FUNDAMENTAL AND APPLIED TOXICOLOGY 30, 4 7 - 5 4 (1996)
ARTICLE NO. 0042
Chronic Dermal Studies of Petroleum Streams in Mice
WILLIAM D. BRODDLE,* MICHAEL W. DENNIS,! DONALD N. KJTCHEN,^ AND EDMOND H. VERNOT§
"Conoco, Inc., 600 North Dairy Ashford Road, Houston, Texas 77079; ^Biology Department, Montana State University-Billings,
1500 North 30th Street, Billings, Montana 59101; \Colorado Pathology Services, Inc., 5101 North Country Road No. 9,
Fort Collins, Colorado 80524; and ^American Petroleum Institute, 1220 Street, N.W., Washington, DC 20005
Received March 20, 1995; accepted September 15, 1995
Chronic Dermal Studies of Petroleum Streams in Mice. BRODDLE, W., DENNIS, M. W., KITCHEN, D. N., AND VERNOT, E. H.
(1996). Fundanu Appl. Toxicol. 30, 47-54.
During petroleum refining, a large number of products are generated which have varying chemical and physical properties. These
are known In the industry as petroleum streams. In order to characterize their carcinogenic activity, a number of these commercially produced streams were administered to C3H/HeJ mice in
chronic dermal bioassays. The bioassays were conducted using one
of two study designs: the first set of test materials was applied for
a lifetime and the second set for 24 months. In the Lifetime study,
the last mice in the test groups survived for periods of 31 to 32
months. Middle distillates, boiling in the range 115-390°C, were
found to decrease the lifespan of exposed mice compared to controls or streams of higher and lower boiling ranges. These middle
distillate streams included straight run kerosine, hydrodesulfurized middle distillate, straight run middle distillate, light catalytic
cracked distillate, and 90/10% and 70/30% mixtures of the last
two. The middle distillate streams also proved to be active as
carcinogens, with tumor incidence ranging from 16 to 67%. Light
alkylate naphtha, heavy catalytic reformed naphtha, vacuum residuum, and unleaded gasoline did not demonstrate significant
carcinogenic potency. Heavy thermal cracked naphtha, heavy catalytic cracked naphtha, and hydrotreated light naphthenic distillate were dermal carcinogens of low potency in this study. Administration of light catalytic cracked naphtha led to a low incidence
of very late developing tumors with a mean latency of 118 weeks.
Application of the 0.1% solution of catalytic cracked clarified oil
in toluene did not result in a significant incidence of tumors, but
the 10% solution caused almost 100% mortality and 100% tumor
incidence in 12 months. There was no correlation between carcinogenic potency and the indices of irritation, alopecia, erythema,
and scabbing. Only two of the streams tested, hydrotreated light
naphthenic distillate and 10% catalytic cracked clarified oil, contain polynuclear aromatic hydrocarbons (PNAs) and may be presumed to be complete carcinogens. The middle distillates and
heavy naphthas are nonmutagenic and essentially free of PNAs.
Their activity may result from promotion of already-initiated skin
sites. Where comparisons could be made, reducing the exposure
period from a lifetime (29-32 months) to 24 months did not change
the evaluations of stream carcinogenicity except in the case of
light catalytic cracked naphtha where six of the seven mice that
developed tumors did so after 24 months, o 1996 soday of Toxico^e
.
47
For a number of years the American Petroleum Institute
(API) has sponsored research into the carcinogenic activity
of the various streams produced in the refining of crude oil.
The refining processes include, among others, distillation
to yield fractions of differing boiling ranges, alkylation to
produce branched-chain saturated hydrocarbons from lowmolecular-weight unsaturated compounds, reforming to provide aromatic hydrocarbons, catalytic cracking to create
more volatile streams from less volatiles ones, hydrotreatment to obtain saturated hydrocarbons from unsaturated
streams, and solvent refining to remove polynuclear aromatic
hydrocarbons (PNAs) from lubricating base oil streams. The
involvement of the API was preceded by reports of human
cancer, especially scrotal cancer, associated with exposure to
"mineral oils" (Bell, 1876; Cruickshank and Squire, 1950;
Henry, 1946, 1947; Lione and Denholm, 1959; VoLkmann,
1875; Waterhouse, 1971). Although the most potent of these
mineral oils were derived from coal tar (Kennaway, 1934)
and shale oil (Leitch, 1922), animal tests (Twort and Twort,
1931) confirmed that some petroleum oils were also active.
Twort and Twort (1939) identified some of the active compounds in petroleum-derived mineral oils as PNAs of characteristic molecular weight and structure. Bingham and Horton
(1966) demonstrated that processing history is a critical determinant of carcinogenic activity in mineral oils and that
those processing steps which might be expected to reduce
or eliminate PNAs also reduced or eliminated carcinogenic
activity. Roy et al. (1988) reported that mutagenicity and
carcinogenicity of petroleum-derived mineral oils with median boiling points of 260-577°C were highly correlated
with concentration of 3- to 7-ring PNAs. Craig (1989) reviewed the good correlations between carcinogenicity of lubricating base oils and PNA content and mutagenicity as
determined using a modified Ames test. He also noted that
middle distillates and heavier naphthas possessed low to
moderate carcinogenic potency even though free of PNAs
and nonmutagenic.
