Arch Environ Contam Toxicol (2008) 54:337–347 DOI 10.1007/s00244-007-9021-5
Chronic Oral Exposure to Bunker C Fuel Oil Causes Adrenal Insufficiency in Ranch Mink (Mustela vison) F. C. Mohr Æ B. Lasley Æ S. Bursian
Received: 24 December 2006 / Accepted: 22 June 2007 / Published online: 1 September 2007 Springer Science+Business Media, LLC 2007
Abstract Animals living in the near-shore marine environment are predisposed to contact with chemical contaminants through land- and ocean-based activities. The release of petroleum hydrocarbons into the marine environment is a stressor to this environment and its resident wildlife. The stress response to chemical threats is dependent on an intact hypothalamic-pituitary-adrenal axis, which also may be a target to the effects of these chemicals. Ranch mink (Mustela vison) were used as surrogates for sea otters (Enhydra lutris) to examine the development of adrenal hypertrophy after chronic, oral exposure to low concentrations of bunker C fuel oil. Animals were fed three different concentrations of fuel oil (48, 520, and 908 ppm) or mineral oil (control) for 60–62 days. At the end of the exposure, blood and fecal samples were collected and organs were weighed and examined microscopically. In all fuel oil groups, exposure resulted in adrenal hypertrophy, an adaptation suggestive of adrenal activation. However, concentrations of serum and fecal glucocorticoids and serum progesterone were not elevated over control values. Hematologic parameters and serum chemistries showed no changes consistent with increased adrenal activity. In
F. C. Mohr (&) Department of Veterinary Pathology, Microbiology and Immunology, School of Veterinary Medicine, University of California, Davis, CA 95616, USA e-mail:
[email protected] B. Lasley Department of Population Health and Reproduction, School of Veterinary Medicine, University of California, Davis, CA 95616, USA S. Bursian Department of Animal Science, Michigan State University, East Lansing, MI 48824, USA
addition, adrenal glands from animals fed the higher concentrations of fuel oil contained large numbers of heavily vacuolated cells. We conclude that petroleum hydrocarbons are inducing an adrenal insufficiency that leads to the adaptive enlargement of the gland. This would increase the susceptibility of fuel oil-exposed animals to the deleterious effects of other environmental stressors. Keywords Adrenal insufficiency Bunker C fuel oil Ranch mink Mustela vision Oral exposure
Introduction The release of petroleum oil into the coastal marine environment is an important stressor to this sensitive ecosystem. A common type of petroleum oil entering into this environment is bunker C fuel oil, a refined, heavy oil, which is transported along marine routes and used to power ships (Irwin et al. 1998). This oil degrades slowly and can contaminate marine environments for years (Vandermeulen and Singh 1994; Irwin et al. 1998). Fuel oil is known to be more toxic than other petroleum oils because of its high polycyclic aromatic hydrocarbon (PAH) content (Anderson et al. 1974; Irwin et al. 1998). Marine animals living in coastal marine ecosystems are at risk for exposure to petroleum hydrocarbons, which can originate from both acute and chronic entry into the environment. The acute effects of petroleum oil contamination have been studied in the sea otter (Enhydra lutris) in detail because this animal is particularly vulnerable to coastal petroleum oil contamination (Lipscomb et al. 1994; Williams et al. 1995). Population and biomarker studies have documented the effects of chronic petroleum oil exposure on Alaskan sea otters (Bodkin et al. 2002). However,
123
338
research on the effects of chronic petroleum oil has been limited in sea otters, in part, because of the threatened status of this species in California and because of the difficulty in performing controlled experiments. We have been studying the mink (Mustela vison) as a surrogate species to understand the effects of bunker C fuel oil on sea otters and other marine foraging mustelids. Chronic, oral exposure to bunker C fuel oil in mink is known to cause hematological, immunological, and reproductive toxicity (Mazet et al. 2000; Mazet et al. 2001; Schwartz et al. 2004a, b). We have observed that the adrenal glands of mink that were chronically exposed to a low concentration of fuel oil were, on average, increased in weight compared to those of control animals. Furthermore, the concentration of glucocorticoids in the feces of fuel oilexposed animals, as a measure of