Versatile Salts as a Strategy to Modify the Biopharmaceutical Properties of Venlafaxine and a Potential Hypoglycemic Effect Study

Article abstract of DOI:10.1021/acs.cgd.0c00007

Venlafaxine (VLF) is a widely prescribed antidepressant thought to have little effect on glucose metabolism, but some clinical reports showed that a VLF overdose may cause hypoglycemia. The raw material of VLF hydrochloride is metastable and may cause problems during the manufacturing and storage process. Here, salt formation was employed to modify the biopharmaceutical properties, and the hypoglycemic potential of these salt forms was also investigated. Four organic acid salts with phthalic acid (PA), 4-chlorobenzoic acid (4-CA), 4-hydroxybenzoic acid (4-HA), and sulfanilic acid (SA) were obtained for the first time. These products were fully characterized by single crystal and powder X-ray diffraction, solid-state nuclear magnetic resonance, differential scanning calorimetry, and thermogravimetric analysis. Stability measurements revealed that most of these forms held superior phase stability against temperature, moisture, and illumination. All salts show significantly improved solubility relative to VLF. The glucose consumption test was also performed in vitro, VLF:HCl could significantly increase glucose consumption and have a higher glucose consumption index. VLF:PA, VLF:4-CA, VLF:4-HA, and VLF:SA demonstrated no hypoglycemia side effect.

Full text of DOI:10.1021/acs.cgd.0c00007

pubs.acs.org/crystal
Article
Versatile Salts as a Strategy to Modify the Biopharmaceutical
Properties of Venlafaxine and a Potential Hypoglycemic Effect Study
Ningbo Gong, Yong Zhang, Ying Wang, Hongmei Yu, Baoxi Zhang, Hailu Zhang,* Yang Lu,*
and Guanhua Du
Cite This: https://dx.doi.org/10.1021/acs.cgd.0c00007
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Venlafaxine (VLF) is a widely prescribed antidepressant thought to have little effect on glucose metabolism, but some
clinical reports showed that a VLF overdose may cause hypoglycemia. The raw material of VLF hydrochloride is metastable and may
cause problems during the manufacturing and storage process. Here, salt formation was employed to modify the biopharmaceutical
properties, and the hypoglycemic potential of these salt forms was also investigated. Four organic acid salts with phthalic acid (PA),
4-chlorobenzoic acid (4-CA), 4-hydroxybenzoic acid (4-HA), and sulfanilic acid (SA) were obtained for the first time. These
products were fully characterized by single crystal and powder X-ray diffraction, solid-state nuclear magnetic resonance, differential
scanning calorimetry, and thermogravimetric analysis. Stability measurements revealed that most of these forms held superior phase
stability against temperature, moisture, and illumination. All salts show significantly improved solubility relative to VLF. The glucose
consumption test was also performed in vitro, VLF:HCl could significantly increase glucose consumption and have a higher glucose
consumption index. VLF:PA, VLF:4-CA, VLF:4-HA, and VLF:SA demonstrated no hypoglycemia side effect.
Venlafaxine may exist in diverse solid forms. The racemate
crystal structure of venlafaxine was reported by Tessler et al. in
2004.13 Three polymorphs of venlafaxine were reported by van
Eupen,14 and one venlafaxine besylate monohydrate was
reported by Corvalan.15 Synthesis and structure of venlafaxine
hydrochloride,16,17 venlafaxine hydrobromide salt,18 and
venlafaxine saccharin salts19 were reported.
Venlafaxine is extremely insoluble, so it is orally adminis-
tered in its hydrochloride form. Venlafaxine hydrochloride has
high solubility. Because of its short half-life, venlafaxine is
administered 2−3 times daily to maintain an effective
therapeutic concentration. Thus, a set of molecular salts of
1. INTRODUCTION
More than 80% of marketed drugs are formulated in the solid
form given the ease and advantage of preparation, manufactur-
ing, processing, administration, and storage.1 Investigators
typically carry out a systematical study of different solid forms
such as polymorphs, hydrates/solvates, salts, and amorphous
forms of an active pharmaceutical ingredient (API) to modify
its physicochemical properties including crystallinity, melting
point, solubility, dissolution, stability, and bioavailability
without changing its core chemical structure.2,3 Among them,
salt formation is one of the most traditional and straightfor-
ward methods in the pharmaceutical industry.
Venlafaxine (1-[2-(dimethylamino)-1-(4-methoxyphenyl)-
ethyl]cyclohexanol, VLF, CAS: 93413-69-5, Figure 1) is a
widely prescribed antidepressant (Effexor) that selectively
inhibits the reuptake of serotonin and norepinephine.4,5 Recent
studies have shown that venlafaxine possesses anti-inflamma-
tory activities,6 alleviating neuropathic pain,7−9 stimulates
neurogenesis,10,11 and disrupts larval behavior in zebrafish.12
Received: January 2, 2020
Revised: March 19, 2020
Published: March 19, 2020
https://dx.doi.org/10.1021/acs.cgd.0c00007
Cryst. Growth Des. XXXX, XXX, XXX−XXX
© XXXX American Chemical Society
A

Crystal Growth & Design
pubs.acs.org/crystal
Article
prepared four new venlafaxine salts with coformers with
phthalic acid (PA), 4-chlorobenzoic acid (4-CA), 4-hydrox-
ybenzoic acid (4-HA), and sulfanilic acid (SA). These forms
were comprehensively characterized by X-ray diffraction. In
addition, their physicochemical and biopharmaceutical proper-
ties were characterized using different techniques including
solid state nuclear magnetic resonance (ssNMR), differential
scanning calorimetry (DSC), thermogravimetric analysis
(TGA), and solubility measurements. Stability of the salts
was evaluated in different conditions. The new contents of
spatial structure analyses, ssNMR, and solubility studies of
venlafaxine maleate (VLF:MA) were also described here.
