Evaluation of alginate dialdehyde cross-linked gelatin hydrogel as a biodegradable sealant for polyester vascular graft.

Evaluation of alginate dialdehyde cross-linked gelatin hydrogel
as a biodegradable sealant for polyester vascular graft
Saraswathy Manju, Chirathodi Vayalappil Muraleedharan, Adathala Rajeev,
Attipettah Jayakrishnan, Roy Joseph
Biomedical Technology Wing, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Poojapura,
Thiruvananthapuram 695012, Kerala, India
Received 28 January 2010; revised 13 January 2011; accepted 10 February 2011
Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.b.31843
Abstract: Vascular grafts are devices intended to replace
compromised arteries in the body and grafts made of poly-
ethylene terephthalate (PET) fabric have been used mainly
for synthetic grafting procedures involving medium to large
diameter vascular grafts. Though porosity of the graft per-
mits tissue in-growth, it would lead to bleeding through the
graft walls immediately after implantation. So it is essential
to seal the pores either by preclotting with patient’s own
blood or by other sealing materials prior to implantation in
order to prevent blood leakage through the graft wall. Biode-
gradable hydrogel materials are ideal candidates for this pur-
pose. Apart from sealing the pores, they offer biocompatible
and low-thrombogenic surfaces when coated on vascular
graft. In the present study, a biodegradable hydrogel, derived
from oxidized alginate and gelatin, has been deposited on
PET grafts by dip coating and were characterized for its effi-
cacy on sealing the pores of the graft. Water permeability in
the static and pulsatile conditions, burst strength, in vitro cell
culture cytotoxicity, hemocompatibility, and endothelial cell
adhesion and proliferation of the coated grafts were investi-
gated. Results showed that the alginate dialdehyde cross-
linked gelatin hydrogel was nontoxic, hemocompatible, and
was efficient in sealing the pores of the graft. Blood perfu-
sion study showed that when hydrogel-coated grafts were
exposed to blood for 30 min, they showed little affinity
toward platelets or leukocytes. Hemolytic potential of PET
was significantly reduced when it was coated with hydrogel.
Improved adhesion and proliferation of endothelial cells
were observed when PET grafts were coated with hydrogel.
Results also showed that coating with hydrogel did not
affect the burst strength of the PET graft. VC 2011 Wiley
Periodicals, Inc. J Biomed Mater Res Part B: Appl Biomater 00B:000–
000, 2011.
Key Words: vascular graft, porosity, periodate oxidation, alginate
dialdehyde cross-linked gelatin hydrogel, hemocompatibility
INTRODUCTION
As the incidence and frequency of vascular procedures
increase over the years, researchers are actively looking for
modified natural materials as a compromise between auto-
grafts and purely synthetic grafts made of materials such as
expanded polytetrafluoroethylene and polyethylene tereph-
thalate (PET).1 Vascular grafts are used for the treatment of
blood vessel aneurysms and fistulas, as well as replacing
diseased arteries in other locations in the body. Whenever
possible, the best choice for vessel replacement is an auto-
graft where sections of the patient’s healthy blood vessels
(usually veins) are harvested and implanted in the required
location. Many patients, however, especially those with pre-
existing vascular disease or patients that have already had
autograft procedures do not have blood vessels that are
healthy enough to adequately serve as replacements. In
these cases, the most common form of treatment has been
the use of synthetic polymeric materials to form either per-
manent or resorbable replacements for the damaged ves-
sels.1 In cases where the graft can be of a large diameter
(greater than 6 mm), the synthetic materials have been
effective. However, in situations where a smaller vessel di-
ameter is required, the synthetic materials cannot be used
due to high rates of stenosis and thrombus formation.2 One
possible solution is to use natural materials like collagen, ei-
ther modified or combined with a synthetic material, to
form a graft that more closely mimics the body’s natural
function and has low thrombogenicity and low incidence of
restenosis.3
It has been recognized that the porosity of vascular graft
plays an important role in their long-term potency and bio-
logical performance.4–6 One main disadvantage of highly po-
rous vascular graft is their high permeability to blood dur-
ing implantation, which may result in severe blood leakage
through the graft wall. Therefore, the pores of the graft
must be sealed before implantation to obtain zero or near
zero permeability. Sealing of the pores are often achieved
either by preclotting with patient’s own blood or by other
sealing materials. Preclotting with patient’s own blood is
not favored as there is often residual clot formation. The
resulting graft hemorrhage is troublesome, increasing opera-
tive time and need for blood transfusion. Many studies have
Correspondence to: R. Joseph; e-mail: rjoseph1965@rediffmail.com
Contract grant sponsor: SCTIMST (TDF-funding)
V
C 2011 WILEY PERIODICALS, INC. 1

been done to develop vascular grafts that are blood tight
during implantation, thus eliminating the need for preclot-
ting of the graft and sufficiently porous to facilitate the tis-
sue in-growth and biological healing.7 Most commonly used
method includes coating or impregnation of porous graft
with a biodegradable component. Coated or impregnated
graft is blood tight during implantation. Owing to the grad-
ual dissolution and degradation within the body, the resorb-
able material creates increasingly large pores in the initially
porous graft, allowing in-growth of endothelial cells (EC).8,9
Various proteins have been used as biodegradable compo-
nents for coating or impregnation of the grafts. They include
fibrin, albumin, gelatin, dextran, and collagen.9–15 The grafts
pretreated with these proteins now represent a high per-
centage of vascular grafts. However, proteins have some
drawbacks, they are generally unstable, hard to obtain in
the pure form, not easy to cross-link and control resorption
rate, expensive, and difficult to render compatible with
standard storage and sterilization procedures. Polysaccha-
ride-based soft materials are a potential solution to avoid
this risk of contamination and can be used as a coating ma-
terial for porous vascular graft prosthesis.16 Hydrogel derived
from natural proteins and polysaccharides are biocompatible
in nature and are widely used for medical applications such
as plasma expanders, blood substitutes, bone healing pro-
moters, wound dressings, and drug delivery.17 They resemble
extracellular matrices of tissue comprised of various amino
acids and sugars based macromolecules thus have the poten-
tial to direct the migration, growth, and organization of cells
during the tissue regeneration.17,18 A hydrogel derived from
alginate dialdehyde (ADA) and gelatin is an in situ forming
biodegradable polymer of well-known biocompatibility and
bioresorbability without employing any extraneous cross-link-
ing agent.19 However, rapid cross-linking and gelation is pos-
sible between oxidized alginate and gelatin in the presence of
borax.20 In this study, hydrogel derived from ADA and gelatin
in the presence of borax was used for coating porous woven
PET grafts with a view to seal the pores and to obtain
implantable grafts with no blood leakage and low-thromboge-
nicity. Coated grafts were studied in vitro to evaluate its
potential to remain impervious to blood while retaining po-
rosity for tissue in-growth and biological healing.
