International Journal of Pharmaceutics 581 (2020) 119232

Contents lists available at ScienceDirect

International Journal of Pharmaceutics

journal homepage: www.elsevier.com/locate/ijpharm

Characterization and therapeutic efficacy evaluation of glimepiride and L- T

arginine co-amorphous formulation prepared by supercritical antisolvent
process: Influence of molar ratio and preparation methods
Heejun Parka,1, Hye Jin Seob,1, Seung-hyeon Hongb, Eun-Sol Haa, Sibeum Leec, Jeong-Soo Kimd,
In-hwan Baeke, Min-Soo Kima, , Sung-Joo Hwangb,
⁎ ⁎

a
College of Pharmacy, Pusan National University, 63 Busandaehak-ro, Geumjeong-gu, Busan 46241, Republic of Korea
b
Yonsei Institute of Pharmaceutical Sciences & College of Pharmacy, Yonsei University, 85 Songdogwahak-ro, Yeonsu-gu, Incheon 21983, Republic of Korea
c
Advanced Drug Delivery Pharma, 25 Tapsil-ro 35 beon-gil, Giheung-gu, Yongin, Gyeonggi 17084, Republic of Korea
d
Dong-A ST Co. Ltd., 21 Geumhwa-ro 105 beon-gil, Giheung-gu, Yongin, Gyeonggi 17073, Republic of Korea
e
College of Pharmacy, Kyungsung University, 309, Suyeong-ro, Nam-gu, Busan 48434, Republic of Korea

ARTICLE INFO ABSTRACT

Keywords: The glimepiride/L-arginine (GA) binary systems were prepared at various molar ratios by using a supercritical
Glimepiride antisolvent (SAS) process. For comparison, the GA system was also prepared by physical mixing (PM), melt
L-arginine quenching (MQ), and solvent evaporation (SE) methods. Analyses by DSC and PXRD showed that only the GA
Co-amorphous binary mixture at 1:1 M ratio prepared by the SAS process was a pure co-amorphous mixture with an excellent
Supercritical antisolvent (SAS)
content uniformity. On the other hand, GA mixture prepared by PM and SE were not pure co-amorphous systems
Dissolution
Therapeutic efficacy
and contained crystalline eutectic mixture, and MQ method at 170 °C induced the decrease in drug content due
to decomposition of glimepiride. The positive deviation of experimentally measured glass transition temperature
(Tg) compared to predicted Tg by the Gordon Taylor equation suggests specific molecular interactions between
glimepiride and L-arginine in solid-state GA co-amorphous (GACA) mixture. The intermolecular interactions
between glimepiride and L-arginine in GACA system were characterized by FT-IR and solid-state NMR analyses.
Improved glimepiride dissolution rate of GACA formulation were confirmed using the solubility test, contact
angle measurement, and dissolution test. Furthermore, the evaluation of pharmacodynamic hypoglycemic effect
demonstrated that GACA prepared by the SAS process significantly improved the therapeutic efficacy of gli-
mepiride.

1. Introduction levels (BGL) (Langtry and Balfour, 1998; Staff, 2011). However, gli-
mepiride has poor aqueous solubility, classified as Biopharmaceutics
Glimepiride is a third-generation sulfonylurea drug effective in the Classification System (BCS) class II, leading to poor dissolution and
treatment of type 2 diabetes mellitus. It acts by stimulating pancreatic limiting drug absorption (Amidon et al., 1995; Massi-Benedetti, 2003).
β-cells to increase insulin production and thus lower blood glucose Many approaches have been explored to increase the solubility of

Abbreviations: ANOVA, analysis of variance; AUC, area under the curve; BCS, biopharmaceutics classification system; BGL, blood glucose level; CP/MAS-TOSS
SSNMR, cross-polarization magic angle spinning and total sideband suppression solid state nuclear magnetic resonance; DCM, dichloromethane; DM, diabetes
mellitus; DSC, differential scanning calorimetry; DW, deionized water; FID, flame ionization detector; FT-IR, fourier-transformed infrared; GA, glimepiride and L-
arginine; GACA, glimepiride and L-arginine co-amorphous; GC, gas chromatograph; HPLC-UV, high performance liquid chromatography-ultraviolet; ICH, interna-
tional council for harmonization of technical requirements for pharmaceuticals for human use; %ΔBGLmax, maximum percentage decrease in BGL; MQ, melt
quenching; MQGA, melt quenching processed glimepiride and L-arginine; PD, pharmacodynamic; PM, physical mixing; PMGA, physical mixing processed glimepiride
and L-arginine; PXRD, powder X-ray diffraction; RSD, relative standard deviation; SAS, supercritical anti solvent; SASGA, supercritical anti solvent processed
glimepiride and L-arginine; SD, Sprague-Dawley; SE, solvent evaporation; SEGA, solvent evaporation processed glimepiride and L-arginine; SEM-EDS, scanning
electron microscopy-energy dispersive X-ray spectrometry; SC-CO2, supercritical carbon dioxide; SNK, Student-Newman-Keuls; STZ, streptozotocin; Tg, glass tran-
sition temperature; tmax, time for maximum PD response; t1/2p, time at which half peak percentage decrease in BGL prevails; VMD, volume mean diameter
⁎
Corresponding authors.
E-mail addresses: minsookim@pusan.ac.kr (M.-S. Kim), sjh11@yonsei.ac.kr (S.-J. Hwang).
1
First author, these authors contribute equally to the study.

https://doi.org/10.1016/j.ijpharm.2020.119232
Received 6 January 2020; Received in revised form 27 February 2020; Accepted 14 March 2020
Available online 30 March 2020
0378-5173/ © 2020 Elsevier B.V. All rights reserved.
H. Park, et al. International Journal of Pharmaceutics 581 (2020) 119232

glimepiride, including the preparation of polymorphs (Bonfilio et al., main objective of the present study was to investigate the solid-state
2012; Endo et al., 2003), cyclodextrin inclusion complexes (Ammar characteristics and perform a biological evaluation of the GACA for-
et al., 2006a, b; Kim et al., 2020), co-crystal (Bian et al., 2019; Kamble mulation at a composition of 1:1 M ratio prepared by the SAS process.
et al., 2018), solid dispersions (Gill et al., 2014; Kiran et al., 2009; These results were additionally compared to those of GA binary mix-
Mehta et al., 2009; Reven et al., 2010), and microencapsulation (Ilić tures prepared by physical mixing (PM), solvent evaporation (SE), and
et al., 2009). However, these techniques have some problems such as melt quenching (MQ). The prepared solid GA binary formulations were
physico-chemical instability, high cost of additives and complexity of characterized by scanning electron microscopy-energy dispersive X-ray
the manufacturing process, hence the difficulty of commercialization. spectrometry (SEM-EDS), particle size analysis, differential scanning
To overcome those limitations, co-amorphous system has attracted in- calorimetry (DSC), powder X-ray diffraction (PXRD), Fourier-trans-
terest over the past two decades. The co-amorphous system is char- formed infrared (FT-IR) spectroscopy, and solid-state nuclear magnetic
acterized as a completely miscible single phase amorphous solid system resonance (SSNMR). Further evaluation included solubility test, contact
composed of binary- or multi-components in solid state (Dengale et al., angle measurement, dissolution test, and evaluation of therapeutic ef-
2016). The main aim of co-amorphous system would be not only the ficacy.
enhancement of solubility and dissolution rate of poorly water soluble
drug, but could be also advantageous for stability due to the higher 2. Materials and methods
glass transition temperature (Tg) than that of each amorphous material.
In co-amorphous system, the components interact on the molecular 2.1. Materials
level at a specific favorable molar ratio (Jensen et al., 2016). In addi-
tion, it has been reported that co-amorphous formulations prepared by Glimepiride was kindly supplied by Hwail Pharmaceutical, Co. Ltd.,
different methods could exhibit significant differences in their physical (Seoul, Korea). L-arginine was purchased from Merck (Darmstadt,
stability and dissolution behavior (Lim et al., 2016; Nair et al., 2019). Germany). Triammonium citrate and streptozotocin (STZ) were pur-
Thus, selecting an appropriate molar ratio and method for the pre- chased from Sigma Aldrich (St. Louis, MO, USA). Dichloromethane
paration of homogeneous powdered co-amorphous formulations is (DCM) was obtained from Junsei Chemical Co. Ltd. (Tokyo, Japan).
crucial to achieving the satisfactory performance of the final product. HPLC grade methanol, ethanol, and acetonitrile were obtained from
There are often used conventional methods for the preparation of Tedia (Fairfield, OH, USA). All other chemicals were of reagent grade.
co-amorphous formulations including three categorized methods: the
thermal method, the solvent evaporation method, and the milling 2.2. Preparation of glimepiride -arginine (GA) binary mixture
method. For these methods, it is very important to use conditions that
will produce unstable amorphous materials, such as fast evaporation 2.2.1. Physical mixing (PM)
rates, fast anti-solvent addition, and fast cooling rates to minimize the Glimepiride (5 g) and L-arginine (1.78 g) were weighed accurately
time for components to recrystallize separately (Chavan et al., 2016; Shi at a molar ratio of 1:1. They were mixed first using mortar and pestle,
et al., 2019). In recent years, a supercritical anti solvent (SAS) process and then, the final physical mixtures of GA (PMGA) were prepared by
using supercritical carbon dioxide (SC-CO2) as an antisolvent has been further mixing with zirconia beads (4 mm diameter) using a 25 ml
successfully used in pharmaceutical applications to improve the solu- geometric tumble mixer for 20 min. This time was determined based on
bility and dissolution rate of poorly water-soluble drugs (Abuzar et al., the processing time before excessive agglomeration of the particles
2018). The SC-CO2 is the most widely used supercritical fluid because of begins due to heat generation following shear stress during mixing. This
its low and easily accessible critical conditions (Tc = 31.1 °C, sample was sieved through a 100 mesh sieve (150 μm) and then stored
Pc = 7.38 MPa). At the critical point, only a single phase exists, which in a vacuum desiccator at a temperature not exceeding 25 °C.
has some properties typical of liquids, such as density, and some of
gases, such as viscosity, compressibility, and mass diffusion coefficient. 2.2.2. Melt quenching (MQ)
The highly dense SC-CO2 is permeable with a high solvent power and The prepared PMGA sample at 1:1 M ratio was placed uniformly in a
diffusion rate; thus it is usually miscible with organic solvents. These closed aluminum pan and heated on a digital hot plate, MSH 20D
properties can cause a high supersaturation and low surface tension, (WiseStir, Wertheim, Germany) to 170 °C for 3 min to ensure complete
hence decreasing the critical energy barrier for nucleation, which melting of both components. The temperature of 170 °C was chosen
should lead to faster nucleation and therefore improved production of based on the experimentally determined eutectic temperature of the
amorphous drug particles (Kim et al., 2008). Further advantages in- 1:1 M ratio eutectic mixture in our previous study (Park et al., 2019).
clude the ability to avoid or minimize the use of organic solvents, re- The molten liquid was cooled quickly by placing the aluminum pan
duction of residual solvents, non-toxicity, non-flammability, low price, over liquid nitrogen. Subsequently, the solidified sample was taken out
and tunability of its solvation power (Pasquali et al., 2008; Wei et al., and powdered using a mortar and pestle. The powdered sample was
2002). Thus, it is expected that these advantages of SAS process could sieved (150 μm) and then stored in a vacuum desiccator at a tem-
be applied successfully to the manufacture of co-amorphous. perature not exceeding 20 °C.
Recently, we reported that glimepiride forms a eutectic mixture
with L-arginine at a 1:1 M ratio and this eutectic mixture showed en- 2.2.3. Solvent evaporation (SE)
hanced aqueous solubility and dissolution rate (Park et al., 2019). In Glimepiride/L-arginine (GA) systems at a molar ratio of 1:1 were
contrast to co-amorphous, the two components in a eutectic mixture are prepared by the SE process (SEGA). Briefly, glimepiride (500 mg) and L-
completely miscible in the liquid state but miscible only to a very arginine (177.5 mg) were dissolved in 100 ml DCM and 10 ml of 90%
limited extent in the solid state (Abdelkader et al., 2014; Cherukuvada ethanol, respectively, and the two solutions were mixed. The drug so-
and Nangia, 2014). Interestingly, it had been reported that co-amor- lution was evaporated by rotary vacuum evaporation (Eyela, NE-series,
phous formulations can be prepared if eutectic mixtures are formed Tokyo, Japan) under reduced pressure at 60 °C for 6 h. This SEGA
from crystalline mixtures (Cruz-Angeles et al., 2019; Löbmann et al., sample was sieved (150 μm) and then kept in a vacuum desiccator at a
2011). In order to manufacture a co-amorphous product using eutectic temperature not exceeding 25 °C.
mixture, the miscibility of the eutectic mixture with molecular inter-
actions between components in the liquid state must be maintained 2.2.4. Supercritical antisolvent (SAS) process
through the solid state, even if the energy level is high. Based on those The experimental equipment used for the SAS process has been
previous reports, further research was carried out to develop a glime- previously described (Kim et al., 2008). The SAS processed GA (SASGA)
piride and L-arginine co-amorphous (GACA) mixture in this study. The systems at molar ratios of 1:0 (SASGA0), 1:0.6 (SASGA1), and 1:1

