pH-Sensitive oral insulin delivery systems using Eudragit microspheres.

Abstract

In this paper, we present in vitro and in vivo release data on pH-sensitive microspheres of Eudragit L100, Eudragit RS100 and their blend systems prepared by double emulsion-solvent evaporation technique for oral delivery of insulin. Of the three systems developed, Eudragit L100 was chosen for preclinical studies. Insulin was encapsulated and in vitro experiments performed on insulin-loaded microspheres in pH 1.2 media did not release insulin during the first 2 h, but maximum insulin was released in pH 7.4 buffer media from 4 to 6 h. The microspheres were characterized by scanning electron microscopy to understand particle size, shape and surface morphology. The size of microspheres ranged between 1 and 40 μm. Circular dichroism spectra indicated the structural integrity of insulin during encapsulation as well as after its release in pH 7.4 buffer media. The in vivo release studies on diabetic-induced rat models exhibited maximum inhibition of up to 86%, suggesting absorption of insulin in the intestine.

Full text

Page 1

977
Introduction
Intensive efforts over the past decade have resulted in
the development of novel types of polymeric carriers
for oral insulin delivery1–5. The main hurdles in develop-
ing oral insulin delivery devices are attributed to poor
intrinsic insulin permeability across the biological mem-
brane. Insulin has high molecular weight, and is easily
degraded by proteolytic enzymes of the stomach or small
intestine6–8. However, insulin denaturation/degradation
can be overcome by designing a suitable carrier that will
protect insulin from harsh environments of the stomach
before it reaches gastrointestinal tract (GIT). Therefore,
to protect insulin in acidic conditions, efforts have been
made to develop pH-sensitive polymers or hydrogels for
an effective oral formulation of insulin9–11. Various strate-
gies have been adopted to achieve insulin delivery, which
include co-administration with absorption enhancers4,
enzyme inhibitors12,13, chemical modifications14,15 or use
of liposomes2,16,17.
In the literature, poly(glycolic acid), poly(lactic
acid), poly(lactic acid-co-glycolic acid), poly[lactic
acid-co -poly(ethylene glycol)], dextran–PEG as well
as pH-sensitive polymers like poly(acrylic acid)
and poly(methacrylic acid), graft copolymers of
poly(methylmethacrylate) with ethylene glycol, chito-
san, cyclodextrin, etc., have been used as carriers in oral
insulin delivery6. In addition, pH-sensitive hydrogels
exhibiting reversible formation of inter-polymer com-
plexes that are insoluble at lower gastric pH, but swell
in alkaline conditions of the intestine, and dissociate
the complexes to release insulin have been used. In this
category of devices, one of the most widely investigated
systems is that of pH-sensitive graft polymers of meth-
acrylic acid and polyethylene glycol7,18,19. Other studies
include the development of complexation polymers20
and pH responsive hydrogels21. Acrylate and methacry-
late copolymers are commercially available as Eudragit
polymers in different ionic forms, since these are widely
employed in the preparation of microspheres to deliver
macromolecules22–24.
In continuation of our ongoing program of research
concerning the development of oral insulin devices, we
RESEARCH ARTICLE
pH-Sensitive oral insulin delivery systems using Eudragit
microspheres
Raghavendra C. Mundargi, Vidhya Rangaswamy, and Tejraj M. Aminabhavi
Reliance Life Sciences, Industrial Biotechnology, Dhirubhai Ambani Life Sciences Centre, Navi mumbai, India
Abstract
In this paper, we present in vitro and in vivo release data on pH-sensitive microspheres of Eudragit L100, Eudragit RS100
and their blend systems prepared by double emulsion-solvent evaporation technique for oral delivery of insulin. Of
the three systems developed, Eudragit L100 was chosen for preclinical studies. Insulin was encapsulated and in vitro
experiments performed on insulin-loaded microspheres in pH 1.2 media did not release insulin during the first 2 h, but
maximum insulin was released in pH 7.4 buffer media from 4 to 6 h. The microspheres were characterized by scanning
electron microscopy to understand particle size, shape and surface morphology. The size of microspheres ranged
between 1 and 40 µm. Circular dichroism spectra indicated the structural integrity of insulin during encapsulation as
well as after its release in pH 7.4 buffer media. The in vivo release studies on diabetic-induced rat models exhibited
maximum inhibition of up to 86%, suggesting absorption of insulin in the intestine.
Keywords: Eudragit L100, Eudragit S100, microspheres, oral delivery, insulin, pH-sensitive
Address for Correspondence: Tejraj M. Aminabhavi, Reliance Life Sciences, Industrial Biotechnology, Dhirubhai Ambani Life Sciences
Centre, Thane Belapur Road, Rabale, Navi Mumbai 400701, India. Tel.: +91 22 6767 8372; Fax: +91 22 6767 8099. E-mail: aminabhavi@
yahoo.com
(Received 28 October 2010; revised 23 December 2010; accepted 04 January 2011)
Drug Development and Industrial Pharmacy, 2011; 37(8): 977–985
© 2011 Informa Healthcare USA, Inc.
ISSN 0363-9045 print/ISSN 1520-5762 online
DOI: 10.3109/03639045.2011.552908

