Interaction of Methylparaben Preservative with Selected Sugars and Sugar Alcohols

Article abstract of DOI:10.1002/jps.10167

The interaction of methylparaben preservative with selected sugars (glucose, fructose, sucrose, lactose, maltose, cellobiose) and sugar alcohols (lactitol, maltitol) were demonstrated in this study. It was observed that the formation of transesterification reaction products between methylparaben and the selected sugars occurred only under mild reaction conditions (e.g., pH 7.4 at 50 deg C ), which were confirmed by HPLC-UV studies and mass spectrometry. On the other hand, under alkaline conditions and high temperature, degradation of the sugars predominated. Because sugars could easily undergo many possible degradation reactions and isomerization including on-column anomerization, the chromatograms of the reaction products were more complicated than those obtained from sugar alcohols. Sucrose, a nonreducing sugar, was much more stable than other selected sugars. The chromatogram of the transesterification reaction products of methylparaben with sucrose clearly showed eight peaks, which were likely to correspond to the same number of hydroxyl groups of sucrose. To compare the rate of the transesterification reaction of methylparaben with sucrose to that with sorbitol, kinetic studies were carried out. Similar rate constants were observed: 5.4 x 10^-7 L mol^-1 s^-1 and 4.9 x 10^-7 L mol^-1 s^-1 for sucrose and sorbitol, respectively.

Full text of DOI:10.1002/jps.10167

Interaction of Methylparaben Preservative with Selected
Sugars and Sugar Alcohols
2
1
1
MINHUI MA, TONY LEE, ELIZABETH KWONG
1
Pharmaceutical Research & Development, Merck Frosst Canada & Co., 16711 Trans Canada Highway, Kirkland,
Quebec, Canada H9H 3L1
2
Analytical Sciences, Amgen Inc., One Amgen Center Drive, Thousand Oaks, California 91320
Received 25 January 2002; revised 7 March 2002; accepted 8 March 2002
ABSTRACT: The interaction of methylparaben preservative with selected sugars
(
glucose, fructose, sucrose, lactose, maltose, cellobiose) and sugar alcohols (lactitol, mal-
titol) were demonstrated in this study. It was observed that the formation of tran-
sesterification reaction products between methylparaben and the selected sugars
occurred only under mild reaction conditions (e.g., pH 7.4 at 508C ), which were confir-
med by HPLC-UV studies and mass spectrometry. On the other hand, under alkaline
conditions and high temperature, degradation of the sugars predominated. Because
sugars could easily undergo many possible degradation reactions and isomerization
including on-column anomerization, the chromatograms of the reaction products were
more complicated than those obtained from sugar alcohols. Sucrose, a nonreducing
sugar, was much more stable than other selected sugars. The chromatogram of the
transesterification reaction products of methylparaben with sucrose clearly showed
eight peaks, which were likely to correspond to the same number of hydroxyl groups of
sucrose. To compare the rate of the transesterification reaction of methylparaben with
sucrose to that with sorbitol, kinetic studies were carried out. Similar rate constants
ꢁ
7
ꢁ1 ꢁ1
ꢁ7
ꢁ1 ꢁ1
s for sucrose and sorbitol,
were observed: 5.4 ꢀ 10 L mol
s
and 4.9 ꢀ 10 L mol
respectively. ß 2002 Wiley-Liss, Inc. and the American Pharmaceutical Association J Pharm Sci
9
1:1715–1723, 2002
Keywords: methylparaben; sugar; sugar alcohol; interaction; transesterification
reaction; sucrose; sorbitol
INTRODUCTION
sorbic acid and benzoic acid, the activity of para-
bens are less pH dependent. They are effective in
2
Parabens, a group of alkyl esters of p-hydroxy-
benzoic acid, are the most-used antimicrobial
preservatives in pharmaceuticals, food, and cos-
a pH range between 4.0 and 8.0. The parabens
increases in activity but decrease in aqueous
solubility with increasing alkyl chain length. A
combination of short and long chain esters is
1
metics. Their use as antimicrobial agents can be
2
4
traced back to the 1920s. Their popularity is
frequently used to provide effective preservation.
largely due to their high levels of safety and broad
Due to the formation of the phenolate anion
2
,3
spectrum of preservative activity.