MATERIALS AND METHODS
Materials
Woven PET vascular grafts were obtained from M/s. TTK
Healthcare, Thiruvananthapuram, India. Sodium alginate
(medium viscosity grade, viscosity of 2% solution ¼ 3500
cps at 25�C), gelatin (Type A, Bloom 300, molecular weight
¼ 100,000), sodium meta periodate, sodium tetraborate
decahydrate (borax), 3-(4,5-dimethylthiazole-2-yl)-2,5, di-
phenyl tetrazolium bromide (MTT), were obtained from
Sigma Chemicals Co., St. Louis, MO. Dialysis tubes (Spectra/
PorVR, MWCO-3500) were from Spectrum Laboratories, CA.
All other chemicals and reagents were procured from Spec-
trochem Pvt. and were of analytical grade.
Periodate oxidation of sodium alginate
Sodium alginate, 20 g (0.10 mol) was dispersed in 100-mL
ethanol. To this dispersion, 10.8 g (0.05 mol) sodium meta
periodate dissolved in 100-mL distilled water was added
and stirred magnetically in the dark at 25�C for 6 h. The
reaction mixture was then dialyzed against distilled water
until the dialyzate was periodate free. Periodate diffusion
into distilled water was checked using dilute silver nitrate
solution. When the presence of periodate was no longer
detectable, dialyzate was freeze dried to get ADA.21
Characterization of ADA
The ADA obtained was characterized by Fourier transform
infrared (FT-IR) spectroscopy and chemical analysis. FT-IR
spectrum was recorded using a Nicolet 5700 (Thermoelec-
tron corporation, USA) spectrophotometer. Chemical analysis
was used for the determination of aldehyde content. Residual
aldehyde content after hydrogel formation was also deter-
mined. The estimation was based on the principle that the
aldehyde group in ADA would react with hydroxyl amine
hydrochloride resulting in the release of hydrochloric acid
(HCl). One mole of HCl would be released per mole of alde-
hyde group reacted. The HCl released could be estimated by
titrating against standard sodium hydroxide (NaOH) solution.
To estimate the aldehyde content in ADA, 0.10 g ADA was
dissolved in 25 mL 0.25N hydroxyl amine hydrochloride pre-
pared in distilled water. About 100 lL methyl orange indica-
tor (0.05% solution) was added and allowed to stand for 2 h.
The mixture was then titrated against 0.1N NaOH solution
taken in the burette. At the end point, the color of the solu-
tion changed from red to yellow and the number of moles of
NaOH reacted was calculated. This is equivalent to the num-
ber of moles of aldehyde groups present in the sample. Resid-
ual aldehyde content in the hydrogel was also estimated. For
this, the coating formulation was separately cast in a Petri
dish. Hydrogel formed was taken out, crushed into lumps,
dried, and powdered. The powder obtained was dispersed in
hydroxylamine hydrochloride solution and the above proce-
dure was repeated for the determination of residual aldehyde
content.
Coating of hydrogel on PET graft
Aqueous solutions of ADA and gelatin were mixed together in
the presence of catalyst (borax) to form ADA cross-linked gel-
atin (ADA-X-G) hydrogel. The hydrogel derived from ADA and
gelatin was coated on the PET graft using a two-step process
as described below: Adequate amounts of ADA and gelatin
were dissolved in 0.1M borax solution and distilled water,
respectively, to make 3% solutions of each. Thoroughly
washed and dried PET grafts were immersed in ADA solution
for 5 min and partly dried in an oven at 40–45�C for 20 min.
It was then immersed in the gelatin solution for about 5 min
and further dried. The process was repeated twice to get uni-
form coating on the graft surface. After thorough washing
with distilled water, grafts were dried in an air oven.
Weight gain on hydrogel coating
The weight of hydrogel coated on the graft was determined
gravimetrically and is expressed as a percentage:
Weight gain on hydrogel coating=% ¼ ðW2 � W1Þ � 100=W1;
2 MANJU ET AL. ALGINATE DIALDEHYDE CROSS-LINKED GELATIN HYDROGEL

where W1 is the weight of the uncoated graft and W2 is the
dry weight of hydrogel-coated graft. Weight gain data were
collected from four samples and the mean weight gain was
determined.
Scanning electron microscopy
Surface morphology of woven PET grafts coated with ADA-
X-G hydrogel was examined using scanning electron micro-
scope (SEM; Hitachi, model S-2400, Japan). The specimens
were coated with gold to facilitate observation of fabric and
hydrogel deposition on the graft surface.