Page 2

978 R. C. Mundargi et al.
Drug Development and Industrial Pharmacy
present here a novel approach of preparing oral insu-
lin delivery formulations based on Eudragit polymers.
Microspheres were prepared by solvent evaporation
technique using light liquid paraffin oil as an external
phase. The developed formulations were characterized
by scanning electron microscopy (SEM) to examine
particle size, shape and surface morphology. In vitro
release of insulin was performed in pH 1.2 and 7.4 buffer
solutions at 37°C. In vivo experiments were conducted
on diabetes-induced rats and by glucose tolerance tests
on healthy rats. Pure insulin, encapsulated insulin and
released insulin from the Eudragit matrices were evalu-
ated by circular dichroism (CD) for assessing the struc-
tural integrity of insulin.
Materials and methods
Materials
Eudragit L100 and S100 were received as gift samples
from Degussa India Ltd., Mumbai, India. Human (human
recombinant expressed in yeast) and bovine insulin were
procured from Sigma Aldrich, St. Louis, MO. Light liquid
paraffin and Span-80 were purchased from Himedia,
Mumbai, India and s.d. fine chemicals, Mumbai, India,
respectively. All other chemicals used were of reagent
grade. Milli Q water was used throughout this study.
Preparation of insulin-loaded Eudragit L100
microspheres by solvent evaporation method
Solvent evaporation method is one of the most popular
and widely used encapsulation techniques for proteins
and peptides25. In this work, solvent evaporation method
is used to produce insulin-loaded microspheres of uni-
form size. Two different approaches used to prepare the
insulin-loaded formulations are given below.
Method I
In this method, 500 mg of Eudragit L100 was dissolved in a
mixture of 5 mL of methanol and 15 mL dichloromethane
(DCM) under constant stirring in a 50 mL beaker. To this,
20 mg of human insulin dissolved in 0.5 mL of 0.1 M HCl
was added and homogenized at 11,000 rpm for 2 min in an
ice bath. The entire solution was transferred to a 250 mL
beaker containing 100 mL of light liquid paraffin oil and
0.5% of Span-80 surfactant. The solution was stirred at
600 rpm for 3 h to evaporate the solvent. After 3 h, the
microspheres were filtered and washed with 50 mL
petroleum ether and 8 mL Milli Q water to remove excess
paraffin oil and Span-80. The completely washed micro-
spheres were freeze dried at−40°C for 20 h and stored at
−20°C before characterization and further analysis.
Similar method as above was employed for the
preparation of blend microspheres of Eudragit L100 and
Eudragit S100. Here, 250 mg of Eudragit L100 was dis-
solved in 10 mL of methanol, whereas 250 mg of Eudragit
RS100 was dissolved in 10 mL of DCM under constant
stirring in a 50 mL beaker. Both the solutions were
mixed with constant stirring, to which 20 mg of human
insulin dissolved in 0.5 mL of 0.1 M HCl was added and
homogenized at 11,000 rpm for 2 min in an ice bath. The
solution was then transferred to a beaker containing
100 mL of light liquid paraffin oil and 0.5% of Span-80
surfactant. The solution was stirred at 600 rpm speed for
3 h to evaporate solvents. After 3 h, microspheres were
filtered and washed with petroleum ether and Milli Q
water to remove excess paraffin oil and Span-80. The
washed particles were freeze dried and stored at−20°C
before characterization.
Method II
In this method, 720 mg of Eudragit L100 was dissolved in
8 mL of methanol under constant stirring in 15 mL falcon
tube. To this, 36 mg of bovine/human insulin dissolved
in 0.5 mL of 0.1 M HCl was added and homogenized at
11,000 rpm for 70 s in an ice bath. The entire solution was
transferred to 1 L plastic beaker containing 100 mL of
light liquid paraffin oil and 0.5% of Span-80. The emul-
sion was stirred at 500 rpm for 4 h to evaporate methanol.
After 4 h, the microspheres were filtered and washed with
50 mL petroleum ether (60–80°C) and 8 mL Milli Q water
to remove excess paraffin oil and Span-80. The micro-
spheres were then dispersed in 2 mL water, freeze dried
at −41°C for 20 h and stored at −20°C before characteriza-
tion and further analysis.
Following the same protocol as above, microspheres
of Eudragit S100 were prepared by taking 720 mg of
Eudragit S100 and dissolving in 8 mL of DCM at constant
stirring in 15 mL falcon tube. To this, 36 mg of bovine
insulin dissolved in 0.5 mL of 0.1 M HCl was added and
homogenized at 11,000 rpm for 70 s. The solution was
transferred to 1 L plastic beaker containing 100 mL of
light liquid paraffin oil and 0.5% of Span-80. The emul-
sion was stirred at 500 rpm for 4 h to evaporate methanol.
After 4 h, the microspheres were filtered and washed
with 50 mL of petroleum ether (60–80°C) and 8 mL Milli
Q water to remove excess paraffin oil and Span-80. The
microspheres were then dispersed in 2 mL water, freeze
dried at −41°C for 20 h and stored at −20°C before charac-
terization and further analysis.
Characterization
Size, shape and morphology analyses
SEM was done on insulin-loaded microspheres mounted
on metal stubs using double-sided adhesive tape, drying
in a vacuum chamber, sputter-coating with a gold layer
and viewing under SEM (JSM-840, Jeol Instruments,
Tokyo, Japan) to characterize the shape and morphology
as well as to confirm the particle size.
Encapsulation efficiency
Microspheres (50 mg) containing insulin were dissolved
in 5 mL methanol. Insulin was extracted using 25 mL
0.1 M HCL solution in a separating funnel, shaken for few
minutes and the solution was filtered using 0.22 µm filter.
Insulin content was analyzed using high-performance
liquid chromatography (HPLC).