Beginning in 1975, the API contracted to have two wellcharacterized crude oils refined by distillation into fractions
with the following boiling ranges: <49, 49-177, 177-288,
288-371, 371-577, and >577°C. These corresponded
roughly to streams produced in commercial refinery distilla0272-0590/96 $18.00
Copyright O 1996 by the Society of Toxicology.
All rights of reproduction in any form reserved.
48
BRODDLE ET AL.
tion operations. The streams were tested for dermal tumorigenic potency and the results were reported by Lewis (1983).
The 371-577°C fraction was the most active, as was expected, since it contained nearly all of the PNAs with carcinogenic potency. The next lower boiling fraction, 288-371°C,
and the residue boiling over577°C were essentially noncarcinogenic as was the fraction boiling under 49°C. However,
the 49-177 and 177-288°C fractions were found to have
moderate tumorigenic potency even though their concentrations of PNAs were negligible. Haider et al. (1984) and
Kane et al. (1984) examined the carcinogenic activity of
refinery streams used to manufacture lubricating oils. They
demonstrated that solvent refining sufficient to remove carcinogenic PNAs eliminated the carcinogenic potency of
these base stocks. Doak et al. (1985) reported that severe
hydrotreatment also eliminated the carcinogenicity of lubricating base stocks. Witschi et al. (1987) and Biles et al.
(1988) exposed mice chronically to petroleum middle distillates with boiling ranges between 140 and 370°C and found
them to be carcinogenic but of relatively weak potency.
Initiation/promotion experiments were carried out on middle
distillates by Gerhart et al. (1988), McKee et al. (1989),
Skisak (1991), and Skisak et al. (1994). All these authors
reported that the middle distillates acted as promoters rather
than initiators of carcinogenesis.
In order to characterize the carcinogenic activity of petroleum refinery streams more fully and to include processing
treatments in addition to distillation, the API sponsored an
extensive series of dermal bioassays of streams representing
a wide spectrum of distillation ranges and physical and
chemical treatments, generating the data presented here and
in Skisak et al. (1994).
MATERIALS AND METHODS
Materials. The test articles were obtained from the API repository (Experimental Pathology Laboratories, Herndon, VA). An API task force of
industry toxicologists selected the test materials which were representative
of streams produced during petroleum refining. The streams have been
characterized for inclusion in the EPA listing of materiaJs covered by Toxic
Substances Control Act regulations and given Chemical Abstract Service
(CAS) identification numbers. Table 1 is a listing of the streams, their
CAS numbers, and their boiling ranges. The streams characterized as light
naphthas are the most volatile ones tested followed by the heavy naphthas
and the middle distillates: straight run kerosine, hydrodesulfurized middle
distillate, straight run middle distillate, and light catalytic cracked distillate.
The 90/10 (%) and 70/30 mixtures of straight run middle distillate and light
catalytic cracked distillate were included in the study because they are
representative of commercial No. 2 fuel oil formulations. Unleaded gasoline
is a mixture of various naphthas as can be inferred from its boiling range.
Hydrotreated light naphthenic distillate is a low viscosity lubricating oil base
stock. Catalytic cracked clarified oil is the residual fraction of distillation of
the products from a catalytic cracking process. The least volatile stream,
vacuum residuum, is the residue of vacuum distillation of the material
remaining after atmospheric distillation of crude oil.
The test articles were stored at 4°C and in 25-ml glass bottles under
nitrogen. At approximately 2-week intervals, fresh bottles were removed
from the refrigerator and held at room temperature for dermal application
TABLE 1
Petroleum Stream Samples Tested in Chronic
Dermal Mouse Studies
Petroleum stream
CAS No.
Light alkylate naphtha"
Light catalytic cracked naphtha*
Heavy catalytic reformed naphtha"
Heavy thermal cracked naphtha"
Heavy catalytic cracked naphtha"
Straight run kerosine"
90% straight run middle distillate/10%
light catalytic cracked distillate*
70% straight run middle distillate/30%
light catalytic cracked distillate*
Straight run middle distillate"
Light catalytic cracked distillate"
Hydrodesulfurized middle distillate*
Hydrotreated light naphthenic distillate"
Catalytic cracked clarified oil, 0.1%*
Catalytic cracked clarified oil, 10%*
Vacuum residuum, 50%*
Unleaded gasoline*
64741-66-8
64741-55-5
64741-68-0
64741-83-9
64741-54-4
8008-20-6
C
Boiling
range (°C)
39-166
34-181
120-185
110-208
121-242
114-271
180-309
181-319
64741-44-2
64741-59-9
64742-80-9
64742-53-6
64741-62^1
64741-62-4
64741-56-6
c
186-391
194-339
261-301
240-424
>350
>350
>495
34-220
° Twenty-four-month application period.