Compared
(pKa ¼ 8.4), the antimicrobial activity of the para-
4
with other well-known preservatives such as
bens decreases at pH > 8.0. The parabens also
undergo hydrolysis in weak alkaline and strong
5
–8
acidic solutions
and the hydrolysis product,
Correspondence to: Elizabeth Kwong (Telephone: 514-428-
113; Fax: 514-428-2677;
E-mail: elizabeth_kwong@merck.com)
p-hydroxybenzoic acid, has practically no anti-
microbial activity. Within a pH range of 3–6,
3
4
however, parabens are stable enough to with-
stand a normal heat sterilization procedure.
Journal of Pharmaceutical Sciences, Vol. 91, 1715–1723 (2002)
ß 2002 Wiley-Liss, Inc. and the American Pharmaceutical Association
8
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 7, JULY 2002
1715

1716
MA, LEE, AND KWONG
Although hydrolysis is the most frequent deg-
radation reaction in aqueous pharmaceutical
formulations, other factors may also decrease
the antimicrobial activity of the parabens such
EXPERIMENTAL
Materials
The following materials were used as supplied:
D-sorbitol (99%), D-fructose (99þ%), p-hydroxyl
benzoic acid (99%) from Aldrich Chemicals Co.
9
–11
12–
as sorption onto solid
and plastic container,
1
4
15–17
solubilization by nonionic surfactants,
1
8
partitioning into flavoring oils, and perhaps
more importantly interaction/reaction with exci-
pients. A wide range of excipients have been
reported to interact/react with parabens, which
(Milwakee, WI); D-(þ)-glucose (99.5%), sucrose
(> 99.5%), maltose (99%), maltitol (98%), a-lactose
(monohydrate SigmaUltra), lactitol (99%), methyl
paraben (99%), D-(þ)-cellobiose (98%) from Sigma
1
9
20
include cellulose derivatives, ethanol, ethano-
(St. Louis, MO); sodium methyl paraben from
2
1
22
lamine, b-cyclodextrin, and sugar alcohols (or
Merck Research Laboratories (West Point, PA);
tetrahydrofuran (THF) (99%) from Anachemia;
trifluoroacetic acid (TFA) from Pierce; and acet-
onitrile (HPLC grade) from EM Science. Deio-
nized water used for solution preparation was
from a milli-Q water system.
2
3–26
polyols).
In this article, we report the inter-
action of methylparaben with selected sugars
through transesterification reaction, which to
our best knowledge has not been reported in the
literature. The sugars selected for this study are
sucrose, maltose, lactose, cellobiose, glucose, and
fructose. Related sugar alcohols, maltitol, lactitol,
and sorbitol, are also used for comparison.
The molecular structures of the selected sugars
and sugar alcohols are shown in Figure 1. The
interaction/reaction products were separated by
HPLC and identified by LC-MS. The kinetics of
the reaction between methylparaben and sucrose
were studied and compared with that between
methylparaben and sorbitol.
Transesterification Reactions of Methylparaben
with Selected Sugars and Sugar Alcohols
Individual sugar or sugar alcohol (D-glucose, mal-
tose, lactose, D-fructose, cellobiose, lactitol, mal-
titol, sucrose, and D-sorbitol) solution at 0.20 M
was prepared by dissolving the compound in
0
.10 M phosphate buffer (pH 7.4) in a tightly
capped vial. An aliquot of 750 mg of sodium
methylparaben was added to each 50 mL of the
sugar or sugar alcohol solutions, which was then
placed in a 508C oven for up to16 h. The sample
solution was then cooled to room temperature and
approximately 1.5 mL of the solution was filtered
through a 0.45-mm PTFE syringe filter and the
last 1 mL was collected in an HPLC vial. The
samples were analyzed by HPLC and LC-MS
to confirm the formation of transesterification
products. The mol ratios of the sugars to sodium
methylparaben in the sample solutions were
approximately 2:1.
Kinetic Study of the Transesterification Reactions
of Methylparaben with Sucrose and Sorbitol
Sucrose and sorbitol solutions containing 0, 0.3,
0
.6, 0.9, 1.2, and 1.5 M were prepared in 0.10 M
phosphate buffer (pH 8.0). An amount of sodium
methylparaben (69.64 ꢂ 0.02 mg) was weighed
into six clean and dry 100-mL volumetric flasks.
Each sodium methylparaben sample was then
made up to volume with one of the above-mention-
ed six sucrose or sorbitol solutions. After the
sample solution was thoroughly mixed, a 5-mL
aliquot was pipetted into a 10-mL glass ampule
Figure 1. Molecular structures of the selected sugars
and sugar alcohols.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 7, JULY 2002

METHYLPARABEN PRESERVATIVE WITH SELECTED SUGARS
1717
that was then sealed using a propane torch. Ten to
5 ampules were prepared for each of the sample
For the quantitation of methylparaben and p-
hydroxybenzoic acid in the kinetic study, one
Zorbax SB-CN (3.5 mm) column (150 ꢀ 4.6 mm)
using isocratic elution at 20% B was used. The
flow rate was set at 0.8 mL/min, with a run time of
17 min.
