Experimental and quantum-chemical study of nucleophilic substitution mechanism in berberine

Article abstract of DOI:10.1007/s10593-016-1810-1

Principal differences in the interaction mechanisms of alkaloid berberine with primary and secondary amines were investigated experimentally and by quantum-chemical calculations according to density functional theory (DFT/B3LYP) with 6-31G** basis set. The nucleophilic substitution of 9-metoxy group with primary amine was shown to proceed through a stage of σ-complex formation and led to 9-alkylamino derivatives of berberine. Analogous substitution with a secondary amine did not occur due to unfavorable thermodynamic parameters. The secondary amine participated in this reaction not as the attacking nucleophile, but rather as a bifunctional catalyst of berberine hydrolysis to berberrubine. The driving force for all these processes was the stabilization of products by hydrogen bonding. Based on the obtained results, we developed a new effective method for the preparation of berberrubine, one of the key intermediates in synthetic transformations of berberine. New 9-monoalkylamino derivatives of berberine containing indole moieties were synthesized.

Full text of DOI:10.1007/s10593-016-1810-1

DOI 10.1007/s10593-016-1810-1
Chemistry of Heterocyclic Compounds 2015, 51(11/12), 997–1007
Experimental and quantum-chemical study
of nucleophilic substitution mechanism in berberine
Oleg N. Burov1*, Mikhail E. Kletskii1, Nikita S. Fedik1,
Sergey V. Kurbatov1, Anton V. Lisovin1
1
Southern Federal University,
7 Zorge St., Rostov-on-Don 344090, Russia;
e-mail: bboleg@gmail.com
Translated from Khimiya Geterotsiklicheskikh Soedinenii,
2015, 51(11/12), 997–1007
Submitted November 4, 2015
Accepted December 14, 2015
H2O
Cl–
O
O
N+
NR2
H2O
Cl–
H2O
Cl–
O
O
O
O
N+
N+
NH2R
NHR2
OMe
NHR
OMe
OMe
OMe
O
O
N+ Cl–
OH
OMe
Principal differences in the interaction mechanisms of alkaloid berberine with primary and secondary amines were investigated
experimentally and by quantum-chemical calculations according to density functional theory (DFT/B3LYP) with 6-31G** basis set. The
nucleophilic substitution of 9-metoxy group with primary amine was shown to proceed through a stage of σ-complex formation and led
to 9-alkylamino derivatives of berberine. Analogous substitution with a secondary amine did not occur due to unfavorable
thermodynamic parameters. The secondary amine participated in this reaction not as the attacking nucleophile, but rather as a
bifunctional catalyst of berberine hydrolysis to berberrubine. The driving force for all these processes was the stabilization of products by
hydrogen bonding. Based on the obtained results, we developed a new effective method for the preparation of berberrubine, one of the
key intermediates in synthetic transformations of berberine. New 9-monoalkylamino derivatives of berberine containing indole moieties
were synthesized.
Keywords: berberine, berberrubine, alkylamination, catalytic hydrolysis, DFT calculations.
The alkaloid berberine (berberine chloride) exhibits a
unique set of medicinally important properties, as recently
described in several review articles.1–6 For example, it reduces
bleeding, shows antispasmodic, antigastric, hypolipidemic,
antiarrhythmic, and hypotensive effects. The antimicrobial,
antitumor, and antileucosis activity of berberine have been
experimentally proven. Berberine is also used in medicinal
increasing acetylcholine concentration by inhibition of
acetylcholinesterase (AChE).8,9 Secondly, these compounds
can take part in π–π-interactions with DNA G-quadruplex
structures.10–12 The introduction of an amino group at
position 9 of berberine molecule resulted in a strong
interaction with the telomeric DNA G-quadruplexes and an
inhibiting effect on telomerase, possibly due to an
increased electrostatic interaction between the terminal
amino group of the side chain and the phosphate backbone
in the G-quadruplex of DNA.13
chemistry as
a fairly reactive molecular scaffold,
containing at least four diversification sites: С-8, С-9, С-13,
and N-7.
The main methods for the preparation of berberine
derivatives have been described in a review.7 Of special
interest among these are the 9-O- and 9-N-alkyl derivatives
of berberine, for two reasons. Firstly, these compounds
show valuable medicinal properties by reducing the
symptoms of Alzheimer's disease and senile dementia, and
It has been reported12 that 9-N-substituted berberine
derivatives show a pronounced inhibitory effect on the
transcription of c-Myc gene in tumor cells and enhanced
cytotoxicity, compared to berberine.11 Thus, there is a
significant interest in the field of medicinal chemistry
towards the development of methods for functionalization
0009-3122/15/51(11/12)-0997©2015 Springer Science+Business Media New York
997