* Lifetime application period.
c
These products have no CAS numbers because they are mixtures of
streams.
Toluene, ACS grade, and benzo-a-pyrene (BaP) were obtained from the
Aldrich Chemical Co. (Milwaukee, WI).
Dosage/Formulations.
Vacuum residuum was applied as a 50% (w/v)
solution in toluene. Two dilutions of catalytic cracked clarified oil, 10 and
0.1 % (w/v) in toluene were formulated for application. All other test articles
were applied neat. BaP was applied as 0.01 and 0.05% (w/v) solutions in
toluene.
Animals, housing, and diet. C3H/HeJ male mice were acquired from
Jackson Laboratories (Bar Harbor, ME). Mice were 7 - 9 weeks of age at
study initiation. Mice were randomized upon receipt, double-housed in
suspended stainless steel cages for a quarantine period, and then singly
housed prior to study initiation. Each mouse was uniquely identified by
toeclips. Some groups were quarantined for 2 weeks and held for I week,
others were quarantined for 1 week and held for 2. Punna Certified Rodent
Chow 5002 was provided ad libitum. Deionized tap water was provided ad
libitum by bottle Animal rooms were maintained at 24.5 ± I.7°C (SD)
and relative humidity at 40 ± 10% (SD). A 12-hr light/dark cycle was
maintained
Experimental design. The mice were randomly divided into groups of
50 mice each. In addition to the test groups, one group received no treatment,
and two groups were administered BaP (0.01 and 0.05%) as positive controls Since two samples were diluted with toluene, a separate toluene group
was included as a vehicle control. The studies were carried out in two
segments with application in the initial one extending over the lifetime of
each animal. Although some mice in this segment survived for 32 months,
the small amount of information available after 24 months did not seem to
justify the considerable costs of a study of this duration. Therefore, in die
next segment, application was limited to 24 months. The test articles and
the positive and solvent controls, BaP and toluene, were applied as 50-^1
doses twice weekly. Positive, solvent, and sham-handled controls were
included for each treatment segment. All mice not dying on study were
killed by ethyl ether inhalation and exsanguinated by puncturing the posterior vena cava.
DERMAL STUDIES OF PETROLEUM STREAMS
Skin preparation and dosing. All materials were applied to the shaved
interscapular region of the back. Mice were clipped approximately every
other week. Materials were applied with a repeating pipet (SMI Instruments
of Hamilton, Inc.) with disposable syringe and/or pipet tip. At dosing, test
materials covered at least 1 cm2.
In-life and postmortem examinations. Physical examinations for dermal irritation and tumors were performed weekly. Dermal irritation was
evaluated by measuring incidence and estimating severity of erythema,
alopecia, and scabbing. A dermal lesion was diagnosed as a tumor when
the mass reached 1 mm in diameter. All mice were subjected to a limited
necropsy upon death or euthanasia. Application site skin was removed and
fixed in 10% neutral-buffered formalin. Skin sections were embedded in
paraffin, sectioned, stained with hematoxylin and eosin, and examined microscopically.
Statistical methods. Only those tumors confirmed by histological examination were used in calculating incidence. Significance of differences between test and control tumor incidences was analyzed by x2 (Snedecor
and Cochran, 1980). The Fisher's Exact Test (one-tailed) was used when
comparing low-tumor incidence groups to the appropriate control. Final
effective number (FEN) is the denominator used in calculating tumor incidence. When median time to tumor was under 60 weeks, the number of
animals alive at that time was used as the FEN. When median time to
tumor was over 60 weeks, the number of animals alive at 60 weeks plus
any mice dying with tumors before 60 weeks was used as the FEN. Latency
was measured as the time in weeks from initiation of dosing to appearance
of the first tumor.
RESULTS
Groups of mice exposed to all test materials except vacuum residuum exhibited some degree of dermal irritation.
Listed in Table 2 are visual estimates of levels of irritative
responses, alopecia, erythema, and scabbing, noted at the
exposure time when they stabilized, usually about a year.
Very mild irritative signs were seen in animals exposed to
light alkylate naphtha, light catalytic cracked naphtha, heavy
catalytic reformed naphtha, hydrotreated light naphthenic
distillate, 0.1 % catalytic cracked clarified oil, and unleaded
gasoline. Irritation classified as mild occurred in groups exposed to heavy catalytic cracked naphtha, straight run kerosine, 10% catalytic cracked clarified oil, and straight run
middle distillate. Exposure to the rest of the streams, heavy
thermal cracked naphtha, light catalytic cracked distillate,
hydrodesulfurized middle distillate, and the 90/10 and 70/
30 mixtures, resulted in moderate irritation.