1
solutions. An aliquot of each of the six sample
solutions remaining in the flask was analyzed
immediately by HPLC as initial. When all of the
ampules were sealed, they were placed in a 608C
oven. Ampules were withdrawn from the oven
one at a time for each sample solution at pre-
determined time intervals.
LC-MS Analyses
Samples withdrawn from the oven were cooled
to room temperature before analysis. An aliquot
The mass spectra of the transesterification reac-
tion products of methylparaben with selected
sugars and sugar alcohols were acquired using a
Finnigan LCQ Deca ion-trap mass spectrometer
(
20 mL) of the sample solution was injected onto
the HPLC to determine the concentration of the
total transesterification reaction products. Ano-
ther 200-mL aliquot of the sample solution was
diluted to 1.0 mL with water and then injected
immediately onto the HPLC to determine the
concentration of methylparaben and p-hydro-
xybenzoic acid using a different HPLC method.
Standard solutions of methylparaben and p-
hydroxybenzoic acid were prepared in water at
(San Jose, CA) with an ESI source operated in
negative ion mode. An HP 1100 HPLC system
with two Zorbax SB-CN (3.5 mm) columns coupled
in series (75 ꢀ 4.6 mm and 150 ꢀ 4.6 mm) were
used. The HPLC conditions were the same as
those described in the LC section using gradient
(I). The following modifications were made: 0.1%
TFA was replaced by 0.1% formic acid in mobile
phase A to reduce the suppression of ionization,
and postcolumn flow splitting was used to reduce
flow rate from 0.6 mL/min to approximately 50 mL/
min for the ESI source. The operating parameters
of the ESI source were as follows: sheath gas flow
rate: 60, auxiliary gas flow rate: 5, spray voltage:
ꢁ
5
the concentrations ranging from 4.0 ꢀ 10 M to
ꢁ
4
8
.0 ꢀ 10 M. Because no standard was available
for the transesterification reaction products, they
were quantified using the same p-hydroxybenzoic
acid standard solutions because they have the
same UV spectra. Standard calibrations were
performed daily.
4
.5 kV, capillary temperature: 3508C , capillary
voltage: ꢁ41 V, and tube lens offset: ꢁ25 V. Mass
spectra were acquired in full-scan MS mode with
a mass scan range of 50 amu. The actual mass
ranges were chosen such that the molecular ions
of the transesterification reaction products would
fall approximately in the center of the 50 amu
mass scan range.
Liquid Chromatography
The HPLC system used in this study was a HP
1
090 liquid chromatograph (Hewlett-Packard,
Palo Alto, CA) equipped with a chemstation, an
autosampler, and a diode array detector. Chro-
matograms were acquired at 255 nm with a
bandwidth of 4 nm. The injection volume was
2
0 mL. To improve the separation of the transes-
RESULTS AND DISCUSSION
terification reaction products (positional isomers),
two Hewlett Packard Zorbax SB-CN (3.5 mm)
columns were coupled in series (75 ꢀ 4.6 mm and
Transesterification Reactions of Methylparaben
with Selected Sugars and Sugar Alcohols
1
4
50 ꢀ 4.6 mm). Column temperature was kept at
08C. For HPLC separation of the transesterifica-
In a recent study on the transesterification
reactions of methylparaben with 12 three- to six-
carbon sugar alcohols, we demonstrated that each
of the hydroxyl groups of a sugar alcohol molecule
could react with methylparaben to form transes-
tion reaction products of methylparaben with
selected sugars and sugar alcohols, the following
gradient (I) was used: [the mobile phase A was
0
.1% (v/v) TFA in water and B was 85:15 (v/v)
2
6
THF/acetonitrile] 0–14 min at 4% B, 16–36 min
at 20% B, 38–40 min at 30% B, 42–51 min at 4%
B. For the quantitation of the transesterification
reaction products in the kinetic study, the follow-
ing gradient (II) was employed: 0–23 min 10% B,
terification reaction products. These reaction
products were positional isomers with the same
m/z ratio as determined by LC-MS. The number of
reaction products that were separated by HPLC
was exactly the same as that expected based on
the number of distinct hydroxyl groups of the
molecule. As a consequence, other common sugar
2
7–29 min at 70% B, 31–40 min at 10% B.
