Content uploaded by Melgardt M De Villiers
Author content
All content in this area was uploaded by Melgardt M De Villiers on Jan 11, 2016
Content may be subject to copyright.
A
vailable online at www.sciencedirect.com
International Journal of Pharmaceutics 345 (2007) 142–153
Pharmaceutical Nanotechnology
Effect of pH on the solubility and release of furosemide from
polyamidoamine (PAMAM) dendrimer complexes
Bharathi Devarakonda
a,b
, Daniel P. Otto
a
, Anja Judefeind
a
,
Ronald A. Hill
b
, Melgardt M. de Villiers
a,∗
a
School of Pharmacy, University of Wisconsin-Madison, Madison, WI 53705, USA
b
Department of Basic Pharmaceutical Sciences, College of Pharmacy, The University of Louisiana at Monroe, Monroe, LA 71209, USA
Received 5 February 2007; received in revised form 18 May 2007; accepted 21 May 2007
Available online 24 May 2007
Abstract
The complexation of the practically insoluble drug furosemide (acidic pK
a
3.22) with lower generation PAMAM dendrimers showed a significant
release dependence on the ionization state of the drug. UV and FTIR studies suggested that the drug was localized in the interior of the dendrimer.
The dendrimer amine, amide and ester groups, demonstrated pH-dependent ionization as did the drug carboxylic acid group and it was proven that
the most efficient drug complexation was achieved in slightly acidic conditions (pH 4.0–6.0). At this pH, amide groups in the dendrimer cavities were
at least partially ionized to expose a positive charge whilst the furosemide carboxylic acid ionized to great extent (pH > pK
a
) resulting in electrostatic
complexation. Conversely, higher release rates were observed in acidic conditions (pH 1.2) where furosemide was virtually unionized, emphasizing
the importance of the drug ionization state in the determination of drug release. Despite the complex interactions between the dendrimer and drug
and its effects on release kinetics, the dendrimers resulted in higher solubility of the drug and contributed significantly to the array of available
techniques to increase the solubility of poorly water-soluble drugs that are very abundant in industry today. Complexation with low generation
PAMAM dendrimers (<generation 4) could provide opportunities to both increase drug solubility and tuning of the release profile for practically
insoluble drugs.
© 2007 Elsevier B.V. All rights reserved.
Keywords: PAMAM dendrimer; Furosemide; Complexation; Ionization; Solubility; Release
1. Introduction
Furosemide, 5-(aminosulfonyl)-4-chloro-2-((2-furanyl-
methyl)amino) benzoic acid (Fig. 1) is a loop diuretic that is
used orally in the treatment of edematous states associated
with cardiac, renal and hepatic failure and the treatment of
hypertension (Al-Obaid et al., 1989; Reynolds, 1989; Murray
et al., 1997). Since furosemide is a weak acid (reported
acidic pK
a
3.48) with a carboxylic acid functional group, its
aqueous solubility increases as function of medium pH from
0.18 mg/ml (pH 2.3) to 13.36 mg/ml (pH 10.0) (Rowbotham et
al., 1976). The major problem associated with the formulation
and effectiveness of the furosemide is its variable oral absorp-
tion (11–90%, Jackson, 1996) due to insufficient aqueous
∗
Corresponding author. Tel.: +1 608 890 0732; fax: +1 608 262 5345.
E-mail address: mmdevilliers@pharmacy.wisc.edu (M.M. de Villiers).
solubility at gastrointestinal pH, thus making solubility the
rate-determining step in the gastric absorption of furosemide
(Hammarlund et al., 1984).
Several techniques have been used to enhance the solu-
bility of drugs in solid dosage forms i.e. solubilization by
surfactants, co-solvents (Shihab et al., 1979), crystal modifi-
cation, pH-control (Doherty and York, 1989), solid dispersions
(Shin and Kim, 2003) and prodrug formation (Suescun et al.,
1998; Mombru et al., 1999). Among these techniques, com-
plexation with cyclodextrins has been widely investigated to
improve the solubility and dissolution properties of furosemide
(
¨
Ozdemir and Ordu, 1998; Ammar et al., 1999; Spamer et al.,
2002; Vlachou and Papaioannou, 2003). These studies reported
1.4–11-fold increases in solubility depending on the type as
well as percentage of the included cyclodextrin to produce the
complexes.
Recently the use of dendrimers, Fig. 1, as drug delivery sys-
tems has gained much attention in the pharmaceutical literature
0378-5173/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.ijpharm.2007.05.039
B. Devarakonda et al. / International Journal of Pharmaceutics 345 (2007) 142–153 143
Fig. 1. Molecular structures of furosemide and PAMAM dendrimers with ester- (G0.5) and amine-terminated (G1) surface functional groups.
and applications include increasing the solubility (Devarakonda
et al., 2004, 2005a,b) and bioavailability of drugs with poor water
solubility (Wiwattanapatapee et al., 2000; Milhem et al., 2000;
Kolhe et al., 2003; Chauhan et al., 2004), the delivery of DNA
and oligonucleotides across cell barriers (Poxon et al., 1996), and
as carriers for gastrointestinal drug delivery (Wiwattanapatapee
et al., 2000).
The aim of this study was to investigate whether EDA
core polyamidoamine (PAMAM) dendrimers could be used to
enhance the aqueous solubility of furosemide. The solubility
of furosemide at different pH was measured in the presence
of amine- and ester-terminated PAMAM dendrimers evolved to
different generations. In addition, the in vitro release of the drug
from complexes at pH’s commonly found in the gastrointestinal
tract is also reported.
2. Materials and methods
2.1. Materials
Furosemide USP (a white to slightly yellow, odorless,
crystalline powder that is practically insoluble in water),
ethylenediamine, methyl acrylate, methanol (HPLC grade),
sodium dihydrogen phosphate, citric acid, tromethamine,
sodium borate, potassium chloride, and standard pH buffers were
obtained from the Spectrum Chemical Company (Gardena, CA,
USA). Double deionized water was used for solubility studies
and HPLC analysis. PAMAM dendrimers were synthesized as
described previously (Esfand and Tomalia, 2001) or purchased
from Sigma–Aldrich (St. Louis, MO, USA).
2.2. Synthesis of PAMAM dendrimers
Ethylenediame (EDA) core PAMAM dendrimers were syn-
thesized using Tomalia’s divergent growth approach (Esfand and
Tomalia, 2001). The synthesis involves two consecutive chain-
forming reactions, the exhaustive Michael additions reaction,
and the exhaustive amidation reaction, repeating alternatively.
Michael addition of methyl acrylate to ethylenediamine in
methanol gives the ester terminated half generation dendrimers
designated, Gn.5. The exhaustive amidation reaction of ester-
terminated dendrimers with large excess of ethylenediamine
in methanol produce amine terminated full generation den-
drimers referred to as Gn. Repetition of Michael addition
and amidation reactions produces the next, higher generation
dendrimers.
In the present study, both amine-terminated full genera-
tion (G0–G3) and ester-terminated half-generation (G0.5–G2.5)
PAMAM dendrimers were used in the solubilization studies
of furosemide. Since the dendrimers are highly hygroscopic,
they were stored as 10% (w/w) solutions in anhydrous
methanol.
