Azo-containing urethane analogues for colonic drug delivery: Synthesis, characterization and in-vitro evaluation

Article abstract of DOI:10.1211/0022357011776063

A urethane-based analogue containing an azo aromatic linkage in the backbone was synthesized for use in colon-specific delivery systems by reacting toluene-2,6-diisocyanate with a mixture of an aromatic azo diol, (bis-4-hydroxyphenyl)-4,4′-diazobiphenyl, poly(ethylene glycol) (Mn = 4000; number-average molecular weight) and 1,2-propanediol (propylene glycol). The resultant compounds (UR-1 and UR-2) were characterized by IR spectroscopy, ¹H NMR spectroscopy, DSC studies, X-ray diffraction studies and molecular weight determination by gel permeation chromatography. The compounds exhibited low molecular weight, lacked film-forming properties and crystallinity in the structure. An in-vitro bacterial degradation test to demonstrate the susceptibility of azo bond to bacterial enzymes was performed using media inoculated with lactobacillus culture. The results indicated degradation of films by azoreductase. In-vitro permeation of 5-aminosalicylic acid was studied in control and lactobacilli-treated films. The permeability of the lactobacilli treated films was significantly increased suggesting the potential of these compounds for application in colonic targeting.

Full text of DOI:10.1211/0022357011776063

Communications
Journal of
Pharmacy and Pharmacology
JPP 2001, 53: 895–900
# 2001 The Authors
Received August 30, 2000
Accepted February 2, 2001
ISSN 0022-3573
Azo-containing urethane analogues for colonic drug
delivery: synthesis, characterization and in-vitro
evaluation
M. S. Chavan, V. P. Sant and M. S. Nagarsenker
Abstract
A urethane-based analogue containing an azo aromatic linkage in the backbone was
synthesized for use in colon-specific delivery systems by reacting toluene-2,6-diisocyanate with
a mixture of an aromatic azo diol, (bis-4-hydroxyphenyl)-4,4h-diazobiphenyl, poly(ethylene
glycol) (Mn l 4000; number-average molecular weight) and 1,2-propanediol (propylene
glycol). The resultant compounds (UR-1 and UR-2) were characterized by IR spectroscopy, 1H
NMR spectroscopy, DSC studies, X-ray diffraction studies and molecular weight determination
by gel permeation chromatography. The compounds exhibited low molecular weight, lacked
film-forming properties and crystallinity in the structure. An in-vitro bacterial degradation test
to demonstrate the susceptibility of azo bond to bacterial enzymes was performed using media
inoculated with lactobacillus culture. The results indicated degradation of films by azoreduc-
tase. In-vitro permeation of 5-aminosalicylic acid was studied in control and lactobacilli-treated
films. The permeability of the lactobacilli treated films was significantly increased
suggesting the potential of these compounds for application in colonic targeting.
Introduction
Improved drug delivery systems are required for drugs currently in use to treat
localized diseases of the colon. Such diseases are most effectively treated by local
delivery of anti-inflammatory agents to the large intestine. However, these agents
are either delivered inefficiently and unpleasantly by large volume rectal enema or
exhibit unacceptable side-effects after oral administration because of absorption in
upper regions of the gastrointestinal tract (Friend 1991). Thus, oral colon-specific
drug delivery is of increasing importance. A number of different physiological
triggers are proposed for targeted release of drugs in the colon. One approach is to
utilize the plethora of microorganisms in the lower gastrointestinal tract (Lloyd et
al 1994). The ability of microorganisms to selectively metabolize certain carbo-
hydrates or azo groups has been effectively exploited in the development of
polymers for colonic drug delivery systems. A universal approach to utilize
bacterial degradation of azo bonds to achieve site-specific release has been the basis
of the synthesis of polymers suitable for coating (Saffran et al 1986), and use of
hydrogels with azo–aromatic cross-links (Bronsted & Kopecek 1992). Earlier
studies in the development of azo polymers have shown that when dosage
forms coated with polymers reached the colon, the azo groups were reduced to
hydrazo groups by the action of azoreductases liberated by intestinal microflora
Department of Pharmaceutics,
Bombay College of Pharmacy,
Kalina, Mumbai 400 098, India
M. S. Chavan, V. P. Sant, M. S.
Nagarsenker
Correspondence: M. S.