Degradation of hydrogel coated on graft surface
To study the degradation profile of hydrogel deposited on
PET grafts, samples of size 1 cm � 1 cm were cut from
ADA-X-G coated PET grafts, weighed, and incubated with 10
mL of phosphate buffered saline (PBS, pH 7.4) at 37�C for
15 days. Graft samples were taken out at definite time inter-
vals, washed, and dried. Weight loss (%) due to degradation
was calculated using the equation:
Weight loss ð%Þ ¼ ½ðW2 � W3Þ=ðW2 � W1Þ� � 100
where W1 is the weight of PET graft before coating, W2 is
the dry weight of hydrogel-coated PET graft, and W3 is the
dry weight of hydrogel-coated PET graft after aging in PBS.
To find out the effect of reduction of Schiff bases pro-
duced during hydrogel formation, hydrogel-coated samples
of size 1 cm � 1cm were immersed in 10 mL of 1M sodium
borohydride solution and incubated at 37oC for 2 h. The
samples were taken out, dried in an oven at 40–45�C, and
weighed. Dried samples were incubated with 10 mL of PBS
at 37�C and degradation studies were conducted for a pe-
riod of 15 days as described above. Mean of five samples
were taken for both systems and the results reported.
Burst pressure measurements
Four samples each of PET grafts (2 cm in length) before and
after coating were used for burst pressure measurements.
Testing was performed on dry grafts. The burst pressure test
set up was made according to the American National standard
for vascular graft prosthesis, ANSI/AAMI VP 20, and testing
was carried out on Instron series IX (model 3345) universal
testing machine. Machine was operated in the compression
mode at a cross-head speed of 125 mm/min and the load at
which sample burst occur was noted. The burst pressure was
calculated using the expression below:
Burst pressure ðKg=cm2Þ ¼ ðBurst load in kN � 1000Þ=
ð9:81 � Area of load application in cm2Þ
Water permeability of graft under steady
physiological pressure
Water permeability was measured as the amount of water
leaked per unit area and time under a physiological pres-
sure of 120 6 5 mmHg and at test temperature 30�C 6 3�C
as given in the section 4.3.1.2 of the American National
Standard for Vascular Graft Prostheses, ANSI/AAMI VP20.
Samples of 3 cm length were flattened and loaded onto the
holder and adjusted the flow rate until a pressure of 120 6
5 mmHg was achieved. From each sample, three flow rate
measurements were taken between 30 and 90 s after ini-
tiating flow through the sample at 120 6 5 mmHg pressure.
The water permeability was calculated using the equation:
Water permeability ðmL=min=cm2Þ ¼ Q � k
where Q is the flow rate through the sample in mL/min and
k the available area for flow in cm2.
Permeability under simulated pulsatile
pressure conditions
Permeability of PBS through the hydrogel-coated PET graft
was tested under simulated pressure conditions experienced
in the body. Grafts of length 200 mm and of diameter 14
mm were used for this study. A custom-designed test set-up
was fabricated and its schematic diagram is shown in Figure
1. In this experiment PBS, maintained at 37�C, was pumped
through the graft and the graft was subjected to a pulsatile
pressure oscillating between 80 and 120 mmHg to simulate
in vivo conditions. The experiment was done at an acceler-
ated frequency of 3 Hz.
In vitro cell culture cytotoxicity
Cytotoxicity evaluation of coated PET graft samples was car-
ried out by testing the extracts of materials with monolayer
of L929 mouse fibroblast cells according to ISO standards
(ISO 10993-5, 1999). Briefly, L929 cells were subcultured
from stock culture (obtained from National Centre for Cell
Sciences, Pune, India) by trypsinization and seeded into
multiwell tissue culture plates. Cells were fed with Dulbec-
co’s minimum essential medium supplemented with bovine
serum and incubated at 37�C in 5% carbon dioxide atmos-
phere. The extract was prepared by incubating hydrogel-
coated PET graft with physiological saline at 37�C 6 2�C for
24 6 1 h and diluted with media containing fetal bovine se-
rum to get an extraction ratio of 3 cm2/mL. Extracts of
ADA-X-G coated PET graft samples, negative control (ultra
high molecular weight polyethylene) and positive control
(diluted phenol) in triplicate were placed on subconfluent
monolayer of L929 mouse fibroblast cells. After incubation
of the cells at 37�C 6 2�C for 24 6 1 h, cell cultures were
examined microscopically for cellular response around the
samples. Cellular responses were scored as 0, 1, 2, and 3,
which means noncytotoxic, mildly cytotoxic, moderately cy-
totoxic and severely cytotoxic, respectively. Cell viability was
further assessed quantitatively by MTT assay that measures
the metabolic reduction of 3-(4,5-dimethylthiazol-2yl)-2,5,
diphenyl tetrazolium bromide to a colored formazan by via-
ble cells. Material extract for MTT assay was prepared as
described above. The extract and control media were
replaced with 200 lL fresh culture medium to which 50 lL
MTT (10 mg/mL in serum-free medium) was added. Cells
were incubated at 37�C overnight. After discarding the MTT
medium, 200 lL of isopropanol was added to all wells and
kept for 20 min in an orbital shaker (Labline Instruments,
Melrose Park, USA) set at 40–60 rpm. The absorbance of
ORIGINAL RESEARCH REPORT
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | MONTH 2011 VOL 000B, ISSUE 00 3

the resulting solution was recorded immediately at 570 nm
using automated micro plate reader (Bio–Tek Instruments,
VT). Results were expressed as O.D after blank (i.e., medium
only) subtraction. Reported values are mean of three
replicates.