Venlafaxine, a serotonin norepinephine reuptake inhibitor, is
thought to have little effect on glucose metabolism, but there
are reports that venlafaxine hydrochloride overdose causes
hypoglycemia.23−25 In order to screen the bioactivities of the
new salts of venlafaxine, we investigated glucose consumption
using human liver cells.
Finally, this new salt of venlafaxine constitutes an interesting
strategy to modify the biopharmaceutical properties and
stability of venlafaxine. VLF:HCl could significantly increase
glucose consumption and have a higher glucose consumption
index. VLF:PA, VLF:4-CA, VLF:4-HA, and VLF:SA demon-
strated no hypoglycemia side effect, which means that further
investigations are required to discover the hypoglycemic
mechanism of the salts of venlafaxine hydrochloride.
Figure 1. Molecular structures of the venlafaxine and the list of
organic acids used in this study.
2. MATERIALS AND METHODS
2.1. Materials. Venlafaxine hydrochloride raw material with a
purity of 99.9%, according to high pressure liquid chromatography
(HPLC), was supplied by Wuhan Yinhe Chemical Co., Ltd. (Hubei,
China). Maleic acid, phthalic acid, 4-chlorobenzoic acid, 4-
hydroxybenzoic acid, sulfanilic acid, and sodium hydroxide were
purchased from Beijing Chemical Reagent Co. (Beijing, China). All
analytical grade solvents were purchased from Sigma-Aldrich (St.
Louis, MO, USA) and were used without further purification.
coumaric, ferulic, oxalic, salicylic, fumaric, and citric acid were
obtained.20 We had already prepared the venlafaxine maleate
and characterized it.21 To propose alternative forms, improve
the manufacturing and extend the range of solid forms,
eliminate or alleviate the side effect, and find new applications
of venlafaxine,22 we had related supramolecular synthesis and
Table 1. Crystallographic Data and Structure Refinement Details
VLF:MA21
VLF:PA
VLF:4-CA
empirical formula
molecular weight
crystal size (mm)
temperature (K)
description
(C17H28NO2) + ·(C4H3O4)−
393.47
0.32 × 0.14 × 0.05
295(2)
plate
(C17H28NO2) + ·(C8H5O4)−
443.51
0.15 × 0.10 × 0.03
295(2)
plate
(C17H28NO2) + ·(C7H4ClO2)−
433.95
0.57 × 0.56 × 0.16
295(2)
block
crystal system
space group
monoclinic
P21/a
triclinic
P1
̅
orthorhombic
Pna21
a (Å)
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
Z
8.543(3)
23.859(7)
10.469(4)
90
100.608(12)
90
8.907(4)
10.562(5)
12.638(6)
85.44(4)
85.46(4)
83.30(4)
2
15.5718(3)
14.4294(3)
10.37708(17)
90
90
90
4
4
volume (Å3)
density (g/cm3)
independent reflections
reflections with I > 2σ(I)
Rint
2097.4(13)
1.246
4005
3704
0.0451
1174.2(10)
1.252
4412
3064
0.1076
2331.65(8)
1.236
3548
3176
0.0306
final R,wR (F2) values [I > 2σ(I)]
goodness-of-fit on F2
completeness
CCDC no.
0.0523, 0.1472
1.017
0.977
0.0876, 0.2186
1.092
0.971
0.0335, 0.0822
1.031
0.997
1960401
1960399
1960400
B
https://dx.doi.org/10.1021/acs.cgd.0c00007
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
pubs.acs.org/crystal
Article
Figure 2. Crystal structure and the 1D, 2D salt bond, and H-bond contacts of venlafaxine salts.
2.2. Preparation of Venlafaxine. To prepare free base of
venlafaxine,26 venlafaxine chloride and NaOH were mixed at a definite
stoichiometric ratio (1:1) in water. The resulting mixture was
sufficiently stirred at room temperature for 40 min to make sure that
the reaction was completed. Then, ethyl acetate was added to extract
the free base, and the organic phase was evaporated slowly under
decompression conditions. The powder was recrystallized in hexane
to acquire venlafaxine crystal at room temperature for a week. The cell
parameters are a = 8.4135(12) Å, b = 8.8669(12) Å, c = 21.7900(30)
Å, β = 92.307(6)°, as described in the literature14 (CCDC refcode:
OCALAG02).
2.5. Single Crystal X-ray Diffraction (SXRD). The SXRD data
of the Three crystals were collected using an microMax 002+ system
equipped with a Cu fine-focus sealed tube (λ = 1.54187 Å) at 295 K.