adrenal function, was not elevated (Schwartz et al. 2004a). However our study was designed to measure immunological parameters, which meant that the animals were frequently handled and bled. All of these manipulations are stressors that could influence adrenal gland size and function. Normal adrenal function is required for health. In addition, the adrenal gland is a major stress-response organ and a part of the hypothalamic-pituitary-adrenal (HPA) axis. Adrenal growth and secretion of glucocorticoids are tightly regulated by elements of this axis. An important regulator of adrenal gland function is adrenocorticotropic hormone (ACTH), which is released from the pituitary gland. The adrenal gland can be a site of toxicity for foreign chemicals that target steroidogenic pathways. Inhibition of these pathways in reproductive organs by toxicants has been extensively studied; whereas, adrenal steroidogenesis as a target for toxicants has received much less attention (Harvey and Everett 2003). Moreover, most of these studies have been done in cell culture. It is well known that some environmental contaminants act as endocrine disruptors by affecting the reproductive system. More attention needs to be directed toward the effects of these contaminants on other endocrine organs such as the adrenal gland. Our previous finding that chronic oral exposure to fuel oil is associated with an increase in adrenal gland weight in mink and that the increase in weight is without an associated increased in the fecal glucocorticoid concentration suggests that components in fuel oil are interfering with the normal function of the HPA axis. The objective of this study is to expose mink chronically to low, environmentally relevant concentrations of fuel oil under experimentally controlled conditions that minimize the input of stressors other than the feeding of fuel oil. Under these conditions we can assess adrenal size and output of steroid hormones in animals fed different concentrations of bunker C fuel oil to understand the targets of fuel oil
123
Arch Environ Contam Toxicol (2008) 54:337–347
toxicity and the role of fuel oil in inducing adrenal hypertrophy.
Materials and Methods Fuel Oil Bunker C fuel oil (fuel oil No. 6) was a gift from Dr. Jonna Mazet, Wildlife Heath Center, University of California, Davis, and originated from British Petroleum, Cherry Point Oil Refinery, Washington. The concentrations of fuel oil used in this exposure were selected to reflect concentrations of petroleum oil that might be found in the environment over a prolonged period of time following an accidental oil spill (Mazet et al. 2000). Other studies have used the same petroleum product to evaluate systemic, hematological, immune, and reproductive changes in mink in response to chronic, oral exposures (Mazet et al. 2000, 2001; Schwartz et al. 2004a, b).
Animals and Experimental Design Mink were housed from mid-January to mid-March in a facility that is operated in accordance with USDA regulations and a protocol approved by the Michigan State University Institutional Animal Care and Use Committee. Forty-eight male, 8-month-old ranch mink random-bred and raised at the Michigan State University Experimental Fur Farm were transferred to an indoor room 8 days prior to the beginning of the exposure. Mink were weighed and housed in individual wired pens with wooden nest boxes attached. Metal screens were placed underneath each cage to collect feces. The room lighting was kept on a 12-h light-dark cycle and the average room temperature over the course of the exposure was 8.9 ± 2.4C (SD). To minimize extraneous stressors, the animals were not handled throughout the exposure. Drinking water was provided to animals ad libitum. The animals were fed once daily a standard ranch mink diet containing either bunker C fuel oil or mineral oil (control diet). A screen of the diet for organochlorine contaminants detected no PCBs and a trace amount of DDE (\10 ng/g). Animals were assigned to four groups corresponding to a targeted concentration of bunker C fuel oil of 0 (mineral oil), 50, 500, and 1000 ppm fuel oil. The assignment was blocked on initial animal weight, in order to ensure a comparable distribution of initial weights in each treatment group. To maintain a similar consistency of oil in the control and fuel oil-containing diets, mineral oil was added to the 50- and 500-ppm fuel oil diets to ensure that the combined total of fuel oil and/or mineral oil in each diet was equal to 1000-ppm. The food was stored at –20C in
