In
vitro
degradation
of
glycineh-lactic
acid
copolymers
J.
Helder,
P.
J.
Dijkstra,
and
J.
Feijen*
Department
of
Chemical Technology, Twente University,
PO
Box
217,
7500
AE,
Enschede,
The Netherlands
The
in
vitro
degradation
of
glycine-DL-lactic
acid copolymers was studied as a function
of the composition. These polydepsipep-
tides were prepared by ring-opening
copolymerization
of
6-methyl-2,5-morpho-
linedione and DL-lactidc. The degradation
of discs
of
the copolymers was performed
in
a
phosphate buffer at pH
7.4
and
37°C.
The decrease in molecular weight and
weight was determined until complete
weight loss had occurred. Poly(DL-lactide)
was used as a reference material.
All
(co)-
polymers show an immediate decrease in
molecular weight, whereas the weight re-
mains almost unchanged during a longer
period of time. Decrease in weight started
earlier
as
the glycine content of the co-
polymer increased. The lactic acid content
of the residual material increased during
the weight loss showing a higher solubil-
ity
of
polymer fragments with a relatively
high content of glycine residues. From the
hydrolysis constants it was concluded that
the degradation was best described by hy-
drolysis of ester bonds via a bulk erosion
process, autocatalyzed by the generated
carboxylic acid end groups. The rate con-
stants varied from
4-7
x
lo-'
(day-') for
all (co)polymers. All (co)polymers show
an increase in the molecular weight distri-
bution upon weight loss.
INTRODUCTION
During the last few decades the use of synthetic polymers for medical
devices has rapidly expanded. Currently the great potential of biodegrad-
able polymers for application in sutures, bone plates, and other surgical
fixation devices and especially drug delivery systems is widely recognized.'
Biodegradable polymers should have labile bonds, which can be cleaved in
the body forming nontoxic degradation products. Several aliphatic poly-
ester~~-~ and poly(a-amino a~ids)~,~ meet these requirements.
The extensive literature concerning the degradation of polyesters has
been reviewed by Holland et a1.' After penetration
of
water into the poly-
mer matrix, degradation starts through
a
bulk hydrolysis
of
ester bonds and
the molecular weight decreases. Weight loss ensues when the molecular
weight
of
the polymer chains has sufficiently decreased.6-'1 In crystalline
polymers like poly(glyco1ic acid)
(PGA),
poly(L-lactic acid)
(PLLA),
and
poly(e-caprolactone)
(PCL)
degradation first occurs in the amorphous re-
gions and later
in
the crystalline
domain^.^,^*-'^
As
a result during the initial
stages
of
degradation the crystallinity increases. The period required for
Journal of Biomedical Materials Research,
Vol.
24, 1005-1020 (1990)
0
1990 John Wiley
&
Sons, Inc.
CCC
002l-9304/90/081005-16$04.00
1006
HELDER,
UIJKSTRA,
AND
FEIJEN
complete degradation depends on the type of polymer, the crystallinity, the
initial molecular weight of the samples, and the
glass
transition tempera-
ture. Although some enzymes were found to affect the degradation
of
PGAI6 and PLLA17
in
vituo,
an enzymatic contribution to the initial stages
of
degradation
of
polyesters has not been observed
in
vim.
18~19
Most authors
come
to
the conclusion that enzymes do not play a role
in
the initial degra-
dation process of polyesters in the glassy state.’~’o~’l However, enzymes are
involved in the subsequent degradation of hydrolysis products formed dur-
ing the initial degradation
of
Only
a
few studies about the degradation of copolymers containing both
ester and amide groups were rep~rted.”~’~ More recently Kaetsu et al.24-26
studied the
in
vivo
degradation of several sequential polydepsipeptides.
Melt pressed films of polydepsipeptides were absorbed after several weeks‘
implantation in rats24,2s and the polydepsipeptides were more readily de-
graded than the corresponding poly(u-amino acids).
