Zwitterionic Molecularly Imprinted Cross-Linked Micelles for Alkaloid Recognition in Water

Article abstract of DOI:10.1021/acs.joc.9b01629

Molecular imprinting within surface/core doubly cross-linked micelles afforded water-soluble nanoparticle receptors for their template molecules. Extremely strong imprinting effects were consistently observed, with the imprinting factor >100:1 in comparison to nonimprinted nanoparticles prepared without the templates. The ionic nature of the cross-linkable surfactant strongly impacted the imprinting and binding process. Imprinted receptors prepared with a zwitterionic cross-linkable surfactant (4) outperformed a similar cationic one (1) when the template was zwitterionic or cationic and preferred their templates over structural analogues regardless of their ionic characteristics. Electrostatic interactions, however, dominated the receptors made with the cationic surfactant. The same micellar imprinting applied to simple as well as complex alkaloids. Imprinted receptors from 4 were also shown to categorize their alkaloid guests according to their structural similarity.

Full text of DOI:10.1021/acs.joc.9b01629

Article
pubs.acs.org/joc
Cite This: J. Org. Chem. XXXX, XXX, XXX−XXX
Zwitterionic Molecularly Imprinted Cross-Linked Micelles for
Alkaloid Recognition in Water
Likun Duan and Yan Zhao*
Department of Chemistry, Iowa State University, Ames, Iowa 50011-3111, United States
S
* Supporting Information
ABSTRACT: Molecular imprinting within surface/core dou-
bly cross-linked micelles afforded water-soluble nanoparticle
receptors for their template molecules. Extremely strong
imprinting effects were consistently observed, with the
imprinting factor >100:1 in comparison to nonimprinted
nanoparticles prepared without the templates. The ionic
nature of the cross-linkable surfactant strongly impacted the
imprinting and binding process. Imprinted receptors prepared
with a zwitterionic cross-linkable surfactant (4) outperformed
a similar cationic one (1) when the template was zwitterionic
or cationic and preferred their templates over structural analogues regardless of their ionic characteristics. Electrostatic
interactions, however, dominated the receptors made with the cationic surfactant. The same micellar imprinting applied to
simple as well as complex alkaloids. Imprinted receptors from 4 were also shown to categorize their alkaloid guests according to
their structural similarity.
to different-sized molecules1,2 and even large entities including
INTRODUCTION
■
viruses and bacteria.3−8 Its simplicity continues to attract many
Molecular recognition is the basis of nearly every biological
researchers’ interest for their applications in a wide range of
process including receptor−ligand binding, enzymatic catalysis,
and transport of molecules across membranes. For a
supramolecular host to bind its guest strongly and selectively,
the two must have a complementary interface where multiple
areas.9−20
Instead of using a macroporous polymeric network, our
group found that micelles of doubly cross-linkable surfactant 1
could serve as an excellent platform for molecular imprinting
noncovalent interactions engage. When the guest is small and
(Scheme 1).21 The process involves a two-stage cross-linking,
symmetrical, a complementary host can be constructed
relatively easily using an organic framework such as a
first on the surface of the micelle using diazide 2 via the click
reaction and then in the micellar core using divinylbenzene via
macrocycle. When the guest becomes more complex in
free-radical polymerization. The cross-linked micelles are
shape and functionality, as with most bioactive molecules,
design and synthesis of a complementary host become much
more challenging. The difficulty increases further with
typically functionalized on the surface using monoazide 3 by
another round of click reaction for increased hydrophilicity and
facile purification. Micellar imprinting can be done directly in
additional requirements such as preorganization of the host’s
water thanks to the ability of micelles to strengthen hydrogen
conformation for guest-binding, solubility in water, and ability
bonds in their nonpolar core.22,23 The number of binding sites
to function under competitive aqueous environments. High
per cross-linked micelle is controlled by the surfactant/
selectivity in binding also becomes a must for biomolecular
template ratio.21 The so-called molecularly imprinted nano-
recognition since closely related biomolecules tend to have
particles (MINPs) are ∼5 nm in diameter and resemble
drastically different biological properties.
proteins in their nanosize, hydrophilic exterior, and hydro-
All of these demands make traditional stepwise total
phobic core. They mimic protein receptors in their binding of
synthesis of supramolecular hosts impractical for biomolecular
common biological guests such as peptides24 and carbohy-
recognition. In this regard, molecular imprinting is an attractive
drates25 and can be converted into highly selective catalysts
alternative for building receptors for complex molecules.1,2 The
method involves first formation of a template−functional
monomer (FM) complex, either spontaneously through
noncovalent interactions or chemically through reversible
covalent bonds. The complex is then polymerized with a
large amount of a cross-linker. Removal of the templates from
the resulting macroporous polymer network affords template-
complementary binding sites in size, shape, and distribution of
functional groups. Molecular imprinting can be applied readily
similar to enzymes.26
The cross-linkable surfactant in micellar imprinting is the
most important component for the synthesis. Not only does it
set the boundary for imprinting by its self-assembled micelle, it
also acts as the functional monomer and cross-linker to interact
Received: June 18, 2019
Published: September 23, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.9b01629
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 1. Preparation of MINP by Surface−Core Double
Cross-Linking of Template-Containing Micelle of 1
Scheme 2. Syntheses of Cross-Linkable Surfactants 4
Dodacanediol 5 is monomethacrylated to afford 6. This step is
the same as that in the synthesis of 1.21 The monoalcohol is
then phosphorylated with ethylene glycol chlorophosphate 7,
and the resulting cyclic phosphate is ring-opened by
commercially available tripropargylamine to afford 4, following
reported literature procedures.36 The surfactant is similar to 1
in many aspectsa C12 carbon chain, a tripropargylammo-
nium group for surface cross-linking, and a methacrylate at the
end of the hydrocarbon tail for core-cross-linking. The
1
compound was fully characterized by H NMR spectroscopy,
13C NMR spectroscopy, and high-resolution mass spectrom-
etry (MS).
with the template through hydrophobic,21 electrostatic,21 and
hydrogen bonds.27 Cationic surfactant 1 was used in the
majority of our micellar imprinting. Although it has allowed a
wide range of biologically interesting molecules to be
imprinted, its cationic nature may not always be desirable for
biological applications. Because biological entities (proteins,
nucleic acids, and lipids) vary in charge, we need to be able to
tune the charge of our materials for specific applications.