Arch Environ Contam Toxicol (2008) 54:337–347
aliquots sufficient for a day’s feeding. Analysis of each diet after the initial mixing and at the end of the exposure was performed to verify the concentration of fuel oil added to each group. The initial and ending concentrations for each fuel oil-containing diet were averaged, yielding mean concentrations of fuel oil in the experimental diets of 48, 520, and 908 ppm. Mink were fed fuel oil for 60–62 days. At the completion of the exposure, the animals were anesthetized by an intramuscular administration of the combination of 22 mg/kg ketamine (Fort Dodge Animal Health, Fort Dodge, IA) and 2 mg/kg xylazine (Ben Venue Laboratories, Bedford, OH). Intracardiac blood samples were taken from all animals and submitted for complete blood counts, analysis of serum chemistries, and determination of the serum concentrations of glucocorticoid and progesterone. To minimize the effects of diurnal variation on serum glucocorticoid concentrations, blood samples were collected from every animal over a 1-h period beginning at 9 AM. Following blood collection, the animals were euthanized by CO2 asphyxiation, and necropsies were performed over a 2-day period to record organ measurements and to complete a gross and histological evaluation. Blood samples and subsequent necropsies were performed in an order designed to ensure that animals representing all four groups were evenly intermixed throughout the procedure. There were no correlations between day or time of necropsy and either body weights, adrenal weights, relative adrenal weights, serum glucocorticoid and progesterone concentrations, or fecal glucocorticoid concentrations.
Serum Hormone Analysis In mink, cortisol is the major glucocorticoid synthesized by the adrenal gland with minor contributions from corticosterone (Aulerich et al. 1999). Concentrations of glucocorticoid in the serum were measured in duplicate (0.02 ml/sample) by a competitive immunoassay using direct chemiluminescent technology (Bayer Diagnostics ADVIA ACS-180 Automated Chemiluminescence System, Norwood, MA). The specificity of the antibody for cortisol and corticosterone was 100% and 2.8%, respectively. The intra-assay coefficient of variation was 3% and the limit of detection was 0.20 lg/dl. No interassay coefficient of variation was calculated because all samples were analyzed in the same assay. We have compared this method with a radioimmunoassay (RIA) procedure (Coat-A-Count; Diagnostic Products Corp., Los Angles, CA), which we have used previously, and found a good correlation between the two assays (r = 0.98). Serum concentrations of progesterone were measured using a commercial RIA kit (Coat-A-Count). Duplicate aliquots (0.05 ml) were assayed. The intra-assay coefficient
339
of variation was 5% and the limit of detection was 0.02 ng/ ml. No interassay coefficient of variation was calculated because all samples were analyzed in the same assay.
Fecal Cortisol Analysis Samples of feces were collected from every animal on days 0, 13, 27, 41, 55, and 59 of the exposure and stored at –20C. The protocol for the extraction of glucocorticoids has been published previously (Schwartz et al. 2004a). Fecal cortisol assay was performed by competitive immunoassay using direct chemiluminescent technology (ACS-180) described above. Each sample (0.02 ml/sample) was measured in triplicate. Because of large number of samples, the extractions and assays were performed in six separate groups. Each group contained samples comprising the complete time series from individual animals that were representative of all treatment and control groups. The intraand interassay coefficients of variation were 9% and 14%, respectively. Statistically no group effect was observed.
Necropsy and Histopathology A necropsy was performed on all animals and organ weights were recorded at the end of the exposure. Tissues from each animal were preserved in 10% neutral buffered formalin, embedded in paraffin, thin sectioned, and stained with hematoxylin and eosin (H&E) for evaluation by light microscopy. The initial evaluation of the microscopic sections was done blind. Transverse sections from the middle of the right adrenal gland were examined in all animals. Microscopic evaluation of the distribution of cellular vacuolation in the zona fasciculata and reticularis of the adrenal glands was scored blindly, with each gland assigned to one of three groups: no to minimal (I), moderate (II), or marked (III) increases in the number of cells with the cytoplasm expanded with vacuoles.