Page 3

Oral insulin delivery system 979
© 2011 Informa Healthcare USA, Inc.
HPLC method I. The amount of insulin released from the
microspheres was collected by taking 100 µL samples
at predetermined time intervals of 1 h up to 4–6 h and
analyzed by HPLC. Insulin was separated on C18 Vydac
218MS54 column (4.6 × 250 mm) having the pore size of
300 A° and a particle size of 5 µm. The buffer solution
for mobile phase was prepared by dissolving 28.4 g of
anhydrous sodium sulfate in 1000 mL double distilled
water and pH of the buffer was adjusted to 2.3 by add-
ing 2.7 mL of orthophosphoric acid. The mobile phase
consisted of (A) 82:18 of acetonitrile:buffer and (B) 50:50
acetonitrile:buffer solutions. HPLC experiments were
performed in gradient mode (see Table 1) at the flow rate
of 1 mL/min, injection volume of 100 µL and detection
wavelength of 210 nm.
HPLC method II. In this method, insulin was analyzed
using MN C18 column (250 × 4.6 mm). HPLC run was
carried out in a gradient mode (see Table 1) at the flow
rate of 1 mL/min, injection volume of 100 µL and detec-
tion wavelength of 210 nm. Mobile phase consisted of
(A) water (0.05% trifluoroacetic acid, TFA v/v) and (B)
acetonitrile: water (80:20) (0.05% TFA v/v).
%Drug
loading
Weight of drug
inmicrospheres
Weight of microsphere
=
s s