Nagarsenker, Department of
Pharmaceutics, Bombay College
of Pharmacy, Kalina, Mumbai
400 098, India. E-mail:
mnagarsenker!hotmail.com
Funding and acknowledgement:
V. P. Sant is thankful to CSIR,
New Delhi for the award of a
Senior Research Fellowship. The
authors acknowledge the help
of Dr K. Iyer, Bombay College of
Pharmacy, Mumbai, India.
895

896
M. S. Chavan et al
(Kimura et al 1992). However, preparation and coating AMX-500 FT NMR at 298K with tetramethylsilane as
of such cross-linked azo polymers was very difficult an internal standard. The DSC scans were obtained
because of the poor solubility of commonly used solv- using a differential scanning calorimeter (Shimadzu
ents. This suggested a need for the development of Thermal Analyser, DT-40) equipped with a monitor
soluble, colon-degradable polymers.
and a computerized thermal analyser system 40-1
This study presents the synthesis of urethane-based printer. The instrument was calibrated using indium as
analogues containing azo aromatic groups that can be a reference standard. Powder X-ray diffraction patterns
degraded by colonic bacteria. With the aim of synthe- of the compounds were recorded using a Phillips X-ray
sizing polymers with film-forming properties and a good diffractometer (PW 1130\00) with a copper target, volt-
solubility profile, azo containing urethane analogues age 45 KV, current 30 mA and a scanning rate of
1
−
were prepared by reaction of (bis-4-hydroxyphenyl)- 2m min . Ultraviolet spectra were recorded on a
4,4h-diazobiphenyl (HPDB), poly(ethylene glycol) Shimadzu UV-160A UV-visible spectrophotometer.
(PEG) and toluene 2,4-diisocyanate (TDI). Films cast
using these analogues in combination with another film-
forming polymer, ethylcellulose (EC) were studied for
in-vitro degradation and in-vitro permeation of 5-
aminosalicylic acid (ASA).
Preparation of HPDB
HPDB was synthesized as indicated in Figure 1.
Benzidine (5 g, 27.13 mmol) was dissolved in a mixture
of concentrated hydrochloric acid (14 mL) and water,
and the solution was cooled to below 5mC. Chilled
solution of sodium nitrite (4.12 g, 59.7 mmol) in water
was added slowly to the benzidine solution under con-
tinuous stirring and the temperature of the reaction
mixture was maintained between 0 and 5mC. The dia-
zonium salt thus formed was maintained in an ice bath
at below 5mC. An alkaline solution of phenol (5.11 g,
54.27 mmol) in sodium hydroxide was prepared and
cooled to 5mC. Chilled diazonium salt was added drop-
wise to cold phenol solution under continuous stirring.
After complete addition, the reaction mixture was al-
lowed to stabilize in the ice bath for 45 min with oc-
casional stirring. The solution was then filtered under
vacuum and the product collected. The product was
purifiedbydissolvingitinanaqueoussolutionofsodium
hydroxide (10% w\v) followed by re-precipitation by
slow addition of concentrated hydrochloric acid. The
precipitated product was recrystallized in a mixture of
ethanol and water (90:10), separated by vacuum filtra-
tion and finally air-dried.
Materials and Methods
Materials
The aromatic azo diol HPDB was synthesized as indi-
cated in Figure 1. PEG-4000 and 1,2-propanediol (pro-
pylene glycol; PG) were procured from S. D. Fine
Chemicals Ltd, Mumbai, India. TDI was a gift sample
from Expanded Incorporation, Mumbai, India. ASA
was a gift sample from Unichem Laboratories Ltd,
Mumbai, India. EC (45 cps) was a gift from Crown
Chemical Company. All other chemicals and solvents
used were of analytical reagent grade. The culture of
lactobacilli was obtained by inoculating curd in fluid
thioglycolatemediumandwasmaintainedanaerobically
at 37mC.