2
H. Park, et al. International Journal of Pharmaceutics 581 (2020) 119232

(SASGA2) were prepared by the SAS process. DCM was selected based 2.5. Powder X-ray diffractometry (PXRD)
on the improved effect of the solvent on crystal growth in a preliminary
study. Drug solution containing glimepiride and L-arginine was pre- PXRD analysis was performed using an X-ray diffractometer
pared as described in the SE method. Compressed CO2 was delivered (Rigaku, D/max-IIIC, Tokyo, Japan) with Cu-Kα radiation as the X-ray
into the particle formation vessel (Internal volume; 1908 cm3, I.D.; source at a voltage of 40 kV and a current of 45 mA. Samples were
9 cm, Length; 30 cm) through the outer capillary of the two-flow spray scanned from 5° to 40° (2θ) with a scanning speed of 3°/min and a step
nozzle (Sonimist® HSS-600-1, Misonix Inc., Farmingdale, NY, USA) size of 0.01°.
using a homemade plunger pump until the desired supercritical con-
dition was reached. Then, a clear drug solution was co-sprayed into a 2.6. Fourier-transformed infrared (FT-IR) spectroscopy
particle formation vessel by an high pressure liquid pump (NP-AX-5,
Nihon Seimitsu Kagaku Co. Ltd., Tokyo, Japan) with SC-CO2 through FT-IR analysis was performed on an FT-IR spectrometer (Thermo
the inner capillary of the two-flow spray nozzle. The flow rates of SC- Fisher, Nicolet 380 FT-IR, Waltham, MA, USA) equipped with the at-
CO2 and drug solution were 30 ml/min and 5 ml/min, respectively. tenuated total reflectance (ATR) accessory. Spectra were collected from
After complete delivery of the drug solution into the precipitation 650 to 4000 cm−1 with a resolution was 4 cm−1. Sixty-four scans were
vessel, SC-CO2 continued to flow into the vessel for an additional performed per sample.
60 min to remove the residual solvent from precipitated particles. The
temperature and pressure of the process was maintained at 60 °C and 2.7. Solid-state nuclear magnetic resonance (SSNMR)
120 MPa by a backpressure regulator (Tescom, model 26-1723-24-194,
Minneapolis, MN, USA) throughout the SAS process. Finally, the CO2 The 13C SSNMR experiment using a combination of cross-polariza-
supply was stopped, and the particle formation vessel was depressur- tion magic angle spinning and total sideband suppression (CP/MAS-
ized to atmospheric pressure. The precipitated particles were collected TOSS) was recorded on a Bruker AVANCE 400 WB (Bruker, Karlsruhe,
from the wall and bottom of the vessel, and kept in an airtight container Germany) at 400 MHz and ambient probe temperature. The MAS rate
at a temperature not exceeding 25 °C. was set at 8 kHz, with a contact time of 1 ms and an acquisition time of
0.01 s.

2.3. Differential scanning calorimetry (DSC) 2.8. Scanning electron microscopy with energy dispersive X-ray
spectrometry (SEM-EDS)
DSC measurements were carried out by DSC S-650 (Scinco Co. Ltd.,
Seoul, Korea) from 20 to 270 °C at a heating rate of 10 °C/min under SEM-EDS analysis of all GA samples was performed using the field
nitrogen purge over. Temperature was calibrated with indium (melting emission scanning electron microscope JSM-7000F (JEOL Ltd., Tokyo,
point = 156.6 °C) standard and enthalpy was calibrated with zinc Japan) equipped with an energy dispersive X-ray spectrometer INCA-
(melting point = 419.5 °C) standard using a 10 °C/min heating rate. 350EDS Microanalysis System (Oxford Instruments, Abingdon, UK) and
Around 2 mg of powder samples were accurately weighed and placed in a backscattered electron detector. Powder samples were mounted on a
aluminum pan, then sealed and used for DSC measurement. carbon fiber film attached to an aluminum sample stub, then sputtered
Thermograms were analyzed with InfinityPro software (version with a gold/palladium alloy in an argon atmosphere for 40 s at beam
4.2.194, Scinco Co. Ltd., Seoul, Korea) to determine peak melting current of 38–42 mA. SEM was operated at an accelerating voltage of
temperature at the apex of each endothermic melting peak and the 20 kV and sulfur (S) was used as an indicator of glimepiride for EDS.
midpoint for the glass transition temperature (Tg).
2.9. Particle size analysis

2.4. Prediction of theoretical Tg value by Gordon-Taylor equation The particle size of all GA samples was determined with a HELOS
laser diffraction analyzer equipped with a RODOS vibrating dry dis-
The Gordon-Taylor Eq. (1) can be used for the calculation of the perser (Sympatec Gmbh, Clausthal‐Zellerfelg, Germany). An air pres-
theoretical Tg for co-amorphous systems consisting of two amorphous sure of 0.3 or 0.5 MPa and vacuum of 5.3 kPa were used to produce
components (Jensen et al., 2014; Löbmann et al., 2013). uniform powder dispersion for each sample.

w1·Tg1 + K·w2·Tg2 2.10. Residual organic solvent concentration

Tg1,2 =
w1 + K·w2 (1)
Residual solvent was measured by a gas chromatograph (GC)
Tg1,2 is the glass transition temperature of the co-amorphous mixture, (Hewlett and Packard, 5890 SERIES II, Palo Alto, CA, USA) interfaced
w1, w2, Tg1, and Tg2 are the weight fractions and the glass transition with a flame ionization detector (FID) and coupled to a head space
temperatures of each of the two amorphous components. K is a constant sampler. Residual solvent was separated using a fused silica capillary
and described by following equation: column (Supelco, SPBTM-1, Bellefonte, PA, USA) with 30 m length,
0.53 mm internal diameter, and 0.5 μm film thickness. GC conditions
Tg1·
K= 1 involved an oven temperature of 40 °C maintained for 8 min. The in-
Tg2· 2 (2) jector was maintained at 180° C, and helium gas was used as the carrier
gas at 65 kPa (approximately 2 ml/min) with a split flow rate of 6 ml/
where ρ1 and ρ1 are the densities of each of two components. The min. Head space conditions were equilibration temperature for 30 min,
crystalline densities of glimepiride and L-arginine were 1.29 and with shaking of the vials at a low agitation speed. Head space samples
1.33 g/cm3, respectively. As glimepiride and L-arginine alone could not were prepared in 10-ml vials filled with 1 ml water in which the sam-
be transformed into amorphous forms by any of the methods used in ples were suspended.
this study, the previously reported Tg of amorphous glimepiride
(24.0 °C) (Cruz-Angeles et al., 2019) and L-arginine (18.4, 42.0 and 2.11. Drug content analysis
55.0 °C) (Izutsu et al., 2005; Jensen et al., 2014; Mattern et al., 1999)
were used for the prediction of theoretical Tg value of a co-amorphous Powdered sample containing about 10 mg of glimepiride was
system. weighed accurately and dissolved in methanol, then diluted to an