The crystallographic data are listed in Table 1. The data were
processed using CrystalClear software (Rigaku, 2009). The structure
was solved by direct methods and refined with full matrix least-squares
on F2 with anisotropic displacement parameters of non-H atoms using
the SHELXS-2016.27 All non-hydrogen atoms were refined with
anisotropic displacement parameters. Especially, H atoms of N−H or
O−H were located from the difference electron density maps, whereas
other H atoms were placed in calculated positions by the riding model
in idealized geometries.
2.6. Powder X-ray Diffraction (PXRD). Powder X-ray diffraction
(PXRD) studies of microcrystalline samples were performed in
Bragg−Brentano geometry on a D/max-2550 (Rigaku, Japan) X-ray
diffractometer with graphite monochomator (40 kV, 150 mA, CuKα,
λ = 1.54187 Å). A scan speed of 8°/min with a step size of 0.02° in
the 2θ range 3−40° was used. The simulated powder patterns were
calculated from the single crystal data using Mercury 3.3.
2.3. Salt Synthesis. Venlafaxine salts were obtained using
different preparation methods as described below.
2.3.1. Liquid Assisted Grinding Method. VLF:MA, VLF:PA, and
VLF:4-CA were prepared by liquid assisted grinding (LAG).
Equimolar amounts of venlafaxine and organic acid were ground
using a mortar and pestle for 10−15 min with the addition of a few
drops of ethyl acetate/ethanol. The resulting powders were collected
for characterization.
2.3.2. Slurry Method. VLF:4-HA and VLF:SA were prepared by a
slurry method. Equimolar amounts of venlafaxine and organic acid
were suspended in ethyl acetate. The suspension was agitated at 600g
for 24 h at room temperature, and the agglomerates were filtered and
dried in a 40 °C oven for 12 h.
2.4. Preparation of Single Crystals. VLF:MA, VLF:PA, and
VLF:4-CA were carried out by dissolving venlafaxine and the acid
coformers in 1:1 molar ratios in the minimum amount of the
appropriate solvent (ethyl acetate or ethanol). The solutions were
allowed to evaporate slowly at room temperature for about 30 days
until crystals suitable for single crystal X-ray diffraction experiments
were obtained.
2.7. Solid State NMR Measurements. 13C Cross-polarization/
magic angle spinning (CP/MAS) spectra of all samples were collected
on a Bruker AVANCE III-500 spectrometer equipped with a 4 mm
double resonance MAS probe. A total sideband suppression (TOSS)
frame was embedded in the standard CP pulse program to remove the
spinning sidebands. 13C CP/MAS TOSS NMR experiments were
performed at an 8 kHz MAS spinning frequency with a 2 ms contact
time. The recycle delay parameters for MA, PA, 4-CA, and 4-HA were
500, 40, 60, and 120 s, and for other samples 8 s, respectively. The 13C
chemical shifts were externally referenced to tetramethylsilane (0
ppm).
C
https://dx.doi.org/10.1021/acs.cgd.0c00007
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
pubs.acs.org/crystal
Article
2.8. Thermal Analysis. The thermal behaviors of differential
scanning calorimetry (DSC) of venlafaxine salts were investigated
using DSC1 Instruments (Mettler Toledo, Greifensee, Switzerland),
Sample measurements were performed at a heating rate of 10 °C
min−1.
The thermal behaviors of TGA were investigated using a Mettler-
Toledo TGA/DSC STARe system (Mettler Toledo, Greifensee,
Switzerland). Samples were heated from 30 to 500 °C in aluminum
oxide cells at a heating rate of 10 °C min −1 under a nitrogen gas flow
of 50 mL min −1. The data were analyzed by using STARe software.
2.9. Stability Studies. For conventional stability studies,
powdered samples of venlafaxine salts were stored in a drug stability
test instrument (SHH-150SD) at three conditions, temperature (60
1 °C), humidity (90 5%, 25 °C), and light (4500 lx 500 lx, 25
°C), Periodically, samples were removed from the instrument and
then measured by the HPLC, PXRD, and DSC method.
2.10. Solubility Studies. Solubility and the dissolution profiles of
venlafaxine salts were investigated in various solutions. Excess
amounts of venlafaxine salts were added into distilled water in a
flask. The salts were placed at 37 °C for 48 h. An aliquot was
centrifuged at 10000g for 10 min. The content of venlafaxine salts in
supernatants, defined as solubility in this study, was assayed by HPLC.
The concentration was determined by HPLC based on a calibration
curve method using an Agilent 1260 HPLC (Agilent Technologies,
USA). Samples were separated by using a Linksil-ODS 5 μm (150
mm × 4.6 mm) column. The mobile phase consisted of acetonitrile
and 0.1 mol L−1 potassium dihydrogen phosphate water solution
(32:68, v/v), and the flow rate was 1.0 mL min −1 with simultaneous
multichannel UV detection at 225 nm. The column temperature was
set to 35 °C, and the injection volume was 5 μL.
3.1.3. VLF:4-CA Salt (1:1). The structure was determined in
space group Pna21. The crystal structure contained one VLF
molecule and one 4-CA molecule in the asymmetric unit.
Different from VLF-MA and VLF-PA, VLF and 4-CA were
bonded through an ionic bond N1−H1···O4 (2.565 Å, 163.18°)
and intermolecular hydrogen bonds O2−H2A···O3 (2.750 Å,
2
175.77°) into a R2 (10) ring (Figure 2-3).