IR spectra of the compounds were recorded on a
Jasco FTIR 5300 spectrometer using the KBr pellet
1
method. H NMR spectra were recorded on a Bruker
Preparation of urethane analogue with an azo
aromatic linkage
Synthesis of the urethane analogue was carried out as
outlined in Figure 2. HPDB (3 g, 27.3 mmol) and PEG
(4 g, 1 mmol) were placed in a 100-mL three-necked
round-bottomed flask equipped with a mechanical
stirrer and a dropping funnel. The flask was evacuated
for 12–16 h and then flushed with dry nitrogen gas. PG
(5.18 g, 245 mmol) was added to the flask and the
contents heated to 120mC in an oil bath with stirring
Figure
(HPDB).
1
Synthesis of (bis-4-hydroxyphenyl)-4,4h-diazobiphenyl

897
Synthesis of azo-containing urethane analogues
Evaluation of film-forming properties and
determination of molecular weight
Urethane analogues (UR-1 and UR-2) were dissolved in
methanol, water, acetone, dimethylformamide (DMF),
THF, methylene chloride and a mixture of methylene
chloride and methanol (7:3) to form a 10% w\w
solution. These solutions were then poured on different
substrates such as glass, teflon and release liner mem-
brane (No. 1022; 3M Company, Germany) and film
formation evaluated after the solvent was allowed to
evaporate at room temperature. The same procedure
was repeated with addition of plasticizers, diethyl phth-
alate and dibutyl phthalate (10% by weight of com-
pound) to the solution of urethane analogue. After
complete evaporation of the solvent, attempts were
made to remove films from the substrates.
Figure 2 Synthesis of the urethane analogues containing an azo
aromatic linkage. HPDB, (bis-4-hydroxyphenyl)-4,4h-diazobiphenyl;
PEG, poly(ethylene glycol); PG, propylene glycol; TDI, toluene 2,4-
diisocyanate. A, B and C represent portions of the polymer resulting
from reaction of TDI with HPDB, PG and PEG respectively.
The molecular weight of compounds was determined
by gel permeation chromatography (GPC) on a Waters
instrument equipped with an R 401 differential refrac-
tometer.
A Styragel HR 1, 3 and 4 column
(7.8 mmi300 mm; Waters) was used with THF as
eluent. Molecular weights were calibrated with poly-
styrene standards.
Table 1 Composition of urethane analogues, UR-1, UR-2 and
UR-3.
Analogue
HPDB (m)
PEG-4000 (m)
PG (m)
In-vitro bacterial degradation test
UR-1
UR-2
UR-3
27.3
27.3
27.3
1.0
2.0
4.0
245.0
244.0
242.0
Preparation of isolated films
The compound prepared did not show any film-forming
properties. Films of the compound were therefore pre-
HPDB, (bis-4-hydroxyphenyl)-4,4h-diazobiphenyl; PEG, poly(ethyl- pared by mixing it with another film-former, EC (45
ene glycol); PG, propylene glycol.
cps). A 10% w\w of the polymer solution (EC:UR, 4:1)
in the mixture of methylene chloride and methanol (7:3)
was prepared. The films were cast on a release liner
membrane (No. 1022; 3M Company). After complete
evaporation of the solvent, the films were removed from
the release liner and dried to a constant weight at 45mC.
under a nitrogen flow. TDI (13.227 g, 273 mmol) was
added to the flask very slowly from a dropping funnel
over a period of 1 h. Stirring was continued until the
viscosity increased. At the end of the reaction, 10 mL Bacterial degradation test
ethanol was added to the reaction mixture to stabilize Lactobacillus species was selected for degradation stud-
the terminal isocyanate groups. Completion of the re- ies as these organisms produce and release azoreductase
action and absence of HPDB and TDI was monitored (Van den Mooter et al 1992). The films were incubated
by thin-layer chromatography (TLC). TDI, HPDB and for 4–7 days at 37mC anaerobically in the V.P.I. diluent
reaction mixture were spotted on a pre-coated silica fluid (Van den Mooter et al 1992) medium inoculated
plate and the chromatogram was developed using a with the lactobacillus culture. Films incubated in media
mixture of chloroform and methanol (9.9:0.1) as mobile withoutmicroorganismsunderidenticalconditionswere
phase. The products were purified by dissolving in used as controls. After incubation, films were rinsed
tetrahydrofuran (THF) and re-precipitation with n- thoroughly with distilled water, dried and evaluated for
hexane. The precipitate was filtered, washed with n- in-vitro degradation by visual inspection, microscopy
hexane and dried at room temperature in air. Urethane and spectroscopy.
analogues with three different compositions were pre-
pared as outlined in Table 1.