3
H. Park, et al. International Journal of Pharmaceutics 581 (2020) 119232

appropriate concentration for quantification. After filtering with a in a clean room at a temperature between 20 and 23 °C with 12 h light
syringe filter (Whatman, GMF of 0.45 μm, Clifton, NJ, USA), 20 μL of and dark cycles and a relative humidity of 50%. Rats were housed in
sample solution was injected into the HPLC system with a flow rate of stainless steel cages with food and water ad libitum. STZ was dissolved
1.0 ml/min to determine the drug concentration at a detection wave- in 0.05 M citrate buffer, pH 4.5, immediately before use. Five rats per
length of 228 nm. Briefly, an HPLC system (Alliance Waters 2690, group were administered STZ (60 mg/kg) by intraperitoneal injection
Milford, MA, USA) equipped with a Waters photodiode array detector (Qinna and Badwan, 2015). On the third day after administration of
and a Luna C18 column (4.6 × 250 mm, 5 μm, Phenomenex, Torrance, STZ, blood samples were obtained from the tail vein and BGLs were
CA, USA) were used to detect glimepiride. The mobile phase consisted measured using a super glucocardTM ΙΙ (GT-1640, Arkray, Kyoto,
of 0.01 M triammonium citrate (pH 6.95)/Acetonitrile/methanol Japan). Rats with a BGL higher than 300 mg/dl were chosen as models
(40:50:10 v/v). The temperature of auto-sampler and column were set of diabetes mellitus (DM). After administration of a single dose of
at 30 and 40 °C, respectively. 10 mg/kg of the drug or its equivalent amount of the GA formulation by
intragastric tubing, fasting BGLs were obtained at different time inter-
2.12. Aqueous solubility test vals. The hypoglycemic response was evaluated as the percent decrease
in BGL calculated as follows:
The solubility of samples was measured in deionized water (DW)
BGL (t = 0) BGL (t )
and at pH 6.8 in a phosphate buffer at 37 ± 0.5 °C. An excess amount % Decrease in BGL = × 100
BGL (t = 0) (3)
of sample (100 mg as a glimepiride) was added to 50 ml of DW and pH
6.8 phosphate buffer solution in test tubes sealed with a stopper, re- where BGL(t = 0) is BGL at 0 min, BGL(t) is BGL at t min. The used
spectively. The test tubes were mixed by vortexer for 1 min and soni- pharmacodynamic (PD) parameters taken into consideration based on
cated for 1 min. Then, test tubes were agitated at 100 rpm in shaking the previously reported papers were as follows (Ammar et al., 2006a):
water bath maintained at a constant temperature of 37 ± 0.5 °C for maximum percentage decrease in BGL (%ΔBGLmax), time for maximum
24 h. A suitable aliquot was withdrawn at predetermined time intervals, PD response (tmax), time at which half peak percentage decrease in BGL
then filtered with a PTFE syringe filter of 0.1 μm pore size (Whatman, prevails (t1/2p) and the area under the percentage decrease in BGL
Clifton, NJ, USA), and diluted with methanol. The concentration of versus time curve (AUC0–24h). These PD parameters were calculated
glimepiride was determined by the HPLC-UV method. using PKSolver software (Zhang et al., 2010).

2.13. Contact angle measurement

2.16. Statistical analysis
The contact angle was measured by the sessile drop technique with
The statistical significance of differences among samples was de-
a drop shape analysis system (EasyDrop, Krüss GmbH, Hamburg,
termined using independent T-test or a one-way analysis of variance
Germany) under controlled conditions of 25 ± 2 °C and 40 ± 5%
(ANOVA) followed by Student-Newman-Keuls (SNK) test using SPSS
relative humidity. All measurements for each sample were made in
12.0 software, (SPSS, Chicago, IL, USA).
triplicate. Two hundred milligrams of powdered sample were com-
pacted by a Perkin Elmer hydraulic press, for 1 min at 2000 psi. A
monochrome CCD camera equipped instrument recorded the experi- 3. Results and discussions
mental procedure with a shooting speed of 25/30 frames/s. The contact
angle between the disc and a single drop of D.W. (3 μL) was determined 3.1. Drug content, mixture uniformity and residual organic solvent
from captured images by drop shape analysis software (DSA1-SW21,
Krüss GmbH, Hamburg, Germany) after the droplet was applied to the Before solid-state characterization, it was first determined whether
disc with a gas chromatography needle and syringe (Hamilton 1750 decomposition due to the preparation method had affected the drug
TLLX with stop). content. As shown in Table 1, the MQGA content was less than 90%,
probably due to chemical degradation occurring during melting at high
2.14. Dissolution test temperature. It can be concluded that there may be limits on using a
method such as hot melt to produce a GACA since the MQ procedure
The dissolution tests were conducted in 900 ml phosphate buffer, caused decomposition of the glimepiride. Accordingly, the MQ method
pH 6.8, as a dissolution medium following the USP paddle method was considered to be inadequate for producing GACA and MQGA was
using the VK 7000 dissolution testing station and VK 750d heater/cir- excluded from further evaluation. The relative standard deviation
culator (Vankel, Cary, NC, USA). Powdered samples containing the (RSD), which can represent the drug content uniformity, are shown in
equivalent of 4 mg glimepiride were accurately weighed then placed in
the dissolution medium. A stirring speed of 50 rpm was used at Table 1
37 ± 0.5 °C. Five-milliliter aliquots from the dissolution medium were Particle size distribution and drug content data of GA binary mixtures prepared
withdrawn at predetermined time intervals and then filtered with PTFE by various methods.
syringe filter of 0.1 μm pore size (Whatman, Clifton, NJ, USA), and Samples VMDa,b ± SD (μm) Spanb,c ± SD Drug contentd (RSDe) (%)
diluted with methanol. The concentration of glimepiride was de-
termined by the HPLC-UV method. PMGA 18.7 ± 2.3 3.5 ± 0.5 98.5 (3.0)
MQGA – – 87.4 (4.2)
SEGA 56.5 ± 3.6 2.7 ± 0.1 99.6 (1.9)
2.15. Hypoglycemic effects after an oral administration of glimepiride/L- SASGA0 78.6 ± 2.7 2.7 ± 0.2 –
arginine systems in diabetic rats SASGA1 20.5 ± 0.3 1.6 ± 0.0 –
SASGA2 5.3 ± 0.2 1.5 ± 0.0 100.4 (0.9)
The hypoglycemic effects after the oral administration of either raw a
Volume mean diameter.
glimepiride or SASGA2 to diabetic rats were studied. Male Sprague- b
Mean value, n = 4.
Dawley (SD) rats (aged 7 weeks, weighing 210–250 g) were purchased c
SPAN = (d90-d10)/d50, where d10, d50 and d90 are the diameter sizes and
from Jung-Ang Lab. Animal, Inc. (Seoul, Korea). The animal study the given percentage value is the percentage of particles smaller than that size.
protocol was in compliance with the institutional guidelines for the care d
Mean value, n = 10.
and use of laboratory animals and was approved by the ethics com- e
Relative standard deviation (RSD) % = (σ/C)·100, where σ is the standard
mittee of Kyungsung University (No. 19-008A). Rats were maintained deviation and C is the mean concentration determined experimentally.

4
H. Park, et al. International Journal of Pharmaceutics 581 (2020) 119232

Table 1. RSDs for all samples are less than 6%, which is generally ac- morphology and size with increasing amounts of L-arginine. The SASGA
ceptable (Sanganwar et al., 2010). Nevertheless, there were significant without L-arginine (SASGA0) had a large needle-like crystal. In contrast
differences in RSD between the GA mixture preparation methods. The to SASGA0, SASGA1 with a mole ratio of G:A = 1:0.6 had short needle-
RSD decreased in the following order: PMGA > SEGA > SASGA2 like crystals. The SASGA2 with a mole ratio of G:A = 1:1 consisted of
with smaller RSD meaning more uniform distribution of drug in the uniform micron-sized particles with VMD of 5.3 μm and exhibited
mixture. This result showed that the glimepiride is most uniformly regular particle morphology with a narrow PSD span value of 1.5 (Fig. 1
distributed in the SASGA2. The uniformity of SEGA is less than that of and Table 1). Smaller span values represent narrow particle size dis-
SASGA2 but better than that of PMGA. These results will be further tributions. Therefore, it can be suggested that the addition of L-arginine
discussed in consideration of the SEM-EDS result. Residual DCM levels can inhibit the growth of glimepiride crystals during precipitation in
in the GA binary mixture samples prepared by SE and SAS methods the SAS process leading to the production of more uniform and smaller
were determined by GC analysis within 3 h after preparation them. The particles, as well as alter the particle morphology (Palazzo et al., 2009).
residual concentrations of DCM in SASGA2 and SEGA were 38 ppm and The mechanism explaining this inhibitory effect on glimepiride crystal
286 ppm, respectively, which were lower than the limit of 600 ppm in growth by L-arginine is likely the molecular interactions between the
articles from the ICH guideline (Guideline, 2005). Although both two substances. In our previous paper, arginine has been shown to
samples meet the residual solvent criteria, it is obvious that the SAS prevent the precipitation of supersaturated glimepiride through mole-
process is more effective at decreasing residual solvent considering that cular interaction in solution (Park et al., 2019). It is assumed that si-
the processing time is much shorter for SAS (3 h) than SE (6 h). milar mechanism may have affected the co-precipitation of GA mixture
during the SAS process (Ojarinta et al., 2017). A detailed explanation
3.2. Morphology and particle size distribution about this molecular interaction will be discussed in a later section
containing solid-state spectroscopic results.
The SEM images of unprocessed raw glimepiride and arginine, EDS images also presented in Fig. 1 show the EDS maps of sulfur
PMGA, SEGA, and SASGA samples are shown in Fig. 1. The volume corresponding to each SEM image of raw glimepiride, L-arginine,
mean diameter (VMD) and span value of those samples are summarized PMGA, SEGA, and SASGA2. The SEM-EDS image provides information
in Table 1. The raw glimepiride is a 20-µm particle with an arrowhead- on the spatial distribution of elements in the crystal. Since sulfur is a
like shape. L-arginine is a prismatic crystal having a columnar char- unique element to the chemical structure of glimepiride, the EDS maps
acteristic with over100 μm size. The SEM image of PMGA consisted of for this element should indicate the distribution of the drug in the GA
not only depositions of glimepiride particles on the surface of L-arginine binary systems. The raw glimepiride has strong signals of sulfur, but L-
crystals but also glimepiride particles distinguished from it. This in- arginine showed no detectable level of sulfur. The EDS maps for GA
directly shows the heterogeneity inherent to the physical mixing binary mixtures revealed that there were significant differences in the
method used in this study. The SEGA was an irregular and coarse crystal uniformity of the glimepiride distribution between PMGA and SASGA2
with a mean particle size of 56.5 μm. samples. In SEM-EDS image of PMGA, it was shown that glimepiride- or
Interestingly, SASGA samples showed remarkable changes in L-arginine-rich areas exist separately in PMGA collected from several