2.3. HPLC analysis of furosemide
The amount of furosemide in the solubility test samples
were analyzed by high performance liquid chromatography
as previously reported (Devarakonda and de Villiers, 2005).
The HPLC used consisted of a Spectrum System, AS 1000
autosampler and P2000 pump, (Thermo Separation Products,
Waltham, MA) equipped with a multiple wavelength UV detec-
144 B. Devarakonda et al. / International Journal of Pharmaceutics 345 (2007) 142–153
tor (UV 3000 detector) set at a wavelength of detection 272 nm.
Chromatographic separation was performed with a Supelco
®
Discovery RP Amide C
16
column (250 mm × 4.6 mm, 5-m
particles, Bellefonte, PA, USA) using a mobile phase of H
2
O:
acetonitrile: acetic acid (60:40:1, v/v); flow rate 1.0 ml/min;
injection volume 20 l. The retention time for furosemide was
approximately 7 min and the limits of detection 1.0 ng/ml.
Results represent the mean of three analyses, and the solu-
tions were protected from light to prevent photodegradation of
furosemide.
2.4. Preparation and characterization of
furosemide–dendrimer complexes
Furosemide was dissolved in methanol and then the den-
drimer was added. The initial molar ratios of furosemide to
dendrimer were 10:1, 20:1, 40:1 and 50:1. The reaction mix-
tures were stirred for 24 h in the dark and then evaporated using
a rotating evaporator to remove the methanol (Kolhe et al.,
2003). The precipitates were dried under vacuum in order to
remove methanol completely, followed by addition of deion-
ized water. Subsequently, this aqueous solution was stirred in
the dark for 24 h to extract the drug–dendrimer complex since
the dendrimer and drug–dendrimer complex is soluble in water
whilst furosemide is not. The solutions were filtered through a
0.2 m hydrophilic PTFE membranes (25 mm diameter, Omni-
pore, Millipore, Bedford, MA, USA) and then lyophilized to
remove water. The drug–dendrimer complex obtained was in
the form of off-white powders. The amount of furosemide in
the complexes was determined by HPLC (Devarakonda and de
Villiers, 2005).
Appropriate quantities (based on the furosemide content)
of the various complexes were dissolved in methanol to pro-
duce a concentration of 6 g/ml furosemide. The UV-spectra
(200–400 nm) of these solutions were obtained with a Shimadzu
MultiSpec spectrophotometer (Shimadzu, Kyoto, Japan). The
wavelengths of maximum absorption of furosemide in methanol
are at 226, 276 and 336 nm. The lower generation PAMAM
dendrimers (<G4), shows strong UV absorbance between
200–240 nm and some weaker absorbance for higher concen-
trations (>100 g/ml) between 240–280 nm (Devarakonda and
de Villiers, 2005). Any shift or suppression of the strong absorp-
tion peaks at 276 and 336 nm (furosemide UV absorption) was
ascribed to the formation of complexes.
G2.5 PAMAM-furosemide and G3 PAMAM-furosemide
complexes in the ratio of 1:1 (w/w, dendrimer/drug) were
prepared by dissolving furosemide and dendrimers for 24 h
in methanol followed by evaporation of the solvent. The
drug–dendrimer complexes, furosemide (alone) and G2.5 as
well as G3 PAMAM dendrimers (alone) were subjected to ATR-
FTIR analysis. The analysis was conducted using a Bruker
Equinox 55/S FTIR spectrophotometer (Bruker Optics, Inc., Bil-
lerica, MA, USA) equipped with a HeNe laser light source and an
ATR sampling accessory with ZnSe crystal. FTIR spectra were
obtained from 32 scans at 4 cm
−1
resolution in a wavenumber
range of 4000–650 cm
−1
. Processing of spectra was performed
with OPUS
TM
5.5 software.
2.5. pH-solubility profile of furosemide
The solubility of furosemide was measured in TRIS-buffer
with pH 2.0–8.0 and pH 10.0–12.0 at 30 ± 1.0
◦
C. The ionic
strength of the buffers was maintained at 0.5 M with potassium
chloride. From the linear portion of the pH-solubility profile
(Fig. 5a), the intrinsic solubility and acidic pK
a
of furosemide
was calculated using the Henderson–Hasselbalch equation for a
weak acid (Eq. (1))(Sinko, 2006).
1
H
3
O
+
=
S
K
a
S
0
−
1
K
a
(1)
where [H
3
O
+
] is the hydronium ion concentration, K
a
the
dissociation constant for weak acid, S the observed molar
solubility of furosemide, and S
0
is the intrinsic molar solubil-
ity of furosemide. By plotting 1/[H
3
O
+
] versus S (Fig. 5b),
K
a
and S
0
can be calculated from the slope (1/K
a
S
0
) and
intercept (−1/K
a
) of the linear regression curve, respec-
tively.
2.6. Solubility measurements
Solubility studies were performed using the Higuchi rotat-
ing bottle method (Higuchi and Connors, 1965). An excess
amount of furosemide was added to 5 ml amber colored vials
containing 3 ml of TRIS buffer (pH 2.0 and 4.0–6.0) with
increasing concentrations of the dendrimers and sealed. The
vials were rotated at 60 rpm while maintained at 30 ± 1.0
◦
C.
Preliminary experiments indicated that 24 h provided suffi-
cient time to achieve equilibrium. After 24 h, samples were
filtered through 0.45 m cellulose acetate filters (Osmon-
ics Inc., Minnetonka, MN, USA), diluted appropriately with
methanol and analyzed by HPLC. Phase solubility diagrams
were constructed by plotting the molar concentration of
furosemide (solubility) versus molar concentration of den-
drimers.
Mathematical analysis of the phase solubility diagrams pro-
vided estimates of the apparent equilibrium stability constants.
It was assumed that a 1:1 furosemide–dendrimer complex was
formed at those dendrimer concentrations where the total sol-
ubility versus dendrimer concentration curve was linear. The
apparent equilibrium stability constant, K
1:1
, was estimated by
regression analysis using the following equation (Higuchi and
Connors, 1965):
S
t
=
K
1:1
S
0
1 + K
1:1
S
0
L
t
+ S
0
(2)
where S
t
is the observed molar solubility of furosemide, K
1:1
is the equilibrium stability constant, S
0
is the intrinsic molar
solubility of furosemide, and L
t
is the total molar dendrimer
concentration.
If the total solubility versus dendrimer concentration curve
was parabolic, it was assumed that higher-order complexes were
formed. These data were analyzed by non-linear regression anal-
ysis using Eq. (3), assuming that only 1:1 and 1:2 complexes
B. Devarakonda et al. / International Journal of Pharmaceutics 345 (2007) 142–153 145
were present.
S
t
= S
0
+ S
0
K
1:1
L
t
+ S
0
K
1:2
L
2
t
(3)
where K
1:2
is the 1:2 equilibrium complexation constant.
