Yaqi Gong1, Shabbir Mohd1, Simei Wu1, Shilin Liu2,3, Ying Pei3, Xiaogang Luo1,3. 1. School of Chemical Engineering and Pharmacy, Wuhan Institute of Technology, LiuFang Campus, No.206, Guanggu 1st road, Donghu New & High Technology Development Zone, Wuhan, 430205 Hubei Province, P.R. China. 2. College of Food Science and Technology, Huazhong Agricultural University, Wuhan, 430205 Hubei Province, China. 3. School of Materials Science and Engineering, Zhengzhou University, No.100 Science Avenue, Zhengzhou City, 450001 Henan Province, P.R. China.
Abstract
Functional modified cellulose microsphere (CMs) materials exhibit great application potential in drug various fields. Here, we designed pH-responsive carboxylated cellulose microspheres (CCMs) by the citric/hydrochloric acid hydrolysis method to enhance oral bioavailability of insulin by a green route. The CMs were high purity cellulose that dissolved and regenerated from a green solvent by the green sol-gel method. The prepared microspheres were characterized by spectroscopic techniques, such as field emission scanning electron microscopy (FE-SEM), Fourier transform infrared spectrum (FT-IR), X-ray diffraction (XPS), etc. The spherical porous structure and carboxylation of cellulose were confirmed by FESEM and FT-IR, respectively. Insulin was loaded into the CCMs by electrostatic interactions, and the insulin release was controlled through ionization of carboxyl groups and proton balance. In vitro insulin release profiles demonstrated the suppression of insulin release in artificial gastric fluid (AGF), while a significant increase at artificial intestinal fluid (AIF) was observed. The insulin release profile was fitted in Korsmeyer-Peppas kinetic model, and insulin release was governed by the Fickian diffusion mechanism. The stability of the secondary structure of insulin was studied by dichroism circular. Excellent biocompatibility and no cytotoxicity of designed CCMs cast them as a potential oral insulin carrier.
Functional modified cellulose microsphere (CMs) materials exhibit great application potential in drug various fields. Here, we designed pH-responsive carboxylated cellulose microspheres (CCMs) by the citric/hydrochloric acid hydrolysis method to enhance oral bioavailability of insulin by a green route. The CMs were high purity cellulose that dissolved and regenerated from a green solvent by the green sol-gel method. The prepared microspheres were characterized by spectroscopic techniques, such as field emission scanning electron microscopy (FE-SEM), Fourier transform infrared spectrum (FT-IR), X-ray diffraction (XPS), etc. The spherical porous structure and carboxylation of cellulose were confirmed by FESEM and FT-IR, respectively. Insulin was loaded into the CCMs by electrostatic interactions, and the insulin release was controlled through ionization of carboxyl groups and proton balance. In vitro insulin release profiles demonstrated the suppression of insulin release in artificial gastric fluid (AGF), while a significant increase at artificial intestinal fluid (AIF) was observed. The insulin release profile was fitted in Korsmeyer-Peppas kinetic model, and insulin release was governed by the Fickian diffusion mechanism. The stability of the secondary structure of insulin was studied by dichroism circular. Excellent biocompatibility and no cytotoxicity of designed CCMs cast them as a potential oral insulin carrier.
Oral administration of
drugs is desirable due to the convenience
and increased compliance to patients, especially for chronic diseases,
such as diabetes,[1] that require frequent
administration. Although, enzyme degradation and hydrolysis in the
gastrointestinal tract could limit the effectiveness of insulin.[2] Insulin is a hormone, which regulates blood sugar,
advised to diabeticpatients, and it also undergoes degradation in
the presence of gastric enzymes and acids due to destruction of disulfide
bonds.[3−5] pH-responsive carriers offer excellent potential
as oral therapeutic systems by enhancing the stability of insulin
delivery in stomach and achieving controlled release in intestines.