The final data collection parameters and refinement statistics
for all structures are summarized in Table 1. The CIF files for
each refinement can be retrieved from the Cambridge
Structural Database (CSD) (CCDC numbers 1960399−
1960401). The ionic bonds and hydrogen-bond geometrical
parameters are given in Tables S1.
Only VLF:4-CA was crystallized in the chiral space group
Pna21, and the next two salts were all crystallized as
enantiomer. In the molecules, the plane of ring 1 (C1, C2,
C3, C4, C5, C6) and the least-squares plane of ring 2 (C8, C9,
C10, C11, C12, C13) were almost perpendicular (95.0°, 86.2°,
90.0° for VLF: MA, VLF:PA, VLF:4-CA, respectively). The
torsion angle of the side-chain substituents also had conforma-
tional differences in salts. The torsion angle of C8−C7−C14−
N1 was 161.6°, 165.5°, 137.2°, and the torsion angle of C2−C1-
C7−C8 was −102.9°, −79.2°, −96.3°, respectively. The
molecule overlay is shown in Figure 3.
2.11. Glucose Consumption Assay. HL-7702 cells were
cultured in 1604 medium plus 20% FBS and 1% antibiotics in an
atmosphere of 5% CO2 at 37 °C. Then, cells were trypsinized and
5
seeded onto 96-well plates with 1.0 × 10 cells per well. Before
experiments, cells were allowed to grow to about 70−80% confluence.
The cells were starved overnight in medium containing 0.5% FBS
before drug treatment. For basal glucose consumption assay, the cells
were treated with DMSO or 100 μM venlafaxine salts or 5 mM
metformin for 12 h, with each treatment in quintuplet. Glucose levels
in the supernatant of medium were assayed with glucose assay kit
(based on glucose oxidase method). Glucose consumption was
calculated as the glucose level of the fresh medium minus glucose
level of the cultured medium. The glucose consumption index,
defined as the ratio of drug to DMSO glucose consumption, was used
to evaluate hypoglycemic activity in vitro.
3. RESULTS AND DISCUSSION
3.1. Single Crystal X-ray Diffraction. 3.1.1. VLF:MA Salt
(1:1). The crystal structure was determined in space group
P21/a. Each asymmetric unit contains one molecule each of
VLF and MA. VLF+ and MA− ions are bonded through ionic
bond N1−H1···O4 (2.684 Å, 157.18°). The salt extended via
ionic bond and intramolecular hydrogen bond O6−H6···O5
(2.418 Å, 166.67°) and intermolecular hydrogen bonds O2−
H2···O5 (2.874 Å, 160.82°), forming into a 1D molecular chain
infinitely extended along the [001] direction (Figure 2-1). The
plane of MA is parallel to the benzene ring in venlafaxine.
3.1.2. VLF:PA Salt (1:1). The structure was determined in
Figure 3. Overlay of molecular conformations (magenta, VLF:MA;
green, VLF:PA; blue, VLF:4-CA).
3.2. Powder X-ray Diffraction. Each crystalline form of a
given substance will produce a characteristic PXRD pattern.
Figure 4 shows the PXRD patterns of venlafaxine, organic acids
and venlafaxine salts, respectively. The experimental powder X-
ray diffraction of VLF:PA, and VLF:4-CA demonstrated good
agreement with their single crystal simulated patterns, which
confirms the purity and homogeneity of the salts in a single
phase. According to the salt forming reaction, the PXRD
patterns of four salts were different with the raw materials and
the physical mixture, indicating toward formation of the new
solid phase. The main powder XRD peaks of salts and their
raw materials are shown in Supporting Information (Table S2).
3.3. 13C Solid-State NMR Analysis. Solid-state NMR is
another often employed technique for analysis of solid forms of
̅
space group P1. The crystal structure contains one VLF
molecule and one PA molecule in the asymmetric unit. The
pack mode was similar to the VLF:MA salt. VLF+ and PA− ions
were bonded through ionic bond N1−H1···O6 (2.693 Å,
149.57°). The salt extended via ionic bond and intramolecular
hydrogen bond O3−H3···O5 (2.594 Å, 167.01°) and
intermolecular hydrogen bonds O2−H2···O3 (2.826 Å,
163.79°) into a 1D chain along the [010] direction (Figure
2-2).
D
https://dx.doi.org/10.1021/acs.cgd.0c00007
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
pubs.acs.org/crystal
Article
Figure 4. Powder X-ray diffraction patterns of venlafaxine salts and the raw materials.
pharmaceuticals. The 13C CP/MAS TOSS NMR spectra of
venlafaxine salts and their starting materials are shown in
Figure 5. The 13C chemical shift assignments of venlafaxine, in
form I, can be found in ref 14. Each chemically distinct carbon
in the form is represented by one single resonance, indicating
that the number of molecules per asymmetric unit (Z′) of this
structure should be equal to 1,28 which can be confirmed by
the reported crystal structure. Similarly, the Z′ = 1 information
for VLF:MA, VLF:PA, and VLF:4-CA can also be learned from
their 13C spectra, which are all in line with the solved structures
in this contribution. For VLF:4-HA and VLF:SA, the signal
splitting at several carbon sites (e.g., 73.9/72.5 and 73.1/72.7
ppm for their C8 signals, respectively) can be observed,
supporting that their Z’ values are both equal to 2.