26
Previously we
showed that the
in
vim
and
in
vitro
degradation behavior
of
some glycine/
DL-lactic acid copolymers and poly(tx4actic acid) appeared c~mparable.’~
In this paper we present the results of studies on the
in
vitro
degradation
of films of nonalternating glyCine/DL-laCtiC acid copolymers (copoly(g1y-DL-
lac)). The degradation was followed by measuring the decrease of the mo-
lecular weight, the weight loss, and the changes in polymer composition as
a function of time. The degradation of the glycine/DL-lactic acid copolymers
was compared with that of pOly(DL-laCtiC acid) (PDLA).
MATERIALS AND METHODS
Materials
Copolymers were synthesized by ring-opening polymerization of
6-methylmorpholine-2,5-dione
(cyclo(G1y-DL-Lac)) and DL-lactide in the melt
at
135°C
as
described previously.” Molar percentages
of
DL-lactide in the
feed of
50,
60,
70,
75,
80,
and
90
were used.
A
low-molecular-weight PDLA
sample (code
100)
was a gift from
CCA
Biochem, Gorinchem, The Nether-
lands. A high-molecular-weight PDLA sample (code
DL)
was prepared
using the procedure applied for the synthesis of the
Preparation
of
polymer
discs
Films were made by casting solutions of the copolymers in tetrahydro-
furan (THE
15%
w/v) onto glass surfaces, which were previously treated
with a 30% v/v solution of dichlorodimethylsilane in toluene. The THF was
slowly evaporated in air at room temperature for at least
3
days and subse-
quently at
40°C
for
24
hours at reduced pressure (0.01 mm
Hg).
The films
were then immersed in demineralized water for
2
h to exchange residual
solvent. After removal from the water, the films were dried over
P,Oj
at re-
DEGRADATION
OF
DL-LACTIC ACID COPOLYMERS
1007
duced pressure
(0.01
mm
Hg)
for 18 h and were then cut into discs with a
diameter of 12 mm. Subsequently, the discs were dried to constant weight
at
reduced pressure (0.01 mm
Hg)
at
40°C
for at least
18
h. The average
weight of 40 discs of each series was 52
t
12, 52
&
7,
52
?
7, 58
?
13,
56
C
4,
59
?
14, and 52
-C
6 mg for (co)polymers
50
to 100, respectively,
and 48
&
3
mg
for polymer
DL.
The average thicknesses of the
40
discs were
530
*
100,
470
k
40,
460
?
70,
460
&
100, 420
2
40,
430
?
120, and
380
*
40
pm for (co)polymers 50 to
100,
respectively, and 350
?
20 pm for
polymer
DL.
Degradation studies
For each (co)polymer
40
discs were used to follow the degradation as a
function of time. Each disc was weighed and placed in a vial containing
5 mL of an aqueous buffer solution (0.1M Na2HP0,/KH,P04, pH 7.4) with
0.03 wt% sodium azide to prevent bacterial growth. The vials were placed
in a thermostatically controlled water bath at 37
?
2°C.
For each (co)poly-
mer one disc was removed after a predetermined immersion time and the
pH of the remaining buffer solution was measured. When the pH was
lower than 7.3 the buffer solutions in the remaining vials containing discs
of
the same (co)polymer were replaced. Using this procedure the buffer solu-
tions never reached a pH value lower than 6.9 even when the weight loss
of
the polymer samples was considerable. After removal from the buffer solu-
tion the discs were washed with distilled water and dried to constant
weight over
P205
at reduced pressure
(0.01
mm
Hg)
for at least 48 h.
The degradation of the (co)polymers was followed by measuring the
weight and the apparent molecular weight of the residual material as a
function
of
time. The weight loss
of
each disc was determined by compar-
ing the dry weight
of
the degraded (co)polymer with the initial weight.
The molecular weight distributions, apparent weight average
(M,
app)
and
number average molecular weight (Mn,
app)
of the undissolved (co)polymer
material were determined using high-performance Gel Permeation Choma-
tography (GPC). The
GPC
unit consisted
of
a Waters model 6000
A
pump
and a Waters
U6K
injector, four Waters mStyragel
(lo5,
lo4,
lo3, and
500
A
in
pore size) columns in series and a Waters
R401
differential refractometer.