For UV measurements, the untreated films (films
without incubation), control films (films incubated in

898
M. S. Chavan et al
absenceof lactobacilli)andtreated films(films incubated (-NlN-) group and an aromatic moiety, respectively.
in presence of lactobacilli) were dissolved quantitatively Figure 4 gives the DSC scan of HPDB, which shows a
in a mixture of methylene chloride and methanol (7:3) peak at 299.7mC. The onset temperature of transition
and the solution diluted appropriately to give an ac- was 259mC, the maximum peak transition temperature
1
−
curate concentration of approximately 100 µg mL . was 299.7mC and the recovery temperature was 323.4mC
Solutions were scanned on a UV-visible spectrophoto- for the peak. The positive value of enthalpy indicates an
meter in the region of 200–800 nm and λmax recorded. exothermic change.
Absorptivity (1%, 1 cm) at the respective λmax was
calculated for all films. Untreated, treated and control
films were also inspected visually and examined under a
Preparation of a urethane analogue with an
azo aromatic linkage
microscope, (Photomicrograph (Hund Wetzlar), with
magnification 10i) for the presence of degradation. Of the three different analogues, UR-3 was obtained as
Reflecting light was used for observations.
a product with elastic consistency (foam-like structure)
which was very difficult to process and therefore not
evaluated further. UR-1 and UR-2 were obtained as
light brown, fine powders, which were further charac-
terized. The absence of HPDB and TDI in UR-1 and
UR-2 was determined by TLC. The TLC plates, when
observed under UV light, (254 nm) showed no spots
corresponding to Rf values of HPDB and TDI indicating
the absence of both these reactants. The solubility of
UR-1 and UR-2 was evaluated in different solvents.
Both compounds were freely soluble in DMF and THF.
UR-2 was freely soluble in methanol, acetone, ether and
methylene chloride, whereas UR-1 was only slightly
soluble in these solvents. Both compounds were insol-
uble in water. The solubility profile of UR-2 was better
compared with UR-1 probably because of the greater
contribution of glycols in the formation of UR-2.
IR spectra of both UR-1 and UR-2 were almost
In-vitro permeation studies
To evaluate the effect of azoreductase on the permea-
bility of polymer films, in-vitro permeation of ASA
across the films was studied in a Keshary-Chien cell.
Films were prepared from a 4% w\w solution of EC
and UR-2 (4:1) in a mixture of methylene chloride and
methanol (7:3). Films were incubated anaerobically for
2 days at 37mC with lactobacillus culture in fluid thio-
glycolate medium. Films were then washed with distilled
water and mounted in Keshary-Chien cells. The receptor
compartment of cells contained phosphate buffer (pH
7.4) and the donor compartment was filled with 2 mL
1
−
ASA solution (1 mg mL ) in phosphate buffer (pH
7.4). Receptor fluid was stirred continuously with a
magnetic needle and maintained at 37p1mC with a
circulatingwater bath. Samples(1 mL)were periodically
withdrawn from the receptor compartment and replaced
with the same volume of fresh buffer. Samples were
analysed spectrophotometrically at 331 nm after ap-
propriate dilution and the amount of ASA permeated
across the film was calculated. The experiments were
performed in triplicate and meansps.d. for each time
point were calculated. Similar permeation studies were
carried out for control films (films incubated in absence
of lactobacillus culture).
similar and showed peaks at 3308 cm 1 (-N-H stretching
−
1
vibrations), 1716 cm (-O- in the urethane linkage),
−
1
−
1597 cm (corresponding to -O-ClO stretching in
1
amide group), 1450 cm (aromatic azo functionality
−
1
1
−
−
-NlN-), 1070 cm and 1225 cm (aryl -O-C linkage
and -O-CH2 linkage), and 768 cm 1 (aromatic ring). The
−
1
IR spectrum of UR-2 is shown in Figure 3. H NMR
spectra of both polymers in d6-dimethylsulfoxide reveal-
ed a doublet for CH3 of the PG unit at δ l 1.21–1.23, a
singlet for -CH2O of the PEG unit at δ l 3.51–3.55, a
doublet for -CH2 of the PG unit at δ l 4.07–4.09, a
quartet for -CH of the PG unit at δ l 5.01 and multiplets
for aromatic rings at δ l 6.95–7.16, 7.51–7.83 and 8.15.