Recently, Grepioni et al. reported six venlafaxine salts along
with their solid state 13C spectra.20 After salt formation, the
chemical shifts of carboxylic signal are altered 3.9, −3.1, 1.9,
0.7, 1.7, and 2.0 ppm, respectively. It seems that the magnitude
and regularity of these chemical shift changes cannot be used
as a clear indicator of salt formation. In this study, similar
phenomena were observed for VLF:MA (from 172.8 to 169.1
ppm), VLF:PA (from 173.1 to 170.2 ppm), VLF:4-CA (from
173.1 to 170.9 ppm), and VLF:4-HA (from 171.8 to 173.3/
173.0 ppm). Meanwhile, we noticed that the 13C chemical shift
changes of the N bonded C14 site may be used as such an
indicator, because this methylene 13C signal is less affected by
molecular packing changes, and the protonation of the
dimethylamino N site should be the main contributor for the
chemical environmental altering. The proton transfer from a
salt former to the dimethylamino group leads to a larger 15N
chemical shift, i.e., a reduced shielding effect on the N site.20
Then, an enhanced shielding effect on the N bonded C14 site
(smaller 13C chemical shift) is desired. These data are
summarized in Table 2. For the two forms of venlafaxine
hydrochloride, similar results, C14 signals at 60 ppm, have
been reported.29
SA molecules exist in the zwitterion form in the solid state.30
For SA, the signals at 134.0 and 145.6 ppm can be assigned to
+
−
−NH3 and −SO3 bonded C, respectively. In VLF:SA, the
+
intermolecular proton transfer should occur from the −NH3
group of SA to dimethylamino N of VLF, which causes a
significant chemical shift change on the −NH3 /−NH2
bonded C from 134.0 to 150.9/152.9 ppm. The ionization
+
state change of SA also causes a large high-field shift on the
−
−SO3 bonded C signal.
3.4. Thermal Analysis Method. Figure 6 shows the DSC
thermograms of raw materials, and four new salts. The sharp
endothermic peaks of VLF:PA, VLF:4-CA, VLF:4-HA, and
VLF:SA were 177.86 °C, 112.85 °C, 146.21 °C, and 143.15
°C, respectively. All four new salts show melting point between
the free base and acid melting point, indicating the possibility
of novel solid forms. The values of melting points indicate that
the thermal stability was VLF:PA > VLF:4-HA ≈ VLF:SA >
VLF:4-CA. At the TG pattern (Figure S1), there is no
evidence of weight loss of solvent or water, which is consistent
E
https://dx.doi.org/10.1021/acs.cgd.0c00007
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
pubs.acs.org/crystal
Article
Figure 5. 13C CP/MAS TOSS NMR spectra of VLF:MA, VLF:PA, VLF:4-CA, VLF:4-HA, VLF:SA, and their starting materials.
Table 2. Chemical Shifts of C14 in Venlafaxine (VLF) and Its Salts
atom
C14
VLF
62.7
VLF:CA
59.4
VLF:PA
59.2
VLF:4-CA
57.5
VLF:4-HA
60.7/59.7
VLF:SA
59.4/59.0
with the high purity of the materials. The weight loss
accompanies the decomposition process.
pure water was evaluated. The solubility of venlafaxine, VLF
MA, VLF:PA, VLF:4-CA, VLF:4-HA, and VLF:SA is 0.39,
288.10, 5.71, 22.61, 47.73, and 36.50 mg mL−1, respectively.
The retention time of venlafaxine and salt are almost the same,
but different from the coformers. The HPLC retention times
were VLF (3.184 min), VLF:HCl (3.158 min), VLF:MA
(1.440 min, 3.168 min), VLF:PA (1.595 min, 3.162 min),
VLF:4-CA (3.161 min, 4.68 2 min), VLF:4-HA (1.978 min,
3.154 min), and VLF:SA (1.292 min, 3.158 min), respectively.
All the five salts have a higher solubility than the free base. A
small amount of the solid phase at the end of solubility
measurement was isolated and dried to a glass plate, and
PXRD and DSC methods were used to analyze the samples.
The results demonstrate that the samples were not completely
identical with the starting polymorphic forms in each case.
3.5. Stability Study. As shown in Table 3, VLF:PA,
VLF:4-CA, and VLF:4-HA remained stable over 10 days under
three conditions, temperature (60 1 °C), humidity (90%
5% RH, 25 °C), and light (4500 lx 500 lx, 25 °C), since there
was no obvious change in the accelerated conditions. VLF:HCl
was stable under high temperature and the light condition for
10 days, but has the polymorphic transformation trend to form
4 (hydrate) under the high humidity condition for 10 days.
VLF:SA was stable under high temperature for 10 days, but
metastable under the high humidity and the light condition for
10 days.
3.6. Solubility Study. The solubility of venlafaxine,
VLF:MA, VLF:PA, VLF:4-CA, VLF:4-HA, and VLF:SA in
F
https://dx.doi.org/10.1021/acs.cgd.0c00007
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
pubs.acs.org/crystal
Article
Figure 6. DSC thermograms of venlafaxine salts and the raw materials.