This motivates the development of an oral insulin transport carrier
to improve its bioavailability, and various studies have been reported.[6−12] Considering the advantages of natural materials, researchers are
looking for nontoxic, biocompatible, biorenewable, and low-cost natural
materials to replace synthetics.Several naturally derived polymers
have been investigated to design
pH-responsive insulin carriers owing to their renewable and biocompatible
nature.[13,14] Some researchers designed natural polymer-based
insulin delivery systems, but most of them utilized toxic reagents
in complex preparation methods.[13,15] Cellulose microsoheres
(CMs), has been widely explored for its high porosity, hydrophilicity,
biocompatibility, and great modification potential.[16] Therefore, cellulose-based microspheres can reasonably
be used as a drug carrier. Although the microsphere system is not
as convenient as the nanometer system, still they have the advantages
of controlling the amount of insulin released in the stomach to protect
insulin from the inhospitable environment of the stomach.[17] Microspheres can reduce the dosing frequency
and thereby improve patient compliance. CMs are easy to prepare by
the sol–gel method from the green solvent (sodium hydroxide/urea
aqueous solution). The high porosity, appropriate specific surface
area, and hydrophilicity of cellulose-based microspheres play key
roles in insulin loading and controlled release ability.[16] In addition, in accordance to the structure
and physicochemical properties of the insulin, the interfacial microstructure
(such as pore size and particle size) of cellulose-based microspheres
can be controlled to design high-performance oral insulin carriers
to improve the bioavailability of insulin.[18] Since cellulose is nondigestible and nondegradable in the human
body, cellulose-based microsphere-loaded insulin can be released continuously
in the body without affecting the carrier’s structure.Insulin is a proteinaceous drug with amino and carboxyl groups
in its structure with an isoelectric point (pI) of 5.5–6.4.[19] The pIof cellulose was 3.0. The carboxylation
of CMs introduces anionic functional groups that reduced the pI of
carboxylated cellulose microspheres (CCMs), which helps to bind with
the insulin drug via electrostatic interactions. The citric/hydrochloric
acid hydrolysis method is a green chemistry approach to fabricate
CCMs, where hydronium ions (H3O+) from HCl dissociation
hydrolyze the amorphous domains of CMs and catalyze the esterification
of hydroxyl groups on the exposed CMs simultaneously.[20] Optimal pH environments promote insulin loading and release
via electrostatic interactions between insulin and CCMs.In
order to design a new oral insulin carrier by a simple green
route, pH-responsive CCMs were prepared though the citric/hydrochloric
acid hydrolysis method to control oral insulin delivery. CCMs’
morphology, insulin loading and controlled release profiles, and the
mechanism involving the interactions between insulin and CCMs were
investigated. The circular dichroism (CD) analysis of native insulin
and released insulin and in vitro cytotoxicity of CMs and CCMs were
also evaluated to establish the effectiveness of designed materials
in drug delivery systems.
Results and Discussion
Morphology and Structure Characterization
FE-SEM images
of CMs, CCMs, and insulin–CCMs are shown in Figure . All the samples
exhibited good spherical shape and a highly crowded porous three-dimensional
(3D) structure. The porous structure was regulated by the gelation
and regeneration processes of cellulose solution when treated with
a large volume of nonsolvent and a relatively high temperature; the
phase separation happened during the process, and the solvent-rich
regions in cellulose solution contributed to the pore formation.[21] The porous structure of CMs is conducive to
the chemical reaction and drug loading, which made cellulose-based
microspheres ideal candidates as insulin carriers. Generally, the
presence of pores facilitates the penetration of water/insulin solution
into CCMs, consequently benefits the diffusion and absorbency of insulin.
Compared with CMs (Figure b), the pores of CCMs (Figure e) increased slightly. It may be due to the electrostatic
repulsion between carboxyl groups of CCMs resulted in a larger pore
of CCMs than that of CMs. Insulin–CCMs exhibited smaller pores
than CMs and CCMs(Figure h). The reason could be that positively charged insulin was
loaded into negatively charged CCMs by electrostatic interaction to
occupy the space of pore of CCMs at pH = 4 PBS. The element content
of the surfaces of CMs, CCMs, and insulin–CCMs were measured
by energy dispersive X-ray spectroscopy (EDS) patterns (Figure c,f,i). Compared to CMs and
CCMs, the presence of sulfur element on the surface of insulin–CCMs
indicated that insulin was loaded into the CCMs.
Figure 1
FESEM images of CMs (a,
b), CCMs (d, e), and insulin-CCMs (g, h);
EDS patterns of CMs (c), CCMs (f), and insulin-CCMs (i).