Blood perfusion experiments
Perfusion experiments were carried out using uncoated and
hydrogel-coated PET grafts. For this study, samples of diam-
eter 10 mm and length 75 mm were used. To avoid loss of
blood by seepage through the walls of the unmodified
grafts, each graft was provided with a sleeve of 15-mm di-
ameter graft. Outer side of the sleeve was spray coated with
the hydrogel so that the blood diffused through the
uncoated graft gets collected within the sleeve. Before expo-
sure to blood, both uncoated and hydrogel-coated PET grafts
were soaked in PBS for 15 min. The grafts were then con-
nected to 3 mL syringes (the hub portion was cut and
removed) on both sides as shown in Figure 2. After rinsing
the lumen with saline, piston of one syringe was removed
and 10 mL blood (taken from human volunteer collected
into anticoagulant CPDA) was added to fill the conduit and
part of the syringe. The piston was placed back and the
blood was pushed back and forth through the lumen gently
using both pistons for a period of 1 min. One milliliter
blood was taken for initial analysis. Keeping the syringes
connected on both sides of the graft, remaining blood (9
mL) was allowed to expose to the graft materials for 30
min under agitation at 75 6 5 rpm using a shaker incubator
thermostated at 35�C 6 2�C. Mixing was ensured by man-
ually pushing the blood back and forth at every 10 min
interval. Five replicates were used and all were treated simi-
larly (Figure 2).
Consumption of platelets and leukocytes by the grafts
was assessed by counting the cells in the blood after 1-min
exposure (initial) and at 30 min exposure (final) using a He-
matology analyzer (Sysmex—K 4500). Partial thromboplas-
tin time (PTT) assay was performed to measure the activa-
tion of plasma by the graft materials. Blood samples
exposed to the grafts for 1 min (initial) and 30 min (final)
were centrifuged at 4000 rpm for 15 min and platelet poor
plasma was aspirated. PTT was determined using a reagent
kit obtained from Diagnostica Stago (France) on start 4
coagulation analyzer. Extent of hemolysis caused by the
unmodified and hydrogel-coated PET grafts were also deter-
mined. The total hemoglobin (Hb) in the initial samples was
measured using automatic hematology analyzer (Sysmex—K
4500). The free Hb liberated into the plasma after 30-min
exposure to the unmodified and hydrogel-coated PET grafts
was measured using Diode array spectrophotometer. The
percentage hemolysis was calculated using the formula: He-
molysis (%) ¼ (Free Hb/Total Hb) � 100.
In vitro cytocompatibility: Cell–material interaction
Endothelial cell adhesion and proliferation. EC isolated
from human umbilical vein was used for the study. Tritiated
thymidine (3H-thymidine) was incorporated into the cells
prior to seeding the cells on materials, by culturing them
with the isotope containing media (5 lCi/mL) in the log
phase of cell cycle. On reaching confluence, the cells were
FIGURE 1. Schematic diagram of the experimental set up used for
measuring water permeability in a simulated in vivo pulsatile pres-
sure condition. [Color figure can be viewed in the online issue, which
is available at wileyonlinelibrary.com]
FIGURE 2. Experimental set-up used for perfusion experiments. Samples kept ready for perfusion experiment (A) and samples during the experi-
ment (B) are shown. In both images, unmodified PET graft within the sleeve is shown in the upper half and hydrogel-coated PET graft is shown
in the lower half. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com]
4 MANJU ET AL. ALGINATE DIALDEHYDE CROSS-LINKED GELATIN HYDROGEL

trypsinized and the cell count was estimated using a Hemo-
cytometer. Known number of cells (2.5 � 104 cells/cm2
area; total cells added was 1.0 � 105 for 4 cm2 in one well)
were seeded on to uncoated PET and hydrogel-coated PET.
Two aliquots of 5 � 104 cells were taken for determination
of radioactivity of cells. After 2 h of seeding, the medium
containing unattached cells were withdrawn and fresh 3H-
thymidine containing medium (2 lCi/mL) was added. The
cells in the withdrawn medium were centrifuged and radio-
activity of 3H-thymidine in the cell pellet was determined.
Radioactivity of an aliquot of known number of cells was
taken as standard, and using the radioactivity of cells in the
withdrawn medium number of unattached cells and number
of cells attached to the material surface were estimated. Cul-
ture was allowed to grow for 72 h and during this period,
complete culture medium contained 2 lCi/mL 3H-thymidine.
At the termination of culture after 72 h, the medium was
removed; cells were washed and lysed for the determination
of radioactivity of 3H-thymidine. The radioactivity was
detected using Triathler Multilabel Tester for all samples.
Number of cells adhered at 2 and 72 h were calculated after
transferring the data to MS Excel using the software Comm-
filer Version 2.068.
Statistical analysis. Statistical analysis of the data was car-
ried out using Student’s t-test: ‘‘paired two sample for
means,’’ and setting hypothesized mean difference as zero
and a as 0.05. The two tailed p values were noted for each
pair. Microsoft Excel was used for the data analysis.
RESULTS
Characterization of ADA–X-G hydrogel
Alginate, a type of polysaccharide, on periodate oxidation
cleaves vicinal glycols to form oxidized alginate (Scheme 1).
Alginate forms a very viscous solution even at very low con-
centrations in aqueous medium and hence the oxidation
reaction was carried out in a heterogeneous medium
[1:1(v/v) ethanol–water mixture] as dispersion. Periodate
oxidation of alginate in dilute solution initially proceeds in a
random manner, that only one monomeric unit in a given
chain is oxidized at a given time and the protection of either
one or both of the neighboring units ensue immediately
before the next oxidative attack on the chain occurs.21 Oxi-
dation is usually accompanied by extensive cleavage of the
chain.