Table 3 presents the survival data measured at 6-month
intervals. One inference from the table is that survival appears to have been higher in the lifetime-exposed groups
than in those exposed for 24 months. When survival in comparable groups at 24 months is examined, it is evident that
more mice were alive in the sham-treated, toluene, and BaP
groups exposed for a lifetime than in those exposed for 24
months. This higher survival rate requires consideration
when making toxicological judgments and comparisons
among the groups. With this proviso, examination of the
information in Table 3 indicates that vacuum residuum, unleaded gasoline, and the naphtha streams are relatively nontoxic when mortality is used as the index. The group exposed
49
to catalytic cracked clarified oil 0.1% also has a low mortality at 24 months, but this is a result of its low concentration
rather than an inherent property of the stream since exposure
to 10% catalytic cracked clarified oil led to 94% mortality
at 12 months with no survivors at 18 months. The middle
distillate streams all showed elevated mortality at 18 months.
With the possible exception of straight run middle distillate,
few mice or none survived 24 months of exposure to these
streams.
Table 4 shows the numbers of mice with histopathologically characterized tumors and the mean latencies of tumor
appearance. Sham-treated controls were free of any skin
tumors in both 24-month and lifetime studies. The streams
that did not develop significant numbers of tumors compared
to their respective controls were light alkylate naphtha,
heavy catalytic reformed naphtha, 0.1% catalytic cracked
clarified oil, vacuum residuum, and unleaded gasoline. The
remaining naphthas, light catalytic cracked naphtha, heavy
thermal cracked naphtha, and heavy catalytic cracked naphtha, demonstrated relatively low numbers of tumors which
were, however, significant when compared to the zero incidence in sham-treated controls. Application of all the middle
distillates led to significant tumor development in mice ranging from 16% incidence in the group exposed to the 90/10
mixture to 67% in the group exposed to straight run kerosine.
For a few samples, the differences in tumor incidence
between exposure for 24 months and for a lifetime could be
compared. These were groups exposed for a lifetime in
which significant numbers of animals survived for 30 months
and the toluene and BaP controls which were run in both
segments. These comparisons are made in Table 5 which
shows general agreement between tumor incidences at 24
months and at lifetime termination. A possible exception
occurred during exposure to light catalytic cracked naphtha
where six mice developed tumors between 24 months and
termination at 32 months. This comparison may not be exact,
however, because the 24-month incidence is based on visual
observation while the lifetime number is histologically determined. Although four mice developed tumors in the lifetime
toluene segment compared with one in the 24-month segment, the significance of this difference is lessened by the
fact that only two mice developed tumors after lifetime exposure to 0.1% catalytic cracked clarified oil, which is 99.9%
toluene.
Although an occasional malignant melanoma or malignant
lymphoma (a total of four each in all the groups combined)
appeared at the application site, they did not affect the statistical significance of total tumor incidence. The most common
tumors were fibromas, fibrosarcomas, squamous cell papillomas, and squamous cell carcinomas. The numbers of animals
in each group developing these tumors are listed in Table
6. It is obvious from the table that the great majority of them
are malignant squamous cell carcinomas and fibrosarcomas
rather than benign papillomas and fibromas. Although the
50
BRODDLE ET AL.
TABLE 2
Irritation Resulting from Exposure to Petroleum Streams0
Stream
Alopecia
Erythema
Scabbing
Light alkylate naphtha
Light catalytic cracked naphtha
Heavy catalytic reformed naphtha
Heavy thermal cracked naphtha
Heavy catalytic cracked naphtha
Straight run kerosinc
90% straight run middle distillate/10%
light catalytic cracked distillate
70% straight run middle distillate/30%
light catalytic cracked distillate
Straight run middle distillate
Light catalytic cracked distillate
Hydrodesulfurized middle distillate
Hydrotreated light naphthenic distillate
Catalytic cracked clarified oil, 0.1%
Catalytic cracked clarified oil, 10%
Vacuum residuum, 50%
Unleaded gasoline
Sham treated*
Toluene*
Toluene'
BaP, 0.01%*
BaP, 0.01 %c
BaP, 0.05%*
BaP, 0.05% c
Low
Low
Low
Moderate
Low
Low-moderate
Moderate
Low
Very low
Low
Moderate
Low-moderate
Low-moderate
Low-moderate
None
Very low
Low
Low
Low
Low
Low
Moderate
Low-moderate
Low
Low
Moderate
High
Low
Low
Low
Very low
Low
None
Low
Low
Low
Low
Low
Low
Low-moderate
Moderate
Moderate
Low
Low
Low—moderate
None
Very low
None
Low
Low
Low
Low
Low-moderate
Low-moderate
Low
Low
Moderate
Low
None
None
None
Very low
None
Low
Low
Low
Low
Low
Low
° Estimated from examinations approximately 1 year into exposure when responses had stabilized.
* 24-month exposure.
c
Lifetime exposure.
squamous cell papillomas and carcinomas are of epithelial
origin and the fibromas and fibrosarcomas are mesothelial
tumors, there seems to be a rough correlation between them,
i.e., groups with large numbers of squamous cell carcinomas
usually also have large numbers of fibrosarcomas.