Chemistry of Heterocyclic Compounds 2015, 51(11/12), 997–1007
Scheme 1
O
O
O
O
O
5
4
1
3
2
6
H2O, 220°C, 20 mm Hg
or
E
N+
7N+ Cl–
N+
Cl–
Cl–
NuH
O
H2O, NHR2 (EtOH), 
8
13
Nu
OMe
OMe
O
– MeOH
– MeOH
9
H
12
10
OMe
OMe
2
1
3
Berberine
Berberrubine
of berberine at position 9, for example, by introducing
amino groups.
Two methods are commonly used for the functionalization
of berberine (Scheme 1): nucleophilic substitution of the
9-metoxy group in berberine (1), forming products with the
general structure 2, and the interaction of berberrubine (3)
(OH-derivative of berberine) with electrophiles.
Primary aliphatic amines are typically used as
nucleophiles (NuH), while alkyl halides14 and halides15
serve as electrophiles. It should be noted that, according to
Chemical Abstracts, 9-dialkylamino derivatives of
berberine have never been described, while the greater
nucleophilicity of secondary amines compared to primary
amines should enhance their reactivity in the substitution of
9-metoxy group.
The goal of this study was to determine the reasons for
such an unusual selectivity of berberine with regard to the
structure of amine. For this, we at first performed a
comparative experimental and quantum-chemical study
(DFT calculations with B3LYP/6-31G** basis set) of
berberine reaction with primary and secondary amines.
leads to the formation of salt 4e, which consists of
methylamine derivative in cation form and a chloride anion
(Scheme 2).
We found three stable topomers for this ion pair, the
most stable of which (topomer 4e) contained a Cl– ion
coordinated with a proton of the amino group. It was more
stable by 6.6 kcal/mol than the topomer 4e', in which the
methyl group was oriented towards the Cl– ion, and more
stable by 17.5 kcal/mol than the topomer 4e'', in which the
chloride ion was coordinated with the H-13 proton (Fig. 1).
For the three systems 4e, we performed a topological
analysis of electron density distribution function according
to the "atoms in molecules" (AIM) model.17 The obtained
data indicate that there were two attractive interactions in
the most stable topomer 4e. We identified critical points (3,–1)
characteristic for the shortened Cl–···H–N contact [ρ(rc) =
1.1·10–2, 2ρ(rc) = 3.1·10–2] and a weak attractive Cl–···H(8)
interaction of similar magnitude [ρ(rc) = 1.1·10–2, 2ρ(rc) =
3.2·10–2]* (Fig. 1). This type of double coordination
stabilized the system by 6.6 kcal/mol (taking into account
the correction for zero-point energy (ZPE)). At the same
time, the topomeric forms 4e' and 4e'' were characterized
by only one attractive interaction of chloride ion with the
berberine molecule (Fig. 1). More detailed data are available
in Table S1 in the Supplementary information file.
The nucleophilic attack by methylamine on the
berberine molecule begins with the barrierless formation of
aggregate 5 of these two molecules, which is more stable
by 5.3 kcal/mol than the isolated reactants. The reaction
further proceeds through the transition states TS 1 and TS 2,
forming the hydrogen bond-stabilized intermediate 6, with
the sequence of low-barrier processes resulting in the
complex 7, which is stabilized by a chain of two hydrogen
bonds. In general, this entire process is thermodynamically
favorable (Scheme 3, Fig. 2; the most important stationary
points on MEP are given in Table S2 in the Supplementary
information file).
1. The reaction of berberine with methylamine
All the previously proposed methods for the synthesis of
9-amino derivatives of berberine gave the target products in
25 to 60% yields (the majority of compounds were
obtained in 40–45% yields). The main procedure used for
these syntheses was refluxing of berberine with the
respective amine in absolute alcohol for many hours.9,11,16
Procedure for the first developed by us involves the
reaction of berberine with amines in the absence of solvent,
giving 9-aminoberberines 4a–d in higher 50 to 60% yields
(Scheme 2).
The interaction of berberine with secondary amines
under analogous conditions did not lead to the formation of
9-dialkylamino derivatives. In order to explain this result,
we calculated the minimum energy pathway (MEP) of
nucleophilic attack by methylamine molecule on berberine
(1). According to the obtained data, this process eventually
* ρ(rc) – electron density; 2ρ(rc) – the Laplacian of electron density.
Scheme 2
O
O
O
O
Cl
Cl
4 a
b
R =
R =
R =
,
,
O
N
N
NH2R
OMe
OMe
NHR
OMe
120°C, 45 h
HO
c
d
R =
,
1
4a d

N
N
H
H
998

Chemistry of Heterocyclic Compounds 2015, 51(11/12), 997–1007
Figure 1. The calculated interatomic distances (Å) in 9-N-methylberberinе (4e). The attractive forces in the three topomers of compound
4е are also shown.
At the same time, in the case of dimethylamine, the
formation of 9-N,N-dimethylberberine (4f) in a bimolecular
nucleophilic process is thermodynamically unfavorable, since
the sum of energy of molecule 4f and methanol at infinite
separation was higher by 1.4 kcal/mol than the energy of
reactants (Scheme 4, Table S3 in the Supplementary
informetion file). The chloride ion in system 4f was shifted
to a peripheral location and practically did not stabilize the
salt by additional attractive interactions, as was found in
the case of primary amine in compound 4e (Fig. 3, the
results of AIM calculations are presented in Figure S10 of
the Supplementary information file).
Scheme 3
999