×100
(1)
%Entrapment
efficiency
Drug loading
Theoretical loading
=


×1 100
(2)
CD
The released insulin samples were filtered (0.22 µm;
Millipore; Ireland) before subjecting to CD analysis (for
understanding the insulin’s integrity) to remove any par-
ticulate matter as well as possible protein aggregates. CD
spectra at 25°C were obtained using a Jasco J-180 spec-
tropolarimeter on a 1 cm path length quartz cell at pro-
tein concentration of 2 mg/mL. Analysis conditions used
were: 0.5 nm bandwidth, 10-mdeg sensitivity, 0.2-nm
resolution, 2s response, 10 nm/min scanning speed
and 200–240 nm measuring range. Each spectrum is the
average of at least three runs and the buffer baseline was
subtracted from the average spectra. Final spectra are
presented in mean residual ellipticity. Deconvolution of
CD spectra was obtained by SELCON method26.
In vitro release of insulin-loaded microspheres
The in vitro release experiments were done on the for-
mulated microspheres prepared by Method I by taking
100 mg of insulin-loaded Eudragit L100 and Eudragit
L100/RS100 blend particles in a flask containing 25 mL of
buffer solution. Dissolution was carried out in an incuba-
tor maintained at 37°C under constant stirring at 100 rpm.
At regular intervals of time, aliquots (2 mL each time)
were withdrawn and analyzed for insulin by HPLC at the
λmax value of 210 nm employing the gradient method as
mentioned in HPLC method I. In order to simulate the
stomach and intestinal environments, all the in vitro
release experiments were performed in solutions of pH
of 1.2 and 7.4 buffer, respectively. The formulations were
kept in 1.2 pH media for the first 2 h and later, in pH of 7.4
media to follow the intestinal environment.
In vitro release experiments were done on the formu-
lated microspheres prepared by Method II. In this case,
50 mg of insulin-loaded microspheres were separately
suspended in 25 mL of pH 1.2 and pH 7.4 release media.
In vitro release was performed in an incubator at 37°C
under constant stirring condition at 100 rpm. Sample
(1 mL aliquot) was withdrawn and replenished with fresh
release media at different time intervals. The release
samples were filtered and analyzed using HPLC as men-
tioned in HPLC method II.
Methods for testing in vivo efficacy of insulin-loaded
microspheres on diabetic-induced rats
Male Wistar rats (250 g) were housed in a 12–12 h light–
dark cycle under the constant temperature environment
of 22°C and relative humidity of 55%; these were allowed
free access to water and food during acclimatization.
To minimize diurnal variance of blood glucose, all the
experiments were performed in the morning. Diabetes
was induced with intravenous injection of 150 mg/kg
alloxan in saline (0.9% NaCl). Ten days after the treat-
ment, rats with frequent urination, loss of weight and
blood glucose levels (BGLs) higher than 300 mg/dL were
included in the experiments and 5% dextrose solution
was given in the feeding bottle for 1 day to overcome the
early hypoglycemic phase. After 72 h, the blood glucose
was measured by a glucometer. Then, the diabetic rats
(glucose level >300 mg/dL) were separated.
In order to investigate the effects of oral insulin-
loaded microspheres, 12 h fasted diabetic rats were
divided into three groups, each group containing six
rats were fed with insulin-loaded microspheres (20 IU)
and placebo microspheres as the control. Group 1 rats
received the placebo microspheres and Group 2 rats
received 20 IU of insulin-loaded Eudragit L100 micro-
spheres, whereas Group 3 rats received 20 IU of insulin-
loaded Eudragit L100/RS100 blend microspheres, both
dispersed in a mixture of 9 mL of 5% carboxy methyl
cellulose and 1 mL of 0.1 M HCl solution through oral
route using the oral feeding needle. The glucometer was
calibrated as per specifications mentioned in the strips
and precision of the glucometer was ±5%. Glucose was
Table 1. Gradient mode HPLC conditions.
Method I
Time (min)Conc. B (v/v)
0 16
2260
24100
2616
3216
Method II
Time (min)
0.10
5.0
20
21
25
Conc. B (v/v)
27
27
60
27
27