Analysis of ADA by FT-IR spectroscopy revealed the
presence of aldehyde groups in the polymer. A major peak
present at 2925 cm�1 and a shoulder peak at 2950 cm�1
indicated CAH stretching vibrations of aldehyde. A shoulder
peak present at 1748 cm�1, corresponding to the CAO
vibrations, also indicated presence of aldehyde group. Chem-
ical analysis of the purified ADA showed that aldehyde con-
tent in ADA was 4.8 � 10�3 mol/g. Estimation of aldehyde
groups in ADA-X-G hydrogel showed that there was no de-
tectable amount of residual aldehyde. This indicates that all
the aldehyde groups were used up during cross linking
reaction involved in the hydrogel formation.
In the two step coating process, ADA-X-G was formed on
the graft surface by a cross-linking reaction (Schiff’s base
formation) between the pendant amino groups of gelatin,
originating from lysine/hydroxylysine, and the aldehyde
group present in oxidized alginate. Borax acts as a catalyst
in the reaction forming complex with hydroxyl groups of
ADA molecule. Here the presence of borax also gives alka-
line pH to the reaction medium and facilitates the Schiff’s
base formation (Scheme 2).
Effect of hydrogel coating on weight gain,
burst pressure, and water permeability
Coating with ADA-X-G resulted in 9% weight gain by the
graft. About 10% increase in burst pressure was observed
when the graft was coated with hydrogel (Table I). Water
permeability determined under steady physiological pres-
sure 120 mmHg for PET grafts before and after coating is
also given in the Table I. It shows that permeability of
hydrogel-coated grafts has been fallen to a range below the
sensitivity level of the instrument. Over 90% reduction in
permeability was observed for the graft after hydrogel
coating.
Morphological features of the coated grafts are shown in
Figure 3. SEM images of uncoated grafts are given for com-
parison [Figure 3(A,B)]. Tightly woven PET fabric structure
is clearly seen under scanning electron microscope. From
Figure 3(C,D), it appears that the quantity of hydrogel de-
posited was not very substantial and at the same time it
was effective in sealing the pores.
Permeability under pulsatile flow conditions
Permeability of PBS buffer through hydrogel-coated PET
graft under pulsatile pressure range of 80/120 mmHg is
shown in Figure 4. It may be seen that negligible quantity
of PBS permeated through the graft during the first 70 h.
The permeability rate slowly increased and after 90 h, a
sudden increase in permeability occurred. Even after this
SCHEME 1. Periodate oxidation of sodium alginate.
ORIGINAL RESEARCH REPORT
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | MONTH 2011 VOL 000B, ISSUE 00 5

increase, the permeability values were still very low and the
graft was as good as a sealed graft. Though this increase in
permeability is not clinically significant, at this point of time
it indicated that the hydrogel network had begun to break-
down due to degradation.
Degradation of hydrogel coated on the PET graft
Weight loss occurred when the hydrogel-coated grafts
were subjected to degradation in PBS at 37�C is shown in
Figure 5. Result shows that about 90% of the coated hydro-
gel had degraded within 15 days. Samples treated with
sodium borohydride are also shown in Figure 5. Data indi-
cate that the degradation profiles of both systems are
almost identical. The initial rate of degradation (measured
as weight loss) is lower for samples treated with sodium
borohydride. However, after 11 days of aging both systems
show almost same weight loss. These results indicate that
SCHEME 2. Schiff’s base formation between amino groups of gelatin and aldehyde groups of oxidized sodium alginate in the presence of
borax.
TABLE I. Weight Gain, Burst Pressure, and Water
Permeability of Control and Hydrogel-Coated Graft
Graft Type
Weight
Gain on
Hydrogel
Coating (%)
Burst
Pressure
(kg/cm2)
Water
Permeability
mL/(min cm2)
Control graft – 29.24 6 2.34 > 450
Hydrogel-coated
graft
9.0 6 0.3 32.15 6 0.63 < 7.6
Average of four measurements is given in the table.
6 MANJU ET AL. ALGINATE DIALDEHYDE CROSS-LINKED GELATIN HYDROGEL

reduction of Schiff base formed in the hydrogel would not
prolong degradation time of hydrogel. Degradation of hydro-
gel caused substantial increase in water permeability under
steady physiological pressure of 120 mmHg. The recorded
permeability at 1 week was 279 6 24 mL/(min cm2) and
data at 1 month showed that the hydrogel-coated graft
becomes as permeable as the control graft. The morphology
of hydrogel-coated PET graft surface after 7 days aging in
PBS revealed that gel degrades into tiny fragments, as seen
in the Figure 6. From the study it is clear that aging is fast
with this system and hydrogel clears off quickly for the tis-
sues to integrate with the vascular graft prosthesis.
In vitro cell culture cytotoxicity studies
Cell culture cytotoxicity studies conducted using mouse
fibroblast cells revealed that neither the gel-coated graft nor
its extract induced any considerable morphological changes
to the cell confirming its nontoxic nature (Figure 7). MTT
assay showed that 98.4% and 74% cells were metabolically
active after contact with 100% extracts of control PET graft
and hydrogel-coated PET graft, respectively.
Perfusion experiments
Platelet count in the blood after exposure to unmodified
and hydrogel-coated PET grafts are shown in Figure 8(A).
Results showed that platelets consumed by the unmodified
PET graft during its 30-min exposure are significantly higher
than that by the hydrogel-coated graft. On the other hand,
hydrogel-coated graft does not show any affinity toward
FIGURE 3. Scanning electron microscopic images of uncoated (A, B) and hydrogel-coated (C, D) PET grafts at different magnifications.