Statistical Analyses Data are expressed as individual values or as means ± SE for each treatment group. Organ weights (wet) are expressed as a percentage of body weight (relative weight) for each animal and then averaged. Data were analyzed after an adjustment for initial weight, using an analysis of covariance model with initial body weight as a covariate and oil dosage as a categorical predictor. When the residuals from an analysis of the raw data failed a Wilk-Shapiro test of normality, the analysis was rerun using log-transformed data. When neither the raw nor the log analysis
123
340
Arch Environ Contam Toxicol (2008) 54:337–347
Table 1 Body and relative organ weights for mink fed different concentrations of fuel oil Fuel oil concentration 0 ppm (n = 12)
48 ppm (n = 11)
520 ppm (n = 11)
908 ppm (n = 12)
Body weight (g)
1918a (55)
1962a (73)
2135a (57)
Left adrenal (%)
a
0.0034 (0.0001)
b
0.0039 (0.0003)
b
0.0038 (0.0003)
0.0046b (0.0003)
Right adrenal (%)
0.0027a (0.0001)
0.0033b (0.0001)
0.0033b (0.0003)
0.0038b (0.0002)
Liver (%) Brain (%)
a
3.47 (0.13) a
0.55 (0.02)
3.66
a,b
0.53
a,b
(0.17) (0.03)
(0.07)
4.48c (0.10)
0.49 (0.02)
0.57a (0.03)
3.68
b,c
1825a (72)
b
Note. Values are mean (SE). Body weights were measured at the end of the exposure. Organ weights are expressed as a percentage of body weight. Mean values in the same row with no superscripts in common are significantly different (p \ 0.05)
passed a test of normality, a Kruskal-Wallis test was used to compare groups. Post hoc comparisons among the exposure groups were based on least squared means (LSMEANS). Spearman correlation coefficients were calculated to determine the relationships between different measured variables. The actual correlation coefficients are included in the text along with the p-values. To establish if exposure to fuel oil had an effect on the correlations between other variables, partial correlation coefficients were also calculated, thereby adjusting for the possible impact of fuel oil exposure. Results were considered statistically significant whenever p \ 0.05. Results Clinical Observations Over the course of the exposure, the mink continued to eat the fuel oil-containing and control diets. Mean body weights of the animals did not differ significantly among the different treatment groups at the beginning (not shown) or at the end of the exposure (Table 1). In addition, body weights did not vary significantly between the initial and final weights for each group. Our observations of the mink throughout the exposure period did not reveal any detectable differences in appearance or changes in the health of the animals within or among the groups. However, two animals (from the 48- and 520-ppm fuel oil groups) were removed from the study because of findings gathered at necropsy and abnormal results of serum chemistry profiles. Both animals had evidence of chronic renal disease with urolithiasis, which was confirmed by microscopic examination. Data collected from these two animals were not included in the final analysis of the different measured parameters.
