GLYCEROL ESTER OF WOOD ROSIN
First draft prepared by
Dr. C.B. Johnson and Dr. M.J. Bonner
Division of Health Effects Evaluation
Office of Premarket Approval
Center for Food Safety and Applied Nutrition
Food and Drug Administration, Washington, DC, USA
Absorption, distribution, and excretion
Acute toxicity studies
Short-term toxicity studies
Long-term toxicity/carcinogenicity studies
Special studies on genotoxicity
Observations in humans
Glycerol ester of wood rosin was previously considered by the
Committee at its eighteenth, twentieth and thirty-third meetings
(Annex 1, references 35, 41 and 83). At its twentieth meeting, the
Committee, in the light of the strong ester bond and anticipated
stability of this material, expressed the view that long-term and
reproductive toxicity studies should be done on the specific substance
before further evaluation. Because of early concerns on the part of
the Committee about the lack of food grade specifications for glycerol
ester of wood rosin, plans for a further toxicological evaluation
had to be postponed. Such specifications were adopted at the
thirty-seventh meeting of the Committee (Annex 1, reference 94). The
specifications define the material as a complex mixture of tri- and
diglycerol esters of resin acids from wood rosin.
This monograph summarizes relevant data in the previous monograph
and data that have become available since the previous evaluation.
2. BIOLOGICAL DATA
2.1 Biochemical aspects
2.1.1 Absorption, distribution, and excretion
Ester gum 8BG, a commercial preparation of glycerol ester of wood
resin, was fed in the diet to groups of 6 male and 6 female F344 rats
under two treatment regimens: (1) for 24 hours at concentrations of
7000 or 28 000 mg/kg and (2) for 10 days at concentrations of 14 000
or 28 000 mg/kg. Food consumption was measured during the treatment
periods and ester gum 8BG intake was calculated for each of the
treatment groups. Faeces were collected during each 24-hour treatment
period and for subsequent 24-hour periods until no ester gum 8BG was
detected. The bulk of ingested ester gum 8BG was excreted in the
faeces within the first 48 h, and only small quantities were present
in the faeces between 48 and 72 h. At the 7000 mg/kg dietary level,
73% of the ingested ester gum 8BG was accounted for in the faeces. At
the 14 000 and 28 000 mg/kg dietary levels, 92% and 89-96%,
respectively, of the ingested ester gum 8BG was recovered in the
faeces. The author postulated that the lower recovery from the
7000 mg/kg group was due to the lack of sensitivity of the analytical
method (HPLC of solvent extract). No necropsy was performed. The
author concluded that in the rat gut, no measurable hydrolysis of
ester gum 8BG took place, and that no absorption from the intestine
was apparent (Blair, 1994).
In recovery experiments, 4 male rats received a single oral dose
of 100 mg (300 mg/kg bw) or 1 mg (3 mg/kg bw) tritiated resin acids
from wood rosin (dehydroabietic, tetrahydroabietic or isopimaric
acids) as a 5% solution in corn oil. After administration of
dehydroabietic acid, the rats excreted on average 80% of the 100 mg
dose in faeces and 7.2% in urine over a 15-day post-treatment period.
Total recoveries in the 4 animals ranged from 71% to 99% at the end of
15 days. Four additional total recovery experiments were conducted,
utilizing a 1mg dose of labelled dehydroabietic acid, and sampling
carried out at various times ranging from 28 to 51 h post-treatment.
In these studies, the amount of dehydroabietic acid recovered averaged
70% in faeces, 8% in urine, 17% in the GI tract, 0.5% in breath, and
1% in the carcass, for a total recovery of 96.5%.
In recovery experiments with tetrahydroabietic acid, 2 rats were
given 45.7 µCi of the tritiated derivative and the radioactivity
monitored at regular intervals after administration (1, 2, 3, 4, 5-8,
9-12 and 13-16 days). Recovery of tritiated tetrahydroabietic acid
averaged 92% in faeces, 5% in urine and 3% in breath for all average
total recovery of 100%. In concurrent studies with isopimaric acid,
recovery also averaged 100% total with 83% found in faeces, 15% in
urine, and 2% in breath.