Table 3. Results of the Stability Study
temperature
humidity
light
no.
sample
5 days
10 days
5 days
10 days
5 days
10 days
1
2
3
4
5
6
VLF (I)31
I + II
I + II
I
I
I
2
A
a
I + III + amorphous
2
A
a
VLF:HCl(2)5
VLF:PA(A)
VLF:4-CA(a)
VLF:4-HA(i)
VLF:SA(α)
2
A
a
i
α
2
A
a
i
α
2 + 4
A
a
i
2 + 4
A
a
i
i
i
α + β
α + β
α + γ
α + γ
a
This indicate that, during the dissolution process, there were
accompanied by transformation and recrystallization.
Table 4. Results of Glucose Consumption Assay
glucose consumption
(mM)
glucose consumption
The dissolution profiles of venlafaxine, VLF:MA, VLF:PA,
VLF:4-CA, VLF:4-HA, and VLF:SA in pure water were also
evaluated (Figure S2). The five venlafaxine salts show the same
dissolution profiles in pure water (pH 7.0) as VLF:HCl, and
the API was released within the first 20 min.
3.7. Glucose Consumption Assay. To evaluate the in
vitro hypoglycemic activity of venlafaxine salts, the glucose
consumption assay was done. HL-7702 cells were treated with
venlafaxine salts for 12 h, and metformin was used as a positive
control. As shown in Table 4, we found that 100 μM of
VLF:HCl or VLF:MA caused a significant increase of glucose
consumption (p < 0.01 or 0.05 vs DMSO) and had a higher
glucose consumption index. Its efficacy was comparable to that
of metformin at 5 mM. But VLF:PA, VLF:4-CA, VLF:4-HA,
and VLF:SA did not show any hypoglycemic activity in vitro.
no.
sample
DMSO
metformin
VLF free base
VLF:HCl
VLF:MA
index
1
2
3
4
5
6
7
8
9
1.90 0.18
2.58 0.24***
1.97 0.28
2.47 0.21**
2.46 0.40*
2.23 0.31
1.97 0.06
2.02 0.01
1.98 0.26
1.00
1.36
1.03
1.30
1.29
1.17
1.03
1.06
1.04
VLF:PA
VLF:4-CA
VLF:4-HA
VLF:SA
a
Values of glucose consumption are the mean SD of quintuplet.
Values of glucose consumption index are the mean of drug treatment
divided by the mean of DMSO treatment. *p < 0.05, **p < 0.01, ***p
< 0.001 vs that of DMSO treatment.
G
https://dx.doi.org/10.1021/acs.cgd.0c00007
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
4. CONCLUSIONS
pubs.acs.org/crystal
Article
China; orcid.org/0000-0002-2274-5703; Phone: +86 10
6315212; Email: luy@imm.ac.cn; Fax: +86 10 63165212
Hailu Zhang − Laboratory of Magnetic Resonance Spectroscopy
and Imaging, Suzhou Institute of Nano-Tech and Nano-Bionics,
Chinese Academy of Sciences, Suzhou 215123, China;
orcid.org/0000-0001-6936-9197; Phone: +86 512
The aim of this study was to investigate structural changes and
salt formation of venlafaxine with versatile acids using different
preparation methods and characterize their physicochemical
properties. In this paper, four new venlafaxine salts were
prepared and characterized. The spatial structure and
interaction information may help to understand the solid
state properties of organic molecules. Crystal structure analysis
shows that there exist reliable N−H···O ionic bonding and O−
H···O hydrogen-bonding interactions between the amino
group and the hydroxyl group of venlafaxine and the hydroxyl
group or/and the carbonyl group of the various acids. The
liquid assisted grinding method and slurry method can be used
to synthesize the salts.
62872713; Email: hlzhang2008@sinano.ac.cn
Authors
Ningbo Gong − Beijing Key Laboratory of Polymorphic Drugs,
Institute of Materia Medica, Chinese Academy of Medical
Sciences and Peking Union Medical College, Beijing 100050,
China; Hainan Medical University, Haikou 571199, China
Yong Zhang − Hainan Medical University, Haikou 571199,
China
Ying Wang − Beijing Key Laboratory of Polymorphic Drugs,
Institute of Materia Medica, Chinese Academy of Medical
Sciences and Peking Union Medical College, Beijing 100050,
China
Hongmei Yu − Beijing Key Laboratory of Polymorphic Drugs,
Institute of Materia Medica, Chinese Academy of Medical
Sciences and Peking Union Medical College, Beijing 100050,
China
Baoxi Zhang − Beijing Key Laboratory of Polymorphic Drugs,
Institute of Materia Medica, Chinese Academy of Medical
Sciences and Peking Union Medical College, Beijing 100050,
China
Although the 4-CA molecule is almost planar and did not
have a chiral center in the chemical structure, the VLF:4-CA
salt can crystallize in the chiral space group. This means 4-CA
maybe used as a selector reagent for racemic venlafaxine or
analogue organic molecules.
SXRD, PXRD, ssNMR, DSC, and TGA can be used to
identify and characterize the different salts. These salts do not
need to always have a high solubility to be useful. The low
solubility maybe has a higher pH suitable for injections and
drops. The five salts all have the lower solubility than that of
VLF:HCl. Among of the salts, VLF:PA has the lowest
solubility. Except for VLF:SA, other venlafaxine salts are stable
under the three test conditions. Since the hydrochloride salt
raw material is metastable at high humidity environment, and
may cause problems during the manufacturing and storage
process, the new salts may provide new API candidates for
venlafaxine commercial production. On the basis of the glucose
consumption assay, VLF:HCl and VLF:MA could significantly
increase glucose consumption and have a higher glucose
consumption index. VLF:PA, VLF:4-CA, VLF:4-HA, and
VLF:SA did not show any hypoglycemic activity in vitro.