Organ Weights Within the data set of animals included in this study, the only observed changes noted at necropsy that were attributed to
123
fuel oil exposure were enlarged adrenal glands and livers. On gross inspection the adrenal glands were normal in shape and color and varied only in size. Mean relative weights of the combined left and right adrenal glands from mink in the fuel oil groups were significantly larger compared to those of the control group (0 ppm) (Fig. 1), representing a 19%, 17%, and 40% increase in the mean relative weights over the control group (mean ± SE: 0.0060% ± 0.0002%, 0.0072% ± 0.0004%, 0.0071% ± 0.0006%, and 0.0084% ± 0.0005% for the 0-, 48-, 520-, and 908-ppm groups, respectively). However, among the fuel oil-exposed groups differences in combined relative adrenal weights were not significant. The large range of relative adrenal weights in the fuel oil-exposed animals compared to the control animals may be attributed to how the animals are feed. The food is semimoist in consistency and placed on top of the cage for the animals to pull into the cage and consume. Depending on the amount of spillage of food, the concentration of fuel oil consumed by each animal will be variable. This can lead to differences in fuel oil intake, which could influence adrenal weights and other parameters. The increases in adrenal relative weights in the fuel oilexposed groups were attributed to increases in both the left and the right adrenal glands (Table 1). Left adrenal relative
Fig. 1 Effect of bunker C fuel oil on adrenal gland relative weight. Animals were exposed to 0, 48, 520 and 908 ppm fuel oil daily in the diet. Circles represent relative weights of the combined adrenal glands from individual animals. Bars indicate mean relative weights for each group. Asterisks indicate significant differences (p \ 0.05) between control and fuel oil groups
Arch Environ Contam Toxicol (2008) 54:337–347
weights were larger than right adrenal relative weights in all four groups of animals (Table 1). This was also true when comparing gross weights between the left and the right adrenal glands in every group (data not shown). Interestingly we have observed similar findings between the left and the right adrenal glands in two other fuel oil exposure studies with ranch mink. Adrenal wet weights and relative weights were positively correlated with fuel oil exposure (0.53301 and 0.50526, respectively; both p’s \ 0.001). Differences in mean relative weights of livers between the fuel oil-exposed and the control animals (Table 1) were also observed. The majority of livers were normal in color, with only a few (n = 5) livers from the 0 (n = 1)-, 48 (n = 3)-, and 520 (n = 1)-ppm groups having notable yellow mottling. Livers from all groups had sharp edges to the lobules, which signified the absence of swelling. Liver relative weights were greater at the two highest fuel oil concentrations compared to those in the 0-ppm group, while no differences in relative weights were detected between the 0-ppm and the 48-ppm fuel oil exposed groups. Within the fuel oil-exposed groups there were also significant differences in relative liver weights between the 48- and the 908-ppm groups. Liver wet weights and relative weights were positively correlated with fuel oil exposure (0.49220 and 0.59995, respectively; both p’s \ 0.001) The brains also showed differences in mean relative weights between groups. There was a significant decrease in brain weights in the 520-ppm fuel oil group compared to the control and 908-ppm groups. Other organs did not show differences in relative weights among the exposure and control groups (heart, left and right kidney, spleen, and mesenteric and left and right prescapular lymph nodes).
341
cell size (Fig. 2A). In the other pattern, the cell’s cytoplasm was expanded, with excessive vacuolation composed of larger, often coalescing, lipid droplets. These cells lost their eosinophilic staining characteristics with H&E staining and had a foamy appearance (Fig. 2B). The three grades (I–III) describing the number of cortical cells in the zona fasciculata and reticularis that showed excessive vacuolization (demonstrated in Fig. 2B) were observed in all groups of mink. In the control and 48ppm fuel oil group the percentages of each grade were similar (Fig. 3). In the 520-ppm group the percentage of grade II adrenals increased, while the percentage of grade I adrenals decreased. The biggest change was observed in the 908-ppm group, with the disappearance of grade I- and a large increase in grade III-classified adrenal glands. In addition, a positive correlation was found between the histological grading scheme and chronic exposure to fuel oil (0.35884; p = 0.01) and adrenal relative weight (0.33859; p = 0.02). If the analysis was adjusted to remove the influence of fuel oil exposure, both correlations became insignificant. This suggests that the increased area of the
Histological Changes Microscopic examination of the adrenal glands in each treatment group showed that the increases in relative weights of the glands in the fuel oil-exposed groups were correlated with enlargement (hypertrophy) of the entire gland. The greatest increase was in the two inner zones of the cortex, the zona fasciculata and reticularis. The junction of the zona fasciculata and reticularis was difficult to distinguish in the mink; therefore, it was not possible to determine if the increase in size of the cortex was due to an individual zone or both zones. Adrenal cortical cells in the zona fasciculata and reticularis had two morphological patterns. In one pattern, the cells contained fine, faintly visible cytoplasmic vacuoles that corresponded to lipid droplets containing cholesterol esters (steroid precursors). The lipid droplets did not alter
Fig. 2 Histological appearance of cortical cells in the zona fasciculata and reticularis. A Representative photomicrograph of cells in the zona fasciculata containing small, faintly visible cytoplasmic vacuoles. B Representative photomicrograph of enlarged cells with cytoplasm filled with larger, coalescing vacuoles. (Original magnifications, 400x.)