FIGURE 4. Permeability of PBS buffer as a function of time under an
accelerated pulsatile pressure flow conditions set in the range of 80/
120 mmHg and at a frequency of 3 Hz. [Color figure can be viewed in
the online issue, which is available at wileyonlinelibrary.com]
FIGURE 5. Degradation profile of hydrogel-coated vascular grafts in
PBS. Hydrogel formed by Schiff base reaction (ADA-X-G hydrogel)
and hydrogel reduced by sodium borohydride [R(ADA-X-G hydrogel)]
are shown for comparative evaluation. [Color figure can be viewed in
the online issue, which is available at wileyonlinelibrary.com]
ORIGINAL RESEARCH REPORT
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | MONTH 2011 VOL 000B, ISSUE 00 7

platelets. With hydrogel-coated graft platelet counts remain
the same even after 30 min of exposure. Similar behavior
was observed when grafts were exposed to leukocytes [Fig-
ure 8(B)]. Significant reduction in leukocyte counts was
observed when unmodified graft was exposed to blood.
Hydrogel coating tends to prevent leukocytes adhesion onto
the PET grafts. PTT assay shows opposing phenomena with
unmodified and hydrogel-coated grafts [Figure 8(C)]. With
unmodified graft, PTT tends to decrease after 30-min expo-
sure whereas with hydrogel-coated graft a significant
increase in PTT was observed. On analyzing data at 30-min
exposure, significant difference in PTT was observed
between unmodified and hydrogel-coated grafts. Hemolysis
data of unmodified and hydrogel-coated grafts are shown in
Figure 8(D). Results showed that hemolysis tendency of PET
were significantly reduced to a fourth of its value when it
was coated with hydrogel.
The number of EC adhered to unmodified and hydrogel-
coated PET grafts after 2 and 72 h are shown in Figure 9.
At both time intervals, slightly higher numbers of EC were
found to be attached on hydrogel-coated graft (12% higher
at 2 h and 15% higher at 72 h). However, the increase is
not significant. Cell proliferation seems to be absent on both
the control and hydrogel-coated graft materials.
DISCUSSION
Sodium alginate derived from brown seaweed is an anionic
linear polysaccharide composed of 1,4-linked b-D-manuro-
nate and 1,4-linked a-L guluronate residues in varying pro-
portions. Periodate oxidation is the simplest route to trans-
form the relatively unreactive hydroxyl groups of
polysaccharides into amine reactive aldehydes. Here, sodium
meta periodate cleaves carbonAcarbon bond that contain
adjacent hydroxyl group and oxidize the same to form
FIGURE 6. SEM images of hydrogel-coated grafts after aging in PBS
for 7 days.
FIGURE 7. Morphology of L929 mouse fibroblast cell after contact
with: (A) dilute phenol (positive control), (B) ultra high density poly-
ethylene (negative control), and (C) extract of hydrogel-coated PET
graft. [Color figure can be viewed in the online issue, which is avail-
able at wileyonlinelibrary.com]
8 MANJU ET AL. ALGINATE DIALDEHYDE CROSS-LINKED GELATIN HYDROGEL

highly reactive aldehydes. Since periodate oxidized alginate
is highly susceptible to biodegradation compared to bare al-
ginate molecule, it has a potential to be used in a number
of biomedical applications wherein biocompatibility and bio-
degradability are essential. Half equivalents of sodium peri-
odate to that of sodium alginate is used to achieve 50% oxi-
dation. ADA of 50% oxidation gives sufficient reactive sites
for Schiff’s base formation. ADA acts as a potential nontoxic
and biodegradable cross-linking agent for gelatin. Cross-link-
ing is due to the Schiff’s base formation between the pend-
ent amino group of gelatin originating from lysine/ hydroxyl
lysine and aldehyde group present in ADA at alkaline pH.
By giving alternate dipping of PET graft in ADA and gelatin
solutions, ADA-X-G hydrogel could be formed on the graft
surface. Chemical analysis of the hydrogel showed that all
these aldehyde groups were consumed during the hydrogel
formation. Vascular prostheses modified utilizing this proce-
dure are impervious and present a suitable level of softness
and flexibility compared to the characteristics of control
graft. Water permeability studies under steady physiological
pressure of 120 mmHg shows that the coating with 3% sol-
utions each of ADA and gelatin is sufficient for the complete
sealing of pores in the woven PET fabric grafts. It is con-
firmed by permeability studies using PBS buffer under a
pulsatile pressure range of 80/120 mmHg. Significant
increase in burst strength was observed when PET graft
was coated with hydrogel. The burst strength measurements
were done on dry grafts. When hydrogel-coated grafts come
in contact with aqueous media, hydrogel would absorb con-
siderable amount of water and hydrated gel may not con-
tribute to the mechanical properties of the graft. So the
increase in burst strength may not be relevant. Hydrogel
coating gives softness to the graft without affecting its flexi-
bility. This coating material is noncytotoxic and would
remain noncytotoxic even on degradation.
In spite of sodium borohydride reduction, 90% of the
hydrogel that was coated on the graft had degraded in 15
days. On treatment with sodium borohydride only the
Schiff’s base formed were reduced and glycosidic linkages
remain unaffected. In the hydrogel system, both glycosidic
and Schiff’s base linkages are susceptible to hydrolysis. It
FIGURE 8. Hemocompatibility data of uncoated and hydrogel-coated PET grafts after perfusion experiments. Data at 1 min and 30 min exposure
are taken as ‘‘initial" and ‘‘final.’’ p values of individual pairs are shown in the figure. (A) Platelet count, (B) leukocyte count, (C) partial thrombo-
platin time, and (D) hemolysis (%).
FIGURE 9. Number of endothelial cell attached on graft materials
determined at 2 and 72 h after seeding the culture.