The flow rate for both gradients was 0.6 mL/min.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 7, JULY 2002

1
718
MA, LEE, AND KWONG
excipients in pharmaceuticals were also investi-
gated. In theory, transesterification reaction could
occur between parabens and any molecules con-
taining hydroxyl groups. However, no such study
on the reaction with sugars has been reported in
the literature except one unsuccessful attempt, in
which no reaction products between methyl-
paraben and aldoses, for example, ribose, xylose,
a- and b-anomers. Anomerization between the
a- and b-anomers was most likely responsible for
the on-column reactions observed in Figure 2.
Similar on-column anomerization has been re-
ported in the literature for the HPLC separation
30
of acyl glucuronides.
Under the same reaction conditions, the selec-
ted sugar alcohols, maltitol and lactitol, were
very stable. The sample solutions of maltitol and
lactitol remained colorless after 16 h of heating
and no extra peaks due to degradation were
observed in the chromatograms of the reaction
products (Figure 2g and h). Maltitol and lactitol
both have three primary and six secondary hy-
droxyl groups. The chromatogram of the reaction
products of methylparaben with maltitol showed
exactly three major and six secondary peaks. In
the case of lactitol, three major and only five
secondary peaks were observed. The missing
secondary peak for lactitol was most likely due
to coelution with the other peak because the
HPLC method shown in Figure 2 was developed
using the transesterification products of maltitol
and was not optimized for the reaction products of
lactitol.
2
4
or cellobiose, were observed.
The failure to observe any reaction products
was most likely due to the harsh reaction condi-
tions that were used in their experiments (908C
2
4
for 3 h at unspecified pH). It is well known that
sugars are much less stable than sugar alcohols.
At high temperature and under alkaline condi-
tions, sugars could undergo many possible degra-
dation reactions such as caramelization,
enolization, followed by fragmentation and addi-
2
7,28
tional secondary reactions.
In a preliminary
experiment of the present study, the reaction
conditions employed for the 12 three- to six-
carbon sugar alcohols (pH ꢃ12 and 908C for 2 h)
were utilized for the reactions of methylparaben
with the selected sugars. Under these conditions,
it was observed that sample solutions turned
brown very quickly upon heating and that the
HPLC chromatograms of the sample solutions
showed no well-defined peaks but several broad
bands of many small overlapping peaks in the
retention region (4–18 min) of the transesterifica-
tion products. Because of the serious degradation
problem, the reactions between methylparaben
and the selected sugars were carried out under
much milder conditions: in phosphate buffer (pH
There was no simple correlation between the
number of peaks and the number of hydroxyl
groups for the selected sugars except for sucrose.
Sucrose has three primary hydroxyl groups—one
on the glucose unit, and two on the fructose unit—
and five secondary hydroxyl groups. Interest-
ingly, five small peaks, two medium size peaks,
and one major peak were observed in the chro-
matogram (Figure 2c). Glucose and fructose are
monosaccharide with five hydroxyl groups, which,
in theory, would generate five transesterification
products (or positional isomers). However, after
7 h of heating at 508C, six peaks (three pairs of
peaks due to on-column reactions) were observed
in the chromatogram of the glucose/methylpar-
aben mixture while only two peaks were observed
in the case of fructose (chromatograms not shown).
After 16 h, several additional peaks/shoulders
were observed (see arrows in Figure 2a and b),
which may be attributed to the isomerization
reaction occuring for glucose and fructose. Due to
enolization, glucose, fructose, and mannose are in
equilibrium through the common 1,2-enediol. Glu-
cose and fructose are the major components at
7.4) at 508C for up to 16 h. Even under these
reaction conditions, the sugar solutions turned
yellow after 16 h of heating and some unidentified
broad peaks, most likely due to sugar degrada-
tion, were observed in the HPLC chromatograms.
The chromatograms of the transesterification
reaction products of methylparaben with the
selected sugars and sugar alcohols are shown in
Figure 2.
Among the selected sugars, sucrose seemed to
be more stable than other sugars, as indicated by
the well-defined peaks of the reaction products
between sucrose and methylparaben in the chro-
matogram (Figure 2c). This is probably due to
the fact that sucrose is a nonreducing sugar. One
common feature of other selected sugars was the
presence of on-column reaction as characterized
28,31
equilibrium.