Fig. 1. SEM-EDS images of raw glimepiride, raw L-arginine, SEGA, PMGA, and SASGA samples.

5
H. Park, et al. International Journal of Pharmaceutics 581 (2020) 119232

Fig. 2. DSC thermograms of raw glimepiride, L-arginine, SEGA, PMGA, and SASGA samples: (a) glimepiride, (b) L-arginine, (c) SASGA0, (d) SASGA1, (e) SASGA2, (f)
PMGA, and (g) SEGA (× number means enlargement ratio).

parts of mixing cylinder, and this un-uniformity particularly pro- planes exposed to the X-ray source would have been altered, producing
nounced in samples taken at the bottom corner. This finding confirmed the change in the relative intensities of the diffraction peaks. This result
that glimepiride is less uniformly distributed in PMGA than in SASGA2 indicates that the SAS process condition used in this study only affected
as shown by RSD of drug content. In contrast, EDS maps of SEGA and particle morphology, and did not induce changes in the crystallinity
SASGA2 showed that the surface of both particles has a homogeneous and polymorphism of pure glimepiride (Brittain, 2016; Park et al.,
distribution of sulfur, which means that glimepiride is uniformly dis- 2007).
tributed in those samples. SEM-EDS analysis did not show a significant In contrast, the thermal behavior of the SAS processed GA binary
difference in uniformity between SEGA and SASGA2. systems with L-arginine significantly changed depending on the amount
of added L-arginine. Interestingly, as shown in Fig. 2, the melting
3.3. DSC and PXRD temperature of glimepiride in SASGA1 decreased and glass transition
temperature (Tg) was appeared compared to raw glimepiride. The
DSC thermograms and PXRD patterns obtained for raw materials PXRD pattern of SASGA1 shows the same peak location as raw glime-
and GA binary mixtures prepared by various methods are shown in piride, but the characteristic peaks of L-arginine disappeared showing
Figs. 2 and 3, respectively. The L-arginine is characterized by a first that L-arginine is distributed in SASGA1 as amorphous state. In addi-
small endothermic peak at 101.8 °C, attributed to a dehydration effect, tion, the intensity of each peak decreased and the background halo was
and second and third melting endothermic peak at 220.7 and 241.7 °C, observed. These results could indicate that only the portion of glime-
followed by decomposition (Mura et al., 2005; Mura et al., 2003). The piride that interacts with the relatively small molar fraction of L-argi-
unprocessed raw glimepiride shows a melting peak at 213.9 °C with nine can form a co-amorphous system and the excess glimepiride is
fusion enthalpy of 176.4 J/g (Endo et al., 2003). SAS processed gli- crystallized to the crystalline state (Jensen et al., 2016; Löbmann et al.,
mepiride without L-arginine (SASGA0) shows a single endothermic 2011). On the contrary, the all of endothermic peaks due to melting
peak at the melting temperature 212.3 °C with fusion enthalpy of disappeared and the only Tg was observed at 59.1 °C in the DSC ther-
188.8 J/g, similar to that of raw glimepiride. In addition, the PXRD mogram of SASGA2 with 1:1 M ratio of glimepiride and L-arginine. In
pattern of SASGA0 is almost identical to that of raw glimepiride. Al- addition, the PXRD pattern of SASGA2 suggests that it is in an amor-
though the PXRD peak locations, 2θ degree, are identical, the relative phous state. In our previous study, it was shown that the eutectic
integrated intensities of some peaks are significantly different. This composition of the GA binary system is a 1:1 M ratio (Park et al., 2019).
difference can be explained by preferred orientation, in which the Unlike in our previous study, in the present study, it was shown that a
distribution of crystal orientation is nonrandom, and the crystal may co-amorphous mixture was uniquely formed by the SAS process at
tend to grow to a greater or lesser degree in some specific orientation, 1:1 M ratio of glimepiride and L-arginine. The predicted theoretical Tg
hence modifying the crystal habit (Haleblian, 1975; Marshall and York, of GA co-amorphous (GACA) mixture from the Gordon Taylor equation
1989). The PXRD pattern of SASGA0 shows that the orientation of a was in the range of 22.6 to 31.8 °C when the Tg values of pure amor-
particular molecular arrangement (corresponding to the peak at 18.1° phous glimepiride and L-arginine reported in previous reports were
2θ) is particularly preferred in the needle-like crystal, when compared used. These predicted theoretical Tg values are lower than the experi-
to raw glimepiride with the highest intensity at 16.8° 2θ. These changes mentally obtained Tg of 59.1 °C. This positive deviation compared to
in orientation of some particular molecular arrangements may be re- predicted Tg suggests the presence of specific molecular interactions
lated to a change in the crystal habit, because the abundance of the between glimepiride and L-arginine in solid state, which are not

6
H. Park, et al. International Journal of Pharmaceutics 581 (2020) 119232

Fig. 3. PXRD patterns of raw glimepiride, L-arginine, SEGA, PMGA, and SASGA samples: (a) glimepiride, (b) SASGA0, (c) SASGA1, (d) SASGA2, (e) SEGA, (f) PMGA,
and (g) L-arginine.

considered in the Gordon Taylor equation (Huang et al., 2017). to induce fast and simultaneous supersaturation of all complexed gli-
Therefore, it can be proposed that the sum and strength of hydrogen mepiride and L-arginine with molecular interactions intact in solution.
bonding between glimepiride and L-arginine in the co-amorphous As a result, there could be differences in supersaturations of each solute
system is higher than the individual amorphous components. In addi- and their complex form during the SE process, so that the co-amorphous
tion, the Tg of SASGA2 was found to be significantly higher than those mixture could be precipitated first, then later crystalline eutectic mix-
of the individual amorphous components, glimepiride and L-arginine, ture could be precipitated separately. Furthermore, recrystallization
previously reported in the literature. This could be an exceptional ad- from the co-amorphous mixture portion might also be induced by in-
vantage because the Tg is generally regarded as a main factor affecting effective evaporation of the solvent over the drying time (Chavan et al.,
physical stability, with a high Tg and low storage temperature being 2016; Kasten et al., 2019).
more amenable to the physical stability of an amorphous system The above results show that PM and SE are inefficient methods for
(Jensen et al., 2014). forming a pure GACA mixture. Therefore, it can be concluded that top-
There was a marked difference between DSC thermograms and down method, particularly SAS process, can be more advantageous for
PXRD patterns between the preparation methods. DSC thermograms of the manufacture of GACA formulations, which solidify the GACA mix-
the PMGA sample showed a broad single peak at 150.4 °C corre- ture from the solution state while maintaining the molecular interac-
sponding to the melting of eutectic mixture. The PXRD pattern of PMGA tions between the two components. This advantage of SAS beyond SE
showed characteristic diffraction peaks of both glimepiride and L-ar- mechanism in co-amorphous formation should be due to efficient su-
ginine but reduced peak intensity and a small background halo were persaturation with fast evaporation and extraction of solvents (Pasquali
also observed, which can mean PMGA includes a minor amount of et al., 2008).
amorphous mixture despite Tg was not observed in the DSC thermo-
gram of PMGA (Figueirêdo et al., 2017). The decreased melting point of 3.4. Molecular interaction analyzed by solid-state spectroscopic methods
glimepiride in PMGA compared to raw glimepiride can be explained by
our previous study suggesting that glimepiride and arginine form a FT-IR spectroscopy was used to study of the molecular interaction
eutectic mixture by physical mixing using a mortar and pestle. Thus, between glimepiride and L-arginine in the GACA system. To help un-
DSC and PXRD results for PMGA shows that PMGA contains a mostly derstand the spectroscopic results, the chemical structures of glime-
eutectic mixture and may contain small amounts of co-amorphous piride and L-arginine with atom numbering are shown in the Fig. 4. The
mixture; definitely, PMGA is clearly not a pure co-amorphous system FT-IR spectra and assigning data are presented in Fig. 5 and Table A.1,
(Lim et al., 2016). respectively. The FT-IR spectrum of raw glimepiride is characterized by
In the GA binary sample prepared using solvent evaporation the N-H stretching vibration in the urea moiety at 3368.0, 3287.0 and
(SEGA), Tg was observed to be 54.3 °C, and broad small endothermic 3134.0 cm−1 and the C=O stretching vibration at 1701.8 and
peaks was also appeared at 155.7 °C on the DSC thermogram of SEGA. 1655.4 cm−1 (Kiran et al., 2009; Özdemi and Gökce, 2018). The FT-IR
This broad endotherm is probably due to the melting of the crystalline spectrum of SASGA0 was almost the same as raw glimepiride, in-
eutectic mixture. A noticeable background halo was observed in the dicating the same polymorphic form as revealed by DSC and PXRD
PXRD pattern of the SEGA sample, but several peaks with low intensity analyses. In contrast, the spectra of SASGA1 and SASGA2 have sig-
arising from crystalline eutectic mixture were observed. These results nificant differences in FT-IR spectra compared to raw glimepiride. In
suggest that both amorphous and crystalline eutectic mixtures are general, the carbonyl stretch of a canonical arginine can be found in the
present in SEGA sample. The amorphous eutectic mixture here is region between 1700 and 1725 cm−1, and the stretch of the respective
commonly a co-amorphous mixture. This incomplete GACA formation ionized carboxyl group can be found in the region between 1500 and
may be due to the solvent removal rate of the SE process being too slow 1650 cm−1 (Kumar and Rai, 2010). In the FT-IR spectrum of SASGA2,