Chemistry of Heterocyclic Compounds 2015, 51(11/12), 997–1007
structural study, compound 4a formed a chloride salt
crystal solvate with unit cell containing two independent
molecules 4a, as well as water and nitrobenzene in 2:3:1
ratio (Fig. 5). The presence of several proton donor and
acceptor groups in compound 4a created a strong hydrogen
bond network. Thus, the crystal contained hydrated
chloride ion chains, featuring an alternating sequence of six-
membered (H2O)4Cl2 and four-membered (H2O)2Сl2 rings.
The Сl···H–O and О···H–О distances characterizing the
strength of hydrogen bonds in the aforementioned chains
were 2.306(4)–2.334(4) Å and 2.002 (2) Å, respectively.
The water molecules involved in the four-membered rings
also formed hydrogen bonds with the amino groups of cations,
with the N–H···O distances ranging from 2.104 to 2.192 Å.
The obtained experimental and computational data
clearly indicate that the presence of crystal hydrate water in
the reaction media had an important effect on the structure
of the starting berberinе (1) and its amination products 4.
For this reason, in the next step we calculated the MEP for
berberine reaction with hydrated methylamine and
dimethylamine. The first calculation was performed for a
direct attack by methylamine hydrate (8) on berberine,
where the water molecule acts as a bifunctional catalyst for
the process (Scheme 5).
The process commences by a barrierless formation of
the initial complex 9, in which a water molecule is linked by a
hydrogen bond to the oxygen atom of methoxy group at the
С-10 atom, while the amine molecule donates a proton
forming a hydrogen bond with a water molecule. The
complex 9 is more stable by 0.8 kcal/mol than the reactants
at infinite separation.
The complex 9 was further transformed over a 10.8 kcal/mol
energy barrier to the Meisenheimer complex 10', in which
1,3-sigmatropic migration of hydroxonium ion (proton
Figure 2. MEP of berberine amination with methylamine. The
relative energy values are given taking into account the
corrections for ZPE. The relative Gibbs free energy values are
given in parentheses. The total energy of isolated reactants is
assumed as reference (see Scheme 3 and Table S2 in the
Supplementary information file).
2. The interaction of berberine
with hydrated amines
2.1. Primary amines
The crystal structure of berberine (1) is relatively
complicated. According to X-ray structural analysis, its unit
cell contains the organic cation, chloride anion, and four
molecules of water that are linked by hydrogen bonds
(Fig. 4).18
The structures of its 9-amino derivatives 4 are at least as
complex. For example, according to the data of X-ray
Scheme 4
Figure 4. The crystal structure of berberine according to X-ray
structural analysis data.18 The results of our DFT calculations are
shown in parentheses. The interatomic distances are given in Å.
Figure 3. The calculated interatomic distances (Å) in 9-N,N-
dimethylberberinе (4f). The Mulliken charge on the chloride ion
is also shown.
1000

Chemistry of Heterocyclic Compounds 2015, 51(11/12), 997–1007
We further considered another sequence of events: an
initial attack by water molecule producing the complex of
berberrubine 14 and only then an attack by amine, resulting
in N-methylberberine 16 (Scheme 6). In this case, the
calculations indicated that the interaction of berberine with
hydrated methylamine 8 occurs strictly stepwise. The
reaction proceeds with the formation of complex 12, in
which a water molecule is linked to methylamine, and the
water molecule donates a proton for the hydrogen bond, in
contrast to the complex 9. The complex 12 is more stable
by 10.1 kcal/mol than the reactants at infinite separation.
In the first step, where the activation energy barrier of TS 5
is equal to 17.9 kcal/mol, a water molecule attacks
berberine at position C-9, forming a Meisenheimer
complex (structure 13', MEP minimum, Fig. 6). After that,
a 1,3-sigmatropic migration of methylammonium (proton
transfer) from the oxygen atom of water molecule to the
oxygen atom of metoxy group occurs in the second step. It
should be noted that this transformation is practically
barrierless, and the energy difference of intermediate
structures and isomers 13' and 13'' is quite insignificant –
within the limits of 1–2 kcal/mol (the MEP plateau, Fig. 6). In
the third step of the process, TS 6 is reached with merely
2.6 kcal/mol energy and the thermodynamically stable
system 14 is formed, representing a complex of three
molecules, methylamine, methanol, and berberrubine.
Remarkably, in the first three stages of the process
presented in Scheme 6, the molecule of methylamine
always participates in a N···H···O hydrogen bridge, with
only slight structural changes. During the 1,3-sigmatropic
shift, the length of O···H bond also changes only slightly
from the structure 12 to berberrubine complex 14 (from
1.790 to 1.768 Å according to DFT calculations), despite
the significant size of the migrating ammonium ion,
correlating with the small activation energy at this stage. In
order to better understand the role of methylamine in the
formation of berberrubine (3) (12→14), we modeled the
reaction of berberine (1) with water dimer – isoelectronic
analog of methylamine hydrate (8) (Scheme 7).
Figure 5. Top: the main geometrical characteristics of 9-N-
benzylberberine (4a) according to X-ray structural analysis. The
calculated values are given in round parentheses. The values
obtained by quantum-chemical calculations for 9-N-methyl-
berberine monohydrate (4e) are given in square parentheses. The
interatomic distances are given in Å. Bottom: hydrated chloride
chains in the crystal of 9-N-benzylberberine (4a) according to
X-ray structural analysis.
transfer by a water molecule from amino group to the
metoxy group) led to the formation of complex 10'' (Fig. 6,
Table S4 in the Supplementary information file). A plateau on
MEP with energy barriers not exceeding 1–2 kcal/mol
corresponded to the proton transfer. It should be noted that
one of the hydrogen atoms in the water molecule was
linked by a stabilizing interaction to the chloride ion during
all process stages. The complex 11, consisting of N-methyl-
berberine, a methanol molecule, water, and chloride ion
linked by hydrogen bonds was formed during the
amination reaction. The formation of complex 11 was
thermodynamically favorable by 17.5 kcal/mol relative to the
pre-reaction complex 9.
An earlier experimental study showed that the
hydrolysis of berberine leads to the formation of
berberrubine (3), when this process occurs under extreme
Scheme 5
Me
N H
H
Cl–
O
O
Cl–
O
H
N
H
Me
O
N
O
H
Me
O
TS 3
1 +
NH2Me H2O
N
H
O
H
H
Me
8
O
O
Me
OMe
10'
9
Cl–
H
H
O
O
H
Cl–
O
O
O
H
H
H
N
TS 4
O
O Me
Me
NH
N
4f + H2O + MeOH
Me
Me
O
N
H
OMe
OMe
10''
11
1001