Additional studies on the pH-dependent solubility and the for-
mation of furosemide–dendrimer complexes were conducted by
adding an excess of furosemide to each of the aqueous solutions
having a defined pH (2.0 and 7.0) and dissolving it (Chauhan
et al., 2004). After equilibration, the suspensions were cen-
trifuged at 8000 rpm for 10 min. The supernatant was filtered
through a membrane filter to obtain saturated furosemide solu-
tions at the respective pH values. The pH of the solutions was
noted, then 5 ml aliquots from the solutions were placed in Spec-
traPor Float-A-Lyzers (5 ml, Spectrum Laboratories, Rancho
Dominguez, CA, USA) with biotech cellulose ester membrane
tubing (MWCO = 2000) that was pre-sealed at one end and
attached to a floatable cap at the other. Before loading the sam-
ples the dialysis tubes were rinsed in DI water to remove the
preservative.
The molecular weight cut-off of the dialysis membrane was
selected because the dendrimers have a significantly higher
molecular weight, ensuring that the dendrimers would remain
inside the dialysis membrane, whereas the small molecular
weight drug would readily diffuse out of the dialysis bag. The
floatable cap facilitated easy loading of the samples to the dialy-
sis bags. The filled dialysis tubes were transferred to 50 ml glass
cylinders containing the saturated drug solutions which were
agitated in the dark using a shaker bath for 24 h to reach equi-
librium (concentration inside and outside the dialysis tube was
the same). Subsequently, 1 ml of the drug solution was removed
from the dialysis tube and replaced with 1 ml of either G2.5 or
G3 PAMAM dendrimer stock solutions in water so that the den-
drimer concentration inside the tubes was either 0.1%, 0.2% or
0.5% (w/v). The solutions were agitated in the dark (to prevent
photochemical degradation) for an additional 24 h preceding
the HPLC analysis of the concentration of furosemide inside
and outside phases of the dialysis tubes (Devarakonda and de
Villiers, 2005).
2.7. In vitro release studies
In vitro release of uncomplexed furosemide (control) and
from drug–dendrimer complexes was performed by the dialy-
sis technique. The dialysis technique made use of SpectraPor
Float-A-Lyzers (5 ml) with biotech cellulose ester membrane
tubing (MWCO = 2000) that was used after being rinsed in
deionized water to remove the preservative. Furosemide was
dissolved in methanol (2 mg/ml) and used as a control and
furosemide complexes were dissolved in deionized water at
a concentration equivalent to 2 mg/ml furosemide. After sam-
ples were taken from the control and dendrimer solutions
(5 ml), they were transferred immediately to the dialysis tubes.
The tubes were promptly placed in 500 ml tall form glass
beakers containing 400 ml of the dissolution medium main-
tained at 37
◦
C. The outer phase was stirred continuously
with a magnetic stirrer and sampling (1 ml) was made at spe-
cific time intervals followed by replenishment with 1 ml fresh
buffer.
The amount of drug in the samples withdrawn from the outer
phase was determined over a 12 h period and analyzed by HPLC
to characterize the release of furosemide. The release studies for
both the control and drug–dendrimer complexes were repeated
in simulated gastric fluid (SGF composed of an aqueous solu-
tion containing 0.2% sodium chloride and 0.7% hydrochloric
acid without pepsin, pH 1.2), simulated intestinal fluid (SIF
composed of an aqueous solution containing 0.68% monobasic
potassium phosphate and sodium hydroxide without pancreatin,
pH 7.4) and the USP dissolution medium for furosemide (phos-
phate buffer pH 5.8). The dialysis apparatus was completely
covered with aluminum foil to prevent photochemical degrada-
tion of furosemide.
2.8. Statistical analysis
To study the effect of generation size of amine-terminated
full-generation dendrimers and pH of the aqueous medium on
the solubility of furosemide, a two-factor factorial design (De
Muth, 1999) with n = 3 replicates was used where the aqueous
solubility of furosemide was measured for four levels of factor
A (pH of the medium; 2.0, 4.0, 5.0, and 6.0) and four levels of
factor B (generation size; G0, G1, G2, and G3). Similarly, the
effect of generation size of ester-terminated half-generation den-
drimers was measured for four levels of factor A (pH) and three
levels of factor B (generation size; G0.5, G1.5, and G2.5). The
effect of the two factors on the aqueous solubility of furosemide
was evaluated at a probability level of p = 0.05 (95% confi-
dence interval) using a commercial software package (Student
Statistix7.0, Analytical Software, Tallahassee, FL, USA). In the
presence of significant interaction among the factors, a one-
way ANOVA was used. Finally, differences between two sample
means were determined by pair-wise comparisons using a least
significant difference (L.S.D.) test performed with SPSS 10.0
for Windows
TM
(SPSS, Chicago, IL, USA).
3. Results and discussion
3.1. Structural characterization of the PAMAM dendrimers
The PAMAM dendrimers were characterized structurally
via
1
H- and
13
C-NMR, and mass spectral analysis (Table 1)
(Devarakonda et al., 2004, 2005a,b). Examples of the molecular
structures of the ester (G0.5) and the amine (G1) terminated
PAMAM dendrimers are shown in Fig. 1 (See also Esfand
and Tomalia, 2001). As shown (Table 1) molecular weight and
number of peripheral groups of dendrimers increase exponen-
tially with each generation, whilst the diameter demonstrated
an approximately linear increase (Esfand and Tomalia, 2001).
This implied that with each ensuing generation, the surface den-
sity of peripheral moieties (primary amines in full generation
dendrimers and ester groups in half generation dendrimers),
increased.
Full generation PAMAM dendrimers have primary amine
groups at each branch end and tertiary amine groups at each
146 B. Devarakonda et al. / International Journal of Pharmaceutics 345 (2007) 142–153
Table 1
Selected characteristics of ethylenediamine core PAMAM dendrimers taken from Esfand and Tomalia (2001) and the maximum number of furosemide molecules
incorporated per dendrimer molecule determined experimentally (see also Fig. 1)
Gn Molecular weight (Da) Diameter (nm) No. of tertiary
nitrogens
No. of surface
groups
No. of furosemide molecules/
dendrimer molecule
NH
2
-terminated
0 517 1.4 2 4 3
1 1430 1.9 6 8 5
2 3256 2.6 14 16 12
3 6909 3.6 30 32 20
COOCH
3
-terminated
0.5 1207 – 6 8 3
1.5 2809 – 14 16 6
2.5 5978 – 30 32 12
–: not available.
branching point. As the dendrimer generation n is increased,
Table 1 shows that the number of primary amines, N, grows as
N = cm
n
and the number of tertiary amines grows as N =cm
n
− 2,
where c is the number of branches of the core and m is the mul-
tiplicity of monomer. For EDA core PAMAM, c = 4 and m =2.
The PAMAM-Gn.5-COOCH
3
dendrimers have N =cm
n
ester
groups and N =cm
n
− 2 tertiary amines.
3.2. Furosemide encapsulation with PAMAM dendrimers
Either furosemide can be encapsulated in the interior or
surface bound with the PAMAM dendrimer amine groups.
Encapsulation/complexation ability of different amounts of
drug in dendrimer (PAMAM-Gn-NH
2
and PAMAM-Gn.5-
COOCH
3
) was studied to estimate the maximum number of
furosemide molecules that can be incorporated in a dendrimer
molecule. The initial molar ratios of furosemide to dendrimer
were 10:1, 20:1, 40:1 and 50:1. These values were selected on the
assumption that there should be a strong interaction between the
amine (both internal tertiary groups and surface groups) func-
tional groups of PAMAM and the COOH groups of furosemide.