This may explain why additional
peaks (see arrows) with the same retention times
as those major peaks found in Figure 2b were
observed in Figure 2a and vice versa. Other
selected sugars (cellobiose, maltose, and lactose)
2
9
by the plateaus between peaks. In aqueous solu-
tion, a phenomenon common for sugars is mutar-
otation, which involves the equilibration of the
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 7, JULY 2002

METHYLPARABEN PRESERVATIVE WITH SELECTED SUGARS
1719
Figure 2. Chromatograms of the transesterification reaction products of methylpar-
aben with (a) glucose, (b) fructose, (c) sucrose, (d) cellobiose, (e) maltose, (f) lactose, (g)
maltitol, and (h) lactitol at pH 7.4 and 508C for 16 h. Arrows in (a) and (b) indicating
enolization reaction products of glucose and fructose, respectively.
are disaccharides that have eight hydroxyl groups
including two primary ones. As can be seen from
Figure 2d, e, and f it is not easy to count the
number of well-defined peaks in the chromato-
grams due to the complication caused by on-
column reaction and sugar degradation. However,
it is important to note that the goal of the present
study was not to achieve ideal separation but to
demonstrate the presence of transesterification
reaction between selected sugars and methylpar-
aben. For this purpose, both the UV and the mass
spectra of the peaks were used to identify the
reaction products. Some of the small peaks that
were likely caused by the reactions of methyl-
paraben with sugar degradates were not the focus
of this study, and no attempts were made to
identify those small peaks.
The UV spectra of all the peaks shown in
Figure 2 were acquired using a diode array detec-
tor. All the peaks showed UV spectra similar to
those of methylparaben or p-hydroxybenzoic acid
with a maximum absorption at 255 nm. Because
all the peaks elute before p-hydroxybenzoic acid
(22.5 min, peak not shown in Figure 2), these
peaks are most likely the transesterification re-
action products of methylparaben with the selec-
ted sugars. This was confirmed by the LC-MS
experiments.
Figure 3 shows some typical total ion chroma-
tograms (a and c) and mass spectra (b and d) of the
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 7, JULY 2002

1720
MA, LEE, AND KWONG
Figure 3. Total ion chromatograms of the transesterification reaction products of
methylparaben with (a) glucose and (c) sucrose. Typical mass spectra of the
transesterification products peaks are shown in b (retention time ¼ 9.25 min, glucose)
and d (retention time ¼ 10.40 min, sucrose), respectively.
transesterification reaction products of methyl-
paraben with glucose and sucrose. Because the
mass scan range was limited to the expected
molecular ions of the reaction products, p-hydro-
xybenzoic acid and methylparaben were not
detected, and therefore excluded from the total
ion chromatograms. All the peaks shown on the
total ion chromatograms were expected to be the
reaction products. It can be seen that the total ion
chromatograms in Figure 3a and c correspond
well with the UV chromatograms in Figure 2a and
c, respectively. It was found that all the peaks on
the same total ion chromatograms had identical
mass spectra with the same m/z ratio for their
molecular ions. The m/z ratio matched what was
expected based on the molecular weight of sugar
monoesters of p-hydroxybenzoic acid. For exam-
ple, the mass spectra of the five peaks shown in
Kinetic Study of the Transesterification Reactions of
Methylparaben with Sucrose and Sorbitol
The kinetics of the reaction between methyl-
paraben and selected sugar alcohols has been
23
reported, where the transesterification reaction
was treated as a simple second-order reaction. In
fact, transesterification reaction between methyl-
paraben and sugar alcohols or sugars involves n
parallel second-order reversible reactions because
of the presence of n hydroxyl groups in the
molecule. The possible interconversion between
the transesterification reaction products (posi-
tional isomers) as demonstrated in a previous
2
6
study could further complicate the situation.
For such a complex system, it would be extremely
difficult if not impossible to derive the rate
equations for all involved reactions. The purpose
of this kinetic study was to determine the overall
transesterification rate of sucrose with methyl-
paraben, and to compare to that of sorbitol. For
this purpose, the simplified approach by Runes-
ꢁ
Figure 3a gave the same m/z ratio of 299 [(M-H) ]
as shown in Figure 3b, and the molecular weight
of glucose monoesters of p-hydroxybenzoic acid is
300. The LC-MS data confirmed that all the peaks
23
separated or partially separated in Figure 2 were
indeed the transesterification reaction products of
methylparaben with sugars.
son and Gustavii
as shown in Figure 4 is
sufficient. The only modification needed is to call
k1 the overall rate constant of the n transester-
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 7, JULY 2002

METHYLPARABEN PRESERVATIVE WITH SELECTED SUGARS
1721
xybenzoic acid agreed well (within 5%) in the
concentration range used, which indicated that
the effect of esterification on the UV absorptivity
of p-hydroxybenzoic acid seemed to be minimal.