FESEM images of CMs (a,
b), CCMs (d, e), and insulin-CCMs (g, h);
EDS patterns of CMs (c), CCMs (f), and insulin-CCMs (i).Correlation between the chemical structures of CMs, CCMs,
and insulin–CCMs
was studied with FT-IR spectra (Figure a). There was a broad peak between 3000–3300
cm–1, which could be attributed to −OH groups
of cellulose.[22] Compared with CMs, a new
carbonyl peak was observed at 1730 cm–1, indicating
the presence of carboxyl groups (−COOH) on CCMs. This could
be caused by the hydronium ions (H3O+) from
HCl catalyzed during the esterification of hydroxyl groups on the
exposed CM chains with carboxyl groups of C6H8O7.[20] The graft yield and grafting
efficiency of CCMs were 12.2 and 4.2%, respectively. The results indicated
that the carboxyl groups of citric acid were successfully grafted
into CMs via the esterification reaction.
Figure 2
FT-IR spectra of CMs,
CCMs, and insulin–CCMs (a); TGA and
DTGA curves of CMs and CCMs (b); zeta potential of CCMs (c).
FT-IR spectra of CMs,
CCMs, and insulin–CCMs (a); TGA and
DTGA curves of CMs and CCMs (b); zeta potential of CCMs (c).Generally, drug carriers require high-temperature
treatment before
using, so it was necessary to evaluate the thermal stability of CMs
and CCMs.[23] The thermogravimetric analysis
(TGA) and differential thermogravimetric analysis (DTGA) for CMs and
CCMs are displayed in Figure b.Interestingly, CCMs exhibited a slightly lower onset
degradation
temperature (111.28 °C) than that of CMs (124.2 °C). This
might be a result of the insertion of carboxyl group that disrupted
the hydrogen bonding of CMs.[20] Moreover,
the small weight loss at the low-temperature range was due to the
evaporation of absorbed water, whereas the two other weight losses
were due to pyrolysis of hydrocarbon chains.[24] Plus, the main degradations of CMs and CCMs did not start at the
same point; the thermographs showed different degradation behaviors
for CMs and CCMs, which indicated that acid hydrolysis could have
increased the degree of crystallinity and thus led to the different
degradation behaviors.[25] With the increase
of temperature, CMs and CCMs showed the degradation at the temperature
range of 150–600 °C. The peak of DTGA represented the
temperature of the maximum weight reduction. Based on the DTGA analysis,
it can be observed that small degradation at 250 °C existed in
CMs and then followed by a relatively small peak. Meanwhile, CCMs
started the main decomposition at about 250 °C, and the degradation
increased until the peak appeared, which suggested that CCMs possessed
higher thermal stability than that of CMs. This result revealed the
successful insertion of the carboxyl group on CMs.[20,25] CCMs can withstand the high temperature of autoclaving and sterilization
before insulin loading, and the structure was not destroyed by the
high temperature, so CCMs were suitable as insulin carriers.[26]To explore insulin loading and release
mechanisms, the zeta potential
of CCMs was measured from pH 1.0 to 13.0. The results are shown in Figure c. When pH > 2.3,
the net negative zeta potential indicated that the ionization of carboxylic
acid (COOH → COO–) increased the absolute
value of the zeta potential, resulting in more negative charges.[24,27] The original pI of insulin fell in the range of 5.5–6.4,[19] and insulin showed a positive zeta potential
at pH = 4 PBS. It indicated that insulin was loaded into CCMs by electrostatic
action. When pH < 2.3, CCM dispersions showed a positive zeta potential,
indicating the protonation of carboxylic acid. The hydrogen bonding
interactions between the protonated carboxylic groups of CCMs might
preserve the structure of CCMs in a compact collapsed state, preventing
the release of insulin in AGF. However, insulin is negatively charged
and CCMs are negatively charged in AIF. The electrostatic repulsion
between COO– of CCM chains enlarged the pore size
of CCMs and promoted water molecules into CCMs. At the same time,
the electrostatic repulsion between insulin and CCMs also promoted
the insulin diffusion and release.The esterification reaction
among the CMs and citric acid can be
verified by using the XPS technique too. Figure shows the peak spectra of C1s. The peaks
at 284.8 and 286.3 eV belong to C–C and C–O–C,
respectively.[28] Compared with Figure b, a new peak appeared
at 288.6 eV in Figure c, which was attributed to the ester (O—C=O) peak.[29] In addition, the change of the peak (centered
at 284.8 and 286.3 eV) area and the appearance of a new peak (O—C=O)
(centered at 288.6 eV) in Figure c indicated the successful esterification reaction
of hydroxyl groups on CMs with carboxyl groups of citric acid. This
further showed the hydrolysis mechanism of citric acid/hydrochloric
acid, which is also in accordance with the FT-IR results.[28,29]
Figure 3
XPS
spectra of CMs and CCMs: C1s of CMs (a), C1s of CCMs (b), and
survey of CMs and CCMs (c).