ORIGINAL RESEARCH REPORT
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | MONTH 2011 VOL 000B, ISSUE 00 9

may be noted that the reduction of Schiff’s base increased
the stability of the bonds involved in cross-linking reaction.
So in ADA-X-G hydrogel, hydrolysis occur both at glycosidic
and Schiff base linkages whereas in R(ADA-X-G) hydrogel
the main site of degradation would be glycosidic bonds.
Because of this reason, degradation rate of R(ADA-X-G)
hydrogel at the initial stages of degradation was lower com-
pared to ADA-X-G hydrogel. However, at the final stages
both systems degraded at the same rate. Degradation stud-
ies of this hydrogel conducted at 37�C in PBS revealed that
complete dissolution of the gel occurred in 5 weeks time.20
Void space between the fibers in the woven PET graft
shows its porous nature that leads to better healing on im-
plantation. In the case of graft coated with ADA-X-G hydro-
gel, complete sealing of pores with more or less uniform
coating was accomplished. This would lead to minimum/no
bleeding of blood through the interstices of the graft follow-
ing post-implantation. Here the graft composed of nonde-
gradable PET coated with rapidly resorbable hydrogel, so
that during endothelialization the resorbable component
would be replaced by EC. It has been demonstrated by di
Marzo et al. that when cross-linked bovine collagen treated
double velour prosthesis was implanted in dogs deposition
of neoendothelial cells was found to occur in 3–10 days
time.22 So in clinical situations degradation of the hydrogel
and deposition of EC could take place as simultaneous
processes.
Cell culture cytotoxicity study revealed the biocompati-
ble nature of ADA-X-G hydrogel. Three-dimensional net-
works of this hydrophilic polymer can absorb large amount
of water without undergoing dissolution due to their cross-
linked structure, because of this they have physical charac-
teristic similar to soft tissues. Hydrogels derived from natu-
ral proteins and polysaccharides resemble the extra cellular
matrices of tissue comprised of various amino acids and
sugars. These macromolecules thus have the potential to
direct the migration, growth, and organization of cells dur-
ing tissue regeneration. The soft rubbery nature of hydrogel
minimizes mechanical and frictional irritation to the sur-
rounding tissues and thus has low or zero interfacial ten-
sion with surrounding biological fluids and tissue, thereby
minimizing the driving force for protein adsorption and cell
adhesion.
Platelet adhesion plays a central role in the short-term
and long-term blood compatibility of the foreign surfaces.
Platelet activity is generally most intense during the first 24
h and subsides to very low level after 1 week. Platelets are
necessary for hemostasis and play an important role in
thrombus formation. Even though unmodified PET showed
thrombogenic behavior, coating with hydrogel caused little
adhesion of platelets and leukocytes onto it. Hydrogel coat-
ing prevented direct exposure of PET to the blood. As a
result thrombogenic and/or hemolytic potential of the PET
were decreased considerably. It may be inferred that ADA-X-
G hydrogel do not activate the coagulation system to any
great extent. Excellent hemocompatibility was demonstrated
for several other hydrogels too.23 Our results are consistent
with the observations made by other authors.
Since the normal vessels are lined by a monolayer of EC,
which are the ideal nonthrombogenic blood contacting sur-
face, it is believed that covering the surface of vascular
prosthesis with EC would lead to a thromboresistant vascu-
lar prosthesis.24 Currently available vascular prosthesis
materials are hydrophobic and do not support EC growth25
without a precoating with adhesive protein, such as fibro-
nectin or collagen or surface modification such as immobili-
zation of cell adhesive peptides or plasma treatment with
gases. Many authors have shown that the adhesion of
human EC on PET and PTFE polymers is always low or
even impossible, when their surface has hydrophobic na-
ture. There are various surface modification techniques
available, which support cell adhesion on the vascular graft
prosthesis.26–28 The results from the cell adhesion and pro-
liferation experiments demonstrated that ADA-X-G could
offer slight improvement to PET surfaces to support endo-
thelial cell growth. Thus the study demonstrates that ADA-
X-G hydrogel-coated PET grafts act as an effective biode-
gradable sealant as well as favor the adhesion of EC.
CONCLUSIONS
This preliminary study provided an attractive approach to
use ADA-X-G hydrogel as a biocompatible and biodegradable
coating material for PET vascular graft prosthesis. Hydrogel
coated on PET graft samples effectively sealed the pores of
the graft leading to reduced water permeability. The coating
did not adversely affect mechanical properties of the graft.
Over 90% reduction in water permeability was achieved by
hydrogel coating under static conditions. The soft rubbery
nature of the hydrogel would minimize mechanical and fric-
tional irritation to the surrounding tissues. A perfusion
experiment revealed that hydrogel-coated PET graft is highly
hemocompatible. Cytocompatibility tests showed that the
coating would enhance the performance of the PET graft
with better adhesion of EC.
ACKNOWLEDGMENTS
The authors thank the Director, SCTIMST for providing facili-
ties for conducting the study. Dr. T.V. Kumary and Dr. Lissy
Krishnan are gratefully acknowledged for providing cytotoxic-
ity data and hemocompatibility data, respectively.
REFERENCES
1. Kannan RY, Salacinski HJ, Butler PE, Hamilton G, Seifalian AM.
Current status of prosthetic bypass grafts: A review. J Biomed
Mater Res B Appl Biomater 2005;74B:570–581.
2. Bos GW, Poot AA, Beugeling T, van Aken WG, Feijen J. Small-di-
ameter vascular graft prostheses: Current status. Arch Physiol
Biochem 1998;106:100–115.
3. Wu H-C, Wang T-W, Kang P-L, Tsuang Y-H, Sun J-S, Lin F-H. Co-
culture of endothelial and smooth muscle cells on a collagen
membrane in the development of a small-diameter vascular graft.