Four rats were given 30 µCi (105 mg) dehydroabietic acid and
quantitative chromatographic analyses were made of faeces and urine
collected over 2 days. Analysis revealed 3 major metabolites, which
were not identified but for convenience were called A, B, and C.
Recovery in faeces amounted to 89% of the administered activity, of
which 14% was dehydroabietic acid, 33% metabolite A, 8% metabolite B,
and 14% metabolite C. Of the 8% of the administered activity recovered
in urine, 0.13% was dehydroabietic acid, 7% was metabolite B, and
0.55% metabolite C. In rats given labelled tetrahydroabietic or
isopimaric acids, the bulk of the radioactivity in faeces and urine
consisted of the unchanged acids.
In tissue distribution studies with dehydroabietic acid, 4 male
and 4 female rats each were given 50 mg (5.5 µCi), and distribution of
radioactivity measured at various time intervals after administration
(1, 2, 4 or 8 h). Radioactivity was distributed among all major
organs, fat, and muscle, with peak levels occurring 4 h after
administration. After 8 h, the highest concentrations of radioactivity
were found in the liver (approximately 12%) and in kidney
(approximately 5%) (Radomski, 1965).
2.2 Toxicological studies
2.2.1 Acute toxicity studies
The results of acute toxicity studies with wood rosin are
summarized in Table 1.
Table 1. Results of acute toxicity studies on pale wood rosin.
Species Route LD50 Reference
Mouse Oral 4100 Hercules, 1974
Rat Oral 8400 "
Guinea-pig Oral 4100 "
2.2.2 Short-term toxicity studies
In a 90-day oral toxicity study, groups of Sprague-Dawley rats
(10/sex) were fed a stock diet containing Ester Gum 8D (prepared as a
30% suspension in corn oil and blended into the diet) at dietary
levels of 0.01, 0.05, 0.2, 1.0 or 5.0%, equal to 6, 31, 120, 630 or
2660 mg/kg bw/day. The diets of controls and all ester gum 8D groups
contained 2.3% corn oil except the diet of the 5.0% group, which
contained 11.7% corn oil. Test parameters for the study included:
clinical observations, mortality, body weights and body-weight gain,
food consumption, food efficiency, haematology, urinalysis, organ
weights and gross and microscopic pathology. During the course of this
study no mortality occurred among treated rats or controls. No
significant effects were noted on body weight, food intake,
haematology, urinalysis, or gross and microscopic histology at dietary
levels up to and including 1.0%. Food consumption at the 5.0% level
was slightly lower than controls. This difference probably reflected
the higher dietary corn oil concentration at this dose level. No
microscopic pathological changes related to treatment were observed in
any organ. The 1.0% treatment level, equal to 630 mg/kg bw/day, was
considered to be the NOEL in this study (Kay, 1960a).
In a 13-week oral toxicity study, groups of Charles River Fischer
344 rats (20/sex), approximately 6 weeks of age, were fed NIH Open
Formula Diet at ester gum 8BG dose levels of 0, 625, 1250 or
2500 mg/kg bw/day. Test parameters included behavioural observations,
ophthalmoscopic examinations, body weights, food consumption,
haematological and clinical chemistry analyses (at the mid-point and
end of the study), organ weights and macroscopic and microscopic
There were no deaths among treated or control rats during the
course of the study and no changes were observed in appearance,
behaviour, or ophthalmoscopic examinations that were attributed to
treatment. There were slight significant decreases noted by the author
in body-weight gain in female rats at the 1250 and 2500 dose levels
during the last few weeks of the study. However, these minor effects
on body-weight gain were negligible and probably can be attributed to
dietary dilution. Dietary dilution can probably also account for a
slight increase in food consumption that was dose-related, present at
all dose levels among groups of both sexes, and sometimes reaching
statistical significance. There were no dose-related statistically
significant differences in mean haematological and clinical chemistry
values between treated and control groups.