The results in Fig. 2 show that the encapsula-
tion/complexation of furosemide in PAMAM-dendrimers
was successfully carried out and most likely occurred in the
interior of the dendrimer. The number of furosemide molecules
incorporated into the dendrimers increased with an increase
in dendrimer size for both NH
2
and COOCH
3
terminated
dendrimers. However, the number of furosemide molecules
associated with a single dendrimer molecule did not correspond
with the number of tertiary or surface amine groups present in
PAMAM-G3-NH
2
and PAMAM-Gn.5-COOCH
3
(values listed
in Table 1).
With the PAMAM-G3 that had 30 tertiary and 32 sur-
face NH
2
, 20 drug molecules were incorporated per dendrimer
molecule to yield the highest amount of bound furosemide
molecules. Correspondingly, only 12 drug molecules were incor-
porated into the PAMAM-G2.5 that had 30 tertiary and no
surface NH
2
groups. This indicated that there was a stronger
interaction between furosemide molecules and NH
2
group ter-
minated dendrimers compared to the COOCH
3
terminated
dendrimers. However, based on the number of furosemide
molecules that was associated with each dendrimer molecule the
complexation was most probably an interaction between interior
dendrimer amide groups and carboxylate ions of furosemide.
Since shifting and suppressions of UV-absorption and
fluorescence quenching have been reported as evidence of com-
plexation for other drug PAMAM dendrimer complexes the
UV-spectra of the drug, dendrimers and drug–dendrimer com-
plexes in methanol were recorded (Kleinman et al., 2000; Kolhe
et al., 2003). Evidence of complexation was observed as there
were significant changes in die UV-spectrum of furosemide in
the complexes as shown in Fig. 3. In addition to the peak maxima
at 276 nm and 330 nm being suppressed in the complexes, the
peak maximum at 276 nm was also shifted slightly to 271 nm.
Furthermore, ATR FTIR analysis supported the com-
plex formation between PAMAM dendrimer and furosemide.
FTIR spectra (2000–650 cm
−1
) of furosemide, G2.5 PAMAM,
furosemide-G2.5 PAMAM are shown in Fig. 4a and for G3
PAMAM and furosemide-G3 PAMAM in Fig. 4b. The charac-
teristic absorption bands of furosemide, C
O stretching of the
carboxylic acid group at 1672 cm
−1
,N H bending at 1592 cm
−1
and 1564 cm
−1
as well as the S O stretching of the sulfonamide
Fig. 2. Furosemide encapsulation or complexation with PAMAM dendrimers.
Molar ratios of furosemide to dendrimer added were 10:1, 20:1, 40:1 and 50:1.
Maximum number of furosemide molecules incorporated was 20 per 1 molecule
G3 PAMAM dendrimer.
B. Devarakonda et al. / International Journal of Pharmaceutics 345 (2007) 142–153 147
Fig. 3. UV-spectra of methanolic solutions: (1) furosemide, 6 g/ml; (2)
furosemide: G1.5 complex; (3) furosemide: G2 complex; (4) furosemide: G2.5
complex; (5) furosemide: G3 complex; (6) 100 g/ml PAMAM G3.
Fig. 4. (a) Expanded ATR FTIR spectra of (1) furosemide (2) G2.5 PAMAM
and (3) furosemide-G2.5-PAMAM. (b) Expanded ATR-FTIR spectra of (1) G3
PAMAM and (2) furosemide-G3-PAMAM.
group at 1321 cm
−1
(asymmetric) and 1141 cm
−1
(symmetric)
(Doherty and York, 1987; Al-Obaid et al., 1989) shifted to dif-
ferent wavenumber positions in the furosemide-G2.5 PAMAM
complex spectrum. The C
O and N H absorption bands shifted
to lower frequencies (1650 cm
−1
and 1553 cm
−1
, respectively)
that could be explained by intermolecular hydrogen bonding
between dendrimer groups and drug.
In contrast the frequency of both S
O absorption bands
shifted to higher frequency values (1330 cm
−1
and 1159 cm
−1
,
respectively). The reason might be the interruption of inter-
molecular hydrogen bonding between furosemide molecules
(Doherty and York, 1987) due to interaction of furosemide
with the dendrimers. Furthermore, shifts to lower frequen-
cies were also obtained for the C
O vibration of the ester
groups (1731–1709 cm
−1
), the amide I band (C O stretch-
ing, 1640–1608 cm
−1
) and amide II band (N H bending/C N
stretching, 1541–1553 cm
−1
) of the G2.5 PAMAM dendrimers
(due to hydrogen bonding between drug and dendrimers). More-
over, the occurrence of the C
O absorption band (ester group
of the G2.5 PAMAM) at the same wavelength (1731 cm
−1
)in
the furosemide-G2.5 PAMAM complex spectrum as in the pure
G2.5 PAMAM spectrum led us to postulate that not all ester
groups at the surface of the G2.5 PAMAM dendrimers were
involved in hydrogen bonding.
A similar pattern was observed for the complexation of
furosemide and G3 PAMAM dendrimers (Fig. 4b). As seen with
the furosemide-G2.5 PAMAM complex, the complexation of G3
PAMAM dendrimers and furosemide yielded frequency shifts to
lower frequencies for the C
O vibration (1636 cm
−1
) and N H
vibration (1545 cm
−1
) of furosemide due to hydrogen bonding
to the dendrimers. Again, the symmetric S
O stretching band
was obtained at a higher wavenumber (1155 cm
−1
) indicating
a reduction in the intermolecular hydrogen bonding between
furosemide molecules. The asymmetric S
O stretching band
could not be allocated precisely in the furosemide-G3 PAMAM
complex spectrum as an overlapping in absorption with the pure
G3 PAMAM occurred. The amide I band of G3 PAMAM also
absorbed at a lower wavenumber (1607 cm
−1
) when complexed
with furosemide indicating that the amide group is involved in
hydrogen bonding with furosemide. The characteristic absorp-
tion bands of PAMAM dendrimers (amide I, amide II and C
O
stretching of ester group) were confirmed by data in literature
(Liu et al., 2004; Popescu et al., 2006).
From the ATR-FTIR results it can be concluded that
changes occurred in a series of intermolecular hydro-
gen bonds (furosemide–furosemide, dendrimer–dendrimer,
furosemide–dendrimer) and that a complexation of furosemide
and dendrimer occurred by hydrogen bonding.
3.3. Effect of pH on the solubility of furosemide in PAMAM
dendrimer solutions
Additional insight into the furosemide–dendrimer interaction
was sought after the increase in solubility was noted upon further
addition of dendrimer to the complex. Therefore, an investiga-
tion was launched to investigate the effect of pH level on the
148 B. Devarakonda et al. / International Journal of Pharmaceutics 345 (2007) 142–153
Fig. 5. (a) pH-solubility profile of furosemide in TRIS buffers at 30 ± 1
◦
C.