123
342
inner cortex containing cells with excessive vacuolization was related to fuel oil exposure. A histological correlate for the enlarged livers in the fuel oil-exposed groups was not evident. In every animal and in all groups, hepatocytes had cytoplasmic lipid accumulation in small vacuoles (microvesicular fatty change) that were diffusely present throughout the hepatic lobules. Although a few livers from the 0-, 48-, and 520-ppm groups grossly had yellow, mottled livers, there was no histological evidence that the color variation was due to increased hepatocellular lipidosis. The cause of mottling pattern to these livers is not known. Another finding observed in both control and fuel oil-exposed groups was mild to moderate accumulations of mononuclear cells in the portal areas of the liver. These accumulations were found in 66%, 36%, 18%, and 42% of the livers in the 0-, 48-, 520-, and 908ppm groups, respectively. There were no significant correlations between the portal mononuclear infiltrates and fuel oil exposure or liver wet weight. In other organs no histological lesions were observed that could be attributed to fuel oil exposure (skin, brain, pituitary, thyroid, heart, lung, esophagus, stomach, small intestine, colon, pancreas, kidney, urinary bladder, prostate, testis, epididymis, spleen, tonsils, thymus and retropharyngeal, sternal, prescapular lymph nodes).
Arch Environ Contam Toxicol (2008) 54:337–347
however, serum glucocorticoid concentrations in the three fuel oil groups were not significantly different from those in the control group (mean ± SE: 1.23 ± 0.49, 0.99 ± 0.29, 0.58 ± 0.16, and 0.42 ± 0.10 lg/ml for the 0-, 48-, 520-, and 908-ppm groups, respectively). Progesterone is another steroid produced by the cortical cells of the adrenal gland and is an intermediate in the synthesis of cortisol and corticosterone. Serum concentrations of progesterone paralleled the serum concentrations of glucocorticoids in all groups of animals (Fig. 4B), with no significant differences among any of the fuel oil exposure groups and the control group (mean ± SE: 0.48 ± 0.15, 0.32 ± 0.05, 0.26 ± 0.05, and 0.25 ± 0.04 ng/ml for the 0-, 48-, 520-, and 908-ppm groups, respectively). Changes in serum glucocorticoid and progesterone concentrations were positively correlated with each other (0.49768; p \ 0.001), and this was independent of fuel oil exposure. Neither the serum cortisol nor the progesterone concentration correlated with the adrenal relative weight or histological grading scheme.
Fecal Glucocorticoid Measuring the concentrations of glucocorticoids in feces is a way of quantifying adrenal glucocorticoid release over multiple time points without stressing the mink. This is
Serum Glucocorticoid and Progesterone To correlate adrenal hypertrophy induced by chronic fuel oil exposure with the release of glucocorticoid, serum concentrations of glucocorticoid and progesterone were measured in all animals at the end of the exposure period. Mean values for the serum glucocorticoid concentration declined with increasing fuel oil concentration (Fig. 4A);
Fig. 3 Effect of bunker C fuel oil on adrenal cortical vacuolation. Animals were exposed to 0, 48, 520, and 908 ppm fuel oil daily in the diet. The graph shows the percentage contribution of each of the three grading classifications for each concentration of bunker C fuel oil and the control. Grade I indicates no to mild, grade II indicates moderate, and grade III indicates marked increases in the number of cells swollen with cytoplasmic vacuoles in the zona fasciculata and zona reticularis
123
Fig. 4 Effect of bunker C fuel oil on serum glucocorticoid and progesterone concentration. Animals were exposed to 0, 48, 520, and 908 ppm fuel oil daily in the diet. Circles represent the serum concentrations of glucocorticoids (A) or progesterone (B) from individual animals. Bars indicate mean concentrations for each group. There were no significant differences among the groups
Arch Environ Contam Toxicol (2008) 54:337–347
343
important when it is necessary to minimize additional stressors to the animals other than the experimental one (fuel oil). Fecal glucocorticoid concentrations did not differ significantly among the three fuel oil groups over the time course of the exposure (Fig. 5). There was a small but significant decrease in fecal cortisol concentration in the 48-ppm fuel oil group compared to the control group (p = 0.04) when these two groups were compared over the entire exposure period. Fecal glucocorticoid concentrations in the two higher fuel oil groups (520 and 908 ppm) were not significantly different from those in the control group.