Page 4

980 R. C. Mundargi et al.
Drug Development and Industrial Pharmacy
measured on a drop of blood collected from the tail
vein spread over on the marked end of the strip before
and at different intervals up to 200 min after the oral
administration. Results were expressed as means ±
standard deviation.
Statistical analyses were done using the SPSS statisti-
cal package. Analysis of variance followed by the least
significant difference procedure was used for compari-
son of BGL and % inhibition from control, Eudragit L100
and Eudragit L100/RS 100 blend groups; the parameter,
p < 0.05 was considered significant.
Results and discussion
In applications involving oral delivery of insulin,
pH-sensitive polymers are used as carriers to protect the
insulin in acidic conditions and to release it in alkaline
medium. Therefore, for developing oral insulin formula-
tion, it is necessary to have a particulate device that will
change its behavior based on the pH of the medium. In
this regard, we have developed a pH-sensitive carrier sys-
tem using Eudragit L100 to effectively encapsulate insu-
lin in order to protect it from the enzymatic/proteolytic
degradation and to release it by the erosion mechanism
in a controlled manner, thus allowing higher concentra-
tion of insulin in the serum.
Some earlier reports are available on using Eudragit
polymers5,12,27–29 loaded with insulin for oral applica-
tions. For instance, Paul and Sharma5 have developed
insulin-loaded tricalcium phosphate microspheres
coated with pH-sensitive Eudragit polymer, wherein
the stability and conformational variations of insulin
as well as its biological activity in diabetic rats were
investigated. Agarwal et al.27 used the co-precipitation
technique to prepare insulin-loaded microspheres of
Eudragit L100 and studied the effect of variables such
as addition of salts in the precipitating medium and
ratio of polymeric solution to volume of precipitating
medium on dissolution and encapsulation efficiencies.
Damgé et al.28 developed the nanoparticles of biode-
gradable poly(ε-caprolactone) with non-biodegradable
Eudragit® RS 100 blend using poly(vinyl alcohol) as a
surfactant to achieve the encapsulation efficiency (EE)
of 96%.
Suitability of Eudragit L100 microspheres prepared
by double emulsion-solvent evaporation method was
evaluated by Jain and Majumdar1. Various parameters
were optimized to attain maximum EE and optimum
in vitro release profile. The microspheres retarded
the insulin release in gastric pH, but provided slow
release in alkaline pH condition of the upper intestine.
Gowthamarajan et al.12 developed the microspheres
of Eudragit L100 and S100 loaded with insulin, pro-
tease inhibitor and bile salts by the solvent diffusion
technique and microspheres of this study have shown
delayed release of insulin. The microspheres prepared
from Eudragit L100 or S100 along with 1% aprotinin
and 1% sodium glycocholate were used for in vivo
evaluation of hypoglycemic effect. The in vivo evalua-
tion of microspheres of Eudragit L100, 1% aprotinin and
1% sodium glycocholate showed a prolonged hypogly-
cemic effect for 3 h compared to intravenous injection
of bovine insulin.
Surface morphology
Surface morphology and size of the microspheres
were examined by SEM. Images of human insulin-
loaded microspheres prepared by method I at 5000×
and method II at 4000× and 2000× magnifications are
depicted in Figure 1A and 1B, respectively. Microspheres
prepared by Method I (2.5% polymer concentration)
are spherical in shape with agglomerations having
the slight rough surfaces. The size of microspheres
prepared by Method I is around 1–5 µm. Microspheres
prepared by Method II (9% polymer concentration) are
also spherical without agglomerations and their sizes
vary from 10 to 40 µm; the microspheres have a smooth
surface. Process parameters like polymer concentra-
tion, insulin loading and solvent play significant roles
in preparing the microspheres29.
EE
Encapsulation of insulin using Eudragit was achieved
by double emulsion-solvent evaporation technique25,
wherein insulin was first dissolved in 0.1 M HCl and
emulsified in Eudragit solution to form the primary
emulsion, which was then emulsified in light paraffin
oil. After the solvent evaporation, microglobules of
primary emulsion were precipitated and microspheres
were solidified. Microspheres produced by Method
I show the rough surfaces with EE of 52% and 26%,
respectively for Eudragit L100 and Eudragit L100/
RS100 (50:50) blend. When insulin in 0.1 M HCl was
homogenized with a mixture of methanol and DCM, a
low EE of 30% was observed. The solubility of insulin in
the external phase was assessed by dissolving 36 mg of
insulin in 0.5 mL HCl (0.1 M) and by adding the insulin
solution under stirring to the beaker containing 100 mL
paraffin oil with 0.5% Span-80. A white turbid solution
was observed, which could be attributed to insulin pre-
cipitation in the external oil phase, revealing the loss
of insulin due to surface bonding, thus resulting in loss
in the washing step. Further, changing the solvent from
methanol-DCM mixture to methanol and increasing
the polymer concentration from 2.5 to 9 wt. % in the
primary emulsion gave 95% yield with a EE of 63% in
case of bovine insulin, whereas for human insulin, the
yield and EE values were 92 and 64%, respectively. With
bovine insulin-loaded Eudragit RS100 microspheres,
the yield was 94%, whereas the EE was 33% (see results
in Table 2).
CD
CD provides qualitative as well as quantitative informa-
tion about the conformation of proteins30. In this work,
CD is used to probe the unfolding and folding of protein