In this work, we designed and synthesized a new cross-
linkable surfactant. Its zwitterionic headgroup was inspired by
natural phosphatidylcholine and was found to strongly impact
the imprinting and binding of charged template molecules. In
comparison to our original 1, the new surfactant allowed
MINPs to differentiate closely related natural products with
varying ionic charges. It also enabled facile preparation of
synthetic receptors for alkaloids, a highly important class of
biomolecules with broad pharmaceutical applications.28,29
An extremely important property of a surfactant is its critical
micelle concentration (CMC), above which micelles start to
form spontaneously in water. There are many ways to
determine the CMC of a surfactant.37 A method of choice
utilizes an environmentally sensitive hydrophobic probe such
as pyrene.38 The method requires very little amounts of
materials and, in our hands, afforded similar CMC values to
those determined from the reduction of surface tension for
analogous surfactants.39 Typically, a series of surfactant
solutions of varying concentrations are prepared using an
aqueous solution containing 0.1 μM pyrene. The emission of
pyrene has five vibronic bands, with the first band (I1) near
372 nm becoming stronger in a more polar environment and
the third (I3) near 384 nm staying nearly constant. The I3/I1
ratio thus increases with decreasing environmental polarity.
Figure 1 plots the I3/I1 ratio of pyrene against the
concentration of 4 in Millipore water. The ratio stayed
unchanged at low surfactant concentration, around 0.80.
Apparently, the hydrophobic probe stayed in the aqueous
phase under such conditions, even though the solution
RESULTS AND DISCUSSION
■
Synthesis and Characterization of Zwitterionic Cross-
Linkable Surfactant. Phosphatidylcholine is a major
component of biological membranes.30,31 Its zwitterionic
headgroup is connected via a glycerol to two fatty acid chains
(Scheme 2). The headgroup is strongly solvated by water and
provides an overall neutral background for a membrane to
interact with its components, including proteins and other
lipids. It is known to cause low structural perturbation to
associated water molecules. Related to this property,
phosphatidylcholine-containing amphiphilic polymers have
strong abilities to resist biofouling, due to their low tendencies
to trigger conformational changes of adsorbed proteins.32−35
Synthesis of our phosphatidylcholine-like cross-linkable
surfactant 4, 12-(methacryloyloxy)dodecyl (2-(tri(prop-2-yn-
1-yl)ammonio)ethyl)phosphate, is shown in Scheme 2. 1,12-
Figure 1. Pyrene I3/I1 ratio as a function of [4]. [Pyrene] = 0.1 μM.
B
DOI: 10.1021/acs.joc.9b01629
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
contained some surfactant molecules. As the concentration of
4 increased beyond a certain point, the I3/I1 ratio began to rise
sharply, with almost a linear dependence on the surfactant’s
concentration beyond the inflection point. The result was
consistent with the formation of an increasing number of
micelles at higher concentrations above the CMC, leading to a
higher partition of the hydrophobic probe into the nonpolar
region of the micelle. The inflection point of the curve is
normally considered the CMC of the surfactant and was 0.20
mM for 4. This number was lower than the 0.55 mM for the
cationic surfactant 1.21 The result was reasonable, given the
repulsion among the cationic headgroups in the micelle of 1
and the overall neutral charge of 4.
Binding Properties of Zwitterionic and Cationic
MINPs. To understand how different headgroups impact
micellar imprinting, we chose to study three model templates
8−10, all of biological origin. 3-Indolepropionic acid 8 is a
natural product with potential therapeutic benefits for
Alzheimer’s diease.40 Tryptophan 9 is a natural aromatic
amino acid. Tryptamine 10 is a monoamine alkaloid able to
bind human trace amine-associated receptors in the central
nervous system.41 The three compounds share the same
hydrophobe (i.e., indole) but have different ionic character-
istics. These features make them particularly suitable for
probing relative contributions of hydrophobic and electrostatic
interactions in the imprinting and binding of MINPs.
Figure 2. (a) Emission spectra of 8 (λex = 300 nm) upon the addition
of different concentrations of MINP4(8) in 50 mM Tris buffer (pH =
7.4). [8] = 2.0 μM. The concentration of MINP was calculated based
on a molecular weight (MW) of 50 000 g/mol determined by DLS.
(b) Nonlinear least-squares fitting of the emission intensity of 8 at
363 nm to a 1:1 binding isotherm.
binding constant of 0.06 × 105 M−1 (Figure S19). The
imprinting factor (IF), thus, was >100:1, indicating an
extremely strong imprinting effect.
Figure 3a,b shows the emission spectra of 8 titrated by
MINP1(8) and the corresponding data fit to the 1:1 binding
Preparation of the MINPs is reported in Experimental
Section. The surface- and core-cross-linking of the micelle was
monitored by 1H NMR spectroscopy. The surface-cross-
linking chemistry had been confirmed independently by mass
spectrometry after the surface-cross-linkages were cleaved by
periodate.42 Dynamic light scattering (DLS) allowed us to
follow the change of size for the micelles during the surface-
cross-linking, surface functionalization, and core-cross-linking.
The DLS size had been confirmed by transmission electron
microscopy.43,44 Since a key difference between surfactants 1
and 4 was their ionic nature, we also determined the ζ
potential of the corresponding MINPs. MINP1, i.e., MINP
prepared with 1 as the cross-linkable surfactant, afforded a ζ
potential of 60.5 mV and MINP with 4 as the surfactant −14.4
mV (Figures S4 and S8).
One benefit of using indole-derived compounds is that their
binding could be studied conveniently by fluorescence
titration. Figure 2a shows the emission spectra of 2.0 μM 8
in 50 mM Tris buffer (pH = 7.4) upon the addition of 0−7.0
μM MINP4(8), i.e., MINP prepared with 4 as the cross-
linkable surfactant and 8 as the template. The higher
concentration of the MINP caused a decrease of the main
emission band at 363 nm. Two isoemissive points were
observed at ca. 336 and 426 nm, suggesting a continuous
transition from the free to the bound 8 during titration. A
blank titration experiment showed that the MINP contributed
negligibly to the observed fluorescence (Figure S18). The
emission intensity at 363 nm fit nearly perfectly to a 1:1
binding isotherm (Figure 2b). Nonlinear least-squares fitting
afforded a binding constant (Ka) of (6.56 0.90) × 105 M−1.
