Studies of the H-D exchange mechanism of malonganenone B

Article abstract of DOI:10.1039/c2ob06926a

Malonganenone B (1) exhibits an unusual H-D exchange of a formyl proton when in deuteric-NMR solvents. Synthetic and kinetic investigations were made to probe the mechanism of this exchange, which appears to occur via an uncommon and transient amine-amide

Full text of DOI:10.1039/c2ob06926a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 1725
www.rsc.org/obc
COMMUNICATION
Studies of the H–D exchange mechanism of malonganenone B†
Peter G. K. Clark,a,b Matthias Leina,c and Robert A. Keyzers*a,b
Received 15th November 2011, Accepted 8th December 2011
DOI: 10.1039/c2ob06926a
H–D exchange of carbonyl-bound protons is very unusual,
Malonganenone B (1) exhibits an unusual H–D exchange of
a formyl proton when in deuteric-NMR solvents. Synthetic
and kinetic investigations were made to probe the mechanism
of this exchange, which appears to occur via an uncommon
and transient amine–amide NHC intermediate.
especially in the absence of any catalyst, but a related exchange
had been observed previously in N,N-dialkylformamide acetals
(2).3 The effects upon the H–D exchange of changing alkyl sub-
stituents, solvent and addition of exogenous acid or base were
examined.3 Carbene, self-protonation and ylide mechanisms
were proposed (see Scheme 1 in ESI†) with the ylide example
deemed the best fit to the data. However, not all of the exper-
imental observations could be reconciled with this mechanism,
especially the putative base inhibition that should in fact
enhance the reaction kinetics.
Since then, progress in the field of N-heterocyclic carbenes
(NHC) has added credence to the alternative carbene mechanism
previously proposed (Fig. 2).3 With this knowledge in mind, we
sought to determine if either the proposed ylide or carbene
mechanisms were responsible for the H–D exchange observed in
1. We report here our evidence supporting an NHC-based mech-
anism for the exchange, which involves a rare mixed amine–
amide NHC intermediate.
The isolation of malonganenone B (1) was first reported from
the gorgonian Leptogorgia gilchristi collected at a reef near
Ponto Malongane, Mozambique,1 and again shortly after from
Euplexaura nuttingi (both Order Alcyonacea) collected from
Uvinage, Tanzania.2 The original structural elucidation of 1 was
problematic as mass spectral analysis, following NMR acqui-
sition in CD3OD, resulted in two quasi-molecular ion peaks dif-
fering by one mass unit, implying substitution of a proton for a
deuteron. Further, 13C NMR data suggested the presence of two
isotopomeric formamide carbonyl carbon resonances that, in
addition to reduced proton integration for this functionality, indi-
cated formyl H–D exchange had occurred (Fig. 1).
Our hypothesis for exchange was the fully reversible attack of
the N-methyl amide of 1 upon the formamide carbonyl followed
by nitrogen-assisted elimination of water that would generate an
NHC precursor in a similar manner to the proposed carbene
mechanism for 2 (Scheme 1).
To test this theory, models of 1 were synthesised for kinetic
studies. Analogue 3 had a methyl group in place of the tetrapre-
nyl tail while 4 contained a phenyl ring instead of the imidazole
(Scheme 2). Hydrolysis of caffeine (5) yielded intermediate 6
that was then formylated using a mixed anhydride to yield 3 in
29% overall yield.4 Difficulties in the purification of 3, however,
prompted the synthesis of a related analogue for further studies.
N-Methylisatoic anhydride (7) was attacked by methylamine to
give intermediate 8,5 which was similarly formylated with the
mixed anhydride to give 4 in an 84% overall yield.
Fig. 1 Formyl H–D exchange in malonganenone B (1).
aSchool of Chemical and Physical Sciences, Victoria University of
Wellington, PO Box 600, Wellington, 6140, New Zealand
bCentre for Biodiscovery, Victoria University of Wellington, PO Box 600,
Wellington, 6140, New Zealand. E-mail: rob.keyzers@vuw.ac.nz; Fax:
+64 4 463 5241; Tel: +64 4 463 5117
These compounds were then subjected to kinetic analyses
using NMR spectroscopy. Upon dissolution in D2O or CD3OD,
the measured integral of the formyl proton resonance of each
cCenter for Theoretical Chemistry and Physics, New Zealand Institute
for Advanced Study, Massey University Auckland, Private Bag 102904,
North Shore City, Auckland 0745, New Zealand
†Electronic supplementary information (ESI) available: Three proposed
mechanisms for H–D exchange in dialkylated formamide acetals. Exper-
imental procedures for the preparation and characterisation of 3, 4, 6, 8,
9, 11 and 13–15. Kinetic analyses and data for 4 and 11. 1H and 13C
NMR spectra for 3, 4, 9, 11, 14 and 15, 1H NMR for 13 and both
detected and simulated MS spectra for 15. Details of molecular model-
ling of pKa values. See DOI: 10.1039/c2ob06926a
Fig. 2 Proposed carbene mechanism of H–D exchange in formamide
acetals.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1725–1729 | 1725