(b) 1/[H
3
O
+
] vs. solubility plot of furosemide plotted according to the
Henderson–Hasselbalch equation (1).(y = 3.7 × 10
6
x − 1.6 × 10
3
, r
2
= 0.9815;
K
a
= 6.1 × 10
−4
M; pK
a
= 3.22; S
0
= 4.4 × 10
−5
M (14.6 g/ml)).
solubility. Fig. 1 shows the unionized and the ionized forms
of furosemide. The formation of the anion is responsible for
the increase in solubility (stronger polar interaction with water)
with increasing pH of the aqueous media as shown in Fig. 5. The
solubility showed a minimum of 10 g/ml at pH 2.0 and a max-
imum of 21.9 mg/ml at pH 8.0 (2000-fold increase), followed
by a marginal decrease in solubility (∼18 mg/ml) as increased
above pH 8.0. The curve demonstrated linearity in the region
pH 2.0–4.0 and the intrinsic solubility and acidic pK
a
of the
furosemide calculated with the Henderson-Hasselbalch Eq. (1),
were 14.62 and 3.22 g/ml, respectively (Fig. 5)(Sinko, 2006).
Solubility measurements of furosemide–dendrimer com-
plexes in pH-controlled media showed a positive, linear
correlation between solubility (Higuchi A
L
-type diagrams,
Fig. 6) and dendrimer concentration except at pH 5.0 with
G3 and at pH 6.0 with G2.5 and G3 dendrimers that showed
Higuchi A
p
-type diagrams (Fig. 7). A summary of the increase
in solubility, calculated as the percentage increase in the aque-
ous solubility of furosemide per mM of added dendrimer, is
given in Table 2. Furosemide is practically insoluble in water,
S
0
= 4.4 × 10
−5
M. The increased solubility of furosemide in the
presence of PAMAM dendrimers could be due to non-covalent
interactions between the drug and the macromolecules involv-
Fig. 6. Linear increase in the aqueous solubility of furosemide in the presence
of increasing concentrations of amine- (closed symbols) and ester-terminated
(open symbols) dendrimers at (a) pH 4.0, (b) pH 5.0 and (c) pH 6.0. Symbols
represent experimentally determined values and lines best fits using Eq. (2).
ing a variety of driving forces such as hydrogen bond formation,
electrostatic interactions, and hydrophobic bonding (Higuchi
and Connors, 1965).
The solubility Higuchi A
L
-type diagrams shown in Fig. 6
indicates that soluble complexes between furosemide and the
dendrimers have 1:1 stoichiometries (Higuchi and Connors,
1965). In the presence of G3 at pH 5.0 and pH 6.0 and G2.5 at
pH 6.0 (Fig. 7) Higuchi A
p
-type diagrams (r
2
= 0.991 ± 0.055)
were observed indicating that these dendrimers form multiple
B. Devarakonda et al. / International Journal of Pharmaceutics 345 (2007) 142–153 149
Fig. 7. Higuchi A
p
-type solubility profiles of furosemide in the presence of
increasing concentrations of dendrimers: G3 at pH 5.0; G2.5 at pH 6.0; G3 at
pH 6.0. Symbols represent experimentally determined values and lines best fits
using Eq. (3).
complexes with furosemide. Table 2 shows the equilibrium sta-
bility constants for the complexation between the drug and these
dendrimers calculated using Eqs. (2) and (3). The fits for Eq.
(2) (Higuchi A
L
-type diagrams) were excellent with a mean
r
2
= 0.989 ± 0.042 for all profiles. At a given pH, the stabil-
ity of the complexes increased with increasing generation size
with amine-terminated dendrimers also proving superior to the
ester-terminated dendrimers with the same number of surface
functional groups.
In addition, the stability constants for complexes of
furosemide with G1 and G0.5 dendrimers were similar to those
Table 2
Percentage increase in the solubility of furosemide per mM of added dendrimer
and equilibrium stability constants (1:1 and 1:2) for dendrimer-furosemide
complexes
pH Gn Increased solubility
(% per mM dendrimer
added)
K
1:1
(M
−1
) K
1:2
(M
−1
)
4 0 0.76 ± 0.02 6.38 –
1 7.72 ± 0.02 83.74 –
2 19.35 ± 0.9 222.3 –
3 111.3 ± 1.6 1720 –
0.5 3.87 ± 0.04 35.94 –
1.5 10.62 ± 0.5 111.5 –
2.5 27.92 ± 1.0 255.4 –
5 0 1.93 ± 0.01 12.21 –
1 5.12 ± 0.02 55.24 –
2 11.28 ± 0.06 133.8 –
3 102.8 ± 0.9 – 241689
0.5 3.70 ± 0.05 45.63 –
1.5 8.63 ± 0.06 114.7 –
2.5 21.88 ± 0.7 389.5 –
6 0 2.87 ± 0.02 36.67 –
1 5.84 ± 0.02 144.6 –
2 15.01 ± 1.0 4205 –
3 98.39 ± 2.0 – 191528
0.5 2.96 ± 0.04 40.24 –
1.5 7.59 ± 0.08 394.0 –
2.5 20.83 ± 0.3 – 199791
reported for furosemide–cyclodextrin complexes which ranged
the same orders of magnitude as the dendrimer complexes
reported here (
¨
Ozdemir and Ordu, 1998; Ammar et al., 1999;
Vlachou and Papaioannou, 2003). Significantly higher stability
constants were observed in the presence of G2, G3, G1.5, and
G2.5 dendrimers indicating the formation of more stable den-
drimer complexes. The fact that no burst release effects were
observed, Fig. 8, additionally suggested that the drug molecules
were encapsulated and not bound to a large extent to amines at
the periphery of the dendrimers.
Full-generation PAMAM dendrimers have primary amines
on the surface and tertiary amines in their internal cavities
whereas half-generation dendrimers expose ester groups on their
surface with internal tertiary amines (Fig. 1). The reported pK
a
values of the primary amines (surface groups) are 7.0–9.0 and
for the interior tertiary amines 3.0–6.0 (Tomalia et al., 1985;
Ottaviani et al., 1996; Chen et al., 2000; Kleinman et al., 2000;
Sideratou et al., 2000; Niu et al., 2003; Maiti et al., 2004, 2005).
At physiological pH 7.4, most of the primary amines are pro-
tonated, and at pH 4.0 all of the tertiary amines are protonated.
Therefore, the protonation level of the PAMAM could be altered
by changing the solution pH, which in turn significantly affected
the ability of the PAMAM dendrimer to interact with furosemide.
Furosemide is a weak acid with an ionizable carboxylic acid
group with an experimentally determined pK
a
of 3.22 and amino
sulfonyl, chlorine, and furonyl methyl amino groups (Fig. 1).
The COOH ionizable functional group might act as a counte-
rion for the dendrimer amine groups thereby participating in
the interaction between furosemide and the dendrimers. This
was substantiated because at pH 2.0, no significant increase
(p > 0.05) in the solubility of furosemide was observed because
furosemide is in the unionized form at this pH and hence could
not interact electrostatically with the ionized dendrimer moi-
eties. Conversely, at pH 4.0–6.0 the furosemide carboxylic group
would be in its ionized form and therefore, the increase in the
solubility of furosemide in the presence of dendrimers between
pH 4.0–6.0 would probably be due to the electrostatic inter-
actions between the positively charged tertiary amines of the
dendrimers and the negatively charged carboxylate anion of
the furosemide (Figs. 6 and 7). In this pH range most of the
surface NH
2
groups of the full generation PAMAM’s will be
protonated with the interior tertiary amines at least partially
protonated.