When the guest was titrated with nonimprinted nanoparticles
(i.e., NINP4), the binding was much weaker, with an estimated
Figure 3. (a) Emission spectra of 8 (λex = 300 nm) upon the addition
of different concentrations of MINP1(8) in 50 mM Tris buffer (pH =
7.4). [8] = 2.0 μM. The concentration of MINP was calculated based
on an MW of 50 000 g/mol determined by DLS. (b) Nonlinear least-
squares fitting of the emission intensity of 8 at 363 nm to a 1:1
binding isotherm.
model, which yielded a Ka value of (13.2 0.3) × 105 M−1.
The guest continued to show weak binding toward non-
imprinted nanoparticles (i.e., NINP1), with Ka ≈ 0.09 × 105
M−1 (Figure S20). The imprinting factor hence was also
>100:1, indicating a strong imprinting effect with the cationic
micelle as well.
In addition to different binding constants, the emission
spectra of the guest differed significantly in shape in the
titrations with the two MINPs (Figures 2a and 3a). Whereas
the cationic MINP1(8) caused a quenching effect of the broad
emission band of the indole ring, multiple vibronic bands
appeared for 8 in the presence of the zwitterionic MINP4(8).
Indole derivatives have two overlapping electronic transitions,
45−47
1
1
1La and Lb.
The La band is broad and sensitive to
environmental polarity, including hydrogen-bonding interac-
tions to the indole nitrogen and charged groups nearby.48 The
1Lb band, on the other hand, tends to have vibronic structures
and, meanwhile, is less sensitive to the environment.49 The
1
relative contribution of the Lb band is known to increase in a
hydrophobic microenvironment.50 Although the complexity of
indole fluorescence and its sensitivity to multiple factors make
it difficult to pinpoint the exact cause of different spectral
shapes in Figures 2a and 3a, the consistency in the difference
(Figures S21−S36) suggests a common origin, most likely the
C
DOI: 10.1021/acs.joc.9b01629
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
stronger hydrophobicity of the zwitterionic micelle (vide
infra).
was obtained from the fluorescence titration. Since it was much
quicker to perform fluorescence titration, we used this method
whenever the template or guest molecule was fluorescent.
Table 1 summarizes the binding constants obtained for
guests 8−10. Since surfactant 1 was good at imprinting anionic
guests, we studied the binding of 8 first. In our hands,
MINP1(8) bound the anionic template more strongly than
MINP4(8) prepared with the same template (Table 1, entry 1).
Although the result was fully expected from the favorable
electrostatic interactions between the cationic micelle and
anionic template, the 2-fold increase in binding represented a
very modest effect. Our previous work showed that electro-
static interactions played important roles in the MINP binding
of anionic guests when 1 was the cross-linkable surfactant.21 It
is known that zwitterionic surfactants tend to pack more tightly
than ionic ones during micellization due to reduced hydration
of their headgroups.51 This suggests that the micelle of 4 is
more hydrophobic in the nonpolar interior than that of 1. The
effect would strengthen the hydrophobic interactions between
MINP4(8) and its template and most likely partly offset the
loss of favorable electrostatic interactions between MINP1(8)
and 8.
Given the complexity of the indole fluorescence, we also
confirmed the bindings by isothermal titration calorimetry
(ITC). This method measures the heat change during titration
and is able to afford a number of parameters, including the
binding enthalpy (ΔH) and number of binding sites per
nanoparticle (N), in addition to the binding constant. After
background subtraction to account for the dilution effect, ITC
showed that the number of binding sites per nanoparticle was
1.14
0.13 for 8 by MINP4(8), consistent with the 1:1
binding suggested by the fluorescence titration (Figure 4).
When zwitterionic 9 was used as the template, MINP4(9)
was able to bind the template with a Ka of 7.75 × 105 M−1 in
Tris buffer (Table 1, entry 5). In contrast, MINP1(9) showed a
much weaker binding for the same template, with a Ka of 0.89
× 105 M−1. The binding was nearly 9 times weaker,
corresponding to a 1.28 kcal/mol difference in binding free
energy (ΔΔG). One possible reason for the stronger ability of
4 to imprint 9 could be the zwitterionic nature of both
compounds. Essentially, both the positively and negatively
charged groups of the template could find suitable groups on
the micelle to ion-pair during imprinting in the case of 4,
whereas only the negatively charged carboxylate of the
template could enjoy favorable interactions with the surfactant
if 1 is used to prepare MINP. Another possible reason is the
“tighter and more hydrophobic micelle” of 4 mentioned above,
which could help the imprinting and binding of any
hydrophobic molecules.
Figure 4. ITC curve obtained at 298.15 K from titration of MINP4(8)
with 8 in 50 mM Tris buffer (pH = 7.4). MINP4(8) = 10 μM in the
cell. The concentration of 8 in the syringe was 0.20 mM.
Because typical MINPs contain ∼50 cross-linked surfactants
per particle, as estimated by DLS, the 1:1 binding was expected
when a 50:1 surfactant/template ratio was used in the
preparation.21 The Ka value obtained was (6.78
0.31) ×
105 M−1, very similar to that from the fluorescence titration
[(6.56
0.90) × 105 M−1]. For the binding of 8 by
MINP1(8), ITC afforded a binding constant of (13.9 0.5) ×
105 M−1 (Figure S41), also in excellent agreement with what
a
Table 1. Binding Constants and Selectivity for MINPs Imprinted against 8−10
b
Ka (×105 M−1
)
CRR
entry
1
template
guest
MINP4
MINP1
MINP4
1
MINP1
1
8
8
6.56 0.90
(6.78 0.31)
3.84 0.51
4.08 0.47
2.95 0.43
7.75 0.87
3.38 0.45
3.80 0.43
3.77 0.65
7.52 0.72
13.2 0.3
(13.9 0.5)
0.91 0.31
0.43 0.10
11.9 3.7
0.89 0.10
0.42 0.02
9.84 2.25
0.67 0.13
0.62 0.14
2
3
4
5
6
7
8
9
8
8
9
9
9
10
8
9
10
8
9
10
8
9
10
0.59
0.62
0.38
1
0.44
0.51
0.50
1
0.07
0.03
13.4
1
0.47
15.9
1.08
1
9
10
10
10
none
none
none
c
c
10
11
12
0.06
0.09
0.04
0.01
c
c
0.05
c
c
0.05
a
The binding constants were obtained from fluorescence titrations performed in duplicate in 50 mM Tris buffer (pH = 7.4). The binding constants
b
in parentheses were obtained from ITC. CRR is the cross-reactivity ratio, defined as the binding constant of a guest relative to that of the template
for a particular MINP. Binding was between the guest and nonimprinted nanoparticles (NINP4 or NINP1) prepared without any templates. The
c
binding was very weak, and the binding constants were estimated from fluorescence titrations (SI).