View Article Online
Scheme 2 Synthesis of N-methyl imidazole (3) and benzene (4)
analogues.
malonganenone B is independent of the tetraprenyl tail of 1.
Addition of exogenous acid or base had opposing effects on the
exchange; triethylamine was catalytic whereas acetic acid-d4 was
inhibitory (Fig. 4). This contradicts the previously reported
observations and is consistent with both the proposed ylide and
carbene mechanisms.3
Scheme 1 Proposed mechanism of H–D exchange for 1.
species decayed with time, exchanging for a deuteron, as shown
by the loss of the formyl proton resonance of 4 (Fig. 3)6 and by
MS. These results confirmed the H–D exchange of
Evidence of cyclisation was sought to confirm our hypothesis
that the proposed mechanisms for H–D exchange of N,N-
Fig. 3 1H NMR (D2O, 600 MHz) spectrum demonstrating H–D exchange of the formyl proton resonance of 4.6
1726 | Org. Biomol. Chem., 2012, 10, 1725–1729
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 5 Determination of the role of base upon the H–D exchange kin-
etics of cation 11.
Fig. 4 Effect of acid or base addition on the rate of H–D exchange of
4.
could not be established as the range of concentrations required
to probe the exchange rate were too dilute for suitable obser-
vation by NMR.
More direct evidence of the proposed NHC forming in the
exchange was sought. Carbenes have historically been trapped
by reaction with an alkene to form stable cyclopropanes;
however, attempts to repeat this feat with 3, 4 or 11 with a
variety of alkene traps all failed. Similarly, attempts to form
dimerised carbenes by direct base deprotonation of 11 were also
unsuccessful.
Fortunately, organometallic methods under full Schlenk con-
ditions were more fruitful. Using silver oxide, a silver-NHC
complex (12) could be generated from 11 (Scheme 4).8 In the
absence of another metal species, fulvalene 13 was observed,
which was believed to result from ligand exchange between two
molecules of 12 followed by reductive elimination. The structure
of 13 was confirmed by MS and NMR (see ESI†), with the
observed linking carbon resonance chemical shift of 106.8 ppm
being comparable to the literature data of enetetramines.8 Given
the lack of steric protection around the electron-rich alkene,
however, this species rapidly degraded to 14 and could not be
purified or characterised further.
Silver-NHC complexes have had great use as transmetallation
agents to generate other organometallic complexes. When 12
was generated in the presence of palladium acetate, transmetalla-
tion occurred to give the transient organometallic complex 15 as
a mix of isomers (Scheme 4). This was confirmed by 2D-NMR,
where a NHC-Pd carbon resonance of δC 199.2 ppm was
detected via HMBC, consistent with the literature,9 and MS data
that demonstrated the characteristic palladium isotope pattern for
a dimerized Pd-system. As before, the lack of protection around
the NHC-complex meant this species also rapidly degraded to 14
before further purification or isolation could be achieved. Syn-
thesis of species related to 3 or 11 with increased steric protec-
tion has yet to yield more stable organometallic products.
The accumulated evidence suggests that a mixed amine–
amide NHC is forming in the H–D exchange of malonganenone
B. The observation of cyclic species, the kinetic evidence and
the formation of 11, 13 and 15 all suggest that an NHC is
Scheme 3 Formation of aminal ester 9 and quinazolinium cation 11.
dialkylformamide acetals could be applied here. Reaction of 4
with CH3OH resulted in the formation of aminal ester 9
(Scheme 3), which was proposed to form via amide attack of the
formamide followed by exchange of the resultant hydroxide for
a methoxide. The lability of the groups around the exchanging
centre in 9 is analogous to that of N,N-dialkylformamide acetals,
which supports the application of the proposed NHC-
mechanism.
Further attempts to trap 4 as the carbonate 10 with ethylchlor-
oformate had a fortunate outcome, with spontaneous decarboxy-
lation yielding the isolable quinazolinium cation 11 instead
(Scheme 3). The apparent stability of this species supports the
NHC-mechanism as it represents the substrate for the rate deter-
mining step of such a mechanism. Gratifyingly, 11 also under-
went H–D exchange in a manner similar to 3 and 4. Kinetic
studies using this compound revealed the exchange is first order
with respect to base at low concentrations,‡ concordant with the
proposed NHC-mechanism (Fig. 5). The pKa of the quinazoli-
nium proton in water was estimated to be 19.8 by use of compu-
tational methods and calibrated using a linear regression model
to match experimentally determined values.7 The pKa of the cor-
responding proton of 3 was determined to be 30.0 using the
same model and would explain the reduced rate of exchange of
3 versus 4 or 11. Unfortunately, the order for exchange of 11
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1725–1729 | 1727