At necropsy, the author noted minor differences between controls
and high-dose groups in male caecum(full)/body weight and in female
liver, thymus and thymus/brain weight. However, these differences were
small and did not appear to be an important effect of treatment. There
were no macroscopic or microscopic changes in any organs that were
related to treatment and none of the organs for which weight changes
were noted had histological abnormalities. The NOEL in this study was
2500 mg/kg bw/day (Blair, 1991, 1992).
In a 90-day study, groups of Sprague-Dawley rats (10/sex) were
fed a stock diet containing 0, 0.01, 0.05, 0.2, or 1.0% N-wood rosin
(added to the diet as a 40% suspension in corn oil), equal to 0, 6.4,
36, 119, or 674 mg/kg bw/day, respectively. Two identical control
groups were utilized. Feeding at a 5.0% level was attempted, but
discontinued early in the study, as all the animals died during the
first 8 days of the treatment period. Final corn oil content was 2.3%
in all test and control diets (except for 11.7% in the 5.0% test
diet). Test parameters included clinical observations, food
consumption, body weights, haematology, urinalyses, organ weights and
gross and microscopic pathology. There was no mortality in controls or
in animals receiving lower doses of wood rosin, and no significant
differences between treated groups and controls were observed in
analyses for haemoglobin, haematocrit, total leucocyte count,
differential leucocyte count or urine analysis parameters.
Body weights were significantly reduced throughout the study for
male and female rats fed 1.0% wood rosin. The decrement in weight gain
at this level occurred during the first 2 weeks of the study;
thereafter, weight gains of this group were comparable to controls.
However, in males fed 1.0% wood rosin, body weight was only
significantly lower when compared to the first control group but not
the second control group. Organ weights revealed statistically
significant increases in both liver to body-weight and brain to
body-weight ratios for male and female rats at the 1.0% wood rosin
level when compared to control groups. However, brain to body-weight
ratio for female rats at the 1.0% level were only significantly
increased compared to those of the second control group but not the
first. No pathological lesions, either macroscopic or microscopic,
related to wood rosin treatment were observed in any of the organs of
treated animals (Kay, 1960b).
2.2.3 Long-term toxicity/carcinogenicity studies
Groups of weanling Sprague-Dawley rats (30/sex, individually
housed) were ted dietary levels of 0, 0.05, 0.2 or 1% rosin in corn
oil for 24 months, equal to 0, 24, 88 or 434 mg/kg bw/day. Final corn
oil content was 2.3% in all test and control diets. At the end of
12 months, 5 animals of each sex were sacrificed for gross and
microscopic pathology studies. All surviving animals were killed at 24
months, organ weights were recorded and pathological examinations were
At both 12 and 24 months, body weights were significantly lower
than controls in both males and females at the 1% diet level. The
decreased body weight may reflect the reduced food consumption also
noted at the 1% diet level, which was attributed to non-palatability.
There were no significant differences between wood rosin treated
groups and controls with respect to mortality, tumour rate,
haematology, urinalysis, gross or microscopic pathology. Elevated
liver to body-weight ratios were noted in high-dose females, with some
sporadic significant differences noted between treated groups and one
or other of the control groups with respect to organ to body-weight
ratios for the kidneys, spleen and gonads (Kohn, 1962a).
Groups of beagle dogs (3/sex) were fed dietary levels of 0.05% or
1.0% (equal to 14 or 260 mg/kg bw/day) N-wood rosin in corn oil for 24
months. A control group consisting of 6 animals of each sex received
the basal diet. Test parameters included body weight, food consumption,
mortality and behavioural changes, haematology and urine analysis,
liver and kidney function tests, and gross and microscopic pathological
examinations. No significant effects were seen on any test parameter
other than weight, except at the 1.0% level, where some increase in
liver and kidney size was noted (although no pathology was present).