Therefore, the standard of p-hydroxybenzoic
acid was used to quantify the transesterification
reaction products.
Figure 4. The transesterification reaction and hydro-
lysis pathways.
The plots of log Cmp,t versus time t are shown in
Figures 5 and 6, respectively, for the degradation
(
i.e., transesterification and hydrolysis) of methyl-
ification reactions instead of the second-order rate
constant for the transesterification.
paraben in sucrose and sorbitol solutions. Accord-
ing to eq. (4), the plots should be linear with a
slope of ꢁkobs/2.303. This was indeed the case up
to certain time points as shown in the figures. For
the reactions in sucrose especially for those with
higher sucrose concentrations, the slopes started
to decrease after ꢃ300 h of reactions. Much less
deviation from linearity was found for the reac-
tions in sorbitol. For those solutions without
sucrose and sorbitol, almost no deviation from
linearity was found. These observations sugges-
ted that the deviation was most likely caused by
the degradation of sucrose and sorbitol after
a long time of heating at 608C in the pH 8.0
phosphate buffer. According to eq. (3), any de-
crease in sucrose or sorbitol concentration would
decrease the observed rate constant, kobs, or the
slope of the plots. Sorbitol is more stable than
sucrose, so less deviation was expected. Because
of the significant deviation, time points after
ꢃ300 h were not included in the linear fitting.
In theory, transesterification reaction is rever-
sible. Because a large excess of sucrose and sor-
bitol (at least 85 times higher concentration than
methylparaben) was used in this study, the
reaction could be treated mathematically as an
irreversible reaction. By following the derivation
2
3
of Runesson and Gustavii, the rate of degrada-
tion of methylparaben can be expressed by
ꢁ
dCmp=dt ¼ k C C
þ k C
ð1Þ
1
s
mp
2
mp
where Cmp and Cs are the concentrations of
methylparaben and sucrose or sorbitol, respec-
tively, k is the pseudofirst-order rate constant
2
for the hydrolysis of methylparaben. Because
C >> C , k C & constant, eq. (1) can be re-
s
mp
1
s
written as
ꢁ
dCmp=dt ¼ kobsCmp
ð2Þ
ð3Þ
ð4Þ
where the observed rate constant
The observed rate constants, k , as a function of
the concentrations of sucrose and sorbitol are
plotted in Figure 7.
kobs ¼ k1 Cs þ k2
Equation (2) can be integrated to give
log Cmp;t ¼ log Cmp;0 ꢁ kobs t=2:303
obs
where C , and Cmp,t are the concentrations of
mp 0
methylparaben at initial and time t. According to
eq. (4), a plot of log Cmp,t versus t should give a
straight line with a slope of ꢁkobs/2.303. The
values of k and k can be evaluated by plotting
1
2
kobs versus C . According to eq. (3), it should give a
s
straight line with a slope of k and an intercept
1
of k2.
In this study, the reaction of methylparaben
with sucrose and sorbitol were carried out at pH
8
.0 and 608C. The concentrations of the total
transesterification reaction products, methyl-
paraben and p-hydroxybenzoic acid in the sample
solutions at different time points were analyzed
by HPLC. Methylparaben and p-hydroxybenzoic
acid were quantified using calibration curves
of their standards. It was found that the two
calibration curves of methylparaben and p-hydro-
Figure 5. Plots of log Cmp versus time for the
hydrolysis and transesterification of methylparaben in
(a) 0 M, (b) 0.3 M, (c) 0.6 M, (d) 0.9 M, and (e) 1.2 M of
sucrose in 0.1 M phosphate buffer (pH 8.0) and at 608C.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 7, JULY 2002

1722
MA, LEE, AND KWONG
CONCLUSIONS
The current study has clearly demonstrated that
under suitable conditions not only sugar alcohols
but sugars can also react with methylparaben to
form transesterification reaction products. Be-
cause sugars are much less stable than sugar
alcohols under alkaline conditions and at high
temperatures (e.g., at pH ꢃ12 and 908C), sugar
degradation predominates and no significant
amount of transesterification reaction products
can be observed in the HPLC chromatograms.