Chemistry of Heterocyclic Compounds 2015, 51(11/12), 997–1007
Scheme 6
conditions – high temperatures and low pressure (Scheme
1).19 Indeed, according to our calculations, the MEP of such
process first involves the formation of complex 17,
stabilized by 24.6 kcal/mol relative to infinitely separated
reactants, and proceeds through three higher energy
transition states. The first of these transition states (TS 9) is
rate limiting – the energy barrier of this stage is maximal,
equal to 33.0 kcal/mol (Table S5 in the Supplementary
information file).
The kinetically and thermodynamically unstable σ-complex
18' was formed in the first step. The complex 18'
underwent further transformations through intermediate
stationary points with insignificant energy differences,
involving the transfer of second (distant) water molecule
from the arriving OH group to the leaving metoxy group
(formation of complex 18'', the plateau in Figure 7). In this
sense, the sequence of steps in the hydrolysis of berberine
(1) by water dimer was not principally different from the
hydrolysis of berberine (1) by methylamine hydrate (8).
According to calculations, the O···H bond length in the
hydrogen bridge also changed only insignificantly from
1.888 to 1.762 Å, when going from the structure 17 to
structure 19 via the transition state TS 10. The final
product, as expected, was the complex of berberrubine with
hydrated methanol 19.
It should also be noted that the relative values of energy
barriers limiting the first stages of berberine hydrolysis
with water dimer and methylamine hydrate (Figs. 6, 7)
Figure 6. MEP of berberine (1) amination with methylamine hydrate (8) through the stage of complex 14 (right side) and direct amination (left
side). The relative energy values include corrections for ZPE. The relative Gibbs free energy values are given in parentheses. The total energy of
isolated reactants is assumed as reference.
1002

Chemistry of Heterocyclic Compounds 2015, 51(11/12), 997–1007
Scheme 7
correlated with the Parr's global electrophilicity indices,
calculated by a known method for both reactants.21 As
expected, the value of ω for the complex 8 was 0.37 eV,
while for the dimer (H2O)2 it was 0.56 eV, indicating a
decrease of nucleophilicity when changing from
methylamine hydrate to dimer of water.
that the further amination of berberrubine (3) according to
the route 14→15→16 must occur under more forcing
conditions, compared to the first three steps. Indeed,
calculations showed that the transformation 14→15
proceeds with a high energy barrier of TS 7 equal to
23.7 kcal/mol (Fig. 6), and the complex 15 thus obtained
must cross another significant energy barrier of TS 8
(27.1 kcal/mol) in order to achieve aromatization.
In aqueous medium this process can lead to two
isomeric complexes 8 and 8'. Obviously, the tricoordinated
nitrogen atom in complex 8 causes amination of berberine,
while the dicoordinated oxygen atom in complex 8' is
responsible for berberine hydrolysis at position С-9 (Fig. 8).
Even though the hydrolysis by dimer of water is
obviously thermodynamically favorable (the energy effect
is equal to 9.9 kcal/mol), practically it can be achieved only
under forcing conditions, according to the published
experimental procedures. Additionally, this provides an
insight into the role of methylamine molecule in the
transformation of berberine (1) to berberrubine (3):
methylamine apparently acts as a bifunctional catalyst,
facilitating the first (rate limiting) step of the process.
Returning to Scheme 6, we should note an important
feature of berberine reaction with methylamine hydrate, in
The formation of amino derivative 16 from the berber-
rubine complex 14 is slightly unfavorable energetically by
merely 0.3 kcal/mol. Additional ab initio (RHF/6-31G**)
computational analysis of the stationary points 14 and 16
showed that the final reaction product 16 is more stable
than the hydrolysis product 14 by 0.8 kcal/mol. It should be
pointed out that the methanol molecule in complex 16 was
not linked by hydrogen bonds to other structures. Despite
the total favorable energy effect of 14.6 kcal/mol relative to
the starting system 12, the process with kinetic control of
steps 14→15→16 can not produce compounds 4a–d under
relatively mild conditions.
Thus, if follows from the results of quantum-chemical
calculations that the reaction with methylamine requires
Figure 7. MEP of berberine hydrolysis with a dimer of water. The relative energy values
include corrections for ZPE. The relative Gibbs free energy values are given in parentheses. The total
energy of isolated reactants is assumed as reference.
1003