XPS
spectra of CMs and CCMs: C1s of CMs (a), C1s of CCMs (b), and
survey of CMs and CCMs (c).
Swelling Properties of CCMs
To examine
the effect of solution pH on water uptake, the swelling ratio (SR)
was determined using CCMs gravimetric ratios in AGF and AIF, respectively.
The swelling property of CCMs was calculated with eq . CCMs showed higher swelling in
AIF (598.3%) than in AGF (455.7%). This observation can be explained
by the ionization and protonation equilibrium of the carboxyl groups
(pKa = 4.3) in CCMs. In AGF, the carboxyl
groups were protonated to promote the formation of intramolecular
hydrogen bonds. Thereby, CCMs exhibited lower swelling ratios. While
in AIF, the swelling ratios were higher than those in AGF. The reason
was that the hydrogen bonds were broken in AIF and generated the electrostatic
repulsion forces among the ionized carboxylic acids,[30,31] benefiting the diffusion of water molecules into CCMs, which therefore
led to quick swelling. CCMs were responsive to changes in pH values,
they can protect insulin against the gastric fluid in AGF while quickly
swelling in AIF to facilitate the diffusion and absorption of insulin.
Insulin Loading and Controlled Release Performance
of CCMs
To study the loading and releasing properties of
insulin, a constant flow pump was used to simulate gastrointestinal
absorption. CCMs were added in a glass column (internal diameter =
1.1 cm). The pH =4 insulin solution was used as eluent solution,
which was collected after passing through the column with CCMs at
predetermined time intervals to calculate insulin loading efficiency
(ILE) and insulin loading quality (LC). The values of ILE and LC depended
on the insulin concentration, insulin/CCMs ratio, solvent pH, etc.[30] The ILE and LC were observed as 2.6% and 335
mg g–1 via eqs and 5, respectively. This may be due
to the opposite charges between insulin and CCMs at pH = 4. Insulin
is positively charged at pH = 4[19] and CCMs
are negatively charged at pH = 4 under room temperature as is shown
by zeta potential of CCMs (Figure c), This promoted insulin loading via electrostatic
interactions.To determine the release properties of insulin
from insulin-CMs and insulin–CCMs, the constant flow pump was
used to simulate gastrointestinal absorption. AGF and AIF were used
as eluent solutions, which were collected after passing through the
column with CMs and CCMs at predetermined time intervals to obtain
the breakthrough curves, respectively. The release behavior of insulin
in AGF and AIF were studied, respectively (Figure ). As shown in Figure a, 68.44% of insulin were released in AGF,
and 65.05% of insulin were released in AIF, respectively. The results
showed that oral insulin delivery could not be controlled by CMs. Figure b shows the insulin
release curves versus time from insulin–CCMs in AGF and AIF,
respectively. In AGF, 32.75% of insulin was released fast, followed
by 17.05% of release. In AIF, the release rate was fast, with 44.5%
of the total insulin release after 10 min, and then an equilibrium
plateau was gradually established at about 85.3%. The burst release
phenomenon was due to the fact that insulin was attached to the surface
of CCMs.[32,33] Controlled release of insulin from insulin–CCMs
can be regulated by the proton balance of carboxyl groups and the
charge of insulin in AGF and AIF. In AGF, the hydrogen bonding interactions
between the protonated carboxylic groups of CCMs might preserve the
structure of insulin in compact collapsed state, preventing the release
of insulin.[13] In AIF, the carboxyl groups
of CCMs existed as COO–. The electrostatic repulsion
between COO– enlarged the pore size of CCMs and
promoted water molecules into CCMs. Insulin, with a molecular weight
of about 6000, is a hydrophilic water-soluble protein,[34] and it is also negatively charged in AIF. The
electrostatic repulsion between insulin and CCMs also promoted the
diffusion of insulin into water molecules. The insulin molecule contains
six positively charged and 10 negatively charged amino acid residues.