7
H. Park, et al. International Journal of Pharmaceutics 581 (2020) 119232

Fig. 4. Chemical structures of glimepiride and L-arginine.

there were remarkable shifts in N–H stretching, C=O stretching and proportion of L-arginine increased, hence the spectral change is most
N–H bending vibration bands assigned to both of glimepiride and ar- significant in SASGA2. Thus, such above FT-IR band shifts suggest the
ginine. In addition, the stretching band due to the ionized carboxyl intermolecular interaction via hydrogen bonds, N–H—O=C, between
group in L-arginine disappeared. These spectral changes can indicate amine and carboxyl group of L-arginine and sulfonyl urea structure of
that arginine loses the ionic interaction between arginine molecules, glimepiride and/or between guanidium structure of L-arginine and
and instead forms a molecular interaction with glimepiride in the co- carboxamide structure of glimepiride. Similar trends in hydrogen
amorphous system (Ammar et al., 2006b). In glimepiride, N-H group bonding of glimepiride have been reported in previous literatures (Bian
and carbonyl oxygen in sulfonyl urea and carboxamide structure can act et al., 2019; Cruz-Angeles et al., 2019; Reven et al., 2010). In addition,
as hydrogen donor and acceptor, respectively (Bian et al., 2019; Endo there were significant changes in aromatic C-H stretching bands of
et al., 2003; Reven et al., 2010). In addition, N-H of guanidium and glimepiride and propyl C-H stretching bands of L-arginine for SASGA2,
amine groups and carbonyl oxygen of carboxyl group in L-arginine can which can indicate the hydrophobic interaction (Das et al., 2007; Li
act as hydrogen donor and acceptor, respectively (Lenz et al., 2017; et al., 2010). These intermolecular interactions can affect the crystal
Lobmann et al., 2013). Furthermore, above spectral changes demon- packing and arrangement of glimepiride, meaning the nucleation and
strated that the shifts of bands become more noticeable as the crystal growth during the crystallization in SAS process could be

Fig. 5. FT-IR spectrum of raw glimepiride, L-arginine, SEGA, PMGA, and SASGA samples: (a) glimepiride, (b) L-arginine, (c) SASGA0, (d) SASGA1, (e) SASGA2, (f)
PMGA, and (g) SEGA.

8
H. Park, et al. International Journal of Pharmaceutics 581 (2020) 119232

13
Fig. 6. C SSNMR spectra of (a) raw glimepiride, (b) L-arginine, and (c) SASGA2.

Table 2
The solubility, contact angle and dissolved drug concentration at predetermined times of raw glimepiride and GA samples (n = 3).
Sample Solubilitya (μg/ml) Contact angle (°)c Dissolved drug concentration (μg/ml)

DW (pH)b pH 6.8 1s 3s 6s 5 min 30 min 60 min

Raw glimepiride 0.6 ± 0.1 (5.4 ± 0.1) 1.8 ± 0.1 (6.8 ± 0.0) 91.8 ± 1.2 88.3 ± 2.7 86.2 ± 1.5 0.11 ± 0.01 0.18 ± 0.02 0.23 ± 0.02
SASGA0 0.7 ± 0.1 (5.4 ± 0.1) 1.7 ± 0.2 (6.8 ± 0.0) 92.1 ± 2.6 89.2 ± 3.1 85.9 ± 2.2 0.10 ± 0.01 0.15 ± 0.01 0.23 ± 0.01
SASGA1 32.6 ± 1.1 (8.6 ± 0.1) 21.2 ± 0.6 (6.8 ± 0.1) 38.2 ± 3.2 22.0 ± 1.9 10.4 ± 2.7 1.11 ± 0.01 1.78 ± 0.04 2.03 ± 0.04
SASGA2 84.2 ± 2.1 (9.3 ± 0.1) 43.7 ± 1.8 (6.9 ± 0.1) 23.3 ± 1.9 13.3 ± 1.2 8.9 ± 1.7 3.72 ± 0.20 4.37 ± 0.08 4.42 ± 0.09
PMGA 38.7 ± 1.3 (9.3 ± 0.1) 20.4 ± 1.0 (6.9 ± 0.1) 37.2 ± 2.1 21.9 ± 1.9 10.2 ± 1.5 0.80 ± 0.01 1.22 ± 0.02 1.47 ± 0.02
SEGA 51.3 ± 1.9 (9.3 ± 0.1) 31.2 ± 1.4 (6.9 ± 0.1) 29.8 ± 1.2 18.1 ± 1.4 9.3 ± 1.8 2.45 ± 0.08 3.35 ± 0.13 3.67 ± 0.18

a
Measured maximum solubility of glimepiride in aqueous medium at 37 ± 0.5 °C within 24 h.
b
Measured pH of medium after 24 h of solubility test.

Fig. 7. Powder dissolution profiles of raw glimepiride, L-arginine, SEGA, PMGA, and SASGA samples in pH 6.8 phosphate buffer solution at 37 °C.

9
H. Park, et al. International Journal of Pharmaceutics 581 (2020) 119232

Reven et al., 2010). Obvious changes in the 13C SSNMR spectra were
observed for SASGA2 consisting of the GACA system compared to pure
glimepiride and L-arginine. Magnified regions of the 13C NMR spectra
show shifts in the carbon signals of the glimepiride carboxamide car-
bonyl group as well as signals from aromatic rings. Moreover, carbon
peaks related to the L-arginine molecule such as the guanidium group
and the 3-carbon aliphatic straight chain were observed to shift. These
results implicate the glimepiride carboxamide group and aromatic ring
as that which interacts with guanidium group and propyl chain of L-
arginine via hydrogen bond and hydrophobic interactions, respectively
(Das et al., 2007; Li et al., 2010). This interaction observed by 13C
SSNMR spectroscopy is well coincident with FT-IR results. On the
contrary, changes in the chemical shift signal due to pyrroline, in gli-
mepiride is relatively small, suggesting that this group may participate
less in the molecular interaction with the L-arginine.