D
DOI: 10.1021/acs.joc.9b01629
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
The benefit of the zwitterionic over cationic surfactant was
also evident in the imprinting of the cationic template (10).
Entry 9 of Table 1 shows that the binding constant for the
template was 7.52 and 0.62 × 105 M−1 by MINP4(10) and
MINP1(10), respectively. Since these numbers do not differ
much from those for the zwitterionic template 9 (entry 5), the
double ion-pairing reason mentioned above must be minor in
micellar imprinting. Otherwise, eliminating the factor should
weaken the binding of 10 by MINP4(10) relative to that of 9
by MINP4(9) significantly. On the other hand, the “tighter and
more hydrophobic micelle” of 4 would still help the imprinting
and binding of any hydrophobic template, including the
cationic 10.
MINP made with the zwitterionic surfactant always bound its
template more strongly than structural analogues (entries 4−
9).
In contrast, when cationic surfactant 1 was used to prepare
MINP, very different behavior was observed. Instead of
favoring the template itself, MINP1(9) showed a strong
preference for guest 8 (Table 1, entries 1−3, under MINP1).
The same happened with MINP1(10). All of these are
considered imprinting failure, when the imprinted material
shows weaker binding for its template than structural
analogues. In fact, when the cationic surfactant was used to
prepare MINPs, regardless of the template, the binding of the
final MINP always followed the order of 8 > 9 > 10. Clearly,
when the hydrophobic group was kept constant, the binding
affinity was mainly controlled by electrostatic interactions for
the cationic MINPs.
The cationic MINP was not completely without merit in
selectivity, however. When the anionic guest (8) was the
template, the selectivity for the template was much higher than
what was obtained by the zwitterionic MINP. The CRR for 9
and 10, for example, was 0.07 and 0.03 toward MINP1(8).
These numbers were much lower than the corresponding
values toward MINP4(8), i.e., 0.59 and 0.62, respectively
(Table 1, entries 2 and 3).
As for the binding between the guests and the nonimprinted
nanoparticles, the Ka values were very similar for NINP4
(0.05−0.06 × 105 M−1) (Table 1, entries 10−12). Thus,
>100:1 IF was consistently obtained for all three guests. The
results also support the minor importance of electrostatic
interactions for the zwitterionic micelles. For the cationic
NINP1, although the bindings were also weak (0.01−0.09 ×
105 M−1), the binding affinity did follow the expected order 8
> 9 > 10, indicating that the electrostatic interactions also
dominated nonspecific binding of the cationic cross-linked
micelles.
Binding of Alkaloids by Zwitterionic MINPs. The
zwitterionic cross-linkable surfactant, thus, gave stronger
binding affinities and higher selectivities when the template
was zwitterionic or cationic. These results encouraged us to use
4 to imprint alkaloids, an extremely important class of
biomolecules containing basic nitrogens and frequently
administered in the cationic salt form.28,29 As a proof of
concept, we first studied three structurally similar alkaloids
11−13. Among the three alkaloids, 1-phenylcyclohexanecar-
boxylic acid derivative 11 (PRE-084) is a sigma receptor
agonist.54 Proadifen hydrochloride 12 has a very similar
structure but completely different biological properties,
showing inhibiting effects on cytochrome P450 enzymes55
and suppressing proliferation of certain cancers.56 Compound
13 is marketed under many different names (Ansmin, Cefadol,
Celmidol, Difenidol, Maniol, Vontrol, etc.). With a slight
variation of structure, this alkaloid is an antiemetic and
antivertigo agent.
Another interesting observation was that the binding
constants for 9 by MINP1(9) and 10 by MINP1(10) were
very similar, 0.89 and 0.62 × 105 M−1, respectively (entries 5
and 9). This might appear quite strange, as the cationic MINP
should repel the cationic template (10) but at least be able to
“tolerate” the zwitterionic one (9). The assumption for this
prediction, however, was that 10 stayed protonated during
binding. The pKa of a primary ammonium group is ∼10. At pH
7.4, under our titration conditions, it should be largely
protonated indeed, at least in solution. Nonetheless, a
protonated amine can shift its pKa dramatically in different
environments. When Isom et al. prepared 25 variants of
staphylococcal nuclease with lysine residues at varying
positions, for example, 19 of the (protonated) lysine residues
have depressed pKa values, some as low as 5.3, due to a strong
environmental effect on the acid−base equilibrium.52 Essen-
tially, when located in a hydrophobic microenvironment
instead of an aqueous solution, the pKa of an ammonium
group is much lower because the protonated, positively
charged ammonium is poorly solvated and unstable, relative
to the neutral amine. The same could happen when the
ammonium has other positively charged ions in the vicinity.
Both situations do exist in the imprinting and binding of 10 by
our cationic micelle. By taking the neutral amino form,
template 10 can migrate deeper into the micelle of 1 and avoid
electrostatic repulsion at the same time. Tryptophan 9,
however, had to stay near the surface of the micelle because
of its two ionic groups.53 Its advantage, of course, is the lack of
electrostatic repulsion with the cationic micelle. In these ways,
each template could optimize its interactions with the cationic
micelle within its own capability and constraints; the end result
was that the two showed similar binding toward the cationic
MINP.
In Table 1, we also included two columns of cross-reactivity
ratio (CRR), defined as the binding constant of a guest relative
to that of the template for a particular MINP. The term is
frequently used in immunology and measures the selectivity of
our “synthetic antibodies” for closely related structural
analogues. Based on the definition, CRR is 1 for the template
and, the smaller the ratio for a particular guest, the weaker is
the binding and the higher the selectivity of that MINP is for
its template molecule.