Due to a handful of positive charge in the insulin molecule and the
formation of the hydrogel barrier after the CCM swelling, insulin
was trapped in CCMs and could not be released completely. This indicated
that the carboxyl groups of CCMs played an important role in the oral
delivery of insulin.
Figure 4
Insulin release curves versus time from insulin–CMs
in AGF
and AIF, respectively (a); insulin release curves versus time from
insulin–CCMs in AGF and AIF, respectively (b).
Insulin release curves versus time from insulin–CMs
in AGF
and AIF, respectively (a); insulin release curves versus time from
insulin–CCMs in AGF and AIF, respectively (b).For the stomach and intestine environments, the calculated
parameters
of the Korsmeyer model are summed up in Table . The n values were less
than 0.45; the correlation coefficients for the model ranged from
0.9397 to 0.9981, indicating that insulin was released according to
the Fickian diffusion.
Table 1
Equilibrium Parameters
of Korsmeyer
Fitting Models
fluid
k
n
R2
AGF
4.960
0.0682
0.9981
AIF
2.705
0.1993
0.9397
Conformation of Insulin Release
The
folding and conformation of insulin correlated with its activity and
can change during the formulation process. Therefore, it is crucial
to prepare controlled release systems that can release insulin in
its active form. The qualitative and quantitative information about
protein conformations were provided by CD spectra.[35] From Figure , the released insulin from insulin–CCMs displayed similar
characteristic peaks at about 208 and 223 nm on CD spectra compared
with native insulin, suggesting that the secondary structure stability
of released insulin was preserved during the loading and release processes.
The CD results confirmed that there were no obvious conformational
changes for the released insulin from insulin–CCMs.
Figure 5
CD spectra
of native insulin and released insulin from insulin–CCMs.
CD spectra
of native insulin and released insulin from insulin–CCMs.
In Vitro Cell Viability
of Samples
Considering the safety of oral preparations, it
is necessary to evaluate
the toxicity of CMs and CCMs.[36] The cytotoxicities
of CMs and CCMs were tested by the MTT method. The biocompatibility
of CMs and CCMs was assessed by A549 cells (Figure. ). It can be seen from Figure that the viability of the cells was above
90%, indicating that both CMs and CCMs were noncytotoxic and had good
biocompatibility. When the concentrations of CMs and CCMs reached
200 μg mL–1, the cell viability exceeded 100%.
This suggested that high concentrations of cellulose-based microspheres
promoted cell growth, which was indicative of low toxicity of cellulose
microspheres.[37,38] Results showed that the CMs and
the reagents used in the preparation of CCM carriers were nontoxic.
CCMs have a potential application to safe oral insulin administration.
Figure 6
In vitro
cell viability of A549 cells against CMs (a) and insulin–CCMs
(b) with different concentrations.
In vitro
cell viability of A549 cells against CMs (a) and insulin–CCMs
(b) with different concentrations.
Design Strategy and Insulin-Loaded and Controlled
Release Mechanisms of CCMs
In this work, CCMs were used as
the carrier to improve the oral bioavailability of insulin. The design
mechanism of CCMs is depicted in Scheme . Based on the design principle of CCMs,
the carboxyl group of citric acid was combined with the hydroxyl group
of CMs through the esterification reaction so that CCMs is pH sensitive.
The pI of insulin is 5.5–6.4, and the pI of the CCMs is 2.3.
Insulin is positively charged and CCMs are negatively charged at pH
= 4 PBS. Insulin was loaded into CCMs by electrostatic interactions. Scheme shows the controlled
release mechanism of oral insulin from CCMs. The controlled release
of insulin can be explained by the ionization and proton equilibrium
of the carboxyl groups (pKa = 4.3) of
CCMs. In AGF, pH = 1.2 PBS was lower than the pKa of the carboxyl groups, which made them positively charged.
The carboxyl groups of CCM existed in the protonated form (−COOH).
Hydrogen bond interactions between −COOH caused CCMs to shrink.
CCMs were in a tight collapsed state, inhibiting the entry of water
molecules and the diffusion of insulin. In AIF, pH = 7.4 PBS was higher
than the pKa of the carboxyl groups, making
them negatively charged. −COOH existed in the form of −COO–, and the electrostatic repulsion between −COO– caused the CCMs to swell. It was beneficial to the
diffusion of water molecules into the CCMs and promoted the release
of insulin. At the same time, the electrostatic repulsion between
insulin and CCMs also promoted insulin diffusion and release.