3.5. Aqueous solubility

Fig. 8. Mean percentage decrease in blood glucose level (%ΔBGL) of diabetic The solubility data of raw glimepiride and all GA mixture samples in
rats after administration of raw glimepiride and SASGA2. DW and pH 6.8 phosphate buffer solution (PBS) were summarized in
Table 2. The maximum solubility in DW increased in the following
Table 3 sequence: raw glimepiride = SASGA0 < SASGA1 < PMGA < SEGA <
Pharmacodynamic (PD) parameters for raw glimepiride and SASGA2 SASGA2 (ANOVA, P < 0.05, ranked by the SNK test). The solubility of
(mean ± S.D., n = 5). glimepiride increased with the addition of L-arginine in SASGA sam-
Samples ΔBGLmax (%) tmax (h) t1/2p (h) AUC0-24h
ples. This shows the ability of L-arginine to solubilize glimepiride, likely
in accordance with the strong basic character of L-arginine and the
Raw glimepiride 29.5 ± 2.2 2.0 ± 0.0 8.4 ± 0.4 346.9 ± 28.3 acidic nature of glimepiride. It was previously reported that the solu-
SASGA2 43.7 ± 1.4* 1.0 ± 0.0* 7.3 ± 0.6 575.6 ± 52.0* bility of glimepiride depended on the pH of the medium (Ning et al.,
2011). This reference shows that the solubility of glimepiride was in-
* Indicates significant (P < 0.05) compared to raw glimepiride.
creased drastically in basic phosphate buffer solution as the pH in-
creased. These facts indicate that the increased pH of the medium due
to the strong basic character of L-arginine could lead to the solubility
influenced. Therefore, it can be supposed that L-arginine has the po- improvement for glimepiride. In order to exclude the effect of in-
tential to interact with some functional groups of glimepiride, and this creasing pH, the solubility test was also performed in a PBS with a pH
feature might contribute to the inhibition of glimepiride crystal growth 6.8. As shown in Table 2, there was no significant increase in pH of
during the recrystallization phase of the SAS process. Consequently, phosphate buffer after solubility test. Nevertheless, the solubility of
although the pure amorphous form of each component, glimepiride and SASGA2 was remarkably increased compared to raw glimepiride,
L-arginine, could not be obtained in this study and the SASGA2 spec- SASGA0 and SASGA1 (P < 0.05). This result is probably mainly due to
trum could not be compared with their spectrum, it is obvious that the the formation of co-amorphous for SASGA2. Amorphous systems do not
increase in spectral changes as the mole fraction of L-arginine increase require an input of energy for the breakage of the crystal lattice and
was due to the formation of co-amorphous through intermolecular in- they have a solubility advantage compared to the solid crystalline
teraction with glimepiride and arginine. SEGA also showed significant forms, hence lower crystallinity can result in a higher dissolution rate
change in FT-IR spectra compared to raw glimepiride and L-arginine, (Leuner and Dressman, 2000; Liu et al., 2018). In addition, the glime-
but the difference was smaller than that of SASGA2. In particular, the piride, in the presence of the basic amino acid that acts as a counter-ion,
N-H stretching band of crystalline glimepiride sulfonyl urea at 3368.0 may give rise to a hydrophilic structure as a result of the potential
and the C=O stretching band of crystalline L-arginine at 1702.5, which ability of the L-arginine to simultaneously interact with glimepiride via
are attributed to sulfonyl urea, confirms once again that the SEGA is not hydrogen bonding (Mura et al., 2005; Vasanthavada et al., 2005). It has
a pure co-amorphous mixture, and crystalline GA eutectic mixture co- been reported that the arginine has a salting-in effect, increasing the
exists in SEGA as revealed by DSC and PXRD. For PMGA, the changes in solubility of a poorly soluble drug (Arakawa et al., 2008; Jensen et al.,
FT-IR bands arising from the sulfonylurea structure are negligible 2014; Serajuddin, 2007). Furthermore, the hydrogen atom of the sul-
compared to raw glimepiride and arginine. This finding from FT-IR fonylurea group reacts with weak bases to form water soluble salts in
analyses reveals that recrystallization methods such as SE and SAS aqueous solutions. In our previous paper, this molecular interaction in
process brought about stronger molecular interactions in solid state solution has been described using various spectroscopic methods (Park
between glimepiride and L-arginine than PM, and the SAS process is et al., 2019). These possible mechanisms could provide reasonable ex-
more useful for producing pure co-amorphous mixture, as was also planation for the improved solubility of glimepiride with L-arginine.
shown by DSC and PXRD analyses. This may be the result of higher Therefore, it is reasonable to hypothesize a specific role for L-arginine
supersaturation and faster nucleation of co-amorphous particles, fol- in the solubility enhancement of glimepiride considering specific mo-
lowing effective solvent removal due to the fast mass transfer efficiency lecular interaction between glimepiride and L-arginine.
of supercritical CO2 during the SAS process (Martin and Cocero, 2008). The difference in solubility between raw glimepiride and SASGA0
In order to further evaluate molecular interactions in detail and gain was not significant, because there were no differences of polymorphism
better understanding on GACA system, SSNMR spectroscopy was per- or crystallinity as described by DSC, PXRD, and FT-IR analyses. The
formed additionally (Park et al., 2010), and results are shown in Fig. 6 solubility of SASGA2 was significantly higher than that of SEGA and
and Table A.2 (Herbert-Pucheta et al., 2011; Özdemi and Gökce, 2018; PMGA (P < 0.05). This result may be due to the formation of pure

10
H. Park, et al. International Journal of Pharmaceutics 581 (2020) 119232

GACA system by SAS process, hence the higher solubilizing ability of L- glimepiride (Ammar et al., 2006a). The better therapeutic efficacy of
arginine caused by the stronger molecular interaction between glime- SASGA2 could be primarily attributed to the improved solubility and
piride and L-arginine compared to incomplete GACA system prepared dissolution associated with the formation of the GACA system through
by SE and PM. intermolecular interactions between glimepiride and L-arginine arising
from the SAS process (Blagden et al., 2007; Singh et al., 2018).
3.6. Wettability determined by contact angle measurement
4. Conclusions
The wettability of a drug can be characterized by contact angle
measurement. It is well known that the better wettability and dis- In this study, solid state characterization showed that the SAS pro-
persibility of a solid drug, the better the chances of avoiding agglom- cess can be successfully applied to prepare a GACA formulation. In
eration of drug particles upon contact with liquid (Buckton et al., 1991; contrast, other comparable formulation methods such as PM and SE
Luu et al., 2019). The contact angle data according to time are pre- failed to produce pure GACA and MQ method induced chemical de-
sented in Table 2. The ANOVA test found significant differences of the composition of glimepiride. It was confirmed that only the SASGA2
contact angles among the GA binary mixtures (P < 0.05). The contact sample at 1:1 M ratio of glimepiride and L-arginine was a pure GACA
angle at 1 s decreased in following sequence: raw glime- mixture. The positive deviation of the experimentally measured Tg
piride = SASGA0 > SASGA1 = PMGA > SEGA = SASGA2 (ranked compared to predicted Tg by the Gordon Taylor equation suggests
by the SNK test). This result indicates that the wettability of glimepiride specific molecular interactions between glimepiride and L-arginine in
was improved by mixing with L-arginine, among the mixture prepara- solid-state GACA mixtures prepared by the SAS process. The detailed
tion methods, the SAS process was the best to get a GACA with the intermolecular interaction between glimepiride and L-arginine in the
highest wettability. The lowering of contact angle suggests changes in GA binary system was characterized by FT-IR and SSNMR analyses. In
surface properties of glimepiride toward a more hydrophilic character addition, the GACA formulation prepared by the SAS process has many
(Hyun et al., 2019; Lee et al., 2018). Therefore, it can be expected that advantageous pharmaceutical properties such as decreased particle
the preparation of a GA system using the SAS process may give rise to a size, uniform drug distribution, improvements in solubility, dissolution
more hydrophilic property of the newly created glimepiride crystal rate, and hypoglycemic effect of glimepiride. In conclusion, it was
surface compared to the SE and the PM methods. Furthermore, it was shown that the better therapeutic efficacy of SASGA2 could be pri-
confirmed that the improvement of wettability is closely related to marily attributed to the improved solubility and dissolution associated
enhancement of solubility in the experimentally used medium when with the formation of the GACA system through intermolecular inter-
comparing the tendency of increase in wettability and solubility actions between glimepiride and L-arginine arising from the SAS pro-
(Ishikawa et al., 2000). cess. Furthermore, it was confirmed that the SAS process is most sui-
table for manufacturing pure GACA. For further studies, research on
3.7. In vitro dissolution physicochemical stability of co-amorphous formulations in relation
with intermolecular interactions are still required.
The dissolution profiles of raw glimepiride, PMGA, SEGA, and
SASGA systems in pH 6.8 phosphate buffer solution are shown in Fig. 7 CRediT authorship contribution statement
and summarized in Table 2. The ANOVA test found significant differ-
ences in dissolution profile among the samples (P < 0.001). The dis- Heejun Park: Conceptualization, Methodology, Investigation,
solved drug concentration at each time point increased in following Writing - original draft. Hye Jin Seo: Conceptualization, Methodology,
order: raw glimepiride = SASGA0 < PMGA < SASGA1 < SEGA < Investigation, Writing - review & editing. Seung-hyeon Hong:
SASGA2 (ranked by the SNK test). It was obvious that the dissolution Methodology, Formal analysis. Eun-Sol Ha: Methodology, Formal
enhancement effect was increased as the content of L-arginine increased analysis. Sibeum Lee: Conceptualization, Methodology, Writing - re-
from 0 to 0.5 mol fraction for SASGA samples. In addition, the SASGA2 view & editing. Jeong-Soo Kim: Conceptualization, Writing - review &
had the highest dissolution rate among GA binary mixture prepared at a editing. In-hwan Baek: Methodology, Formal analysis. Min-Soo Kim:
molar ratio of 1:1 by various methods. This result should be due to the Conceptualization, Investigation, Writing - review & editing,
improved solubility and wettability via formation of pure GACA by SAS Supervision, Funding acquisition. Sung-Joo Hwang:
process (Kim et al., 2010). Thus, it was shown that co-amorphous for- Conceptualization, Investigation, Writing - review & editing,
mation with L-arginine is very effective in improving dissolution rate of Supervision, Funding acquisition.
glimepiride (Ning et al., 2011).

Declaration of Competing Interest

3.8. Evaluation of pharmacodynamics (PD) therapeutic efficacy

Fig. 8 shows the mean percentage decrease in BGL of diabetic rats The authors declare that they have no known competing financial
after administration (% ΔBGL) of raw glimepiride or SASGA2, and their interests or personal relationships that could have appeared to influ-
PD parameters are presented in Table 3. The differences in AUC0-24h, ence the work reported in this paper.
maximum percentage decrease in BGL (%ΔBGLmax), and time for
maximum percentage decrease in BGL (tmax) between raw glimepiride Acknowledgements
and SASGA2 sample are significant (p < 0.05). With respect to these
PD parameters presented in Table 3, it was clear that the AUC0-24h and This study was supported by a grant from Basic Science Research
%ΔBGLmax of SASGA2 have higher values than those of raw glime- Program (NRF-2019R1F1A1056350), and Key Research Institutes in
piride. Furthermore, the tmax of SASGA2 is lower than that of raw gli- Universities (NRF-2018R1A6A1A03023718) through the National
mepiride. These results indicate that the SASGA2 system has better Research Foundation of Korea (NRF) funded by the Ministry of Science
therapeutic efficacy and faster onset of action compared to raw and ICT & Future Planning, Republic of Korea.

11
H. Park, et al. International Journal of Pharmaceutics 581 (2020) 119232

Appendix A

See Tables A1 and A2.