DISCUSSION
One generalization which can be drawn from the data is
that the middle distillates appear to be more toxic on chronic
dermal administration than the other streams tested except
for 10% catalytic cracked clarified oil. Using the 24-month
survival data in Table 3 for comparison, it can be seen that
only one of the six middle distillate streams tested, straight
run middle distillate, had higher than 8% survival. Three
middle distillate streams, the two mixtures and hydrodesulfurized middle distillate, did not have any surviving animals.
The reasons for this higher toxicity are unknown, but may
be related to the molecular size ranges of the components
of the streams. The low-molecular-weight naphthas are quite
volatile and may not reside on the skin long enough for
systemic absorption of toxic quantities while higher molecular-weight streams may be too large to be efficiently ab-
sorbed. The low toxicity of 0.1 % catalytic cracked clarified
oil is due to its extreme dilution since application of a 10%
solution of this material led to only 6% survival at 12 months
and 100% mortality before 18 months. Skisak et al. (1994)
reported that chronic dermal application of 1% catalytic
cracked clarified oil led to significantly higher mortality in
treated mice over controls at 24 months.
Biles et al. (1988) conducted a series of lifetime dermal
carcinogenesis studies on middle distillates. They did not
observe reduced survivability in the test groups compared
to mineral oil controls. Their administration schedule was
different from that in the present study, however, consisting
of three applications per week of 25 /xl. Therefore their total
weekly application volume was 75 /xl and ours was 100
/xl, and this difference may account for the higher toxicity
observed in our study. Alternatively, there may be a threshold daily dose necessary to achieve toxicity which was exceeded by the 50 /xl used in this study but not by the 25 /xl
applied by Biles et al.
The incidence rates of application site tumors identified
at termination and mean latencies for their appearance are
given in Table 4. The streams exhibiting statistically insignificant numbers of tumors included light alkylate naphtha,
51
DERMAL STUDIES OF PETROLEUM STREAMS
TABLE 3
Percentage Survival in Mouse Groups Chronically
Exposed to Petroleum Streams
Months
6
12
18
24
30
Light alkylate naphtha"
Light catalytic cracked naphtha*
100
98
94
94
56
88
24
62
18
Heavy catalytic reformed naphtha"
100
94
86
36
Heavy thermal cracked naphtha"
94
94
70
24
Heavy catalytic cracked naphtha"
98
92
78
34
Straight run kerosine"
90% straight run middle
light catalytic cracked
70% straight run middle
light catalytic cracked
94
90
70
8
96
92
54
0
0
94
90
60
0
0
96
94
68
18
Light catalytic cracked distillate"
Hydrodesulfurized middle distillate*
100
98
86
80
50
20
6
0
0
Hydrotreated light naphthenic distillate"
Catalytic cracked clarified oil, 0.1%'
Catalytic cracked clarified oil, 10%*
Vaccum residuum, 50%*
Unleaded gasoline*
98
98
90
100
96
82
96
6
98
96
54
88
0
96
88
22
56
0
60
70
10
0
10
10
Sham treated"
Sham treated*
100
90
100
90
96
86
52
62
18
96
96
94
94
80
76
30
52
10
BaP, 0.01%'
BaP, 0.01%*
92
100
92
98
78
84
14
38
0
BaP, 0.05%"
BaP, 0.05%*
96
100
66
86
6
2
0
0
Stream
distillate/10%
distillate''
distillate/30%
distillate*
Straight run middle distillate"
Toluene"
Toluene*
0
" Twenty-four-month exposure, no 30-month data.
* Lifetime exposure.
heavy catalytic reformed naphtha, 0.1% catalytic cracked
clarified oil, vacuum residuum, and unleaded gasoline. The
mice exposed to light catalytic cracked naphtha developed
tumors very late, with six of the seven tumors appearing
after 24 months. Heavy thermal cracked naphtha and heavy
catalytic cracked naphtha demonstrated low tumor incidence. These results show some agreement with those of
Lewis (1983) who reported tumorigenic activity in experimental distillate fractions of crude oil with boiling ranges of
49-177 and 177-288°C. There were important differences
between the streams examined by Lewis (1983) and those in
this study. The fractions examined by Lewis were untreated
except for distillation, whereas those reported here represent
complex physical and chemical treatments as well as distillation. The absence of tumors in mice exposed to heavy catalytic reformed naphtha is evidence of the probably noncarcinogenic character of the alkyl-substituted benzenes which
make up this stream.
Hydrotreated light naphthenic distillate is a low viscosity
oil which has been hydrogenated. Hydrotreatment is intended to reduce or eliminate unsaturation and aromaticity
of PNAs and to cleave heterocyclic compounds with consequent reduction or elimination of carcinogenicity (Haider et
al., 1984). The International Agency for Research on Cancer
(1984) has determined that sufficient evidence exists that
mildly hydrogenated oils are carcinogenic to animals, but
that evidence is inadequate to evaluate severely hydrotreated
oils. The degree of hydrotreatment of the stream used in this
study was undetermined, but the 15% incidence of tumors
obtained indicates that the tumorigenic potency of the stream
is low.