View Article Online
Scheme 4 Transient formation of fulvalene 13 and organometallic complex 15.
g. nuttingins C–E) have all been reported as marine natural pro-
ducts, with the stable quinazolinium cation nuttingin F (e.g.
motif 17) noted as a degradation product. These isolable com-
pounds support our proposed mechanism as plausible intermedi-
ates and products, and also imply the key role of cation 17 as a
core metabolite linking all of the structural variants of the malon-
ganenone and nuttingin families.
The kinetic data and transient formation of organometallic
species using similar analogues of malonganenone B have indi-
cated that an NHC-based carbene mechanism is the likely route
for H–D exchange in this molecule. This is at odds with the pre-
viously proposed mechanism of exchange in related N,N-dialky-
lated formamides. The formation of cation 11 indicates the
stability of the direct precursor to the NHC and adds weight to
our argument. Further investigations of the chemical impli-
cations, both structural and catalytic, and biological activities
related to this mechanism are currently underway.
Scheme 5 Relationships between malonganenones and nuttingins.
generated from these models, which in turn suggests an NHC
may be formed as a transient species in solutions of the natural
product. To the best of our knowledge, stable mixed NHCs of
this nature are rare with only a few recent examples being
reported.10 There are equally few examples of naturally occur-
ring NHC-based reactions or biosyntheses, although thiamine-
based carbenes have been implicated in several transketolase
reactions.11 It is therefore possible that further NHC-based mech-
anisms beyond thiamine may exist in nature that could have pro-
found biosynthetic and bioactive implications, although the
reaction kinetics in water would presumably quench such a
carbene rapidly and therefore a purely defensive bioactive role
for such an NHC is unlikely.
These observations also shed new light on the relationships
between the alkaloid moieties of the related malonganenone and
nuttingin families (Scheme 5).1,2 As proposed for the biogenesis
of 1 from malonganenone A,1 alkylation of 16 would yield 17,
the NHC-precursor and common ancestor to the other species.
Formation of an NHC from this and subsequent oxidative degra-
dation would yield 18, while enzymatic reduction could give 19.
Alternatively, hydrolysis of 17 would yield 20, as found in mal-
onganenone B. It should be noted that the structural motifs of 16
(e.g. malonganenone A), 18 (e.g. nuttingins A and B) and 19 (e.
Acknowledgements
The authors thank Drs Mattie Timmer, Joanne Harvey, Bradley
Williams and John Spencer for significant and useful discus-
sions, as well as for the generous donation of Pd(OAc)2 and
degassed NMR solvents (J. Spencer). Ms. Melanie Nelson is
thanked for technical assistance. PGKC also gratefully acknowl-
edges the financial support of a VUW Science Faculty Strategic
Grant and the Curtis Gordon Scholarship.
Notes and references
‡As analysed by a half-life based method, which is outlined in the sup-
porting information; by B. Cox, in Modern Liquid Phase Kinetics,
Oxford University Press, Oxford, 1994.
1 R. A. Keyzers, C. A. Gray, M.
H Schleyer, C. E. Whibley,
D. T. Hendricks and M. T. Davies-Coleman, Tetrahedron, 2006, 62, 2200.
2 H. Sorek, A. Rudi, Y. Benayahu, N. Ben-Califa, D. Neumann and
Y. Kashman, J. Nat. Prod., 2007, 70, 1104.
3 G. Simchen and W. Kantlehner, Tetrahedron, 1972, 28, 3535.
4 R. Hoskinson, Aust. J. Chem., 1968, 21, 1913.
5 O. S. Tee and G. V. Patil, J. Org. Chem., 1976, 41, 838.
1728 | Org. Biomol. Chem., 2012, 10, 1725–1729
This journal is © The Royal Society of Chemistry 2012

View Article Online
6 This species exists as a pair of rotamers, resulting in two sets of similar
resonances. The formyl resonance of each rotamer disappears at an equal
rate.
7 (a) D. P. Dissanayake and R. Senthilnithy, J. Mol. Struct. (THEOCHEM),
2009, 910, 93; (b) B. Ghalami-Choobar, H. Dezhampanah, P. Nikparsa
and A. Ghiami-Shomami, Int. J. Quantum Chem., 2011, DOI: 10.1002/
qua.23211.
8 (a) H. M. J. Wang and I. J. B. Lin, Organometallics, 1998, 17, 972;
(b) J. Kamplain, V. Lynch and C. Bielawski, Org. Lett., 2007, 9, 5401.
9 I. Ozdemir, S. Demir, O. Sahin, O. Buyukgungor and B. Cetinkaya,
J. Organomet. Chem., 2010, 695, 1555.
10 For example, L. Benhamou, N. Vujkovic, V. César, H. Gornitzka,
N. Lugan and G. Lavigne, Organometallics, 2010, 29, 2616.
11 D. Enders and A. A. Narine, J. Org. Chem., 2008, 73, 7857.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1725–1729 | 1729