Table A1
FT-IR spectrum data of raw glimepiride, L-arginine, SEGA, PMGA, and SASGA samples.
Assignment of glimepiride Wavenumber (cm−1) Assignment of L-arginine

Glimepiride SASGA0 SASGA1 SASGA2 SEGA PMGA L-arginine

a
Sulfonyl urea ν N-H 3368.0 3367.1 3365.5 3366.2 3368.0 ν N-H Guanidium

3329.9 3331.9 3331.1 3358.8

3287.0 3287.0 3291.8 3293.6 3291.1 3288.1 3290.3
3132.8 3134.1 3140.9 3156.8 3139.6
3090.3 3092.1 3088.9
3051.5 3049.2 ν N-H Amine
ν C=O 1702.2 1703.2 1702.5 ν C=O free Carboxyl
1701.8 1701.4 1698.2 1698.6 1699.1 1699.3
δb N-H 1671.4 1670.4 1672.0 1672.7 δ N-H Amine
Carboxamide ν C=O 1655.4 1654.2 1650.0 1650.2 1651.7
−
1622.2 ν COO Carboxyl
Aromatic or hexyl ν C-H 2943.4 ν C-H Propyl
2929.6 2930.2 2937.6 2937.2 2932.8
2928.3 2924.9 2927.8 2926.3
2860.6 2859.5 2875.6 2875.3 2874.0 2863.0 2862.9
2842.0 2841.6 2856.9 2854.6 2854.1 2842.1

a,b
Stretching and bending vibration, respectively.

Table A2
13
Solid state C CP/MAS-TOSS SSNMR peaks of raw glimepiride, L-arginine, and SASGA3.
13
Assignment Carbon No. C-chemical shift (δ, ppm) Δδa

Glimepiride SASGA2 L-arginine

Glimepiride Pyrroline C=O 2 172.9 173.2 −0.2

Carboxamide C=O 6 153.1 153.2 −0.1
152.0 −1.2
Phenyl C-C 10 145.1 145.1 0.0
144.0 1.1
C-S 13 139.3 141.4 −2.1
139.3 0.0
Pyrroline C 3 132.8 133.2 −0.5
Phenyl C-H 12,14 131.0 130.1 0.9
11,15 127.5 127.4 0.1
Pyrroline C-H2 5 50.6 51.1 −0.5
Arginine Carboxyl COOH 1 183.4 180.7 −2.7
182.6
181.8 179.5 −2.3
Guanidium C-N 7 164.8 −6.2
158.1 158.7 0.6
Carboxyl C-N 2 55.8 56.1 0.2
Propyl C-N 8 41.4 43.4 2.1
Propyl C-H2 3 33.8 32.8 −1.0
C-H2 4 24.9

a
Chemical shift differences (Δδ) was calculated as follows: Δδ = δ (pure glimepiride or L-arginine) – δ (SASGA2).

References Ammar, H., Salama, H., Ghorab, M., Mahmoud, A., 2006b. Implication of inclusion
complexation of glimepiride in cyclodextrin–polymer systems on its dissolution,
stability and therapeutic efficacy. Int. J. Pharm. 320, 53–57.
Abdelkader, H., Abdallah, O.Y., Salem, H., Alani, A.W., Alany, R.G., 2014. Eutectic, Arakawa, T., Kita, Y., Koyama, A.H., 2008. Solubility enhancement of gluten and organic
monotectic and immiscibility systems of nimesulide with water-soluble carriers: compounds by arginine. Int. J. Pharm. 355, 220–223.
phase equilibria, solid-state characterisation and in-vivo/pharmacodynamic evalua- Bian, X., Jiang, L., Gan, Z., Guan, X., Zhang, L., Cai, L., Hu, X., 2019. A glimepiride-
tion. J. Pharm. Pharmacol. 66, 1439–1450. metformin multidrug crystal: synthesis, crystal structure analysis, and physico-
Abuzar, S.M., Hyun, S.-M., Kim, J.-H., Park, H.J., Kim, M.-S., Park, J.-S., Hwang, S.-J., chemical properties. Molecules 24, 3786.
2018. Enhancing the solubility and bioavailability of poorly water-soluble drugs Blagden, N., de Matas, M., Gavan, P.T., York, P., 2007. Crystal engineering of active
using supercritical antisolvent (SAS) process. Int. J. Pharm. 538, 1–13. pharmaceutical ingredients to improve solubility and dissolution rates. Adv. Drug
Amidon, G.L., Lennernäs, H., Shah, V.P., Crison, J.R., 1995. A theoretical basis for a Deliv. Rev. 59, 617–630.
biopharmaceutic drug classification: the correlation of in vitro drug product dis- Bonfilio, R., Pires, S.A., Ferreira, L.M., de Almeida, A.E., Doriguetto, A.C., de Araújo,
solution and in vivo bioavailability. Pharm. Res. 12, 413–420. M.B., Salgado, H.R., 2012. A discriminating dissolution method for glimepiride
Ammar, H., Salama, H., Ghorab, M., Mahmoud, A., 2006a. Formulation and biological polymorphs. J. Pharm. Sci. 101, 794–804.
evaluation of glimepiride–cyclodextrin–polymer systems. Int. J. Pharm. 309, Brittain, H.G., 2016. Polymorphism in Pharmaceutical Solids. CRC Press.
129–138. Buckton, G., Bulpett, R., Verma, N., 1991. Surface analysis of pharmaceutical powders: X-