Comprehensive consideration of the solubility, stability and
safety, VLF:4-HA was the most appropriate salt for improving
the safety of venlafaxine by minimizing the hypoglycemic
effect.
Guanhua Du − Beijing City Key Laboratory of Drug Target
Identification and Drug Screening, Institute of Materia Medica,
Chinese Academy of Medical Sciences and Peking Union
Medical College, Beijing 100050, China
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.cgd.0c00007
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We acknowledge the financial support from Key National
Research and Development Programmes (No.
2016YFC1000901), CAMS Innovation Fund for Medical
Sciences (Grant No. 2017-I2M-1-010), Construction and
Application of Technology Integration System for Efficient
Identification of Natural/Effective Active Small Molecules
(No. 2018ZX09711001-001), and the Youth Innovation
Promotion Association of CAS (No. 2012242).
ASSOCIATED CONTENT
* Supporting Information
■
sı
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.cgd.0c00007.
The ionic bonds and hydrogen-bond geometry, the main
powder XRD peaks of salts and their raw materials, the
TGA patterns of VLF salts, the dissolution profiles of
VLF salts in pure water (PDF)
REFERENCES
■
(1) Couillaud, B. M.; Espeau, P.; Mignet, N.; Corvis, Y. State of the
Art of Pharmaceutical Solid Forms: from Crystal Property Issues to
Nanocrystals Formulation. ChemMedChem 2019, 14, 8−23.
(2) Berry, D. J.; Steed, J. W. Pharmaceutical cocrystals, salts and
multicomponent systems; intermolecular interactions and property
based design. Adv. Drug Delivery Rev. 2017, 117, 3−24.
(3) Pindelska, E.; Sokal, A.; Kolodziejski, W. Pharmaceutical
cocrystals, salts and polymorphs: Advanced characterization techni-
ques. Adv. Drug Delivery Rev. 2017, 117, 111−146.
Accession Codes
CCDC 1960399−1960401 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
(4) Holliday, S. M.; Benfield, P. Venlafaxine. Drugs 1995, 49, 280−
294.
(5) Roy, S.; Aitipamula, S.; Nangia, A. Thermochemical Analysis of
Venlafaxine Hydrochloride Polymorphs 1−5. Cryst. Growth Des. 2005,
5, 2268−2276.
(6) Hajhashemi, V.; Minaiyan, M.; Banafshe, H.; Mesdaghinia, A.;
Abed, A. The anti-inflammatory effects of venlafaxine in the rat model
AUTHOR INFORMATION
Corresponding Authors
■
Yang Lu − Beijing Key Laboratory of Polymorphic Drugs,
Institute of Materia Medica, Chinese Academy of Medical
Sciences and Peking Union Medical College, Beijing 100050,
H
https://dx.doi.org/10.1021/acs.cgd.0c00007
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
pubs.acs.org/crystal
Article
of carrageenan-induced paw edema. Iran J. Basic. Med. Sci. 2015, 18,
654−658.
(7) Aiyer, R.; Barkin, R. L.; Bhatia, A. Treatment of Neuropathic
Pain with Venlafaxine: A Systematic Review. Pain Med. 2016, 18,
1999−2012.
(26) Liu, Z. J.; Liu, H.; Chen, X. W.; Lin, M.; Hu, Y.; Tuo, X.; Yuan,
Z. Y.; Sun, X. X. Efficient resolution of venlafaxine and mechanism
study via X-ray crystallography. Chirality 2018, 30, 268−274.
(27) Sheldrick, G. M. Crystal structure refinement with SHELXL.
Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71, 3−8.
(28) Li, P.; Chu, Y.; Wang, L.; Wenslow, R. M.; Yu, K.; Zhang, H.;
Deng, Z. Structure determination of the theophylline-nicotinamide
cocrystal: a combined powder XRD, 1D solid-state NMR, and
theoretical calculation study. CrystEngComm 2014, 16, 3141−3147.
(29) Bernardi, L. S.; Ferreira, F. F.; Cuffini, S. L.; Campos, C. E. M.;
Monti, G. A.; Kuminek, G.; Oliveira, P. R.; Cardoso, S. G. Solid-state
evaluation and polymorphic quantification of venlafaxine hydro-
chloride raw materials using the Rietveld method. Talanta 2013, 117,
189−195.
(8) Sumpton, J. E.; Moulin, D. E. Treatment of neuropathic pain
with venlafaxine. Ann. Pharmacother. 2001, 35, 529−532.
(9) Trouvin, A. P.; Perrot, S.; Lloret-Linares, C. Efficacy of
venlafaxine in neuropathic pain: a narrative review of optimized
treatment. Clin. Ther. 2017, 39, 1104−1122.
(10) Xu, H.; Chen, Z.; He, J.; Haimanot, S.; Li, X.; Dyck, L.; Li, X.
Synergetic effects of quetiapine and venlafaxine in preventing the
chonic restraint stress-induced decrease in cell proliferation and
BDNF expression in rat hippocampus. Hippocampus 2006, 16, 551−
559.