For a given generation size, the solubility of furosemide did
not increase significantly (p > 0.05) when the pH of the aqueous
medium was increased from pH 4.0–6.0 (Table 2), since the den-
drimers would have the same low surface charge density in this
pH-range. At a given pH; however, the solubility of furosemide
increased significantly with an increase in PAMAM dendrimer
generation size.
This could be due to an increase in the number of inte-
rior binding sites available with an increase in generation size
(Tables 1 and 2). The solubility of furosemide also increased
significantly (p < 0.05) in the presence of ester-terminated half-
generation dendrimers (Table 2). Conversely, at a given pH,
when compared to the full-generation dendrimers, the increase
in the solubility of furosemide was significantly lower in the
150 B. Devarakonda et al. / International Journal of Pharmaceutics 345 (2007) 142–153
Fig. 8. In vitro release of furosemide from furosemide: PAMAM dendrimer
complexes compared with the diffusion of a furosemide solution: (a) SGF, pH
1.2; (b) furosemide tablet dissolution medium (USP) pH 5.8; (c) SIF pH 7.4
(n = 3).
presence of the half-generation dendrimers. This was because
these dendrimers have non-reactive surface groups and the inte-
rior tertiary amines of the half-generation dendrimers are less
susceptible to protonation (Ottaviani et al., 1996). Therefore,
these dendrimers would interact with furosemide mainly via
hydrogen bond formation and only to a diminished extent by
electrostatic interactions.
Maiti et al. (2005) found that in the presence of a polar solvent
like water, significant penetration of water throughout the inte-
rior of the dendrimer causes the dendrimer structure to swell.
The extent of swelling depends on the extent of protonation.
At high pH > 12.0 there was no protonation, however the mere
presence of a good solvent such as water increased the size of
the dendrimer by almost 10–15% (Maiti et al., 2005). This trend
was observed in all the generations reported in this paper. At
neutral pH, all the primary amines are protonated and there was
no significant change in the dendrimer size as was evident from
very small increase in radius of gyration, indicating a limited
degree of swelling (Maiti et al., 2004, 2005).
As the pH was lowered further (pH < 4.0), almost all ter-
tiary amines were protonated and at that protonation level, the
dendrimer size increased almost 30–40% in water compared
to the case when no solvent is present. This results from the
favorable interaction of the solvent with the primary and ter-
tiary amines. It was concluded that, depending on the degree of
protonation, a significant portion of the counterions condensed
within the dendrimer, residing very close to the protonated sites
with some moieties solvated by the water. It is therefore possible
to assume that for ionized furosemide as the protonation level
of the dendrimer increased, more furosemide molecules would
reside inside the dendrimer since the dendrimer swelled more
at higher protonation levels (Kleinman et al., 2000; Maiti et al.,
2005).
In another study, Beezer et al. (2003) observed that all
drug/dendrimer complexes were unstable at pHs less than pH
7 and the bound substrate begins to precipitate after only 10 min
at pH 6. This suggests that it is the nitrogens within the den-
drimer that are important with respect to binding, for as they are
protonated (at pH 6 and below), the ability of the dendrimer to
bind its guest is lost. This implied that the binding was proba-
bly due to a simple ion-pairing mechanism. The internal tertiary
nitrogens are strongly basic, the pK
a
of an aqueous solution
of PAMAM dendrimers was measured as 9.5, and are there-
fore capable of deprotonating the acidic guest molecules. The
ensuing quaternized nitrogens can then bind to the resulting car-
boxylate counter-ions. From the observed pH dependence of
binding, we concluded that it was only possible for acidic guest
molecules to bind within the dendrimers interior, as nifedipine
and other small non-polar molecules could not be retained within
the dendrimer (Beezer et al., 2003; Devarakonda et al., 2004).
Further studies on pH-dependent furosemide solubility were
conducted by preparing saturated solutions of furosemide in
aqueous medium at pH 2.0 and 7.0 and the effect of G2.5 and G3
PAMAM addition was observed. The solubility of furosemide
was increased from 9.9 to 9640 g/ml (1000-fold increase) by
increasing the pH from 2.0 to 7.0. A contrasting trend was
observed when different concentrations of the dendrimers were
added to the saturated solutions (devoid of excess undissolved
drug) (Table 3), i.e. furosemide solubility was decreased with
increasing PAMAM dendrimer concentrations due to the inter-
action between the drug and dendrimer to form complexes that
did not diffuse out of the dialysis tubes. This effect resulted in an
B. Devarakonda et al. / International Journal of Pharmaceutics 345 (2007) 142–153 151
Table 3
Effect of the addition of G2.5 and G3 PAMAM dendrimers to saturated aqueous furosemide solutions at different pH on the solubility of furosemide
pH Solubility (g/ml) Dendrimer added (%) G2.5 PAMAM G3 PAMAM
pH
a
Solubility (g/ml) pH
a
Solubility (g/ml)
2 9.9 ± 1.2 0.1 3.0 8.8 ± 0.4 3.1 8.4 ± 0.4
0.2 3.4 8.5 ± 0.5 3.6 8.1 ± 0.2
0.5 3.6 7.9 ± 0.2 3.7 7.5 ± 0.1
7 9640 ± 78 0.1 8.3 2345 ± 21 8.4 878 ± 13
0.2 8.2 1141 ± 11 8.5 439 ± 8
0.5 8.0 469 ± 8 8.6 177 ± 4
a
Change in pH upon addition of dendrimer.
increase in the concentration of furosemide inside the dialysis
tube whilst the concentration outside the tube decreased.
For both dendrimers the relative decrease in furosemide sol-
ubility was significantly higher at pH 7.0 (75–98%) than at pH
2.0 (11–24%) since more drug molecules dissolved in the pH
7.0 solution (more ionized), resulting in more pronounced phase-
separation after addition of dendrimer solution. This observation
was consistent with the change in solubility seen with a change
in the protonation level of the drug and dendrimer as a function
of pH seen in Figs. 6 and 7. The NH
2
terminated PAMAM den-
drimer at pH ∼ 8.5 (pH changed from 7 to 8.5 after the addition of
the dendrimer) interacted strongly with the ionized drug because
the amount of free drug in solution decreased by 90–98%
depending on the amount of dendrimer added (0.1–0.5%) as
seen in Table 3.
At pH ∼ 8 the COOCH
3
terminated G2.5 dendrimer also
significantly interacted with ionized furosemide because the
solubility decreased by 75–95% with the addition of the
same concentrations of dendrimer. However, the decrease in
furosemide solubility, Table 3, was significantly greater for the
G3 compared to the G2.5 dendrimer.