Our data show that the zwitterionic MINP4 was able to
distinguish its template from closely related structural
analogues, regardless of the ionic characteristics of the
template. For example, when the anionic template 8 was
used as the template, zwitterionic 9 and cationic 10 gave a
CRR of 0.59 and 0.62, respectively, for MINP4(8) (Table 1,
entries 2 and 3). Although the exact values differed somewhat
for MINP4(9) and MINP4(10), the same trend persisted, i.e.,
Among the three alkaloids, 11 and 13 are quite similar and
12 is more different. For example, although there is a difference
of ester versus hydroxyl in the center of the structure for 11
and 13, the two are similar in size, shape, and amphiphilicity.
Compound 12, on the other hand, has an extra butyl near the
E
DOI: 10.1021/acs.joc.9b01629
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
a
Table 2. Binding Data for MINPs Obtained by ITC
entry
template
guest
N
Ka (105 M−1
)
CRR
−ΔG (kcal/mol)
−ΔH (kcal/mol)
TΔS (kcal/mol)
1
2
3
4
5
6
7
8
9
11
11
11
11
11
11
12
12
12
14
11
12
13
8
0.8 0.2
0.8 0.1
0.7 0.1
3.15 0.33
0.516 0.048
1.97 0.14
0.11 0.03
0.12 0.02
0.12 0.03
0.360 0.028
4.60 0.67
0.222 0.034
7.81 0.31
1
7.5
6.4
7.2
5.5
5.6
5.6
6.2
7.7
5.9
26.6 2.4
9.8 1.3
31.5 3.1
−19.1
−3.4
0.16
0.63
0.03
0.04
0.04
0.08
1
−24.3
b
b
b
−
−
−
−
−
b
b
b
9
−
−
b
b
b
10
11
12
13
14
−
−
1.3 0.2
0.7 0.1
1.1 0.1
0.8 0.1
2.5 0.3
11.1 3.1
4.3 0.5
3.3 0.6
3.7
−3.4
1.6
0.05
10
8.0
4.7
a
b
Binding was measured in 50 mM Tris buffer (pH = 7.4) by ITC unless indicated otherwise. Binding was measured in 50 mM Tris buffer (pH =
7.4) by fluorescence titration. The titrations were performed in duplicate, and the errors between the runs were <20%.
two phenyls and also has a diethylamino instead of six-
membered ring on the other end of the structure. To see
whether our imprinted micelles could differentiate the two
groups of compounds, we prepared MINPs with 4 as the cross-
linkable surfactant for 11 and 12 and studied their binding for
all three compounds. Because the compounds were not
fluorescent, we measured the binding by ITC.
of these applications, design and synthesis of molecular
receptors become too difficult as the guest gets more
complicated in structure and larger in size. This work shows
that micellar imprinting affords a convenient method to create
protein-sized, water-soluble organic nanoparticles for natural
products/drugs. It also revealed that zwitterionic cross-linkable
surfactant 4 outperformed the cationic one (1) in most cases
when the guests carried identical hydrophobes but different
ionic groups. Imprinted receptors prepared with this surfactant
was able to categorize and recognize closely related alkaloids.
The one-pot, 2-day synthesis of MINP at room temperature
was straightforward to perform for complex as well as simple
alkaloids. Given that many important alkaloids are isolated
from natural sourcese.g., vinblastine from the leaves of
Catharanthus roseus57,58a facile preparation of their receptors
is significant for their isolation and purification.
Table 2 shows that both MINP4(11) and MINP4(12)
bound their template more strongly than other alkaloids. The
binding constant for 11 and 12 was 3.15 and 4.60 × 105 M−1,
respectively, by their own MINP receptor (entries 1 and 8).
When 11 was used as the template, CRR was 0.16 for 12 and
0.63 for 13. Thus, 13 was more similar to 11 than 12, as far as
the binding was concerned. When 12 was used as the template,
CRR was 0.08 for 11 and 0.05 for 13. Thus, once again, 11 and
13 were very similar and differed significantly from 12, on the
basis of their interactions with the MINP. Consistent with
these results, when the guests differed more substantially,
binding became even weaker. The binding constants of
MINP4(11) for 8−10 (Table 2, entries 4−6), for example,
were only about twice as much as those of nonimprinted
nanoparticles for the same guests (Table 1, entries 10−12).
Unlike molecular synthesis of receptors that gets more
difficult with increasing complexity of ligands, molecular
imprinting of complex guests can be as straightforward as
that of simple ones. Vinblastine 14 is a dimeric indole-based
alkaloid with a complex structure. It is able to inhibit the
formation of microtubules57,58 and normally administered
together with other anticancer drugs against a wide range of
cancers. In our hands, it could be imprinted as easily as the
other much simpler templates. We also measured the binding
of the alkaloid by the MINP. The binding constant was 7.81 ×
105 M−1 (Table 2, entry 7), in line with those for other
alkaloids (by their own MINP receptors).