Elution was performed at 25°C with a flow rate of 2.0 mL/min using THF as
eluent. The columns were calibrated with polystyrene standards having
narrow molecular weight distributions. It has to be realized that the
Mn,,pp
and MW,app values measured depend on the composition of the molecules.
During the degradation the molecular weight of the polymers
is
changing
and the chemical composition may change also.
To
obtain a better estimation
of
the real values of the initial molecular
weights of the copolymers
Mn,,,l,
values were determined. The M,,,,,, value
of, e.g., copolymer
75
was obtained by multiplying the
Mn,,pp
value of
copolymer 75 with the ratio
M,*/Mn,.pp*
determined for a copolymer with
1008
HELDER,
DIJKSTRA,
AND
FEIJEN
the same composition but a different molecular weight. the
M,*
value was
measured by membrane osmometry and the
Mn,
value was measured
by GPC.
The overall composition of partly degraded samples was determined by
'H-NMR spectroscopy. The spectra were recorded on a Nicolet
200
MHz
NMR apparatus using CDC1, as a solvent and tetramethylsilane
(TMS)
as an
internal reference. The copolymers gave distinct signals for the glycine
methylene protons at
6
4.0
and for the lactic acid methine protons at
6
5.1,
respectively. The composition was calculated from the methylene and
methine proton integrations.
To
investigate the microstructure of the (co)polymer discs the cross sec-
tions and surfaces were examined by scanning electron microscopy (SEM)
with a JEOL JSM
35
CF
scanning electron microscope. The samples were
coated with a charge conducting layer of gold using a Balzer sputter unit.
RESULTS
Degradation profiles
Table
I
shows that the initial molecular weights (M,,&
of.
the copolymers
and polymer
100
are
in
the same order of magnitude. The apparent weight-
average molecular weight
(M,
app),
number-average molecular weight
(M,,,
app),
and the residual weight (%wt)
of
the (co)polymer samples as a function of
immersion time in buffer solution are shown in Figure
1.
This figure shows
that for each (co)polymer the decrease of the molecular weight starts imme-
diately after immersion
of
the film
in
the buffer solution. After the start of
weight
loss
the MW,app and M,,,pp values
of
the (co)polymers remain almost
constant (copolymers
50-80)
or are even higher ((co)polymers
90,
100,
and
DL)
than the values of the samples measured just before weight loss.
TABLE
I
Composition and Initial Molecular Weights
of
the
(Co)polvmers
Mol
Fraction
of
DL-Lactide
Sample
M,
appa
M",
Cdcb
M,
appa
Code In the
Feed
In the Copolymer
(x
(X (X
50
60
70
75
80
90
100
DL
0.50
0.60
0.70
0.75
0.80
0.90
1.00
1.00
0.55
2
0.03
0.64
&
0.03
0.74
4
0.03
0.78
2
0.03
0.82
4
0.03
0.92
2
0.03
1.00
1.00
0.5
1.1
2.0
2.9
3.8
4.8
3.8
16.7
3.9
2.7
3.2
4.2
6.5
4.0
3.2
14.2
1.5
3.0
4.3
5.4
6.7
9.7
8.2
42.6
'Apparent molecular weights were measured
by
GPC.
?he
M,
values
of
these samples were calculated (see
degradation
sfudies).
DEGRADATION
OF
DL-LACTIC ACID COPOLYMERS
1009
20,
160
i'
time
(days)
70
0
0
F
%wt
mcb
ttme
(days)
DL
~000000
0
M.
.PP
mwt
40
20
0
204060801m
time
(days)
time
(days)
Figure
1.
MW,app
(o),
M,,,pp
(o),
and residual weight
(0)
of
copolymer
50
(a),
60
(b),
70
(c),
75
(d), 80 (e),
90
(0,
100
(g),
and
DL
(h)
as
a
function
of
time.
All (co)polymers show a similar pattern for the residual weight
as
a func-
tion
of
immersion time. During the first
3
days a fast, but small decrease
of
the weight is observed. Thereafter the weight
of
the samples remains