The 10% solution of catalytic cracked clarified oil in toluTABLE 4
Tumor Incidence" In Mice After chronic Dermal Treatment
with Petroleum Streams
Mean latency
(weeks)
Stream
Incidence (%)
Light alkylate naphtha*
Light catalytic cracked naphtha'
Heavy catalytic reformed naphtha'
Heavy thermal cracked naphtha*
Heavy catalytic cracked naphtha*
Straight run kerosine*
90% straight run middle distillate/10%
light catalytic cracked distillate'
70% straight run middle distillate/30%
light catalytic cracked distillate1"
Straight run middle distillate*
Light catalytic cracked distillate*
Hydrodesulfurized middle distillate'
Hydrotreated light naphthenic
distillate*
Catalytic cracked clarified oil, 0.1 %c
Catalytic cracked clarified oil, 10%'
Vacuum residuum, 50%'
Unleaded gasoline'
Sham treated*'
Toluene*
Toluene'
BaP, 0.01%*
BaP, 0.01%'
BaP, 0.05%*
BaP, 0.05%'
2
15'
0
13"
13"
67"
104
118
16"
86
28"
23'
63'
63'
80
93
79
73
15'
4
100
4
4
0
0
9'
53
94
113
22
120
123
65
98
100
72
77
111
87
86
47
49
" Number of mice with histologically confirmed tumors/FEN x 100.
* 24-month exposure.
' Lifetime exposure.
'Significantly different from sham-treated control, p < 0.05.
52
BRODDLE ET AL.
TABLE 5
Comparison of Mouse Dermal Tumor Incidence at 24 Months
and at Termination in Lifetime Studies
Incidence (%)
Stream
24 months
Lifetime
Light catalytic cracked naphtha'
Catalytic cracked clarified oil, 0.1%°
Vacuum residuum, 50%'
Unleaded gasoline"
Toluene*
BaP, 0.01%*
BaP, 0.05*
2
4
2
0
2
50
98
15
4
4
4
9
65
100
" Lifetime study: histopathologicaJly identified tumors at termination are
compared to visually identified tumors at 24 months in the same study.
* Histopathologically identified tumors in the 24-month study are compared to such tumors in the lifetime study.
ene proved to be a stronger carcinogen than 0.05% BaP.
Although exposure to each resulted in 100% incidence of
skin tumors, mean latency was only 22 weeks in the catalytic
cracked clarified oil groups compared to 49 weeks in those
exposed to 0.05% BaP. Catalytic cracked clarified oils may
contain 5% or more of 4- and 5-member condensed ring
aromatic hydrocarbons, and possibly aromatic heterocyclic
compounds as well, and this may explain the potency of
this poorly characterized stream. Biles et al. (1988) utilized
dilutions of catalytic cracked clarified oil as positive controls
in their examination of middle distillate carcinogenicity. Skisak et al. (1994) reported that chronic application of a 1%
solution of catalytic cracked clarified oil to mice led to 100%
tumor incidence but with a mean latency of 72 weeks. Further dilution to 0.1% appears to essentially eliminate the
activity of the stream.
The rest of the materials tested were middle distillate
streams or binary mixtures of middle distillate streams. They
were all tumorigenic, although potency was quite variable
with a tumor incidence of 16% after treatment for the 90/
10 mixture to 67% for straight run kerosine. The potency of
the mixture is somewhat in doubt since it is lower than
that of either of its two components. There is no obvious
correlation of potency with boiling range since the most
volatile middle distillate stream, straight run kerosine, is
among the most potent while the highest boiling stream,
straight run middle distillate, is of relatively low potency.
The positive findings for the middle distillates are in
agreement with previous investigations on such materials
(Biles etal., 1988; Lewis, 1983; Witschi et ai, 1987), which
have consistently found middle distillates to be carcinogenic
in chronic dermal experiments. Biles etal. (1988) discounted
PNAs as a cause of the activity because their concentration
was too low to have contributed to the activity. Analysis of
our middle distillate streams revealed similarly low concentrations of PNAs.
Recent papers (Gerhart et al., 1988; Skisak, 1991; Skisak et al., 1994) have reported on the results of initiation/
promotion studies on middle distillates. In these investigations, middle distillates resulting from various treatments
have been consistently shown to be tumor promoters and
not tumor initiators. Skisak et al. (1994) hypothesized that
middle distillates might act as carcinogens by selectively
inducing cellular proliferation at skin sites previously initiated by background mutagens. The role of irritation in
promotion of carcinogenesis has been debated (Frei and
Stephens, 1968; Gerhart et al., 1988; Scribner and Suss,
1978; Skisak, 1991). The irritation hypothesis usually
identifies compensatory reparative hyperplasia following
irritation and injury as the event critical to promotion.