Biomaterials 2007;28:1385–1392.
4. Sheehan SJ, Rajah SM, Kester RC. Effect of preclotting on the po-
rosity and thrombogenicity of knitted Dacron grafts. Biomaterials
1989;10:75–79.
5. White RA. The effect of porosity and biomaterials on the healing
and long term mechanical properties of vascular prosthesis.
ASAIO J 1988;34:95–100.
10 MANJU ET AL. ALGINATE DIALDEHYDE CROSS-LINKED GELATIN HYDROGEL

6. Guidoin R, King M, Marceau D, Cardou A, de la Faye D, Legendre
JM, Blais P. Textile arterial prostheses: Is water permeability
equivalent to porosity? J Biomed Mater Res 1987;21:65–87.
7. Contreras MA, Quist WC, Logerfo FW. Effect of porosity on small-
diameter vascular graft healing. Microsurgery 2000;20:15–21.
8. Izhar U, Schwalb H, Borman JB, Hellener GR, Hotoveli-Salomon
A, Marom G, Stern T, Cohn D. Novel synthetic selectively degrad-
able vascular prostheses: A preliminary implantation study. J
Surg Res 2001;95:152–160.
9. Guidoin R, Snyder R, Martin L, Botzko K, Marois M, Awad J, King
M, Domurado D, Bedros M, Gosselin C. Albumin coating of a knit-
ted polyester arterial prosthesis: An alternative to preclotting. Ann
Thorac Surg 1984;37:457–465.
10. Jonas RA, Schoen FJ, Levy RJ, Castaneda AR. Biological sealants
and knitted Dacron: Porosity and histological comparisons of vas-
cular graft materials with and without collagen and fibrin glue
pretreatments. Ann Thorac Surg 1986;41:657–663.
11. Haverich A, Walterbusch G, Borst HG. The use of fibrin glue for
sealing vascular prostheses of high porosity. Thorac Cardiovasc
Surg 1981;29:252–254.
12. Machy D, Carteron P, Jozefonvicz J. A new vascular polyester
prosthesis impregnated with cross-linked dextran. J Biomater Sci
Polym Ed 2002;13:963–975.
13. Marois Y, Chakfe N, Deng X, Marois M, How T, King MW,
Guidoin R. Carbodiimide cross-linked gelatin: A new coating for
porous polyester arterial prostheses. Biomaterials 1995;16:1131–
1139.
14. Zilla PP, Bezuidenhout D, Theron JP. Film with highly porous vas-
cular graft prostheses. U.S Patent No.7022135, 2006.
15. Haimovich B, Difazio L, Katz D, Zhang L, Greco RS, Dror Y, Free-
man A. A new method for membrane construction on ePTFE vas-
cular grafts: Effect on surface morphology and platelet adhesion.
J Appl Polym Sci 1997;63:1393–1400.
16. Lee JH, Kim WG, Kim SS, Lee JH, Lee HB. Development and char-
acterization of an alginate-impregnated polyester vascular graft. J
Biomed Mater Res 1997;36:200–208.
17. Rowley JA, Madlambayan G, Mooney DJ. Alginate hydrogels as
synthetic extracellular matrix materials. Biomaterials 1999;20:
45–53.
18. Tibbitt MW, Anseth KS. Hydrogels as extracellular matrix mimics
for 3D cell culture. Biotechnol Bioeng 2009;103:655–663.
19. Cai K, Zhang J, Deng L, Yang L, Hu Y, Chen C, Xue L, Wang L.
Physical and biological properties of a novel hydrogel composite
based on oxidized alginate, gelatin and tricalcium phosphate for
bone tissue engineering. Adv Eng Mater 2007;9:1082–1088.
20. Balakrishnan B, Jayakrishnan A. Self-cross-linking biopolymers as
injectable in situ forming biodegradable scaffolds. Biomaterials
2005;26:3941–3951.
21. Painter T, Larsen B. Formation of hemiacetal between neighbour-
ing hexuronic acid residues during the periodate oxidation of al-
ginate. Acta Chem Scand 1970;24:813–833.
22. di Marzo L, Hunter WJ, Schultz RD, Risavy J. In vivo study of col-
lagen-impregnated double velour prosthesis. Vasc Endovascular
Surg 1989;23:168–174.
23. Faxa¨lv L, Ekblad T, Liedberg B, Lindahl TL. Blood compatibility of
photografted hydrogel coatings. Acta Biomater 2010;6:2599–2608.
24. Stansby G, Berwanger C, Shukla N, Hamilton G. Endothelial cell
seeding of vascular graft: Status and prospects. Cardiovasc Surg
1994;2:543–548.
25. Pratt KJ, Jarrell BE, Williams SK, Carabasi RA, Rupnik MA, Hub-
bard FA. Kinetics of endothelial cell surface attachment forces. J
Vasc Surg 1988;7:591–599.
26. Pratt KJ, Williams SK, Jarrell BE. Enhanced adherence of human
endothelial cells to plasma discharge modified polyethylene ter-
ephthalate. J Biomed Mater Res 1989;23:1131–1147.
27. Ertel SI, Ratner BD, Horbett TA. Radiofrequency plasma deposi-
tion of oxygen containing films on polystyrene and polyethylene
terephthalate substrates improves endothelial cell growth. J
Biomed Mater Res 1990;24:1637–1659.
28. Ramires PH, Mirengli L, Romano AR, Paiumbo F, Nicolardi G.
Plasma treated PET surfaces improve the biocompatibility of
human endothelial cells. J Biomed Mater Res 2000;51:535–539.
ORIGINAL RESEARCH REPORT
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | MONTH 2011 VOL 000B, ISSUE 00 11