The DSC scans of UR-1 and UR-2 are shown in
Figure 4. These thermograms were similar showing
characteristic peaks at 277.2mC and 279.1mC in the case
of UR-1 and UR-2, respectively. Onset transition temp-
Results and Discussion
Preparation of HPDB
HPDB was obtained as a black, shiny, fine powder (mp eratures were 222.5mC (UR-1) and 223.1mC (UR-2),
285–288mC). Figure 3 shows the IR spectrum of HPDB maximum peak transition temperatures were 277.2mC
1
−
with characteristic bands at 3408 cm (a broad peak), (UR-1) and 279.1mC (UR-2), and recovery temperatures
1
1
−
−
1468 cm and 725 cm , that are indicative of the pres- were 326.8mC (UR-1) and 328.1mC (UR-2). The nega-
ence of a phenolic hydroxy functionality aromatic azo tive value of enthalpy for UR-1 and UR-2 indicated an

899
Synthesis of azo-containing urethane analogues
HPDB
51
43
35
2h
Figure 5 X-ray diffractogram of urethane analogue UR-2.
27
19
11
UR-2
short range of order and low degree of crystallinity of
the analogue.
Evaluation of film-forming properties and
determination of molecular weight
No films could be formed by these analogues on any of
the substrates used (glass, teflon and release liner) even
aftertheadditionofplasticizers. Evaporationofsolvents
resulted in deposits of weak conglomerates which was
suggestive of low molecular weights. This was supported
by molecular weight studies. The number- and weight-
average molecular weights of UR-1 and UR-2 were
determined by GPC. UR-1 was found to have a number-
average molecular weight (Mn) of 1663.2 and weight-
average molecular weight (MW) of 2085.74. UR-2 had
a Mn of 1620.47 and MW 1989.12.
4000
3000
2000
1500
1000
400
Wavenumber
Figure 3 IR spectra of (bis-4-hydroxyphenyl)-4,4h-diazobiphenyl
(HPDB) and UR-2.
299.7
HPDB
UR-1
In-vitro bacterial degradation test
Films incubated in a lactobacillus culture showed a
colour change from brown to yellow, whereas control
films did not show any colour change. Schacht et al
(1996) reported a colour change from brown to pink for
films composed of polyamides and polyurethanes. The
EC:UR (4:1) films were also observed under a micro-
scope after anaerobic incubation for 5 and 7 days.
Photomicrographs of films after incubation for 5–7 days
indicated loss of integrity, which could be attributed to
degradation of the azo linkages of UR-1 or UR-2. As
these films were prepared with an impermeable polymer
(EC) of a high viscosity grade (45 cps), the films formed
were relatively hydrophobic and thick. This could be the
UR-2
277.2
279.1
20
180
340
Temp (°C)
Figure 4 DSC scans of (bis-4-hydroxyphenyl)-4,4h-diazobiphenyl
(HPDB) and urethane analogues UR-1 and UR-2.
endothermic change in contrast to HPDB which showed reason for the longer time required for the degradation
a positive peak with exothermic change. Thus DSC of the azo linkages of the UR-1 and UR-2 components
scans of urethane analogues were markedly different of the films.
from the DSC scan of HPDB indicating its absence in
UV spectrophotometric studies of untreated, treated
these analogues. The XRD pattern of urethane ana- and control films were also carried out. All the films
logue, UR-2 (Figure 5) showed a diffused spectrum with showed a peak at λmax 397 nm. The UV scans of EC:UR
the absence of any significant peaks. This indicated the (4:1) films incubated with or without lactobacilli are

900
M. S. Chavan et al
60
50
1.00
Without lactobacilli
With lactobacilli
40
30
20
10
0
Control
Treated
0.0
200
0
2
4
6
8
10
600
Wavelength (nm)
Time (h)
Figure 6 UV scan of EC:UR (4:1) film incubated in presence
(treated) and absence (control) of lactobacilli.