Chemistry of Heterocyclic Compounds 2015, 51(11/12), 997–1007
complex 20, more stable by 10.5 kcal/mol than the
reactants at infinite separation, in which the dimethylamine
molecule participated in an N···H···Cl hydrogen bond.
However, the initial attack by amine at the C-9 atom was
not thermodynamically favored, since the postreaction
complex 21 was less stable by 11.5 kcal/mol than the
reactants (Table S6 in the Supplementary information file).
The latter fact can be explained by the absence of
stabilizing (attractive) interactions in the product 21
between the chloride ion and hydroxyl group protons (Fig.
S10 in the Supplementary information file). Besides that,
the water and methanol molecules in complex 21 are not
far removed from the reaction site. The water molecule in
Figure 8. The structures of methylamine–water clusters (8)
according to B3LYP/6-31G** calculations. The most important
interatomic distances are in Å.
less energy and leads directly to 9-amino derivatives of
berberine. The presence of crystal hydrate water in the
starting berberinе (1) facilitates its direct amination, but
can lead to the formation of hydrolysis product –
berberrubine (3) in small amounts.
this case is linked by a bridging proton to the
dimethylamino group, while the methanol molecule in
complex 21 is not linked to the rest of the system by
hydrogen bonds according to AIM calculations.
The conclusions are entirely different when the sequence
of reaction steps is changed – berberine (1) is first attacked
by water, then by dimethylamine (Scheme 9).
2.1. Secondary amines
We also investigated experimentally the interaction of
berberine with secondary aliphatic amines: piperidine,
morpholine, piperazine, N-piperonylpiperazine, and homo-
piperazine. The reactions were continued for 1–24 h at
reflux conditions in anhydrous solvents: ethanol, butanol, and
acetonitrile. The reactions with anhydrous piperidine and
morpholine were also performed without solvent at the
reflux temperatures of amines. In none of the cases we
were able to detect the formation and to isolate the
respective 9-N-dialkyl derivatives of berberine from the
reaction mixtures. It was found that the reaction of
berberine with secondary amines gave 23–82% yields of
berberine hydrolysis product – berberrubine (3) in each
experiment (Scheme 1).
Thus, according to calculations, the first two steps of
berberine (1) hydrolysis by dimethylamine hydrate
(Scheme 9) are analogous to those of berberine (1) reaction
with methylamine hydrate (8) (Scheme 6). Initially the comp-
lex 22 is formed, which is more stable by 9.9 kcal/mol than
the isolated reactants (Fig. 9). Then the position C-9 is
attacked by dimethylamine hydrate, leading to the
formation of thermodynamically stable complex 23'. This
particular step is rate limiting for the whole process, with
the energy barrier of TS 11 equal to 16.6 kcal/mol (Fig. 9,
Table S7 in the Supplementary information file).
This step is followed by 1,3-sigmatropic shift of
dimethylammonium ion from one oxygen atom to the other
(23'→23''). Such a stepwise migration is reflected,
similarly to the earlier example, by a plateau on MEP,
where the energy changes due to local minima and
conformational barriers do not exceed 2–3 kcal/mol. The
reaction site of the molecule does not significantly
rearrange during these steps (Fig. 9).
In the third step, an associate of methanol and
dimethylamine molecules leave the complex 23'', taking
only 3.0 kcal/mol of energy (transition state TS 12). The
obtained product 24 is significantly (by 25.5 kcal/mol)
more stable than the the starting system of infinitely
separated reactants.
It was shown experimentally and by quantum-chemical
calculations that direct amination at the C-9 position of
berberine is preferred in the case of primary amines, instead
of hydrolysis to berberrubine followed by its amination.
The amination reaction with dimethylamine in the presence
Since the observed hydrolysis in anhydrous media can
be explained only by the presence of crystal hydrate water
in berberine itself, we prepared completely anhydrous ber-
berine by heating berberine hydrate under vacuum in the
presence of P2O5.22 No hydrolysis was observed upon inter-
action of dehydrated berberine free of crystal hydrate water
with secondary amines in anhydrous alcohols, as the reac-
tion mixture contained only unchanged reagents. At the same
time, the formation of berberrubine (3) occurred after the addi-
tion of even catalytic amounts (0.2 equiv) of secondary amine.
The simplest model selected by us for the quantum-
chemical study of berberine amination with secondary
amines was the process involving berberine complex with
dimethylamine hydrate (Scheme 8).
The amination of berberine (1) with dimethylamine
proceeded through the thermodynamically favored
H
O
H
N Me
Scheme 8
Me
O
H
Me
Cl
O
Cl
O
O
H
N
O
Me
N
N
1 +
4f + NHMe2 + MeOH
NHMe2H2O
Me
OMe
OMe
H
O
H
OMe
21
20
1004