Hematology and Serum Chemistry The hematopoietic system of mink was affected by oral exposure to fuel oil. Erythrocytic parameters that differed (p \ 0.05) between the fuel oil and the control groups included the hematocrit, erythrocyte count, hemoglobin concentration, mean corpuscular volume (MCV), and plasma total solids (Table 2). The first three of these parameters were all slightly, but significantly, decreased in the 520- and 908-ppm fuel oil exposure groups compared to the control and 48-ppm groups, whereas the MCV was elevated in animals exposed to the two highest concentrations of fuel oil. In addition, plasma total solids were also decreased and followed a similar pattern in significance as the first three erythrocytic parameters. Leukocyte and platelet numbers were not significantly different among the control and the three fuel oil exposure groups. The results of serum chemistry analysis in fuel oilexposed and control mink are summarized in Table 3. Serum chemistry values that differed among the four groups were the serum sodium, calcium, total protein, globulin, cholesterol and total bilirubin concentrations, and alkaline phosphatase (ALP) and alanine aminotransferase (ALT) activities. These parameters were significantly different in the 908-ppm fuel oil group compared to the control group with the exception of ALT. In the 908-ppm group ALT was significantly
Fig. 5 Effect of bunker C fuel oil on fecal glucocorticoid concentration. Animals were exposed to 0, 48, 520, and 908 ppm fuel oil daily in the diet. Samples from individual animals were collected over 59 days of fuel oil exposure. Fecal glucocorticoid concentrations, as nanograms per gram dry weight feces (dwf), are expressed as mean ± SE
higher than in the two lower fuel groups but not the control group. Except for sodium and ALT, all other serum values decreased in value in the fuel oil-exposed animals compared to the controls. This decrease was most notable for ALP, total bilirubin, and cholesterol. This is further substantiated by a significant negative correlation between the fuel oil concentration and the serum ALP (–0.39760; p = 0.006) and total bilirubin (–0.47808; p \ 0.001). No other serum values correlated with fuel oil concentration. Certain serum chemistry parameters are indicators of liver function and injury. Relative liver weights were positively correlated with ALT (0.37171; p = 0.01) and negatively correlated with total bilirubin (–0.33607; p = 0.02), both of which were dependent on fuel oil exposure. Serum ALT and ALP correlated positively with mononuclear infiltrates in the portal regions of the liver (0.49395, p = 0.005, and 0.45269, p = 0.002, respectively), and these correlations were not dependent on fuel oil exposure. No correlations were found between the hepatic infiltrates and aspartate transferase or total bilirubin. Both glucose and cholesterol showed no correlation with other liver-related parameters, histological appearance of the liver, or relative liver weights.
