New synthetic technology for the synthesis of hindered α-diazoketones via acyl mesylates

Article abstract of DOI:10.1021/ol990790l

(equation presented) A mild and reliable one-pot protocol for the elaboration of sterically demanding carboxylic acids into α-diazoketones via acyl mesylates has been developed. Aside from delineating the reaction parameters which render this strategy quite general for hindered carboxylic acids, we have directly proven the existence of the fleeting acyl mesylate group as the reactive species in these reactions and shed light onto the differing mechanisms which are operative in the activation of hindered and simple carboxylic acids with methanesulfonyl chloride.

Full text of DOI:10.1021/ol990790l

ORGANIC
LETTERS
1999
Vol. 1, No. 6
883-886
New Synthetic Technology for the
Synthesis of Hindered r-Diazoketones
via Acyl Mesylates
K. C. Nicolaou,* Phil S. Baran, Yong-Li Zhong, Ha-Soon Choi, Kin Chiu Fong,
Yun He, and Won Hyung Yoon
Department of Chemistry and The Skaggs Institute for Chemical Biology, The Scripps
Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037, and
Department of Chemistry and Biochemistry, UniVersity of California, San Diego,
9500 Gilman DriVe, La Jolla, California 92093
kcn@scripps.edu
Received July 7, 1999
ABSTRACT
A mild and reliable one-pot protocol for the elaboration of sterically demanding carboxylic acids into r-diazoketones via acyl mesylates has
been developed. Aside from delineating the reaction parameters which render this strategy quite general for hindered carboxylic acids, we
have directly proven the existence of the fleeting acyl mesylate group as the reactive species in these reactions and shed light onto the
differing mechanisms which are operative in the activation of hindered and simple carboxylic acids with methanesulfonyl chloride.
R-Diazoketones are versatile synthetic intermediates with a
plethora of uses including dipolar cycloadditions, alkene,
C-H, and X-H insertions, ylide formation, homologation
of carbonyl compounds, and the classic Arndt-Eistert
synthesis.1 Although there are numerous methods for the
preparation of these useful synthons,2 we encountered several
instances during our total synthesis of the CP molecules3
wherein the synthesis of extremely hindered R-diazoketones
from the corresponding carboxylic acids was inefficient or
even impossible using previously described synthetic meth-
ods. Herein, we present a new method which is specifically
suited for the one-pot preparation of sterically encumbered
R-diazoketones in high yield.
The current method takes advantage of the highly reactive
nature of the acyl mesylate species in order to activate
sterically hindered acids for attack by diazomethane. Al-
though a few scattered reports on the use of acyl mesylates
are present in the literature, there is no definitive evidence
that the intermediates of these reactions are actually acyl
mesylates.4c Aside from demonstrating the generality of this
new one-pot protocol, we have accumulated spectroscopic
evidence which directly proves the existence of acyl mesy-
lates for the first time. We have also reinvestigated some
previous reports4 which claimed to employ acyl mesylates
and have determined that, in most cases, symmetrical
anhydrides are the actual intermediates present as a result
of sulfonyl chloride-mediated dehydration.
(1) Askani, R.; Taber, D. F. In ComprehensiVe Organic Synthesis; Trost,
B. M., Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 6, p 126.
(2) Reference 1, p 126. See also: Atkinson, R. S. In ComprehensiVe
Organic Synthesis; Barton, D. H. R., Ollis, W. D., Eds.; Pergamon Press:
Oxford, 1979; Vol. 2, p 246.
The need for a mild and reliable procedure for the
preparation of R-diazoketones arose from our studies on the
total synthesis of the CP molecules3 in which the venerable
(3) Part 1: Nicolaou, K. C.; Baran, P. S.; Zhong, Y.-L.; Choi, H.-S.;
Yoon, W. H.; He, Y.; Fong, K. C. Angew. Chem. Int. Ed. 1999, 38, 1669.
Part 2: Nicolaou, K. C.; Baran, P. S.; Zhong, Y.-L.; Fong, K. C.; He, Y.;
Yoon, W. H.; Choi, H.-S. Angew. Chem. Int. Ed. 1999, 38, 1676. For an
additional rationale for the employment of acyl mesylates to construct
hindered diazoketones, see Scheme 9 of Part 1.
(4) (a) Dunn, A. D.; Mills, M. J.; Henry, W. Org. Prep. Proced. Int.
1982, 14, 396. (b) Chandrasekaran, S.; Turner, J. V. Synth. Commun. 1982,
12, 727. (c) Ja´szay, Z. M.; Petneha´zy, I.; To¨ke, L. Synthesis 1989, 745. For
work on acyl fluorosulphonates, see: Olah, G. A.; Narang, S. C.; Garcia-
Luna, A. Synthesis 1981, 790.
10.1021/ol990790l CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/31/1999