Chemistry of Heterocyclic Compounds 2015, 51(11/12), 997–1007
Scheme 9
of crystal hydrate water proceeds as a stepwise hydrolysis of
the O–Me bond at the C-9 atom, with the eventual formation
of berberrubine. Direct amination at the same atom does
not occur for energy reasons. In order to explain why the
reaction is more facile in the case of attack by methylamine
hydrate at the C-9 atom of berberine, we compared the
The results of this work showed that the methods for
synthesis of berberine 9-dialkylamino derivatives,
analogous to those developed for monoalkylamino
derivatives, are not feasible and the search for new
synthetic approaches is necessary. One of the options is
changing the anion and including the target molecule in
solvate cage or in a suitable complex.
Based on this study, a completely new approach can be
proposed for the preparation of berberine 9-hydroxy
derivatives, using much milder conditions than the
previously described reactions and giving higher yields,
thus it is particularly applicable to the hydrolysis of less
stable derivatives of berberine.
–
local nucleophilicity indices ωk of the nitrogen atoms.
–
According to calculations, the ωk (N) index in the
amination reaction increased from 0.18 to 0.32 eV when
going from methylamine to its monohydrate.
The same effect was observed also when comparing
–
isolated dimethylamine and its hydrate: ωk (N) for the reac-
tion of berberine amination increased from 0.14 to 0.26 eV.
However, the determining factor in choosing the
mechanism of nucleophilic attack at the C(9) atom – either
amination or hydrolysis, was the thermodynamics of the
process. In the case of amination, the reaction
thermodynamics were obviously unfavorable, and
hydrolysis occurred at the reaction site. Besides that, we
can conclude that the driving force of all aforementioned
reactions is the stabilization of product due to the formation
of hydrogen bonds.
Experimental
1Н and 13С NMR spectra were acquired on a Bruker
DPX-250 spectrometer (250 and 63 MHz, respectively) in
1
DMSO-d6, internal standard TMS. Н NMR signals were
assigned by using two-dimensional COSY and NOESY
spectroscopy data (mixing time 0.6–1.3 s). Several of
the 13С NMR signals overlapped, thus the observed number
of signals was smaller than expected theoretically. High-
Figure 9. MEP of nucleophilic attack by dimethylamine hydrate on berberine (1). The
relative energy values include corrections for ZPE. The relative Gibbs free energy values
are given in parentheses. The total energy of isolated reactants is assumed as reference.
1005