Skisak (1991) was able to neutralize the dermal carcinogenic activity of a middle distillate by concurrent treat-
TABLE 6
Number of Mice with Specific Tumors after Chronic Dermal
Treatment with Petroleum Streams"
Stream
Fibroma"
SCP*
see
Fibres.*
Light alkylate naphtha
Light catalytic cracked naphtha
Heavy catalytic reformed naphtha
Heavy thermal cracked naphtha
Heavy catalytic cracked naphtha
Straight run kerosine
90% straight run middle distillate/
10% light catalytic cracked
distillate
70% straight run middle distillate/
30% light catalytic cracked
distillate
Straight run middle distillate
Light catalytic cracked distillate
Hydrodesulfurized middle distillate
Hydrotreated light naphthenic
distillate
Catalytic cracked clarified oil, 0.1%
Catalytic cracked clarified oil, 10%
Vacuum residuum, 50%
Unleaded gasoline
Sham treated
Toluene'
Toluene'
BaP, 0.01%'
BaP, 0.019k/
BaP, 0.05%'
BaP, 0.05%/
0
0
0
0
0
0
2
0
2
0
3
1
2
1
I
2
0
1
1
25
3
0
0
0
7
5
0
0
0
1
0
3
3
0
5
13
20
5
2
0
0
0
0
0
0
10
4
1
0
0
9
1
0
0
0
0
8
5
2
0
0
45
1
1
0
0
3
2
15
16
22
" 50 mice per treatment group.
* Squamous cell papilloma.
' Squamous cell carcinoma.
* Fibrosarcoma.
' 24-month exposure.
1
Lifetime exposure.
2
0
2
1
7
2
5
12
3
1
0
5
0
0
0
0
1
6
15
31
34
53
DERMAL STUDIES OF PETROLEUM STREAMS
ment with dexamethasone which inhibits cellular replication. However, dexamethasone has a number of other
physiological effects which may determine or influence
the anti-promoting activity. Thus far, no study has correlated degree of irritation with carcinogenic potency, and
the results in Table 2 also show a lack of correlation. Biles
et al. (1988) found that groups demonstrating the greatest
degree of epidermal degeneration and necrosis produced
the lowest tumor yields. However, it may be that such
severe responses inhibit reparative hyperplasia rather than
increase it.
With some exceptions as indicated in Table 2, the naphtha streams were generally less irritating than the middle
distillates and were also less potent as carcinogens. Within
each of the groups of streams classified as naphthas and
as middle distillates, however, there did not appear to be
a correlation between carcinogenic and irritative activity.
Vacuum residuum, a mixture of very high molecular
weight compounds with very low volatility which did not
possess significant carcinogenic activity, was the only
stream tested which resulted in no signs of irritation in
treated mice.
Although the naphtha streams are defined as being more
volatile than the middle distillates, there is actually a continuum of boiling ranges among the various streams, and
there is considerable overlap between the upper ranges of
the heavy naphthas and the lower ranges of the middle
distillates. It is not unexpected, therefore, that the heavy
naphthas should have the low carcinogenic activity shown
in this study. In fact, the 49-177°C distillate fraction of
the crude oils in the Lewis (1983) study, which was in
the boiling range of naphtha streams, was tumorigenic.
Most refinery streams, including naphthas, are mixtures
of various organic compound classes. Because of this, it
is difficult to associate carcinogenic activity of the streams
with any particular type of compound. However, some of
the naphthas are products of treatments which lead to one
compound type predominantly. More than 90% of our
sample of heavy catalytic reformed naphtha consisted of
alkyl-substituted benzenes, and light alkylate naphtha is
made up almost completely of branched-chain aliphatic
hydrocarbons. The lack of activity of these streams suggests that these classes of compounds are not dermal carcinogens, at least in the molecular weight ranges found
in the naphthas. Horton et al. (1957) and Sice (1966)
showed that aliphatic straight-chain hydrocarbons of particular size, such as dodecane, were accelerators or promoters of carcinogenesis. Varying concentrations of these
or similar compounds may be expected to occur in naphthas, particularly heavy naphthas, and middle distillates.
These may determine or contribute to the activity of these
streams.
ACKNOWLEDGMENT
We thank Ms. Re'Naye Williams for her diligent editing and administrative help in preparing the manuscript.
REFERENCES
Bell. J. (1876) Paraffin epithelioma of the scrotum. Edinburgh Med. J. 22,
135.
Biles, R. W., McKee, R. H., Lewis, S. C , Scala, R. A., and DePass, L. R.
(1988) Dermal carcinogenic activity of petroleum-derived middle distillate fuels. Toxicology 53, 301-314.
Bingham, E., and Falk, H. (1969) Environmental carcinogens: The modifying effect of carcinogens on the threshold response. Arch. Environ,
Health 19, 779-783.
Bingham, E., and Horton, A. E. (1966). Environmental carcinogenesis:
Experimental observations related to occupational cancer. Adv. Biol. Skin
7, 183-193.
Bradley, J. V. (1968). Distribution Free Statistical Tests, pp. 195-200.
Prentice Hall International, Englewood Cliffs, NJ.
Craig, P. H. (1989). Skin carcinogenicity bioassays of petroleum refinery
streams: Issues of interpretation. Prog. Clin. Biol. Res. 298, 381-391.