(11) Longo, F. M.; Yang, T.; Xie, Y.; Massa, S. M. Small Molecule
Approaches for Promoting Neurogenesis. Curr. Alzheimer Res. 2006, 3
(6), 5−10.
(12) Thompson, W. A.; Arnold, V. I.; Vijayan, M. M. Venlafaxine in
embryos stimulates neurogenesis and disrupts larval behaviorinzebra-
fish. Environ. Sci. Technol. 2017, 51, 12889−12897.
(30) Low, J. N.; Glidewell, C. A quasi-diamondoid hydrogen-bonded
framework in anhydrous sulfanilic acid. Acta Crystallogr., Sect. C: Cryst.
Struct. Commun. 2002, 58, o209−o211.
(31) Van Eupen, J. T. H.; Westheim, R.; Deij, M. A.; Meekes, H.;
Bennema, P.; Vlieg, E. The solubility behaviour and thermodynamic
relations of the three forms of Venlafaxine free base. Int. J. Pharm.
2009, 368 (1−2), 146−153.
(13) Tessler, L.; Goldberg, I. Venlafaxine, an antidepressant drug.
Acta Crystallogr., Sect. E: Struct. Rep. Online 2004, 60, o1868−o1869.
(14) van Eupen, J. T. H.; Elffrink, W. W. J.; Keltjens, R.; Bennema,
P.; de Gelder, R.; Smits, J. M. M.; van Eck, E. R. H.; Kentgens, A. P.
M.; Deij, M. A.; Meekes, H.; Vlieg, E. Polymorphism and Migratory
Chiral Resolution of the Free Base of Venlafaxine. A Remarkable
Topotactical Solid State Transition from a Racemate to a Racemic
Conglomerate. Cryst. Growth Des. 2008, 8, 71−79.
(15) Corvalan, C. H.; Vega, D. R. Venlafaxine besylate monohydrate.
Acta Crystallogr., Sect. E: Struct. Rep. Online 2013, 69, o1737−o1738.
(16) Vega, D.; Fernandez, D.; Echeverria, G. 1-[2-(1-Hydroxycy-
clohexyl)-2-(4-methoxyphenyl)ethyl]dimethyl-ammonium chloride
(venlafaxine hydrochloride). Acta Crystallogr., Sect. C: Cryst. Struct.
Commun. 2000, 56, 1009−1010.
(17) Saravanan, M.; Satyanarayana, B.; Reddy, P. P. An Improved
and Impurity-Free Large-Scale Synthesis of Venlafaxine Hydro-
chloride. Org. Process Res. Dev. 2011, 15, 1392−1395.
(18) Yardley, J. P.; Husbands, G. E. M.; Stack, G.; Butch, J.; Bicksler,
J.; Moyer, J. A.; Muth, E. A.; Andree, T.; Fletcher, H. 2-Phenyl-2-(1-
hydroxycycloalkyl)ethylamine derivatives: synthesis and antidepres-
sant activity. J. Med. Chem. 1990, 33, 2899−2905.
(19) Banerjee, R.; Bhatt, P. M.; Ravindra, N. V.; Desiraju, G. R.
Saccharin Salts of Active Pharmaceutical Ingredients, Their Crystal
Structures, and Increased Water Solubilities. Cryst. Growth Des. 2005,
5, 2299−2309.
(20) Spinelli, F.; Dichiarante, E.; Curzi, M.; Giaffreda, S. L.;
Chierotti, M. R.; Gobetto, R.; Rossi, F.; Chelazzi, L.; Braga, D.;
Grepioni, F. Molecular salts of the antidepressant venlafaxine: an
effective route to solubility properties modifications. Cryst. Growth
Des. 2017, 17, 4270−4279.
(21) Wang, Y.; Gong, N. B.; Yang, D. Z.; Du, G. H.; Lu, Y.
Preparation, characterization and stability of venlafaxine maleate 2018,
37, 1099−1102.
(22) Ali, L.; Ahmad, M.; Aamir, M. N.; Minhas, M. U.; Rasul, A.;
Yousuf, M.; Hussain, H.; Khan, J. A.; Sohail, M. Venlafaxine-loaded
sustained-release poly(hydroxyethyl methacrylate-co-itaconic acid)
hydrogel composites: their synthesis and in vitro/in vivo attributes.
Iran. Polym. J. 2019, 28, 251−258.
(23) Ling, Y.; Khara, L.; Alvarez, G.; McBeth, P. B. Persistent
Hypoglycemia in Venlafaxine Overdose. Arch. Clin. Med. Case Rep.
2019, 03, 94−99.
(24) Francino, M. C.; Deguigne, M. B.; Badin, J.; Turcant, A.;
Perrotin, D. Hypoglycaemia: A little known effect of Venlafaxine
overdose. Clin. Toxicol. 2012, 50, 215−217.
(25) Meertens, J. H. J. M.; Monteban-Kooistra, W. E.; Ligtenberg, J.
J. M.; Tulleken, J. E.; Zijlstra, J. G. Severe Hypoglycemia Following
Venlafaxine Intoxication. J. Clin. Psychopharmacol. 2007, 27, 414−415.
I
https://dx.doi.org/10.1021/acs.cgd.0c00007
Cryst. Growth Des. XXXX, XXX, XXX−XXX