Chemistry of Heterocyclic Compounds 2015, 51(11/12), 997–1007
resolution mass spectra were recorded on a Bruker
3.95 (3H, s, 10-OCH3); 4.70–4.78 (4H, m, 6-CH2, NCH2);
6.16 (2H, s, OCH2O); 6.19 (1H, d, J = 2.8, H-3'); 6.29
(1H, dd, J = 2.8, J = 2.5, H-4'); 6.76 (1H, br. s, NH); 7.08
(1H, s, H-4); 7.53–7.58 (2H, m, H-12, H-5'); 7.77 (1H, s,
H-1); 7.94 (1H, d, J = 8.8, H-11); 8.74 (1H, s, H-13); 10.09
(1H, s, H-8). 13C NMR spectrum, δ, ppm: 27.0; 44.0; 55.5;
57.3; 102.5; 106.7; 107.7; 108.9; 110.8; 118.3; 120.3; 121.0;
125.3; 130.7; 133.5; 136.0; 136.4; 142.7; 146.5; 148.1; 150.0;
135.9. Found, m/z: 401.1486 [M–Cl]+. C24H21N2O4.
Calculated, m/z: 401.1496.
micrOTOF II instrument, with electrospray ionization. The
measurements were performed in positive ion mode (capil-
lary voltage 4500 V), the mass scanning range was 50–
3000 Da. Melting points were determined in glass
capillaries on a PTP apparatus. Merck Silicagel 60 (70–
230 μm) was used for column chromatography.
Commercially available berberine chloride hydrate, trypt-
amine, serotonin, piperazine, homopiperazine, morpholine,
piperonylpiperazine (Alfa Aesar), benzylamine (Lancaster
Synthesis), and furfurylamine (Acros Organics) were used
for the syntheses.
N-[2-(1H-Indol-3-yl)ethyl]berberine chloride (4c).
Yield 298 mg (60%), red needles, mp 279–280°С
1
Computational methodology. Gas phase quantum-
chemical calculations were performed by density functional
theory (DFT) using 6-31G** basis set and B3LYP func-
tional, including the Becke three-parameter exchange func-
tional23,24 and the Lee–Young–Parr correlation functional.25
The selection of a basis set that is widely applied to the
study of processes involving nucleophilic attack26–28 was
justified by the complexity of alkaloids, as well as the
reaction mechanisms. When needed, ab initio calculations
were also performed with the same basis set.
Complete geometry optimization for the structures
corresponding to stationary points on MEP was continued
until the gradient value reached 10–7 hartree/bohr, using
Gaussian 03 software suite29 running on a Silver cluster at
the Chemistry Department of the Southern Federal
University.
The nature of stationary points was established by
calculation of the normal vibrational frequencies (Hesse
matrices).30 The MEP of reactions were found by following
the gradient from transition states in the direct and reverse
directions of the reaction coordinate.31 The analysis of
electron density distribution function according to the AIM
model was performed according to a published method.17
Synthesis of berberrubine chloride (3). Secondary amine
(3.0 mmol) was added to a solution of berberine chloride (1)
(372 mg, 1.0 mmol) in ethanol (25 ml). The mixture was
refluxed for 4–5 h, then ethanol was removed by distil-
lation. The residue was separated by chromatography on
silica gel (eluent 10:1 CHCl3–EtOH), collecting the
fraction with Rf 0.40, which was then recrystallized from
THF or acetone. Yield 65–82%, red needles, decomp. 279–
282°С (decomp. 270–285°С19). The obtained berberrubine
chloride was identical to a sample reported in the
literature.19
Synthesis of 9-N-substituted berberine derivatives
4a–d (General method). The appropriate amine (5.0 mmol)
was added to berberine chloride (372 mg, 1.0 mmol), the
mixture was heated for 4–5 h at 120°С, the reaction
mixture was washed with acetone (3×10 ml) to remove the
remaining amine. The dry residue was separated by
chromatography on silica gel (eluent 10:1 CHCl3–EtOH), the
obtained compounds were recrystallized from acetone or THF.
N-Benzylberberine chloride (4a) matched a sample
reported in the literature.8 Yield 289 mg (65%).
(acetone). H NMR spectrum, δ, ppm (J, Hz): 3.05 (2H, t,
J = 5.0, NCH2CH2); 3.18–3.22 (2H, m, 5-CH2); 3.89–3.94
(2H, m, NCH2CH2); 3.96 (3H, s, OCH3); 4.70–4.76 (2H, m,
6-CH2); 6.18 (2H, s, OCH2O); 6.48 (1H, m, 9-NH); 6.96
(1H, t, J = 7.4, H-5'); 7.03–7.06 (1H, m, H-6'); 7.09 (1H, s,
H-2'); 7.20 (1H, s, H-4); 7.35 (1H, d, J = 7.9, H-7'); 7.47–
7.54 (2H, m, H-12, H-4'); 7.77 (1H, s, H-1); 7.88 (1H, d,
J = 8.7, H-11); 8.68 (1H, s, H-13); 9.95 (1H, s, H-8); 10.92
(1H, s, 1'-NH). 13C NMR spectrum, δ, ppm: 24.5; 27.8; 53.1;
56.3; 102.1; 102.6; 105.2; 108.8; 110.5; 112.0; 117.9;
118.6; 118.8; 120.3; 121.5; 122.2; 123.7; 127.4; 129.8;
132.5; 133.9; 136.8; 146.6; 147.8; 148.9; 149.9; 166.3.
Found, m/z: 464.1966 [M–Cl]+. C29H26N3O3 Calculated, m/z:
464.1969.
N-[2-(5-Hydroxy-1H-indol-3-yl)ethyl]berberine chlo-
ride (4d). Yield 288 mg (56%), red needles, decomp. 256–
1
258°С (THF). H NMR spectrum, δ, ppm (J, Hz): 2.94–
3.05 (6H, m, N(CH2)2, 5-CH2); 3.75 (3H, s, 10-OCH3);
4.50–4.54 (2H, m, 6-CH2); 6.09 (2H, s, OCH2O); 6.52 (1H,
d, J = 7.9, H-12); 6.60–6.64 (2H, m, H-6', 9-NH); 6.83 (1H,
s, H-2'); 6.97 (1H, s, H-4); 7.11 (1H, s, H-4'); 7.14 (1H, d,
J = 8.8, H-7'); 7.32 (1H, d, J = 7.9, H-11); 7.63 (1H, s,
H-1); 8.09 (1H, s, H-13); 8.76 (1H, br. s, OH); 9.16 (1H, s,
H-8); 10.68 (1H, s, 1'-NH). 13C NMR spectrum, δ, ppm:
23.7; 27.8; 39.7; 53.3; 102.1; 102.5; 105.2; 108.8; 109.0;
112.0; 112.3; 118.0; 120.1; 121.8; 122.2; 124.1; 128.0;
129.9; 131.3; 132.4; 134.0; 146.5; 147.8; 148.9; 149.7;
+
150.8 Found, m/z: [M–Cl]+ C29H26N3O4 . Calculated, m/z:
480.1918.
X-Ray structural study of compound 4a. Crystals
suitable for X-ray structural analysis were obtained by slow
evaporation of a nitrobenzene solution of compound 4а.
The composition of the colorless crystals was
C26H23ClN2O3·0.5C6H5NO2·1.5H2O (M 535.49). The X-ray
structural study was performed on an automated Bruker
APEX II diffractometer (graphite monochromator, λ(MoKα)
0.71073 Å, ω-scanning) at 120 K. Empirical accounting for
absorption and the correction of systematic errors were
performed with the SADABS software. The structure was
solved by a direct method and refined by full matrix
method of least squares by F2 with anisotropic thermal
hkl
parameters for all non-hydrogen atoms. The positions of
hydrogen atoms in amino groups and water molecules were
revealed by differential Fourier synthesis, the hydrogen
positions at the carbon atoms were calculated. The solving
and refinement of structure were performed with the
SHELX 2009-9.13 software suite.32 The complete
N-(Furan-2-ylmethyl)berberine chloride
(4b).
Yield 240 mg (58%), red needles, mp 270–272°С (THF). 1H
NMR spectrum, δ, ppm (J, Hz): 3.15–3.19 (2H, m, 5-CH2);
1006