Cruikshank, C. N. D., and Squire, J. R. (1950). Skin cancer in the engineering industry from the use of mineral oil. Br. J. Indust. Med. 7, 1 —
11.
Doak, S. M. A., Brown, V. K. H., Hunt, P. F., Smith, J. D., and Roe,
F. J. C. (1983). The carcinogenic potential of twelve refined mineral oils
following long-term topical application. Br. J. Cancer 48, 429—436.
Frei, J. V , and Stephens, P. (1968). The correlation of promotion of tumor
growth and of induction of hyperplasia in epidermal two-stage carcinogenesis. Br. J. Cancer 22, 83-92.
Gerhart, J. M., Hatoum, N. S., Haider, C. A., Warne, T. M., and Schmitt,
S. L. (1988). Tumor initiation and promotion effects of petroleum streams
in mouse skin. Fundam. Appl. Toxicol. 11, 76-90.
Haider, C. A., Wame, T. M , Little, R. Q., and Garvin, P. J. (1984). Carcinogenicity of petroleum lubricating oil distillates: Effects of solvent refining,
hydropressing, and blending. Am. J. Ind. Med. 5, 265-274.
Henry, S. A. (1946). Cancer of the Scrotum in Relation to Occupation.
Oxford Univ. Press, London.
Henry, S. A. (1947). Occupational cutaneous cancer attributable to certain
chemicals in industry. Br. Med. Bull. 4, 389-401.
Horton, A., Denman, D. T., and Trosset, R. P. (1957). Carcinogenesis of the
skin. II. The accelerating properties of aliphatic and related hydrocarbons.
Cancer Res. 17, 758-774.
International Agency for Research on Cancer (1984). Monogr. Vol. 33, pp.
150-151.
Kane, M., LaDov, E , Holdsworth, C , and Weaver, N. (1984). Toxicological
characteristics of refinery streams used to manufacture lubricating oils.
Am. J. Ind. Med. 5, 183-200.
Kennaway, E. L. (1934). On cancer producing tars and tar fractions. J. Ind,
Hyg. 5, 462-488.
Leitch, A. (1922). Paraffin cancer and its experimental production. Br Med.
J. 2, 1104-1109.
Lewis, S. C. (1983). Crude petroleum and selected fractions: Skin cancer
bioassays. Prog. Exp. Tumor Res. 26, 68-84.
Lione, J. G., and Denholm, J. S. (1959). Cancer of the scrotum in wax
pressmen. AMA Arch. Ind. Health 19, 530-539.
McKee, R. H., Plutnick, R. T., and Przygoda, R. T. (1989). The carcinogenic
initiating and promoting properties of a lightly refined paraffinic oil.
Fundam. Appl. Toxicol. 12, 748-756.
Roy, T. A., Johnson, S. W., Blackburn, G. R., and Mackerer, C. R. (1988).
Correlation of mutagenic and dermal carcinogenic activities of mineral
oils with polycyclic aromatic compound content. Fundam. Appl. Toxicol.
10, 466-476.
54
BRODDLE ET AL.
Scribner, J. D., and Suss, R. (1978). Tumor initiation and promotion. Int.
Rev. Exp. Pathol. 18, 158-163.
Twort, C. C , and Twort, J. M. (1931). The carcinogenic potency of mineral
oils. J. Ind Hyg. Toxicol. 13, 204-226.
Sice, J. (1966). Tumor promoting activity of n-alkanes and 1-alkanols.
Toxicol. Appl. Pharmacol. 9, 70-74.
Twort, S. W., and Twort, C. C. (1939). Comparative activity of some
Skisak, C. M. (1991). The role of chronic acanthosis and subacute inflammation in tumor promotion and CD-I mice by petroleum middle distillates. Toxicol. Appl. Pharmacol. 109, 399-411.
Volkmann, R. (1875). Ubev Theer-Paraffin und Russkregs (Shornsteinfeg-
Skisak, C. M., Furedi-Machacek, E. M., Schrrutt, S. S., Swanson, M. S.,
and Vemot, E. H. (1994). Chronic and initiation/promotion skin bioassays
of petroleum streams. Environ. Health Perspect. 102, 82-87.
Snedecor, G. W., and Cochrane, W. G. (1980). Statistical Methods, 7th ed.,
pp. 470-471, Iowa State Uruv. Press, Ames.
carcinogenic hydrocarbons. Am. J. Cancer 25, 80-85.
erkrebs). In Beitrage zur Chimgie. Breitkopf & Hartel, Leipzig.
Waterhouse, J. A. H. (1971). Cutting oils and cancer. Ann. Occup. Hyg.
14, 161-170.
Witschi, H. P., Smith, L. H., Frome, E. L., Pequet-Goad, M. E., Griest,
W. H., Ho, C.-H., and Guerin, M. R. (1987). Skin tumorigenic potential
of crude and refined coal liquids and analogous petroleum products.
Fundam. Appl. Toxicol. 9, 197-303.