Arndt-Eistert homologation5 was utilized to install the
quartenary center present in the natural products. The
resistance of the extremely hindered (concave situated)
carboxylic acid group of 1 to conversion to the corresponding
acid chloride with Vilsmeir reagent6 even under forcing
conditions (Scheme 1) sparked our efforts to develop an
Table 1. CP Diazoketone Intermediates Synthesized
Scheme 1a
a Reagents and conditions: i, Vilsmeir reagent (excess), THF:
CH3CN (2:1), -20 to 25 °C, 10 h, then 1 h sonication, ca. 30%
decomposition, ca. 70% methyl ester after addition of CH2N2; ii,
MsCl (5.0 equiv), Et3N (10 equiv), THF, -50 to -10 °C, 5 min,
then CH2N2 (solution in anhydrous Et2O), 1 h, 86%.
operationally simplistic protocol for hindered carboxyl
activation. We reasoned that a highly reactive reagent such
as sulfene (produced from methanesulfonyl chloride and a
suitable base) might rapidly engage even the most hindered
carboxylic acids such as 1 to yield an activated species (acyl
mesylate) which would rapidly react with diazomethane to
produce the desired R-diazoketone. In the event, treatment
of acid 1 with MsCl/Et3N at -10 °C for 3 min followed by
addition of CH2N2 at 0 °C afforded the R-diazoketone 2 in
86% yield.
While traveling through the CP “synthetic labyrinth” 3 we
had the opportunity to perform this reaction on several
substrates in our search for a final synthetic route (Table 1).
Acid labile functionalities (as seen in examples 4-7), base-
sensitive functionalities (as seen in examples 3-5 and 8),
and the presence of the maleimide functionality (6) are
compatible with the developed conditions.
a Reactions were performed at 0.02-0.2 M in THF. Reactions were
complete after ca. 30 min. Substrate 8 required 4 h for complete reaction.
The synthesis of compounds 4 and 5 is reported in ref 3. Other compounds
will be reported in the full account of ref 3. Temperatures as low as -50
°C worked; however, the reaction was usually carried out at -10 °C.
b Isolated yield of pure diazoketone. Small amounts (1-8%) of methyl ester
could be separated by flash chromatography (SiO2, hexane:EtOAc 5:1 to
2:1).
this method, coupled with its extremely mild nature, was a
pivotal element of our eventual completion of the CP
molecules.3
Curiously, we found that in virtually all cases the reactions
of hindered carboxylic acids furnished a small amount (1-
8%) of the corresponding methyl ester regardless of the
number of equivalents of MsCl/Et3N (5 and 20 equiv lead
to the same result) or the time allotted for acyl mesylate
generation (3 and 15 min lead to the same result). We
reasoned that the methyl ester byproduct must arise from
small amounts of symmetrical anhydride which are produced
in the first step of the reaction. Verification of this hypothesis
evolved from experiments on more simple acids (vide infra).
In most cases, the crude diazoketones were suitable for
Wolff rearrangement after aqueous workup, thus completing
the Arndt-Eistert homologation. Indeed, the reliability of
(5) Arndt, F.; Eistert, B. Ber. Dtsch. Chem. Ges. 1935, 68B, 200. We
had identified the Arndt-Eistert reaction as the most prudent method to
complete the quartenary center after a series of model studies which will
be delineated in a full account of ref 3.
(6) For leading references, see: Marson, C. M.; Giles, P. R. Synthesis
Using Vilsmeir Reagents; CRC Press: Boca Raton, FL, 1994.
884
Org. Lett., Vol. 1, No. 6, 1999

