hz), 1.98–1.90
(m, 1H), 1.79 (q, 1H, J = 7.6 Hz), 1.44–1.43
(m, 1H), 1.34 (d, 3H, J = 6.8 Hz). EIMS (m/e, relative intensity): 324 (M+˙, 91), 306 (11), 295 (77), 281 (20), 253 (15), 223 (30), 209
(21), 196 (37), 183 (100), 168 (41), 144 (28), 115 (14). HRMS (EI-trisector) m/z: calcd for C20H24N2O2 324.1838, found 324.1818. No further characterization
was carried out.
Macrosalhine Nb-Oxide (60)
Dess–Martinperiodinane
(184 mg, 0.434 mmol) was
added to a solution of the primary alcohol 54 (100 mg,
0.207 mmol) in CH2Cl2 (30 mL) in one portion
at 0 °C. The reaction mixture was then allowed to warm to rt
and stirred for 2 h at this temperature. The reaction mixture was
quenched with a saturated solution of aq NaHCO3 (9 mL),
and a saturated solution of aq Na2S2O3 (9 mL) was added, after which the mixture was stirred for 10 min
at 0 °C. The aq layer was extracted with additional amounts of
CH2Cl2 (3 × 20 mL), and the combined organic
layer was washed with brine (20 mL) and dried (MgSO4).
The solvent was removed under reduced pressure to provide the crude
product, which was separated by preparative TLC [silica gel, CH2Cl2/MeOH (v/v 20:1) with 1% NH4OH(14%)]
to provide the aldehydeNb-oxide 59 (69 mg, 67%). 1H NMR (300 MHz, CDCl3): δ 10.1 [(s, 1H (CHO)]. EIMS (m/e, relative intensity): 496 (M+˙, 100),
467 [25 (M – CHO)], 451 (15), 437 (24), 409 (12), 308 (17),
279 (14), 248 (23), 211 (43), 183 (97), 170 (71), 75 (20). HRMS (EI-trisector) m/z: clcd for C29H44N2O3Si 496.3121, found 496.3097.HRMS (ESI-TOF) m/z: (M + H)+ calcd for C29H45N2O3Si 497.3199, found
497.3193. No further characterization was carried out.To a
solution of 59 (5 mg, 0.01 mmol) in THF (2 mL) was added
1 N aqHCl (0.8 mL, 0.6 mmol). The mixture which resulted was heated
to reflux for 2 days, after which it was cooled to rt and brought
to pH = 7 with 10% aq NH4OH. The mixture was extracted
with CH2Cl2. The CH2Cl2 layer was washed with brine, dried (Na2SO4), and removed under reduced pressure to afford an oil. Macrosalhine Nb-oxide 60 (2.6 mg, 75%) was isolated
by preparative TLC (silica gel) as an oil. 1H NMR (300
MHz, CDCl3): δ 7.48 (d, 1H, J =
7.7 Hz), 7.30 (d, 1H, J = 9.5 Hz), 7.19 (t, 1H, J = 7.9 Hz), 7.09 (t, 1H, J = 7.7 Hz),
5.10 (s, 1H), 4.28 (d, 1H, J = 8.7 Hz), 4.03 (d,
1H, J = 10.5 Hz), 3.66–3.61 (m, 4H), 3.45
(dd, 1H, J = 10.9, 1.9 Hz), 3.22–3.05 (m,
2H), 2.67–2.51 (m, 3H), 2.07–2.05 (m, 2H), 1.56 (d,
2H, J = 5.2 Hz), 1.40 (d, 3H, J =
7.3 Hz). 13C NMR (75 MHz, CDCl3): δ 120.8,
118.7, 118.0, 108.6, 92.8, 69.9, 63.5, 61.8, 49.6, 46.1, 38.3, 31.4,
29.1, 26.5, 25.2, 12.1 (the quaternary carbons are not reported).
EIMS (m/e, relative intensity):
340 (M+˙, 82), 339 (91), 322 (40), 296 (32), 265
(19), 239 (15), 221 (18), 196 (13), 183 (100), 168 (18), 97 (18),
83 (14), 57 (14). HRMS (EI-trisector) m/z: calcd for C20H24N2O3 340.1787, found 340.1779. HRMS (ESI-TOF) m/z: (M + H)+ calcd for C20H25N2O3 341.1860, found 341.1837.
Authors: M Toufiqur Rahman; Jeffrey R Deschamps; Gregory H Imler; Alan W Schwabacher; James M Cook Journal: Org Lett Date: 2016-08-16 Impact factor: 6.005
Authors: Md Toufiqur Rahman; Veera Venkata Naga Phani Babu Tiruveedhula; Michael Rajesh Stephen; Sundari K Rallapalli; Kamal P Pandey; James M Cook Journal: Molecules Date: 2022-03-07 Impact factor: 4.411
Figure 1
Scheme 1
Figure 2
Figure 3
Figure 4
Scheme 2
Scheme 3
Scheme 4
Scheme 5
Scheme 6
Scheme 7
Scheme 8
Scheme 9
Scheme 10
Scheme 11
Scheme 12
Scheme 13
Scheme 14
Scheme 15