Under milder reaction conditions used in this
study, the chromatograms clearly show the for-
mation of transesterification reaction products
between methylparaben and the selected sugars,
which is confirmed by LC-MS. However, the chro-
matograms obtained are not as clean as those
obtained with sugar alcohols because of the
occurrence of sugar degradation. Isomerization
and on-column reaction (most likely anomeriza-
tion) further complicated the chromatograms of
the transesterification reaction products. Sucrose
seems to be an exception, which gives well-defined
chromatogram of the transesterification reaction
products with eight peaks. The kinetic study
indicates that the rates of the transesterification
reactions of methylparaben with sucrose and
sorbitol are very similar and that the rate of
hydrolysis of the methylparaben is not affected by
the sugar or sugar alcohol molecule.
Figure 6. Plots of log Cmp versus time for the
hydrolysis and transesterification of methylparaben in
(
a) 0 M, (b) 0.3 M, (c) 0.6 M, (d) 0.9 M, (e) 1.2 M, (f) 1.5 M
of sorbitol in 0.1 M phosphate buffer (pH 8.0) and at
08C.
6
According to eq. (3), the overall rate constants
for the transesterification reactions, k , and the
1
rate constants for the hydrolysis of methylpar-
aben, k , were obtained from the slopes and
2
ꢁ
7
the intercepts of the plot: k ¼ 5.4 (ꢂ 0.3) ꢀ 10
1
ꢁ
1 ꢁ1
ꢁ7
ꢁ1 ꢁ1
ꢁ1
Lmol
s
and k ¼ 8.7 (ꢂ 0.2) ꢀ 10
s
for
and
2
ꢁ
7
sucrose and k ¼ 4.9 (ꢂ 0.5) ꢀ 10 Lmol
s
1
ꢁ
7
ꢁ1
k ¼ 8.5 (ꢂ 0.4) ꢀ 10
s
for sorbitol. The results
2
showed that the rate of the transesterification
reaction of methylparaben with sucrose was very
similar to that with sorbitol, although sucrose has
two more hydroxyl groups. Additionally, the
results also showed that the hydrolysis of methyl-
paraben is independent of the sugar or sugar
REFERENCES
alcohol molecules because the k values for both
2
sucrose and sorbitol were very similar.
1
2
. Martin C. 1997. The art of preservation. Chem
Britain March 35–38.
. Aalto TR, Firman MC, Rigler NE. 1953. p-Hydro-
xybenzoic acid esters as preservatives I: Uses,
antibacterial and antifugal studies, properties
and determination. J Am Pharm Assoc 42:449–457.
. Matthews C, Davidson J, Bauer E, Morrison JL,
Richardson AP. 1956. p-Hydroxybenzoic acid esters
as preservatives. II. Acute and chronic toxicity in
dogs, rats and mice. J Am Pharm Assoc 45:260–267.
. Rieger MM. 2000. Methylparaben. In: Kibbe AH,
editor. Handbook of pharmaceutical excipients, 3rd
ed. Washington, DC: American Pharmaceutical
Association; London: The Pharmaceutical Press.
pp. 340–344.
3
4
5
. Raval NN, Parrott EL. 1967. Hydrolysis of methyl-
paraben. J Pharm Sci 56:274–275.
Figure 7. The observed rate constants for the degra-
dation of methylparaben at varying concentrations of
6. Kamada A, Yata N, Kubro K, Arakawa M. 1973.
Stability of p-hydroxybenzoic acid esters in acidic
medium. Chem Pharm Bull 21:2073–2076.
(*) sucrose and (&) sorbitol in phosphate buffer
pH 8.0) and at 608C.
(
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 7, JULY 2002

METHYLPARABEN PRESERVATIVE WITH SELECTED SUGARS
1723
7
8
. Blaug SM, Grant DE. 1974. Kinetics of the degra-
dation of the parabens. J Soc Cosmet Chem 25:
of methyl and ethyl 4-hydroxybenzoates in ethanol-
water systems. Int J Pharmaceut 27:1–15.
21. Juenge EC, Gurka DF, Kreienbaum MA. 1981.
Analysis of drug contamination from parabens
in Theophylline Olamine. J Pharm Sci 70:589–
593.
22. Chan LW, Kurup TTR, Muthaiah A, Thenmozhiyal
JC. 2000. Interactionofp-hydroxybenzoicesterswith
beta-cyclodextrin. Int J Pharmaceut 195:71–79.
23. Runesson B, Gustavhii K. 1986. Stability of para-
bens in the presence of polyols. Acta Pharm Suec
23:151–162.
24. Hensel A, Leisenheimer S, Muller A, Busker E,
Wolf-Heuss E, Engel J. 1995. Transesterification
reaction of parabens (alkyl 4-hydroxybenzoates)
with polyols in aqueous solution. J Pharm Sci
84:115–118.