EXPERIMENTAL SECTION
■
General Method. Routine 1H and 13C NMR spectra were
recorded on a 400 and 600 MHz NMR spectrometer. Electrospray
ionization mass spectrometry (ESI-MS) data were recorded on
Shimadzu LCMS-2010 mass spectrometer. Dynamic light scattering
(DLS) data were recorded at 25 °C using PDDLS/CoolBatch 90T
with PD2000DLS instrument. Isothermal titration calorimetry (ITC)
was performed using a MicroCal VP-ITC Microcalorimeter with
Origin 7 software and VPViewer2000 (GE Healthcare, Northampton,
MA). Syntheses of compounds 1−3 and 6 were reported previously.21
Compound 4. N,N-Diisopropylethylamine (84 mg, 0.65 mmol)
was added slowly to a stirred solution of 6 (135 mg, 0.5 mmol) and 7
(71 mg, 0.5 mmol) in anhydrous tetrahydrofuran (2 mL) at 0 °C. The
reaction mixture was warmed up to room temperature. After 2 h, the
mixture was filtered through a pad of celite and the filtrate was
concentrated under reduced pressure. The colorless oil was dissolved
in anhydrous methylene chloride (3 mL) at 0 °C, followed by
addition of trimethylsilyl triflate (111 mg, 0.5 mmol) and
tripropargylamine (200 mg, 1.5 mmol). The reaction mixture was
warmed to room temperature and stirred for 12 h. The reaction
mixture was diluted with chloroform (10 mL) and neutralized with
saturated sodium bicarbonate solution. The organic layer was
separated, dried over anhydrous Na2SO4, and concentrated under
reduced pressure. The residue was purified by column chromatog-
raphy over silica gel using 4:1 methylene chloride/methanol and then
55:45:3 methylene chloride/methanol/water as eluents to afford a
colorless oil (139 mg, 55%). 1H NMR (600 MHz, CDCl3, δ): 6.09 (s,
1H), 5.54 (s, 1H), 4.78 (s, 6H), 4.40 (s, 2H), 4.13 (t, J = 6.7 Hz, 2H),
3.99 (s, 2H), 3.84 (m, 2H), 3.42 (s, 3H), 1.94 (s, 3H), 1.65 (m, 4H),
1.19 (m, 16H). 13C{H} NMR (100 MHz, CDCl3, δ): 167.6, 136.5,
125.2, 83.4, 70.8, 66.0, 60.20, 58.8, 51.2, 30.9, 30.9, 29.72, 29.70,
29.63, 29.57, 29.3, 28.6, 26.0, 25.91 18.3. HRMS (ESI/QTOF) m/z:
[M + H]+ calcd for C27H43NO6P, 508.2823; found 508.2835.
CONCLUSIONS
■
Synthetic receptors have potential applications in the isolation,
extraction, and monitoring of biologically interesting mole-
cules, including drugs and natural products. They could also be
used to probe or intervene biological interactions. Despite all
F
DOI: 10.1021/acs.joc.9b01629
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Preparation of MINPs and NINPs. Preparation of MINP1
followed previously reported procedures, and typical yields were
70−80%.21 In the preparation of MINP4, copper chloride used in the
click reaction (for surface-cross-linking and functionalization)
precipitated in the presence of 4. The click reaction was performed
with copper chloride as the precatalyst with additional 4 equiv of a
tridentate ligand N,N,N′,N″,N″-pentamethyldiethylenetriamine.59 For
the purification of MINP4, the final reaction mixture was poured into
acetone (8 mL). The precipitate was collected by centrifugation and
washed with a mixture of methanol (5 mL) three times. The off-white
powder was dried in air to afford the final MINPs in typically 70%
yield. The preparation of nonimprinted nanoparticles (NINP1 and
NINP4) followed the same procedures, except that the template was
not used.
Determination of Binding Constants by Fluorescence
Titration. A typical procedure is as follows. A stock solution
containing MINP4(8) (200 μM) was prepared in 50 mM Tris buffer
(pH = 7.4). Aliquots (2 μL) of the MINP stock solution were added
to 2.00 mL of the solutions of 8−10 in the same buffer (2.0 μM).
After each addition, the sample was allowed to sit for 1 min at room
temperature before the fluorescence spectrum was collected. The
excitation wavelength (λex) was 300 nm. The excitation slit width was
10 nm, and the emission slit width was 10 nm. The binding constant
was obtained by nonlinear least-squares fitting of the emission
intensity at 363 nm to 1:1 binding isotherm.
(8) Pan, J.; Chen, W.; Ma, Y.; Pan, G. Molecularly Imprinted
Polymers as Receptor Mimics for Selective Cell Recognition. Chem.
Soc. Rev. 2018, 47, 5574−5587.
(9) Lakshmi, D.; Bossi, A.; Whitcombe, M. J.; Chianella, I.; Fowler,
S. A.; Subrahmanyam, S.; Piletska, E. V.; Piletsky, S. A. Electro-
chemical Sensor for Catechol and Dopamine Based on a Catalytic
Molecularly Imprinted Polymer-Conducting Polymer Hybrid Recog-
nition Element. Anal. Chem. 2009, 81, 3576−3584.
(10) Hoshino, Y.; Koide, H.; Urakami, T.; Kanazawa, H.; Kodama,
T.; Oku, N.; Shea, K. J. Recognition, Neutralization, and Clearance of
Target Peptides in the Bloodstream of Living Mice by Molecularly
Imprinted Polymer Nanoparticles: A Plastic Antibody. J. Am. Chem.
Soc. 2010, 132, 6644−6645.
(11) Hoshino, Y.; Kodama, T.; Okahata, Y.; Shea, K. J. Peptide
Imprinted Polymer Nanoparticles: A Plastic Antibody. J. Am. Chem.
Soc. 2008, 130, 15242−15243.
(12) Urraca, J. L.; Aureliano, C. S. A.; Schillinger, E.; Esselmann, H.;
Wiltfang, J.; Sellergren, B. Polymeric Complements to the Alzheimer’s
Disease Biomarker B-Amyloid Isoforms Aβ1−40 and Aβ1−42 for
Blood Serum Analysis under Denaturing Conditions. J. Am. Chem.
Soc. 2011, 133, 9220−9223.
(13) Ma, Y.; Pan, G. Q.; Zhang, Y.; Guo, X. Z.; Zhang, H. Q.
Narrowly Dispersed Hydrophilic Molecularly Imprinted Polymer
Nanoparticles for Efficient Molecular Recognition in Real Aqueous
Samples Including River Water, Milk, and Bovine Serum. Angew.
Chem., Int. Ed. 2013, 52, 1511−1514.
(14) Kuwata, T.; Uchida, A.; Takano, E.; Kitayama, Y.; Takeuchi, T.
Molecularly Imprinted Polymer Arrays as Synthetic Protein Chips
Prepared by Transcription-Type Molecular Imprinting by Use of
Protein-Immobilized Dots as Stamps. Anal. Chem. 2015, 87, 11784−
11791.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.joc.9b01629.
■
S
Fluorescence and ITC titration curves, additional
figures, and NMR spectra of key compounds (PDF)
(15) Liu, Y. B.; Fang, S. M.; Zhai, J. Q.; Zhao, M. P. Construction of
Antibody-Like Nanoparticles for Selective Protein Sequestration in
Living Cells. Nanoscale 2015, 7, 7162−7167.