As we turned our attention to other sterically hindered
carboxylic acids in order to establish the generality of the
protocol, we discovered a pronounced solvent effect on the
reaction. Scheme 2 illustrates the effect of different solvents
Table 2. Selected Hindered Diazoketones (12-17) Synthesized
via the Corresponding Acyl Mesylates
Scheme 2. Solvent Effects on the Formation of
Diazoketone 10a
a All reactions were performed on a 1.0 mmol scale at 0.2 M in
the indicated solvent. Entry 1 was performed without added Et3N.
b Determined by 1H NMR spectroscopy of the crude reaction
mixture. c Total isolated yield of 10 + 11.
on the ratio of desired R-diazoketone 10 to the methyl ester
byproduct 11 in the reaction of (1S)-(-)-camphanic acid 9.
Not surprisingly, in the cases in which triethylamine is used
as the base (entries 2-4), a general trend is observed in
which more polar solvents favor formation of the diazo-
ketone, presumably due to a stabilizing effect on the polar
acyl mesylate intermediate (vide infra). Table 2 further
demonstrates the generality of this condition utilizing ac-
etonitrile as solvent with a variety of hindered acids.
Particularly striking is the selective conversion of the
vancomycin derivative 177 to the corresponding diazoketone,
containing no less than nine amides and a multitude of
sensitive and easily epimerizable centers. This task was also
accomplished without dehydration of the more easily acces-
sible primary amide to the corresponding nitrile.8 We have
also found that TMSCHN2 may be used as an alternative to
CH2N2 to attack acyl mesylates; however, reaction times are
considerably longer (ca. 8 h at 0 °C).9
a Reactions were performed at 0.2 M in acetonitrile at 0 °C except for
acid 17. Diazoketone generation was complete in ca. 1 h. Substrates 12-
16 were commercially available. b Isolated yields of pure diazoketones.c See
refs 7 and 8. 1,2-Dichloroethane was employed as solvent at -26 °C.
byproduct frequently observed in these reactions. We were
indeed surprised to observe exclusive formation of the methyl
ester accompanied by the corresponding symmetrical anhy-
dride upon application of our protocol to several simple
substrates. Performing the reaction under the conditions
prescribed by Dunn et al.4a for the preparation of the acyl
mesylate of cinnamic acid (19) (Scheme 3) led only to the
corresponding symmetrical anhydride 20 along with some
Although the formation of the acid chloride of carboxylic
acids is the most logical route to R-diazoketones from simple,
sterically unblocked acids, we sought to understand the
mechanism of the reaction by using these acids as it might
explain the unpredictable appearance of the methyl ester
(7) For a related synthesis of compound 17, see: Nicolaou, K. C.;
Mitchell, H. J.; Jain, N. F.; Winssinger, N.; Hughes, R.; Bando, T. Angew.
Chem. Int. Ed. 1999, 38, 240; Nicolaou, K. C.; Mitchell, H. J.; Jain, N. F.;
Winssinger, N.; Hughes, R.; Bando, T., Koumbis, A, Natarajan, S. Chem.
Eur. J. 1999, in press. Special thanks to N. Winssinger for a generous sample
of 17.
(8) In this instance, 1,2-dichloroethane proved to be a superior solvent
to acetonitrile. The dehydration of primary amides to the corresponding
nitriles with methanesulphonyl chloride is known: see ref 4a. Acid 17 is
completely unreactive with conventional esterification methods due to the
extreme steric shielding present in its vicinity.
1
unreacted starting material (observed by H NMR and IR
spectroscopy of the crude reaction mixture).10
(10) The use of methanesulfonyl chloride in the synthesis of symmetrical
anhydrides is known: Nangia, A.; Chandrasekaran, S. J. Chem. Res., Synop.
1984, 100.
(11) For the synthesis of 25, see: Nicolaou, K. C.; Pfefferkorn, J. A.;
Kim, S.; Wei, H.-X. J. Am. Chem. Soc. 1999, 121, 4724. J. Pfefferkorn is
gratefully acknowledged for providing a generous sample of 25.
(12) Brewster, J. H.; Ciotti, C. J., Jr. J. Am. Chem. Soc. 1955, 77, 6214.
(9) The acid chlorides which we have tried are unreactive to TMSCHN2.
Org. Lett., Vol. 1, No. 6, 1999
885