495–506.
. Sunderland VB, Watts DW. 1984. Kinetics of the
degradation of methyl, ethyl and n-propyl 4-hydro-
xybenzoate esters in aqueous solution. Int J
Pharmaceut 19:1–15.
. Yousef RT, El-Nakeeb MA, Salama S. 1973. Effect
of some pharmaceutical materials on the baterici-
dal activities of preservatives. Can J Pharm Sci 8:
9
54–56.
1
1
0. Allwood MC. 1982. The adsorption of esters of p-
hydroxybenzoic acid by magnesium trisilicate. Int J
Pharmaceutics 11:101–107.
1. Myburgh JA, McCarthy TJ. 1980. The influence of
suspending agents on preservative activity in
aqueous solid/liquid dispersions. Pharm Weekbl
(
Sci) 2:143–148.
25. Thompson MJ, Fell AF, Clark BJ, Robinson ML.
1993. LC studies on the potential interaction of
paraben preservatives with sorbitol and glycerol. J
Pharm Biomed Anal 11:233–240.
26. Ma M, DiLollo A, Mercuri R, Lee T, Bundang M,
Kwong E. 2002. High-performance liquid chroma-
tography and liquid chromatography-mass spectro-
metry studies on the transesterification reaction of
methylparaben with twelve 3- to 6-carbon sugar
alcohols and propylene glycol and the isomerization
of the reaction products by acyl migration. J
Chromatogr Sci 40:170–177.
27. Whistler RL, Daniel JR. 1985. Carbohydrates. In:
Fennema OR, editor. Food chemistry, 2nd ed. New
York: Marcel Dekker. pp. 69–137.
28. Belitz H-D, Grosch W, Burghagen MM. 1999. Food
chemistry. New York: Springer Verlag. pp. 237–
318.
1
1
1
2. Patel NK, Nagabhushan N. 1970. Drug-plastic
interactions. II. Sorption of p-hydroxybenzoic acid
esters by capran polyamide and in vitro biological
activity. J Pharm Sci 59:264–266.
3. Kakemi K, Sezaki H, Arakawa E, Kimura K, Ikeda
K. 1971. Interactions of parabens and other phar-
maceutical adjuvants with palstic containers.
Chem Pharm Bull 19:2523–2529.
4. Bin T, McCrosky L, Kulshreshtha AK, Hem SL.
2
000. Adsorption of Esters of p-hydroxybenzoic acid
by filter memberanes: Mechanism and effect of
formulation and processing parameters. Pharm
Dev Tech 5:95–104.
1
1
1
5. Poprzan J, de Navarre MG. 1959. The influence of
nonionic emulsifiers with preservatives VIII. J Soc
Cosmet Chem 10:81–87.
6. Blaug SM, Ahsan SS. 1961. Interaction of parabens
with non-ionic macromolecules.
0:441–443.
7. Yamaguchi M, Asaka Y, Tanaka M, Mitsui T, Ohta
S. 1982. Antimicrobial activity of butylparaben in
relation to its solubilization behavior by nonionic
surfactants. J Soc Cosmet Chem 33:297–307.
J
Pharm Sci
29. Trapp O, Schurig V. 2001. Approximation function
for the direct calculation of rate constants and
Gibbs activation energies of enatiomerization of
racemic mixtures from chromatographic para-
meters in dynamic chromatography. J Chromatogr
A 911:167–175.
5
1
1
2
8. Chemburkar PB, Joslin RS. 1975. Effect of flavor-
ing oils on preservative concentrations in oral
liquid dosage forms. J Pharm Sci 64:414–417.
9. Kurup TRR, Wan LSC, Chan LW. 1995. Interaction
of preservatives with macromolecules. Part II. Cellu-
lose derivatives. Pharm Acta Helva 70:187–193.
0. Sunderland VB, Watts DW. 1985. Alkaline ethano-
lysis of methyl 4-hydroxybenzoate and hydrolysis
30. Mortensen RW, Corcoran O, Cornett C, Sidelmann
UG, Troke J, Lindon JC, Nicholson JK, Hansen SH.
1
2001. LC- H NMR used for determination of the
elution order of S-naproxen glucuronide isomers in
two isocratic reversed-phase LC-systems. J Pharm
Biomed Anal 24:477–485.
31. Jones M Jr. 1997. Organic chemistry. New York:
W.W. Norton & Company. pp. 1246–1249.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 7, JULY 2002