Both mean body weight and food consumption in high-dose males were
approximately 30% less than in low-dose males, which would be
consistent with lack of diet palatability. The author concluded that
the NOEL in this study was 1.0% (Kohn, 1962b).
2.2.4 Special studies on genotoxicity
The results of genotoxicity assays on glycerol ester of wood
rosin and resin acids present in wood rosin are summarized in Tables 2
and 3, respectively.
2.3 Observations in humans
Medical doctors reported that a 22-year old woman developed
papules, dryness and pigmentation on the lips after application of a
lipstick several times a day. Patch tests showed that the woman tested
positive for only one of the ingredients in the lipstick, ester gum at
a level of 0.1%. In further patch testing she did not show positive
reactions for rosin, balsam of Peru, nor oil of turpentine. The
subject was diagnosed as sensitized to ester gum in the lipstick
(Ogino et al., 1989).
A medical doctor reported a case of an 8-year old boy who
suffered from recurring perioral dermatitis for 18 months; the subject
frequently chewed gum before each episode of dermatitis developed.
Patch testing was positive to cobalt, rosin, fragrance-mix, oakmoss,
and isoeugenol as well as chewing gum and bubble gum. The perioral
dermatitis improved but did not disappear after the child stopped
chewing gum. Possible sensitivity to allergens other than rosin could
not be ruled out (Satyawan et al., 1990).
Table 2. Results of genotoxicity assays on glycerol ester of wood rosin.
Test System Test Object Concentration Results Reference
Ames test (1) S. typhimurium 10 000 µg/plate Negative Ishidate et al.,
TA92, TA94 1984
Ames test (1) S. typhimurium 2.5-500 µg/plate Negative Jagannath,
TA98, TA100 1988
Chromosome Chinese hamster 8000 µg/ml Negative Ishidate et al.,
aberration fibroblast 1984
CHO/Cytogenetic Chinese hamster 127-507 µg/ml Negative Murli, 1988
assay (1) ovary
Unscheduled DNA rat primary 5.1-102 µg/ml Negative Cifone, 1988
Note that application of treatment in above experiments was single dose.
(1) Both with and without rat liver S-9 fraction.
Table 3. Results of genotoxicity assays on resin acids present in wood rosin
Test System Test Object Concentration Results Reference
Ames test (1) S. typhimurium 250-1000 Positive for Nestmann
TA98, TA100 µg/plate neoabietic acid in et al., 1979
TA1535, TA1537 absence of S-9
Mutagenicity Yeast strains D7, 50-2000 Positive for Nestmann &
test XV185-14C µg/ml neoabietic acid in Lee, 1983
(1) Both with and without rat liver S-9 fraction.
A dentist reported a case of a 33-year old man with contact
allergy to rosin from a periodontal dressing. Periodontal surgery was
performed with no post-operative complications. One week after the
first operation a new surgical dressing was applied. Four days later
the patient began to experience both oral and dermatological symptoms,
but the symptoms disappeared 24 h after the periodontal dressing was
replaced with a wax packing. Patch tests showed that the patient had
contact allergy to rosin, but not to eugenol or zinc oxide which were
also components of the original periodontal dressing (Lysell, 1976).
Patch testing in dental patients who exhibited stomatitis after
repeated applications of periodontal dressing revealed some
sensitization to colophony (rosin). Out of 18 patients (6 men and 12
women, aged 33-71 years), 3 (17%) had a positive reaction to colophony
(Koch et al., 1971).
A total of 133 dental patients who were preoperatively negative
to medication and materials used in dentistry showed negligible
sensitization to colophony, with only one (0.8%) positive result in
patch tests (Koch et al., 1973).
A patch testing study in 150 women investigated contact allergy
caused by cosmetics and toiletries (including those containing rosin).
The type of rosin was not identified in the report. Only one positive
reaction to rosin (0.7%) was observed out of the 150 women tested
(De Groot et al., 1988).