Scheme 1
Synthesis of CCMs with the pH-Responsive Property and Controlled
Release Mechanism of Insulin from CCMs
Conclusions
In this work, CCMs were prepared
by a green route via the citric/hydrochloric
acid hydrolysis method from CMs, which were prepared through the sol–gel
transition method from green cellulose solvent. CCMs were chosen to
be used as a candidate material of an oral insulin carrier.The oral bioavailability of insulin was insured by the pH sensitivity
of CCMs. In vitro release studies demonstrated that release of insulin
was 48.87 and 85.12% in AGF and AIF, respectively. The release curve
of insulin from CCMs conformed to Fickian diffusion by the Korsmeyer–Peppas
model fitting. The CD results that the secondary structure stabulity
of released insulin was preserved during the loading and release process.
Cell viability revealed that the CCMs were noncytotoxicity.These results indicated that the encapsulation of insulin into
CCMs was a key factor in the improvement of its oral absorption and
oral bioactivity; the designed CCMs had the potential for oral insulin
delivery.
Materials and Methods
Materials
Cellulose (cotton linter
pulp; α-cellulose > 95%) was purchased from Hubei Chemical
Fiber
Group Ltd. (Xiangfan, China). Its viscosity average molecular weight
(Mv) was 12.5 × 104 Da.
Citrate was purchased from Aladdin. Insulin (porcine pancreatic) was
provided by Xuzhou WanBang Co. Phosphate buffer saline (PBS), AIF,
(pH = 7.4 PBS), and AGF (pH = 1.2 PBS) were prepared to simulate the
pH environment of the human stomach and intestines.[39] The pH = 1.2 PBS was prepared with 7 mL of hydrochloric
acid (37 wt %) and 2.0 g of sodium chloride dissolved in 1 L of deionized
water (DI water). Potassium dihydrogen phosphate (6.8 g) was dissolved
in DI water, and the pH value was adjusted to 7.4, followed by diluting
the solution to 500 mL to prepare pH = 7.4 PBS.
Preparation of CCMs
CMs were prepared
through the sol–gel transition method according to our previous
work.[40] CCMs were fabricated according
to the citric/hydrochloric acid hydrolysis method.[20] Briefly, 8 g of CMs and the acid mixture (90% citric acid/10%
hydrochloric acid (v/v)) were added to a 500 mL three-necked flask
and stirred constantly (300 r min–1) at 80 °C
for 4 h. After the reaction was completed, the suspension was quickly
cooled to room temperature. Then, CCMs and mixed acids were separated
by filtration processes, and CCMs were further washed three times
with DI water to remove unreacted acid mixtures. Finally, CCMs were
dried in a vacuum oven over at 60 °C for 48 h.
Characterizations
The morphology
and structure of CMs, CCMs, and insulin–CCMs were investigated
by FE-SEM (GeminiSEM 300), and the surface elements were analyzed
by EDS. Samples were crushed into powder for other characterizations,
and pellets were pressed with potassium bromide for FT-IR analysis.
FTIR spectra were recorded on an FT-IR spectrometer (170-SX, Thermo
Nicolet Ltd., U.S.A.). The CMs and CCMs chemical states of the elements
were measured by XPS (Kratos, U.K.). Thermal properties of the samples
were assessed usingTGA instrument (STA449F3 Germany) under nitrogen
atmosphere from 10 to 600 °C at rate of 10 °C min–1. The surface charge of the samples was determined by the Malvern
nanoparticle size potentiometer (Zetasizer Nano S90). A secondary
structure of insulin was evaluated by Chirascan Plus (Applied Photophysics,
United Kingdom).
Graft Yield and Graft Efficiency
of CCMs
The graft yield and graft efficiency of CCMs were
calculated by
the gravimetric method using the following relationships.where W0 is the
CMs before reacting, Wg represents the weight of the CCMs
after reacting, and Wm denotes the weight of the citric
acid used for reacting.[41]
pH-Responsive Swelling
CCMs were
immersed in AGF and AIF until the swelling equilibrium is reached
to study the pH swelling properties of CCMs by the gravimetric method.