Scheme 3. Reinvestigation of the Chemistry of Acyl Mesylates
Scheme 4. Synthesis and Characterization of the Stable Acyl
Mesylate 26
Previous investigators speculated that they were producing
stable acyl mesylates (rather than simple anhydrides) because
the nucleophiles which were employed (R-OH,4b R-NH2,4a,c
R-NHNH24c) may react with the symmetrical anhydrides
before reacting with any remaining reagent (the remaining
reagent then recycles the newly formed acid). In 1955,
Brewster and Ciotti reported a similar phenomenon of
complementary reactivity using tosyl chloride.12 Thus, the
dehydrative properties of sulfonyl chlorides can be easily
mistaken for the intermediacy of an acyl sulfonate. In the
case of hindered carboxylic acids (Tables 1 and 2), the
formation of symmetrical anhydrides is disfavored Vis-a`-Vis
the formation of an acyl mesylate species for steric reasons.
This fortuitous reactivity difference allows rapid formation
of R-diazoketones via these highly reactive intermediates.13
In conclusion, we have developed a very mild one-pot
synthesis of R-diazoketones from extremely hindered car-
boxylic acids, directly proven the intermediacy of acyl
mesylates in these reactions, and shed light on the substrate-
dependent mechanisms which are operative in the synthesis
of acyl mesylates.
In the case of acids 18 and 22-24, the corresponding
symmetrical anhydride reacted slowly or not at all with
diazomethane. Aqueous workup of the reaction hydrolyzed
the anhydride, thus liberating the corresponding carboxylic
acid which was quickly trapped with diazomethane to furnish
the methyl ester. It should also be mentioned that despite
repeated attempts using a variety of different conditions, we
were unable to synthesize the acyl mesylates of compounds
18 and 22-24.
To directly prove the existence of the acyl mesylate
species, we employed the extremely hindered acid 25.11 Thus,
treatment of 25 with Et3N/MsCl at 0 °C for 10 min followed
by filtration through Celite, evaporation, and dissolution in
CDCl3 allowed the first spectrocopic measurements to be
taken of an acyl mesylate (26) (Scheme 4). Treatment with
CH2N2 at 0 °C for 20 h led to the corresponding diazoketone
in 30% yield (unoptimized).
Acknowledgment. We thank Drs. D. H. Huang and G.
Siuzdak for NMR spectroscopic and mass spectrometric
assistance, respectively. This work was financially supported
by the National Institutes of Health (USA), The Skaggs
Institute for Chemical Biology, a postdoctoral fellowship
from the Korea Science and Engineering Foundation (H.-
S.C.), doctoral fellowships from the National Science
Foundation (P.S.B.) and Boehringer Ingelheim (Y.H.), and
grants from Pfizer, Glaxo, Merck, Schering Plough, Hoff-
mann-LaRoche, DuPont, and Abbott Laboratories.
(13) All compounds reported were fully characterized spectroscopically.
In addition, all diazoketones exhibited a medium to strong intensity band
in the infrared at 2105-2120 cm-1. General Procedure: To a solution of
of (1S)-(-)-camphanic acid 9 in CH3CN (0.2 M) was added Et3N (10 equiv)
at 0 °C. Freshly distilled methanesulphonyl chloride (5.0 equiv) was then
added dropwise over 1 min. After 10 min an excess of CH2N2 (anhydrous
ether solution) was added so that the color of the solution remained a
constant light yellow. After 30 min, the reaction was complete as ascertained
by TLC. The reaction was quenched upon addition of H2O, and the product
was extracted with Et2O, washed with brine, and dried over MgSO4.
Concentration followed by flash chromatography (SiO2, hexanes:EtOAc 5:1)
furnished an inseparable mixture of diazoketone 10 and methyl ester 11
(6:1) in 86% yield.
OL990790L
886
Org. Lett., Vol. 1, No. 6, 1999