A patch testing study in 1785 patients investigated contact
sensitivity to several suspected allergens, including colophony
(rosin). The specific type of rosin tested was not identified. A total
of 50 patients (2.8%) tested positive for colophony 48 or 72 h post
application. On a gender basis, males experienced a 1.8% incidence
(11/613) and females a 3.3% incidence (39/1172); this difference in
sex distribution was not considered to be significant. Sensitivity to
colophony occurred at a higher incidence (4.4%) in patients aged 50 or
older (Young et al., 1988).
Several recent studies, including a metabolic study in rats, a
13-week toxicity study in rats and mutagenicity studies, have been
conducted on glycerol ester of wood rosin. In the 13-week toxicity
study, the NOEL was 2500 mg/kg bw/day, the highest dose tested.
Mutagenicity studies were negative.
The results of the recent metabolic study showed that glycerol
ester of wood rosin given to rats in the diet was, for the most part,
recovered unchanged in the faeces, and suggested that it was not
hydrolyzed in the gut to a significant extent and was largely
unabsorbed. However, the lack of sensitivity of the analytical method
used was such that a firm conclusion could not be reached as to the
non-bioavailability of glycerol ester of wood rosin and/or its
component resin acids.
Absorption studies with tritiated resin acids from wood rosin
(e.g., dehydroabietic, tetrahydroabietic and isopimaric acids)
indicated that more than 90% were recovered in urine or faeces within
2 weeks (most within 4 days) after oral administration. The small
amount of dehydroabietic acid absorbed appeared to have been
metabolized in the liver to three or four uncharacterized metabolites,
which were then excreted in the bile and urine. Little evidence was
found to show that tetrahydroabietic and isopimaric acids were
The Committee noted the absence of adequate long-term toxicity/
carcinogenicity studies and reproductive toxicity studies on glycerol
ester of wood rosin. Because of the limited toxicological information
available, the Committee was unable to establish an ADI. It is
considered that, as a minimum, studies demonstrating the metabolic
stability and non-bioavailability of glycerol ester of wood
rosin under conditions resembling those present in the human
gastrointestinal tract would be required to permit further evaluation
of this material.
BLAIR, M (1991). Ester Gum 8BG. 13-week dietary toxicity study in
rats. Unpublished report No. 548-007 from International Research and
Development Corporation, Mattawan, MI, USA. Submitted to WHO by
Hercules Inc., DE, USA.
BLAIR, M. (1992). Ester Gum 8BG. 13-week dietary toxicity study in
rats. Amendment to the Final Report. Unpublished report No. 548-007
from International Research and Development Corporation, Mattawan,
MI, USA. Submitted to WHO by Hercules Inc., DE, USA.
BLAIR, M. (1994). A dietary excretion study with Ester Gum 8BG in
Fischer 344 Rats. Unpublished report No. 3352.1 from Springborn
Laboratories, Inc., Spencerville, OH, USA. Submitted to WHO by
Hercules Inc., DE, USA.
CIFONE, M.A. (1988). Mutagenicity test on Ester Gum 8BG in the rat
primary hepatocyte unscheduled DNA synthesis assay. Unpublished Report
No. 10349-0-447 from Hazleton Laboratories America, Inc., Kensington,
MD, USA. Submitted to WHO by Hercules Inc., DE, USA.
DE GROOT, A.C., BEVERDAM, E.G.A., AYONG, C.T., COENRAADS, P.J. &
NATER, J.P. (1988). The role of contact allergy in die spectrum of
adverse effects caused by cosmetics and toiletries. Contact
Dermatitis, 19: 195-201.
HERCULES POWDER CO (1974). Unpublished report submitted to WHO.