Chemistry of Heterocyclic Compounds 2015, 51(11/12), 997–1007
[Russian translation]; Mir, Moscow, 2001.
crystallographic dataset was deposited at the Cambridge
Crystallographic Data Center33 (deposit CСDC 1421105).
18. Kariuki, B. M.; Jones, W. Acta Cryst. Sect. C 1995, 51, 1234.
19. Iwasa, K.; Kamigauchi, M.; Uek, M.; Taniguch, M. Eur. J.
Med. Chem. 1996, 31, 469.
The Supplementary information file containing the
geometrical characteristics of structures 5–24 and the
transition states, as well as the full and relative energy (in
tabular form) for all stationary points on the reaction MEP
is available online at http://link.springer.com/journal/10593.
20. Gorb, L.; Podolyan, E.; Dziekonski, P.; Sokalski, W. A.;
Leszczynski, J. J. Am. Chem. Soc. 2004, 126, 10119.
21. Jovené, C.; Jacquet, M.; Marrot, J.; Bourdreux, F.; Kletsky, M. E.;
Burov, O. N.; Gonçalves, A.-M.; Goumont, F. Eur. J. Org.
Chem. 2014, 6451.
22. Elliott, I. W. J. Heterocycl. Chem. 1967, 4, 639.
This work was performed within the framework of
Project part of the State Assignment No.4.129.2014/K from
the Ministry of Education and Science of the Russian
Federation.
23. Becke, A. D. Phys. Rev. A. 1988, 38, 3098.
24. Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
25. Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B. 1988, 37, 785.
26. Burov, O. N.; Kletskii, M. E.; Gulevskaya, A. V. Russ. Chem.
Bull. 2013, 62, 1156. [Izv. Akad. Nauk, Ser. Khim. 2013, 1155.]
27. Kletskii, M. Е.; Burov, О. N.; Dalinger, I. L.; Shevelev, S. А.
Comp. Theor. Chem. 2014, 1033, 31.
References
1. Saha, S. K.; Khuda-Bukhsh, A. R. Eur. J. Pharmacol. 2013,
714, 239.
28. Suzdalev, K. F.; Den'kina, S. V.; Starikova, A. A.;
Dvurechensky, V. V.; Kletsky, M. E.; Burov, O. N.
Mendeleev Commun. 2011, 21, 231.
2. Siow, Y. L; Sarna, L; Karmin, O. Food Res. Int. 2011, 44, 2409.
3. Khazir, J.; Mir, B. A.; Pilcher, D. L.; Riley, D. L. Phytochem.
Lett. 2004, 7, 173.
29. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A.;
Vreven, Jr. T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.;
Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.;
Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.;
Klene, M.; Li, X.; Knox, J. E., Hratchian, H. P.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.;
Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.;
Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.;
Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Martin, R. L.; Fox, J. D.; Keith, T.;
Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.;
Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03;
Gaussian, Inc.: Wallingford, 2004.
4. Crycova, L.; Dostal, J.; Marek, R. Phytochem. 2007, 68, 150.
5. Ji, H.-F.; Shen, L. Molecules 2011, 16, 6732.
6. Krivogorsky, B.; Pernat, J. A.; Douglas, K. A.; Czerniecki, N. J.;
Grundt, P. Bioorg. Med. Chem. Lett. 2012, 22, 2980.
7. Nechepurenko, I. V.; Salakhutdinov, N. F.; Tolstikov, G. A.
Khimiya v Interesakh Ustoichovogo Razvitiya 2010, 18, 1.
8. Shan, W.-J.; Huang, L.; Zhou, Q.; Meng, F.-Ch.; Li, X.-Sh.
Eur. J. Med. Chem. 2011, 46, 5885.
9. Huang, L.; Shi, A.; He, F.; Li, X. Bioorgan. Med. Chem.
2010, 18, 1244.
10. Ma, Y.; Ou, T.-M.; Hou, J.-Q.; Lu, Y.-J.; Tan, J.-H.; Gu, L.-Q.;
Huang, Zh.-Sh. Bioorg. Med. Chem. 2008, 16, 7582.
11. Ma, Y.; Ou, T.-M.; Tan, J.-H.; Hou, J.-Q.; Huang, Sh.-L.;
Gu, L.-Q.; Huang, Zh.-Sh. Eur. J. Med. Chem. 2011, 46, 1906.
12. Huang, L.; Luo, Z.; He, F.; Shi, A. Bioorg. Med Chem. Let.
2010, 20, 6649.
13. Ma, Y.; Ou, T.-M.; Tan, J.-H.; Hou, J.-Q.; Huang, Sh.-L.
Bioorg. Med. Chem. Let. 2009, 19, 3414.
14. Bodiwala, H. S.; Sabde, S.; Mitra, D.; Bhutani, K. K.; Singh, I. P.
Eur. J. Med. Chem. 2011, 46, 1045.
30. Schlegel, H. B. Theor. Chim. Acta 1984, 66, 333.
31. Hirsh, M.; Quapp, W. Chem. Phys. Lett. 2004, 395, 150.
32. Sheldrick, G. M. Acta Crystallogr., Sect. A: Found.
Crystallogr. 2008, A64, 112.
15. Burov, O. N.; Kurbatov, S. V.; Morozov, P. G.; Kletskii, M. E.;
Tatarov, A. V. Chem. Heterocycl. Compd. 2015, 51, 772.
[Khim. Geterotsikl. Soedin. 2015, 51, 772.]
16. Lam, Y. Int. J. Pharma Sci. Res. 2014, 5, 350.
17. Bader, R. F. W. Atoms in Molecules. Quantum Theory
33. Allen, F. H. Acta Crystallogr., Sect. B: Struct. Sci. 2002, B58, 380.
1007