3.4. Release studies
The in vitro release of furosemide from the furosemide–
PAMAM dendrimer complexes was determined in SGF with-
out pepsin (pH 1.2), SIF without pancreatin (pH 7.4) and the
USP dissolution medium for furosemide (phosphate buffer pH
5.8). The results are shown in Fig. 8. The release profiles of the
furosemide–dendrimer complexes were compared with that for
pure furosemide using a similarity factor (Eq. (4))(Moore and
Flanner, 1996) to see whether the rate and extent of furosemide
release was different from that of the solutions.
f
2
= 50 log
⎧
⎨
⎩
1 +
1
n
n
i
(R
i
− T
i
)
2
−0.5
100
⎫
⎬
⎭
(4)
where n is the number of time points, R
t
is dissolution value of
the reference sample at time t, and T
t
is the dissolution value
of the test sample at time t. The similarity factor is a sim-
ple model independent approach using mathematical indices to
define differences and similarities between dissolution profiles.
These factors are divided from Minkowski-differences (average
absolute differences) and mean-square difference respectively.
When f
2
= 100 then the test and reference mean profiles are iden-
tical. The test and reference products are not equivalent when
there is larger than 10% difference between dissolution pro-
files, indicated by a similarity factor f
2
<50(Moore and Flanner,
1996).
The f
2
values of the control solutions were different at the dif-
ferent pH values. The order of release was pH 7.4 ≥ pH 5.8 > pH
1.2. Although release at pH 1.2 seemed not to be the same as
shown in Fig. 8(a) calculated f
2
values ranged from 53 to 66
indicating that the release profiles were within 10% over the
extent of the release profiles. Therefore, the ranking of the con-
trol solution, drug-G2.5-PAMAM and drug-G3-PAMAM were
equivalent at pH 1.2. This was perhaps not surprising considering
that furosemide would be virtually unionized at this pH.
At pH 5.8 f
2
values indicated that the release from the G2.5
complex was similar to that of the G3 complex (f
2
= 65), however
these profiles differed by more than 10% from that of the control
solution (f
2
= 36 and 32). The release from the G2.5 (f
2
= 16) and
G3 (f
2
= 13) complexes at pH 7.4 was significantly slower than
from the control solution with the release from the G2.5 and
G3 complex showing very similar release profiles (f
2
= 61). In
addition, the release from the complexes at pH 7.4 was even
slower than from the complexes at pH 5.8 and pH 1.2 (f
2
< 50).
The release results indicate that in addition to solubilizing
furosemide the PAMAM dendrimers could be utilized to con-
trol the release of this drug. At low pH, the dendrimers release
the drug very fast and the diffusion out of the dialysis tube is the
same as for the control solution of the drug. In contrast at neutral
pH the interaction between the drug and dendrimer was signifi-
cantly enhanced and decelerated the release of the drug. Again
the differences in release rate can be correlated to a combina-
tion effect of the ionization state of the drug and the PAMAM
dendrimers.
4. Conclusion
In this study, the increase in the solubility of the practically
insoluble drug furosemide in combination with low generation
water-soluble polyamidoamine (PAMAM) dendrimers (G <4)
was demonstrated for the first time. FTIR and solubility studies
of the complexes suggested that furosemide was encapsulated
in the dendrimer cavity. The increase in solubility of the drug
by the dendrimers was determined by the pH of the aqueous
medium, generation size, and type and number of internal ter-
152 B. Devarakonda et al. / International Journal of Pharmaceutics 345 (2007) 142–153
tiary amine groups. The increase in the solubility of furosemide
in the presence of dendrimers was primarily due to the electro-
static interactions between the positively charged tertiary amines
of the dendrimers and negatively charged carboxylate anion of
furosemide, resulting in a favorable solubilization effect.
Overall, the results showed that the PAMAM dendrimers
could be exploited to improve the solubility and dissolution
of furosemide, however that the enhancement depended on the
choice of dendrimer, generation size, and surface functional
group of the dendrimer. Careful selection of the counterion
species exposed by the dendrimer markedly influenced the
drug–dendrimer interaction during the drug release process at a
given pH. In acidic dissolution media, both drug and dendrimer
were fully protonated and abolished the electrostatic associa-
tion with a resultant high drug release. Conversely at neutral
pH, the ionization states of both the drug and dendrimer favored
electrostatic interaction with resultant slower release.
From this study we conclude that the ionization state of the
drug molecule dominated the solubility and release properties of
the drug–dendrimer complex and could be exploited in future for
the formulation of pharmaceutical products containing poorly
water-soluble drugs. In addition, to provide even more effec-
tive encapsulation, higher generations of dendrimers could be
employed to compare it to the dendrimers studied here.
Acknowledgements
We thank University of Louisiana-Monroe, University of
Wisconsin-Madison and NSF (#0210298 “Nanoengineered
Shells”) for supporting this work.
References
Al-Obaid, A.M., Al-Shammary, F.J., Al-Rashood, K.A.M., Mian, M.S., 1989.
Analytical profile of furosemide. In: Florey, K. (Ed.), Analytical Profiles of
Drug Substances, vol. 18. Academic press, New York, pp. 153–193.
Ammar, H.O., Ghorab, M., El-Nahhas, S.A., Emara, L.H., Makram, T.S., 1999.
Inclusion complexation of furosemide in cyclodextrins. Part 1. Effect of
cyclodextrins on the physicochemical characteristics of furosemide. Phar-
mazie 54, 142–144.
Beezer, A.E., King, A.S.H., Martin, K., Mitchel, C.J., Twyman, L.J., Wain,
C.F., 2003. Dendrimers as potential drug carriers; encapsulation of acidic
hydrophobes within water soluble PAMAM derivatives. Tetrahedron 59,
3873–3880.
Chauhan, A.S., Jain, N.K., Diwan, P.V., Khopade, A.J., 2004. Solubility
enhancement of indomethacin with poly(amidoamine) dendrimers and tar-
geting to inflammatory regions of arthritic rats. J. Drug Target. 12, 575–583.
Chen, W., Tomalia, D.A., Thomas, J.L., 2000. Unusual pH-dependent polarity
changes in PAMAM dendrimers: evidence for pH-responsive conforma-
tional changes. Macromolecules 33, 9169–9172.
De Muth, J.E., 1999. Basic Statistics and Pharmaceutical Statistical Applica-
tions. Marcel Dekker, New York, 596 pp.
Devarakonda, B., de Villiers, M.M., 2005. Development and validation of an
HPLC method involving solid-phase extraction for the analysis of hydropho-
bic drugs in the presence of polyamidoamine (PAMAM) dendrimers. J. Liq.
Chromatogr. Relat. Technol. 28, 2325–2338.
Devarakonda, B., Hill, R.A., de Villiers, M.M., 2004. The effect of PAMAM den-
drimer generation size and surface functional group on the aqueous solubility
of nifedipine. Int. J. Pharm. 284, 133–140.
Devarakonda, B., Hill, R.A., Liebenberg, W., Brits, M., de Villiers, M.M.,
2005a. Comparison of the aqueous solubilization of practically insoluble
niclosamide by polyamidoamine (PAMAM) dendrimers and cyclodextrins.
Int. J. Pharm. 304, 193–209.
Devarakonda, B., Li, N., de Villiers, M.M., 2005b. Effect of polyamidoamine
(PAMAM) dendrimers on the in vitro release of water insoluble nifedip-
ine from aqueous gels. AAPS Pharm. Sci. Technol. 6 (Article 63),
(http://www.aapspharmscitech.org).