(16) Liu, J.; Yin, D.; Wang, S.; Chen, H. Y.; Liu, Z. Probing Low-
Copy-Number Proteins in a Single Living Cell. Angew. Chem., Int. Ed.
2016, 55, 13215−13218.
AUTHOR INFORMATION
Corresponding Author
*E-mail: zhaoy@iastate.edu.
■
(17) Horikawa, R.; Sunayama, H.; Kitayama, Y.; Takano, E.;
Takeuchi, T. A Programmable Signaling Molecular Recognition
Nanocavity Prepared by Molecular Imprinting and Post-Imprinting
Modifications. Angew. Chem., Int. Ed. 2016, 55, 13023−13027.
(18) Panagiotopoulou, M.; Salinas, Y.; Beyazit, S.; Kunath, S.; Duma,
L.; Prost, E.; Mayes, A. G.; Resmini, M.; Bui, B. T. S.; Haupt, K.
Molecularly Imprinted Polymer Coated Quantum Dots for Multi-
plexed Cell Targeting and Imaging. Angew. Chem., Int. Ed. 2016, 55,
8244−8248.
ORCID
Yan Zhao: 0000-0003-1215-2565
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The authors acknowledge NIGMS (R01GM113883) for the
financial support of this research.
■
(19) Bertolla, M.; Cenci, L.; Anesi, A.; Ambrosi, E.; Tagliaro, F.;
Vanzetti, L.; Guella, G.; Bossi, A. M. Solvent-Responsive Molecularly
Imprinted Nanogels for Targeted Protein Analysis in Maldi-Tof Mass
Spectrometry. ACS Appl. Mater. Interfaces 2017, 9, 6908−6915.
REFERENCES
■
(1) Wulff, G. Enzyme-Like Catalysis by Molecularly Imprinted
Polymers. Chem. Rev. 2002, 102, 1−28.
̌
̇
(20) Pan, G.; Shinde, S.; Yeung, S. Y.; Jakstaite, M.; Li, Q.; Wingren,
A. G.; Sellergren, B. An Epitope-Imprinted Biointerface with Dynamic
Bioactivity for Modulating Cell−Biomaterial Interactions. Angew.
Chem., Int. Ed. 2017, 56, 15959−15963.
(2) Haupt, K.; Mosbach, K. Molecularly Imprinted Polymers and
Their Use in Biomimetic Sensors. Chem. Rev. 2000, 100, 2495−2504.
(3) Aherne, A.; Alexander, C.; Payne, M. J.; Perez, N.; Vulfson, E. N.
Bacteria-Mediated Lithography of Polymer Surfaces. J. Am. Chem. Soc.
1996, 118, 8771−8772.
(4) Ren, K.; Zare, R. N. Chemical Recognition in Cell-Imprinted
Polymers. ACS Nano 2012, 6, 4314−4318.
(5) Borovicka, J.; Stoyanov, S. D.; Paunov, V. N. Shape Recognition
(21) Awino, J. K.; Zhao, Y. Protein-Mimetic, Molecularly Imprinted
Nanoparticles for Selective Binding of Bile Salt Derivatives in Water. J.
Am. Chem. Soc. 2013, 135, 12552−12555.
(22) Nowick, J. S.; Chen, J. S.; Noronha, G. Molecular Recognition
in Micelles - the Roles of Hydrogen-Bonding and Hydrophobicity in
Adenine Thymine Base-Pairing in Sds Micelles. J. Am. Chem. Soc.
1993, 115, 7636−7644.
(23) Nowick, J. S.; Chen, J. S. Molecular Recognition in Aqueous
Micellar Solution - Adenine Thymine Base-Pairing in Sds Micelles. J.
Am. Chem. Soc. 1992, 114, 1107−1108.
(24) Awino, J. K.; Gunasekara, R. W.; Zhao, Y. Sequence-Selective
Binding of Oligopeptides in Water through Hydrophobic Coding. J.
Am. Chem. Soc. 2017, 139, 2188−2191.
̌
of Microbial Cells by Colloidal Cell Imprints. Nanoscale 2013, 5,
8560−8568.
(6) Shen, X.; Bonde, J. S.; Kamra, T.; Bulow, L.; Leo, J. C.; Linke,
̈
D.; Ye, L. Bacterial Imprinting at Pickering Emulsion Interfaces.
Angew. Chem., Int. Ed. 2014, 53, 10687−10690.
(7) Bao, H.; Yang, B.; Zhang, X.; Lei, L.; Li, Z. Bacteria-Templated
Fabrication of a Charge Heterogeneous Polymeric Interface for
Highly Specific Bacterial Recognition. Chem. Commun. 2017, 53,
2319−2322.
G
DOI: 10.1021/acs.joc.9b01629
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(25) Gunasekara, R. W.; Zhao, Y. A General Method for Selective
Recognition of Monosaccharides and Oligosaccharides in Water. J.
Am. Chem. Soc. 2017, 139, 829−835.
(26) Hu, L.; Arifuzzaman, M. D.; Zhao, Y. Controlling Product
Inhibition through Substrate-Specific Active Sites in Nanoparticle-
Based Phosphodiesterase and Esterase. ACS Catal. 2019, 9, 5019−
5024.
(27) Arifuzzaman, M. D.; Zhao, Y. Water-Soluble Molecularly
Imprinted Nanoparticle Receptors with Hydrogen-Bond-Assisted
Hydrophobic Binding. J. Org. Chem. 2016, 81, 7518−7526.
(28) Brossi, A. The Alkaloids. Chemistry and Pharmacology; Academic
Press: New York, 1983.
(29) Aniszewski, T. AlkaloidsSecrets of Life: Alkaloid Chemistry,
Biological Significance, Applications and Ecological Role; Elsevier:
Amsterdam, 2007.
(30) Lee, A. G. Biomembranes: A Multi-Volume Treatise; JAI Press:
Greenwich, 1995.
(31) Yeagle, P. The Structure of Biological Membranes, 2nd ed.; CRC
Press: Boca Raton, 2005.
(32) Ishihara, K.; Ueda, T.; Nakabayashi, N. Preparation of
Phospholipid Polylners and Their Properties as Polymer Hydrogel
Membranes. Polym. J. 1990, 22, 355−360.