Figure 7 In-vitro permeation of 5-aminosalicylic acid across films
of EC:UR-2 (4:1) incubated in fluid thioglycolate medium.
given in Figure 6. Values of absorptivity (1%,1 cm) for
untreated films (films without incubation), films incu-
bated in media containing lactobacilli and films incub-
ated in media without lactobacilli were 32.58, 9.16 and
27.46, respectively. The results indicate a small decrease
in absorptivity of films incubated in media inoculated
without lactobacilli compared with untreated films. Sig-
nificant reduction in the absorptivity of treated films, to
the extent of 66%, was observed compared with control
films. The absorptivity of the film is ascribed to the
characteristic absorption of the azo aromatic chromo-
phore of the urethane analogues. The reduction of azo
linkages by microorganisms would therefore decrease
this absorptivity (Kimura et al 1992). Thus, UV spec-
troscopy results of the in-vitro bacterial degradation
tests indicated that azo bonds in the film were reduced
by the bacterial enzyme, azoreductase.
acrylate polymers was significantly enhanced by bac-
terial degradation of the azo polymer (Van den Mooter
et al 1992). Further, Shantha et al (1995) showed that
the release of 5-fluorouracil from azo polymeric hydro-
gels was faster and was dependent on the concentration
of azo polymer in the hydrogel.
In conclusion, the urethane analogues synthesized in
the present study, although devoid of film-forming
properties on their own, still possessed significant po-
tential for use in colonic drug delivery in combination
with other standard film-forming polymers.
References
Bronsted, H., Kopecek, J. (1992) Hydrogels for site specific drug
delivery to the colon: in vitro and in vivo degradation. Pharmacol.
Res. 9: 1540–1545
Friend, D. (1991) Colon specific drug delivery. Adv. Drug Deliv. Rev.
7: 149–201
In-vitro permeation studies
Kimura, Y., Makita, Y., Kumagai, T., Yamane, H., Kitao, T., Sastani,
H., Kim, S. I. (1992) Degradation of azo containing polyurethane
by the action of intestinal microflora: its mechanism and application
as a drug delivery system. Polymer 33: 5294–5299
The permeation of ASA, an anti-inflammatory agent
used in the treatment of ulcerative colitis, was performed
across the EC:UR-2 (4:1) film in Keshary-Chien cells.
Results of the permeation studies are shown in Figure 7.
Incontrol films(incubatedwithoutlactobacillus culture)
only 5.16p0.26% ASA (n l 3) permeated across the
films in 8 h, whereas 43.51p5.77% ASA (n l 3) per-
meated across treated films (incubated in presence of
lactobacillus culture). These results suggest that during
incubation in presence of lactobacilli, azoreductase ac-
tivity on the UR-2 component of the films resulted in a
significant increase in permeability. These results are in
accordance with previous reports which have shown
that permeability of caffeine from azo-containing meth-
Lloyd, A. W., Martin, G. P., Soozandehfar, S. H. (1994) Azopoly-
mers: a means of colon specific drug delivery Int. J. Pharm. 106:
255–260
Saffran, M., Kumar, G. S., Savariar, C., Burnham, J. C., Williams, F.,
Neckers, D. C. (1986) A new approach to the oral administration of
insulin and other peptide drugs. Science 233: 1081–1084
Schacht, E., Gevaert, A., Gevaert, E., Kenaway, E. R., Molly, K.,
Verstraete, W., Adriaensens, P., Carleer, R., Gelan, J. (1996)
Polymers for colon specific delivery. J. Control. Rel. 39: 327–338
Shantha, K. L., Ravichandran, P., Panduranga Rao, K. (1995) Azo
polymeric hydrogels for colon targeted drug delivery. Biomaterials
16: 1313–1318
Van den Mooter, G., Samyn, C., Kinget, R. (1992) Azo polymers for
colon specific drug delivery. Int. J. Pharm. 87: 37–46