Doherty, C., York, P., 1987. Evidence for solid- and liquid-state interactions
in a furosemide-polyvinylpyrrolidone solid dispersion. J. Pharm. Sci. 76,
731–737.
Doherty, C., York, P., 1989. Microenvironmental pH control of drug dissolution.
Int. J. Pharm. 50, 223–232.
Esfand, R., Tomalia, D.A., 2001. Laboratory Synthesis of
Poly(amidoamine)(PAMAM) dendrimers. In: Frechet, J.M.J., Toma-
lia, D.A. (Eds.), Dendrimers and Other Dendritic Polymers. John Wiley,
New York, pp. 587–604.
Hammarlund, M.M., Paalzow, L.K., Odlind, B., 1984. Pharamcokinetics of
furosemide in man after intravenous and oral administration. Application
of moment analysis. Eur. J. Clin. Pharmacol. 26, 197–207.
Higuchi, T., Connors, K.A., 1965. Phase-solubility techniques. In: Reilly, C.N.
(Ed.), Advances in Analytical Chemistry and Instrumentation. John Wiley,
New York, pp. 117–212.
Jackson, E.K., 1996. Goodman and Gilman’s the Pharmacological Basis of
therapeutics. McGraw-Hill, New York, pp. 698.
Kleinman, M.H., Flory, J.H., Tomalia, D.A., Turro, N.J., 2000. Effect of pro-
tonation and PAMAM dendrimer size on the complexation and dynamic
mobility of 2-naphthol. J. Phys. Chem. B 104, 11472–11479.
Kolhe, P., Misra, E., Kannan, R.M., Kannan, S., Lieh-Lai, M., 2003. Drug
complexation, in vitro release and cellular entry of dendrimers and hyper-
branched polymers. Int. J. Pharm. 259, 143–160.
Liu, D., Gao, J., Murphy, C.J., Williams, C.T., 2004. In situ attenuated total reflec-
tion infrared spectroscopy of dendrimer-stabilized platinum nanoparticles
adsorbed on alumina. J. Phys. Chem. B 108, 12911–12916.
Maiti, P.K., Cagin, T., Lin, S.-T., Goddard, W.A., 2005. Effect of solvent and
pH on the structure of PAMAM dendrimers. Macromolecules 38, 979–
991.
Maiti, P.K., Cagin, T., Wang, G., Goddard, W.A., 2004. Structure of
PAMAM dendrimers: generations 1 through 11. Macromolecules 37, 6236–
6254.
Milhem, O.M., Myles, C., McKeown, N.B., Attwood, D., D’Emanuele, A., 2000.
Polyamidoamine starburst dendrimers as solubility enhancers. Int. J. Pharm.
197, 239–241.
Mombru, A.W., Mariezcurrena, R.A., Suesclin, L., Pardo, H., Manta, E., Prandi,
C., 1999. Two absorption furosemide prodrugs. Acta Crystallogr. C 55,
413–416.
Moore, J.W., Flanner, H.H., 1996. Mathematical comparison of dissolution
profiles. Pharm. Technol. 20, 64–74.
Murray, M.D., Haag, K.M., Black, P.K., Hall, S.D., Brater, D.C., 1997. Variable
furosemide absorption and poor predictability of response in elderly patients.
Pharmacotherapy 17, 98–106.
Niu, Y., Sun, L., Crooks, R.M., 2003. Determination of intrinsic proton binding
constants for poly(amidoamine) dendrimers via potentiometric pH titration.
Macromolecules 36, 5725–5731.
Ottaviani, M.F., Montalti, F., Romanelli, M., Turro, N.J., Tomalia, D.A., 1996.
Characterization of starburst dendrimers by EPR. 4. Mn(II) as a probe of
interphase properties. J. Phys. Chem. 100, 11033–11042.
¨
Ozdemir, N., Ordu, S., 1998. Improvement of dissolution properties of
furosemide by complexation with beta cyclodextrins. Drug Dev. Ind. Pharm.
24, 19–25.
Popescu, M.C., Filip, D., Vasile, C., Cruz, C., Rueff, J.M., Marcos, M., Serrano,
J.L., Singurel, G., 2006. Characterization by Fourier transform infrared spec-
troscopy (FT-IR) and 2D IR correlation spectroscopy of PAMAM dendrimer.
J. Phys. Chem. B 110, 14198–14211.
Poxon, S.W., Mitchell, P.M., Liang, E., Hughes, J.A., 1996. Dendrimer delivery
of oligonucleotide. Drug Deliv. 3, 255–261.
Reynolds, J.E.F. (Ed.), 1989. Martindale: the Extra Pharmacopoeia, 29th ed. The
Pharmaceutical Press, London, pp. 987–991.
Rowbotham, P.C., Stanford, J.B., Sugden, J.K., 1976. Some aspects of the pho-
tochemical degradation of frusemide. Pharm. Acta Helv. 51, 304–307.
B. Devarakonda et al. / International Journal of Pharmaceutics 345 (2007) 142–153 153
Shihab, F.A., Ebian, A.R., Mustafa, R.M., 1979. Effect of polyethylene glycol,
sodium lauryl sulfate and polysorbate-80 on the solubility of furosemide.
Int. J. Pharm. 4, 13–20.
Shin, S.-C., Kim, J., 2003. Physicochemical characterization of solid dispersion
of furosemide with TPGS. Int. J. Pharm. 251, 79–84.
Sideratou, Z., Tsiourvas, D., Paleos, C.M., 2000. Quarternized
poly(propylenimine) dendrimers as novel pH-sensitive controlled–release
systems. Langmuir 16, 1766–1769.
Sinko, P.J. (Ed.), 2006. Martin’s Physical Pharmacy and Pharmaceutical Sci-
ences, 5th ed. Lippincott, Williams and Wilkens, New York, pp. 231–265.
Spamer, E., M
¨
uller, D.G., Wessels, P.L., Venter, J.P., 2002. Characterization
of the complexes of furosemide with 2-hydroxypropyl--cyclodextrin and
sulfobutyl ether-7--cyclodextrin. Eur. J. Pharm. Sci. 16, 247–253.
Suescun, L., Mariezcurrena, R.A., Mombru, A.W., Gonzalez, O.A., Manta, E.,
Prandi, C., 1998. Three isostructural furosemide prodrugs. Acta Crystallogr.
C 54, 1911–1915.
Tomalia, D.A., Baker, H., Dewald, J., Hall, M., Kallos, G., Martin, S., Roeck,
J., Ryder, J., Smith, P., 1985. A new class of polymers: starburst-dendritic
macromolecules. Polym. J. 17, 117–132.
USP XXIX, 2006. US Pharmacopeial Convention, Rockville, MD.
Vlachou, M., Papaioannou, G., 2003. Preparation and characterization of the
inclusion complex of furosemide with hydroxypropyl-beta-cyclodextrin. J.
Biomater. Appl. 17, 197–206.
Wiwattanapatapee, R., Carreno-Gomez, B., Malik, N., Duncan, R., 2000.
Anionic PAMAM dendrimers rapidly cross adult rat intestine in vitro: a
potential oral delivery system? Pharm. Res. 17, 991–998.