(33) Ishihara, K.; Nomura, H.; Mihara, T.; Kurita, K.; Iwasaki, Y.;
Nakabayashi, N. Why Do Phospholipid Polymers Reduce Protein
Adsorption? J. Biomed. Mater. Res. 1998, 39, 323−330.
(34) Patel, J. D.; Iwasaki, Y.; Ishihara, K.; Anderson, J. M.
Phospholipid Polymer Surfaces Reduce Bacteria and Leukocyte
Adhesion under Dynamic Flow Conditions. J. Biomed. Mater. Res.,
Part A 2005, 73A, 359−366.
(35) Chen, M.; Briscoe, W. H.; Armes, S. P.; Klein, J. Lubrication at
Physiological Pressures by Polyzwitterionic Brushes. Science 2009,
323, 1698−1701.
(36) Seuring, J.; Reiss, P.; Koert, U.; Agarwal, S. Synthesis,
Characterization and Properties of a New Polymerisable Surfactant:
12-Methacryloyl Dodecylphosphocholine. Chem. Phys. Lipids 2010,
163, 367−372.
(37) Rosen, M. J. Surfactants and Interfacial Phenomena, 2nd ed.;
Wiley: New York, 1989; pp 108−168.
rescence of Proteins. Methods in Enzymology; Academic Press, 1997;
Vol. 278, pp 113−150.
(49) Valeur, B.; Weber, G. Resolution of the Fluorescence Excitation
Spectrum of Indole into the 1la and 1lb Excitation Bands*. Photochem.
Photobiol. 1977, 25, 441−444.
(50) Gasymov, O. K.; Abduragimov, A. R.; Glasgow, B. J. Probing
Tertiary Structure of Proteins Using Single Trp Mutations with
Circular Dichroism at Low Temperature. J. Phys. Chem. B 2014, 118,
986−995.
́
(51) Seredyuk, V.; Alami, E.; Nyden, M.; Holmberg, K.; Peresypkin,
A. V.; Menger, F. M. Micellization and Adsorption Properties of
Novel Zwitterionic Surfactants. Langmuir 2001, 17, 5160−5165.
̃
(52) Isom, D. G.; Castaneda, C. A.; Cannon, B. R.; García-Moreno,
B. E. Large Shifts in pKa Values of Lysine Residues Buried inside a
Protein. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 5260−5265.
(53) The pKa shift is less significant when the ammonium group is
on the surface of the micelle, strongly solvated by water.
(54) Rubin, M.; Wishinsky, H. 1-Arylcyclohexanecarboxylic Acids. J.
Am. Chem. Soc. 1946, 68, 828−832.
(55) Marshall, F. N.; Williamson, H. E. Natruretic Response During
Infusion of Beta-Diethylaminoethyl-Diphenylpropyl Acetate Hydro-
cloride (Skf 525-a). J. Pharmacol. Exp. Ther. 1964, 143, 395−400.
̌
̌
́
̌
̌
́
̌
́
(56) Jendzelovsky, R.; Koval, J.; Mikes, J.; Papcova, Z.; Plsíkova, J.;
̌
Fedorocko, P. Inhibition of Gsk-3beta Reverses the Pro-Apoptotic
Effect of Proadifen (Skf-525a) in Ht-29 Colon Adenocarcinoma Cells.
Toxicol. In Vitro 2012, 26, 775−782.
(57) Na, G. C.; Timasheff, S. N. Stoichiometry of the Vinblastine-
Induced Self-Association of Calf Brain Tubulin. Biochemistry 1980, 19,
1347−1354.
(58) Jordan, M. A.; Wilson, L. Microtubules as a Target for
Anticancer Drugs. Nat. Rev. Cancer 2004, 4, 253−265.
(59) Zhang, S.; Zhao, Y. Controlled Release from Cleavable
Polymerized Liposomes Upon Redox and pH Stimulation. Bio-
conjugate Chem. 2011, 22, 523−528.
(38) Kalyanasundaram, K.; Thomas, J. K. Environmental Effects of
Vibronic Band Intensities in Pyrene Monomer Fluorescence and
Their Application in Studies of Micellar Systems. J. Am. Chem. Soc.
1977, 99, 2039−2044.
(39) Zhang, S.; Zhao, Y. Facile Synthesis of Multivalent Water-
Soluble Organic Nanoparticles Via “Surface Clicking” of Alkynylated
Surfactant Micelles. Macromolecules 2010, 43, 4020−4022.
(40) Bendheim, P. E.; Poeggeler, B.; Neria, E.; Ziv, V.; Pappolla, M.
A.; Chain, D. G. Development of Indole-3-Propionic Acid (Oxigon)
for Alzheimer’s Disease. J. Mol. Neurosci. 2002, 19, 213−217.
(41) Khan, M. Z.; Nawaz, W. The Emerging Roles of Human Trace
Amines and Human Trace Amine-Associated Receptors (Htaars) in
Central Nervous System. Biomed. Pharmacother. 2016, 83, 439−449.
(42) Li, X.; Zhao, Y. Protection/Deprotection of Surface Activity
and Its Applications in the Controlled Release of Liposomal Contents.
Langmuir 2012, 28, 4152−4159.
(43) Fa, S.; Zhao, Y. Peptide-Binding Nanoparticle Materials with
Tailored Recognition Sites for Basic Peptides. Chem. Mater. 2017, 29,
9284−9291.
(44) Fa, S.; Zhao, Y. Water-Soluble Nanoparticle Receptors
Supramolecularly Coded for Acidic Peptides. Chem. − Eur. J. 2018,
24, 150−158.
1
(45) Slater, L. S.; Callis, P. R. Molecular Orbital Theory of the la
and 1lb States of Indole. 2. An Ab Initio Study. J. Phys. Chem. A. 1995,
99, 8572−8581.
(46) Pierce, D. W.; Boxer, S. G. Stark Effect Spectroscopy of
Tryptophan. Biophys. J. 1995, 68, 1583−1591.
(47) Lami, H.; Glasser, N. Indole’s Solvatochromism Revisited. J.
Chem. Phys. 1986, 84, 597−604.
(48) Callis, P. R. [7] 1La and 1Lb Transitions of Tryptophan:
Applications of Theory and Experimental Observations to Fluo-
H
DOI: 10.1021/acs.joc.9b01629
J. Org. Chem. XXXX, XXX, XXX−XXX