12
H. Park, et al. International Journal of Pharmaceutics 581 (2020) 119232

ray photoelectron spectroscopy (XPS) related to powder wettability. Int. J. Pharm. naproxen–cimetidine co-amorphous systems prepared by quench cooling, coprecipi-
72, 157–162. tation and ball milling. J. Pharm. Pharmacol. 68, 36–45.
Chavan, R.B., Thipparaboina, R., Kumar, D., Shastri, N.R., 2016. Co amorphous systems: a Liu, X., Feng, X., Williams, R.O., Zhang, F., 2018. Characterization of amorphous solid
product development perspective. Int. J. Pharm. 515, 403–415. dispersions. J. Pharm. Invest. 48, 19–41.
Cherukuvada, S., Nangia, A., 2014. Eutectics as improved pharmaceutical materials: Löbmann, K., Grohganz, H., Laitinen, R., Strachan, C., Rades, T., 2013. Amino acids as co-
design, properties and characterization. Chem. Commun. 50, 906–923. amorphous stabilizers for poorly water soluble drugs–Part 1: Preparation, stability
Cruz-Angeles, J., Videa, M., Martínez, L.M., 2019. Highly soluble glimepiride and and dissolution enhancement. Eur. J. Pharm. Biopharm. 85, 873–881.
Irbesartan co-amorphous formulation with potential application in combination Löbmann, K., Laitinen, R., Grohganz, H., Gordon, K.C., Strachan, C., Rades, T., 2011.
therapy. AAPS PharmSciTech 20, 144. Coamorphous drug systems: enhanced physical stability and dissolution rate of in-
Das, U., Hariprasad, G., Ethayathulla, A.S., Manral, P., Das, T.K., Pasha, S., Mann, A., domethacin and naproxen. Mol. Pharm. 8, 1919–1928.
Ganguli, M., Verma, A.K., Bhat, R., 2007. Inhibition of protein aggregation: supra- Lobmann, K., Laitinen, R., Strachan, C., Rades, T., Grohganz, H., 2013. Amino acids as co-
molecular assemblies of arginine hold the key. PLoS ONE 2. amorphous stabilizers for poorly water-soluble drugs - Part 2: Molecular interactions.
Dengale, S.J., Grohganz, H., Rades, T., Löbmann, K., 2016. Recent advances in co- Eur. J. Pharm. Biopharm.: Off. J. Arbeitsgemeinschaft für Pharmazeutische
amorphous drug formulations. Adv. Drug Deliv. Rev. 100, 116–125. Verfahrenstechnik e.V 85, 882–888.
Endo, T., Iwata, M., Nagase, H., Shiro, M., Ueda, H., 2003. Polymorphism of glimepiride: Luu, T.D., Lee, B.-J., Tran, P.H., Tran, T.T., 2019. Modified sprouted rice for modulation
crystallographic study, thermal transitions behavior and dissolution study. STP of curcumin crystallinity and dissolution enhancement by solid dispersion. J. Pharm.
Pharma Sci. 13, 281–286. Investigat. 49, 127–134.
Figueirêdo, C.B.M., Nadvorny, D., de Medeiros Vieira, A.C.Q., Sobrinho, J.L.S., Neto, P.J.R., Marshall, P., York, P., 1989. Crystallisation solvent induced solid-state and particulate
Lee, P.I., Soares, M.F.d.L.R., 2017. Enhancement of dissolution rate through eutectic modifications of nitrofurantoin. Int. J. Pharm. 55, 257–263.
mixture and solid solution of posaconazole and benznidazole. Int. J. Pharm. 525, 32–42. Martin, A., Cocero, M.J., 2008. Micronization processes with supercritical fluids: funda-
Gill, B., Kaur, T., Kumar, S., Gupta, G., 2014. Formulation and evaluation of glimepiride mentals and mechanisms. Adv. Drug Deliv. Rev. 60, 339–350.
solid dispersion tablets. Asian Journal of Pharmaceutics (AJP): Free full text articles Massi-Benedetti, M., 2003. Glimerpiride in type 2 diabetes mellitus: a review of the
from Asian J Pharm 4. worldwide therapeutic experience. Clin. Ther. 25, 799–816.
Guideline, I.H.T., 2005. Impurities: guideline for residual solvents Q3C (R5). Current Step Mattern, M., Winter, G., Kohnert, U., Lee, G., 1999. Formulation of proteins in vacuum-
4, 1–25. dried glasses. II. Process and storage stability in sugar-free amino acid systems.
Haleblian, J.K., 1975. Characterization of habits and crystalline modification of solids and Pharm. Dev. Technol. 4, 199–208.
their pharmaceutical applications. J. Pharm. Sci. 64, 1269–1288. Mehta, A., Vasanti, S., Tyagi, R., Shukla, A., 2009. Formulation and evaluation of solid
Herbert-Pucheta, J.-E., Colaux, H., Bodenhausen, G., Tekely, P., 2011. Disentangling dispersion of an antidiabetic drug. Curr. Trends Biotech. Pharm. 3, 76–84.
crystallographic inequivalence and crystallographic forms of l-arginine by one- and Mura, P., Bettinetti, G.P., Cirri, M., Maestrelli, F., Sorrenti, M., Catenacci, L., 2005. Solid-
two-dimensional solid-state NMR spectroscopy. J. Phys. Chem. B Condensed Matt. state characterization and dissolution properties of naproxen–arginine–hydrox-
Mater. Surfaces Interf. Biophys. 115, 15415–15421. ypropyl-β-cyclodextrin ternary system. Eur. J. Pharm. Biopharm. 59, 99–106.
Huang, Y., Zhang, Q., Wang, J.-R., Lin, K.-L., Mei, X., 2017. Amino acids as co-amorphous Mura, P., Maestrelli, F., Cirri, M., 2003. Ternary systems of naproxen with hydroxypropyl-
excipients for tackling the poor aqueous solubility of valsartan. Pharm. Dev. Technol. β-cyclodextrin and aminoacids. Int. J. Pharm. 260, 293–302.
22, 69–76. Nair, A., Varma, R., Gourishetti, K., Bhat, K., Dengale, S., 2019. Influence of preparation
Hyun, S.-M., Lee, B.J., Abuzar, S.M., Lee, S., Joo, Y., Hong, S.-H., Kang, H., Kwon, K.-A., methods on physicochemical and pharmacokinetic properties of co-amorphous for-
Velaga, S., Hwang, S.-J., 2019. Preparation, characterization, and evaluation of cel- mulations: the case of co-amorphous atorvastatin: naringin. J. Pharm. Innov. 1–15.
ecoxib eutectic mixtures with adipic acid/saccharin for improvement of wettability Ning, X., Sun, J., Han, X., Wu, Y., Yan, Z., Han, J., He, Z., 2011. Strategies to improve
and dissolution rate. Int. J. Pharm. 554, 61–71. dissolution and oral absorption of glimepiride tablets: solid dispersion versus mi-
Ilić, I., Dreu, R., Burjak, M., Homar, M., Kerč, J., Srčič, S., 2009. Microparticle size control cronization techniques. Drug Dev. Ind. Pharm. 37, 727–736.
and glimepiride microencapsulation using spray congealing technology. Int. J. Ojarinta, R., Heikkinen, A.T., Sievänen, E., Laitinen, R., 2017. Dissolution behavior of co-
Pharm. 381, 176–183. amorphous amino acid-indomethacin mixtures: The ability of amino acids to stabilize
Ishikawa, T., Watanabe, Y., Takayama, K., Endo, H., Matsumoto, M., 2000. Effect of the supersaturated state of indomethacin. Eur. J. Pharm. Biopharm. 112, 85–95.
hydroxypropylmethylcellulose (HPMC) on the release profiles and bioavailability of a Özdemi, T., Gökce, H., 2018. FT-IR, Raman, and NMR spectroscopy and DFT theory of
poorly water-soluble drug from tablets prepared using macrogol and HPMC. Int. J. Glimepiride molecule as a Sulfonylurea compound. J. Appl. Spectrosc. 85, 560–572.
Pharm. 202, 173–178. Palazzo, B., Walsh, D., Iafisco, M., Foresti, E., Bertinetti, L., Martra, G., Bianchi, C.L.,
Izutsu, K.-I., Fujimaki, Y., Kuwabara, A., Aoyagi, N., 2005. Effect of counterions on the Cappelletti, G., Roveri, N., 2009. Amino acid synergetic effect on structure, morphology
physical properties of l-arginine in frozen solutions and freeze-dried solids. Int. J. and surface properties of biomimetic apatite nanocrystals. Acta Biomater. 5, 1241–1252.
Pharm. 301, 161–169. Park, H.J., Kim, M.-S., Kim, J.-S., Cho, W., Park, J., Cha, K.-H., Kang, Y.-S., Hwang, S.-J.,
Jensen, K., Löbmann, K., Rades, T., Grohganz, H., 2014. Improving co-amorphous drug 2010. Solid-state carbon NMR characterization and investigation of intrinsic dis-
formulations by the addition of the highly water soluble amino acid, proline. solution behavior of fluconazole polymorphs, anhydrate forms I and II. Chem. Pharm.
Pharmaceutics 6, 416–435. Bull. 58, 1243–1247.
Jensen, K.T., Larsen, F.H., Löbmann, K., Rades, T., Grohganz, H., 2016. Influence of Park, H.J., Kim, M.-S., Lee, S., Kim, J.-S., Woo, J.-S., Park, J.-S., Hwang, S.-J., 2007.
variation in molar ratio on co-amorphous drug-amino acid systems. Eur. J. Pharm. Recrystallization of fluconazole using the supercritical antisolvent (SAS) process. Int.
Biopharm. 107, 32–39. J. Pharm. 328, 152–160.
Kamble, R., Bothiraja, C., Mehta, P., Varghese, V., 2018. Synthesis, solid state char- Park, H., Ha, E.-S, Kim, J.-S., Kim, M.-S., Hwang, S.-J., 2019. Preparation and char-
acterization and antifungal activity of ketoconazole cocrystals. J. Pharm. Invest. 48, acterization of glimepiride eutectic mixture with L-arginine for improvement of
541–549. dissolution rate. International journal of pharmaceutics, submitted.
Kasten, G., Duarte, Í., Paisana, M., Löbmann, K., Rades, T., Grohganz, H., 2019. Process Pasquali, I., Bettini, R., Giordano, F., 2008. Supercritical fluid technologies: an innovative
optimization and upscaling of spray-dried drug-amino acid co-amorphous formula- approach for manipulating the solid-state of pharmaceuticals. Adv. Drug Deliv. Rev.
tions. Pharmaceutics 11, 24. 60, 399–410.
Kim, D.-H., Lee, S.-E., Pyo, Y.-C., Tran, P., Park, J.-S., 2020. Solubility enhancement and Qinna, N.A., Badwan, A.A., 2015. Impact of streptozotocin on altering normal glucose
application of cyclodextrins in local drug delivery. J. Pharm. Invest. 50, 17–27. homeostasis during insulin testing in diabetic rats compared to normoglycemic rats.
Kim, M.-S., Jin, S.-J., Kim, J.-S., Park, H.J., Song, H.-S., Neubert, R.H., Hwang, S.-J., 2008. Drug Des. Dev. Therapy 9, 2515.
Preparation, characterization and in vivo evaluation of amorphous atorvastatin cal- Reven, S., Grdadolnik, J., Kristl, J., Žagar, E., 2010. Hyperbranched poly (esteramides) as
cium nanoparticles using supercritical antisolvent (SAS) process. Eur. J. Pharm. solubility enhancers for poorly water-soluble drug glimepiride. Int. J. Pharm. 396,
Biopharm. 69, 454–465. 119–126.
Kim, M.-S., Kim, J.-S., Hwang, S.-J., 2010. Enhancement of wettability and dissolution Sanganwar, G.P., Sathigari, S., Babu, R.J., Gupta, R.B., 2010. Simultaneous production
properties of cilostazol using the supercritical antisolvent process: effect of various and co-mixing of microparticles of nevirapine with excipients by supercritical anti-
additives. Chem. Pharm. Bull. 58, 230–233. solvent method for dissolution enhancement. Eur. J. Pharm. Sci. 39, 164–174.
Kiran, T., Shastri, N., Ramakrishna, S., Sadanandam, M., 2009. Surface solid dispersion of Serajuddin, A.T., 2007. Salt formation to improve drug solubility. Adv. Drug Deliv. Rev.
glimepiride for enhancement of dissolution rate. Int. J. Pharm. Tech. Res. 1, 822–831. 59, 603–616.
Kumar, S., Rai, S., 2010. Spectroscopic studies of L-arginine molecule. Shi, Q., Moinuddin, S.M., Cai, T., 2019. Advances in coamorphous drug delivery systems.
Langtry, H.D., Balfour, J.A., 1998. Glimepiride. Drugs 55, 563–584. Acta pharmaceutica sinica B 9, 19–35.
Lee, K.-H., Park, C., Oh, G., Park, J.-B., Lee, B.-J., 2018. New blends of hydro- Singh, D., Bedi, N., Tiwary, A.K., 2018. Enhancing solubility of poorly aqueous soluble
xypropylmethylcellulose and Gelucire 44/14: physical property and controlled re- drugs: critical appraisal of techniques. J. Pharm. Invest. 48, 509–526.
lease of drugs with different solubility. J. Pharmaceut. Invest. 48, 313–321. Staff, P., 2011. Physicians’ desk reference. PDR Network.
Lenz, E., ouml, bmann, K., Rades, T., Knop, K., Kleinebudde, P., 2017. Hot Melt Extrusion Vasanthavada, M., Tong, W.-Q.T., Joshi, Y., Kislalioglu, M.S., 2005. Phase behavior of
and Spray Drying of Co-amorphous Indomethacin-Arginine With Polymers. J. Pharm. amorphous molecular dispersions II: Role of hydrogen bonding in solid solubility and
Sci.: A Public. Am. Pharm. Assoc. 106, 302–312. phase separation kinetics. Pharm. Res. 22, 440–448.
Leuner, C., Dressman, J., 2000. Improving drug solubility for oral delivery using solid Wei, M., Musie, G.T., Busch, D.H., Subramaniam, B., 2002. CO2-expanded solvents: un-
dispersions. Eur. J. Pharm. Biopharm. 50, 47–60. ique and versatile media for performing homogeneous catalytic oxidations. J. Am.
Li, J., Garg, M., Shah, D., Rajagopalan, R., 2010. Solubilization of aromatic and hydro- Chem. Soc. 124, 2513–2517.
phobic moieties by arginine in aqueous solutions. J. Chem. Phys. 133, 054902. Zhang, Y., Huo, M., Zhou, J., Xie, S., 2010. PKSolver: An add-in program for pharma-
Lim, A.W., Löbmann, K., Grohganz, H., Rades, T., Chieng, N., 2016. Investigation of cokinetic and pharmacodynamic data analysis in Microsoft Excel. Comput. Methods
physical properties and stability of indomethacin–cimetidine and Programs Biomed. 99, 306–314.

13