Page 5

Oral insulin delivery system 981
© 2011 Informa Healthcare USA, Inc.
secondary structure either kinetically or at equilibrium.
Figure 2 displays different CD spectra of (A) pure bovine
insulin; (B) released bovine insulin from the developed
formulation at pH 7.4. Figure 3 shows the CD spectra
of (A) pure human insulin, while (B) shows that of the
released human insulin from Eudragit L100 at pH 7.4. The
CD spectra in basic pH revealed no significant difference
in the secondary structure of released insulin compared
to native insulin.
In vitro release study
In vitro release profiles of insulin-loaded Eudragit L100
microspheres at pH 1.2 and 7.4 shown in Figure 4 indicate
that at pH 1.2, the polymer shrinks, whereas at pH 7.4,
it swells to release insulin. At pH 1.2, Eudragit L100 has
released nearly 9% insulin, but in pH 7.4, almost 100%
release of insulin occurred in about 5 h. To minimize the
amount of insulin release in pH 1.2, the insulin-loaded
microspheres were formulated in the form of tablet
using 25 mg of microspheres and 12 mg of poly(vinyl pyr-
rolidone), 4 mg of magnesium stearate, and 184 mg of
microcrystalline cellulose as excipients. Further, the for-
mulated tablet was coated with 5% Eudragit L100 solution
in isopropanol by dip-coating method. By following this
procedure, insulin release in pH 1.2 was negligible, but
maximum insulin release occurred in pH 7.4 as shown in
Figure 5. On the other hand, blending of Eudragit RS100
with Eudragit L100 prevented the release of insulin at pH
1.2 followed by a maximum release in pH 7.4 as shown
in Figure 6.
Figure 7 shows in vitro release of bovine insulin-
loaded Eudragit L100 microspheres prepared by Method
II. In this formulation, 0.3% of insulin is released in pH
1.2, whereas in pH 7.4, 93% is released in 3 h. In case of
human insulin-loaded Eudragit L100 formulation, insu-
lin release in pH 1.2 is 0.8%, whereas in pH 7.4, it is as
high as 76% in 3 h (see Figure 8). In order to reduce the
initial burst release in pH 7.4 media, another formulation
was prepared using Eudragit RS100. However, the in vitro
release of bovine insulin indicated a slow release in pH
7.4, but high release of insulin was observed in pH 1.2 as
shown in Figure 9.
In vivo efficacy studies
Figure 10 displays the in vivo efficacy results of oral
insulin-loaded formulations. These results are plotted
as average BGL (mg/dL) versus time for placebo micro-
spheres and experimental formulations viz., insulin-
loaded Eudragit L100 microspheres and insulin-loaded
Eudragit L100/RS100 blend microspheres with dose of
A
B
Figure 1. (A) SEM image of group of microspheres of Eudragit L100
prepared by method I. (B) SEM image of group of microspheres of
Eudragit L100 prepared by method II.
Table 2. % Yield and encapsulation efficiency of the
microspheres for different formulations.
Insulin type Polymer type
Method I
HumanEudragit L100
HumanEudragit L100 + Eudragit
RS100 blend
Method II
BovineEudragit L100
HumanEudragit L100
BovineEudragit RS100
aEE, encapsulation efficiency.
% Yield% EEa
86
80
52
26
95
92
94
63
64
33