ISHIDATE, M., Jr., SOFUNI, T., YOSHIKAWA, K., HAYASHI, M., NOHMI, T.,
SAWADA, M. & MATSUOKA, A. (1984). Primary mutagenicity screening of
food additives currently used in Japan. Fd. Chem. Toxic.,
JAGANNATH, D.R. (1988). Mutagenicity test on Ester Gum 8BG in the Ames
salmonella/microsome reverse mutation assay. Unpublished report
No. 10349-0-401 from Hazleton Laboratories America, Inc.,
Kensington, MD, USA. Submitted to WHO by Hercules Inc., DE, USA.
KAY, J.H. (1960a). Ninety-day subacute oral toxicity of Ester Gum 8D.
Unpublished Report (no study no. given) from Industrial Bio-Test
Laboratories Inc., Northbrook, IL, USA. Submitted to WHO by Hercules
Inc., DE. USA.
KAY, J.H. (1960b). Ninety-day subacute oral toxicity of N-wood rosin.
Unpublished report (no study no. given) from Industrial Bio-Test
laboratories, Inc., Northbrook, IL, USA. Submitted to WHO by Hercules
Inc., DE, USA.
KOCH, G., MAGNUSSON, B. & NYQUIST, G. (1971). Contact allergy to
medicaments and materials used in dentistry (II). Odont. Revy,
KOCH, G., MAGNUSSON, B., NOBRÉUS, N., NYQUIST, G. & SÖDERHOLM, G.
(1973). Contact allergy to medicaments and materials used in dentistry
(IV). Odent. Revy, 24:109-114.
KOHN, F.E. (1962a). Two-year chronic oral toxicity of N-wood rosin -
albino rats. Unpublished report (no study no. given) from Industrial
Bio-Test Laboratories, Inc., Northbrook, IL, USA. Submitted to WHO by
Hercules Inc., DE, USA.
KOHN, F.E. (1962b). Two-year chronic oral toxicity of N-wood rosin -
dogs. Unpublished report (no study no. given) from Industrial Bio-Test
Laboratories, Inc., Northbrook, IL, USA. Submitted to WHO by Hercules
Inc., DE, USA.
LYSELL, L. (1976). Contact allergy to rosin in a periodontal dressing.
J. Oral Medicine, 31: 24-25.
MURLI, H. (1988). Mutagenicity test on Ester Gum 8BG OSR in an
in vitro cytogenetic assay measuring chromosomal aberration
frequencies in Chinese hamster ovary (CHO) cells. Unpublished report
No. 10349-0-437 from Hazleton Laboratories America, Inc., Kensington,
MD, USA. Submitted to WHO by Hercules Inc., DE, USA.
NESTMANN, E.R., LEE, E.G., MUELLER, J.C. & DOUGLAS, G.R. (1979).
Mutagenicity of resin acids identified in pulp and paper mill
effluents using the Salmonella/mammalian-microsome assay. Environ.
Mutagen., 1: 361-369.
NESTMANN, E.R. & LEE, E.G. (1983). Mutagenicity of constituents of
pulp and paper mill effluent in growing cells of Saccharomyces
cerevisiae. Mutat. Res., 119: 273-280.
OGINO, Y., HOSOKAWA, K., SUZUKI, M., MATSUNAGA, K., HIROSE, O.,
ARIMA, Y. & HAYAKAWA, R. (1989). Allergic contact dermatitis due to
ester gum in a lipstick. Skin Research, 31(Suppl. 6): 180-184.
RADOMSKI, J.L. (1965). The absorption, fate and excretion of
dehydroabietic acid, isopimaric acid and tetrahydroabietic acid in
rats. Unpublished report (no Study No. given) from University of Miami
School of Medicine, Coral Gables, FL, USA. Submitted to WHO by
Hercules Inc., Wilmington, DE, USA.
SATYAWAN, I., ORANJE, A.P. & VAN JOOST, T. (1990). Perioral dermatitis
in a child due to rosin in chewing gum. Contact Dermatitis,
YOUNG, E., VAN WEELDEN. H. & VAN OSCH, L. (1988). Age and sex
distribution of the incidence of contact sensitivity to standard
allergens. Contact Dermatitis, 19: 307-308.