Dry CCMs (Gi) (10 mg) were added to 50
mL of AGF and AIF soaking for 24 h, respectively. The immersing fluid
was removed by centrifugation, and the weight of the wet CCMs (Gs) was recorded. The SR was calculated using the following
equation:where Gs is the weight of wet CCMs, and Gi is the weight of dried CCMs.[26]
In Vitro Insulin Loading and Release Properties
Insulin
was loaded into CCMs by electrostatic action. CCMs (0.4
g) were added in glass column (internal diameter = 1.1 cm). The column
was pretreated with PBS eluent water of pH = 4. pH = 4 insulin solution
was then used as the eluent solution. The eluent solution was collected
after passing through the column with CCMs at predetermined time intervals
to calculate the insulin loading efficiency and insulin loading quality.
The concentration of insulin was detected by UV/vis spectrophotometer
at wavelength of 276 nm and quantified by comparing with a standard
curve.[42] When the concentrations (C) of insulin at pH = 4 PBS were 0.0, 0.2, 0.4, 0.6, 0.8,
and 1.0 mg mL–1, the absorbance (A) at the wavelength of 276 nm were 0.000, 0.212, 0.395, 0.620, 0.820,
and 1.046, respectively. Then the standard curve equation A = – 0.004 + 1.040C (R2 = 0.999) was obtained by linear fitting. Finally, the
concentration of insulin was calculated according to the standard
curve equation. ILE (%) and LC(mg g–1) were calculated
using the following equations:where C0 and C1 are the initial insulin
concentration (mg mL–1) and the final or equilibrium
insulin concentration (mg mL–1), respectively. V represents the volume of insulin solution (mL), and m denotes the dry weight of CCMs (g).In order to
assess the insulin release property, constant flow pump was used to
simulate gastrointestinal absorption. AGF and AIF were used as eluent
solutions, which were collected after passing through the column with
CCMs at predetermined time intervals to obtain the breakthrough curves,
respectively. The concentration of insulin was calculated according
to the standard curve equation. The cumulative release percentage
of insulin was calculated according to the equation as follows:[13]where m represents
the mass of insulin in insulin–CCMs before release, V denotes the volume of the eluted water in the collector
at a predetermined time interval, and C means the concentration of insulin in the nth predetermined time interval.The insulin release data
were fitted by the Korsmeyer–Peppas
equation to study the kinetics of insulin release from insulin-CCMs:[43]Here, M is the mass of insulin released
from the start of release to time t, M∞ means the mass of insulin in insulin–CCMs before release,
and K and n denotes the rate constant
and release exponent, respectively. The mechanism of insulin release
depends on the value of n: n ≤
0.45 indicates the Fickian diffusion model, 0.45 < n < 0.89 corresponds to the non-Fickian diffusion model, and n > 0.89 corresponds to the case II diffusion model.
Cytotoxicity of CCMs and Insulin–CCMs
To evaluate the potential toxicity of CMs and CCMs, the MTT assay
was performed to measure the cell cytotoxicity of CMs and CCMs in
A549 cells.[44,45] The A549 cells were seeded into
a 96-well plate for 24 h, and then CMs and CCMs with a concentration
gradient (0, 50, 100, 150, and 200 μg mL–1) were added to the culture medium and cultured at 37 °C for
24 h. After removing the culture medium, 3-(4, 5-dimethylthiazol-2-yl)-3,5-diphenyltetrazolium
bromide (MTT) solution was added to A549 cells and cultured for 4
h. Then, 200 μL of DMSO was added to the A549 cells after the
removal of the MTT solution. The crystals were sufficiently dissolved
by low-speed shaking. The absorbance was measured at 495 nm with microplate
reader (SpectraMax M5, Molecular Devices).[46,47]
Circular Dichroism (CD) Tests of Released
Insulin
The activity of released insulin was evaluated by
analysis of the structural stability by using CD spectrophotometer
according to Hong et al.[48] The released
insulin and native insulin were measured at 25 °C with a cell
path length of 1.0 cm, a bandwidth of 1.0 nm, response time of 0.25
s, and constant nitrogen flow of 5 L min–1. The
native insulin was prepared in pH = 7.4 PBS. The samples were scanned
from 190 to 250 nm at resolution of 1.0 nm. For CD tests, each sample
was tested for three times, and the final data took the average of
three measurements.