Table 2 Hematological parameters for mink fed different concentrations of fuel oil Fuel oil concentration
Hematocrit (%) 6
Erythrocytes (·10 /ll) Hemoglobin (g/dl) MCV (fl)
0 ppm (n = 12)
48 ppm (n = 11)
56 ± 0.7a
55 ± 1.5a
52 ± 1.0b
51 ± 1.0b
a
a
b
8.56 ± 0.1b
b
16.46 ± 0.3b 62 ± 0.8b
9.26 ± 0.2
a
17.45 ± 0.2 61 ± 0.6a
9.07 ± 0.2
a
17.31 ± 0.4 61 ± 0.6a
520 ppm (n = 10)
8.55 ± 0.2
16.65 ± 0.3 63 ± 0.7b
908 ppm (n = 12)
Note. MCV, mean corpuscular volume.Values are mean ± SE. Mean values in the same row with no superscripts in common are significantly different (p \ 0.05)
123
344
Arch Environ Contam Toxicol (2008) 54:337–347
Table 3 Serum chemistry values for mink fed various concentrations of fuel oil Fuel oil concentration
Sodium (mmol/L)
0 ppm (n = 12)
48 ppm (n = 11)
520 ppm (n = 10)
908 ppm (n = 12)
150 ± 0.4a
152 ± 0.5a,b
152 ± 0.8a,b
152 ± 0.4b
a,b
9.6 ± 0.1b
6.3 ± 0.1b,c
6.1 ± 0.1c
a
a
Calcium (mg/dl)
10 ± 0.1
10 ± 0.1
Total protein (g/dl)
6.7 ± 0.1a
6.4 ± 0.1a,b
a
a,b
Globulin (g/dl) ALP (U/L) ALT (U/L) Total bilirubin (mg/dl) Cholesterol (mg/dl)
3.0 ± 0.1 a
68 ± 7
2.8 ± 0.1 51 ± 3
384 ± 96
a,b
a
0.38 ± 0.2 379 ± 28a
b
9.8 ± 0.1 2.6 ± 0.1
b
2.7 ± 0.1b
b
51 ± 6b
48 ± 4
206 ± 50
a
a
0.20 ± 0 316 ± 25b
a
442 ± 109b
172 ± 30
b
0.16 ± 0.02 234 ± 11c
0.15 ± 0.02b 239 ± 10c
Values are mean ± SE. Mean values in the same row with no superscripts in common are significantly different (p \ 0.05). Parameters measured but not significantly different among the groups: albumin, amylase, aspartate aminotransferase, blood urea nitrogen, chloride, creatine kinase, creatinine, direct and indirect bilirubin, iron, magnesium, phosphorous, and potassium
Discussion Chronic exposure to fuel oil causes bilateral adrenal hypertrophy in ranch mink. Also, at the completion of the exposure the serum concentrations of glucocorticoids and progesterone were no different among the fuel oil-exposed groups and control groups. This observation has now been demonstrated under experimental conditions minimizing the influence of other stressors. Fuel oil at 50 ppm may not be the lower limit for causing adrenal hypertrophy because there were no significant differences among the relative adrenal weights in the three fuel oil exposure groups. Blood sample measurements of glucocorticoids give point-in-time assessments of adrenal activity, while fecal samples measure a composite of adrenal activity and glucocorticoid secretion over longer periods corresponding to food transit times. For this exposure the benefits of measuring fecal glucocorticoid concentrations were in collecting samples with minimal stress to the animals and in following the pattern of the fecal glucocorticoid concentration over time. Our finding of no or slightly decreased changes in the fecal glucocorticoid concentrations strengthens our observation that serum glucocorticoid concentrations are not changed with chronic fuel oil exposure or adrenal enlargement. Although not statistically significant, there was a trend in the temporal pattern of fecal glucocorticoid concentrations that showed a decrease in glucocorticoid concentrations at the last three time points in the control and 48-ppm groups and little to no change in glucocorticoid concentrations in the two other fuel oil groups. This observation suggests that hypertrophied adrenal glands in animals exposed to high concentrations of fuel oil could be releasing more adrenal steroids than the control animals. Ranch mink exposed to weathered crude oil (100 and 1000 ppm) over 4 months showed no evidence of adrenal
123
hypertrophy (Beckett et al. 2002). Ferrets (Mustela putorius) were exposed orally to naturally weathered Alaskan North Slope crude oil for 5 days. Adrenal/body weight ratios from animals receiving this oil at 500, 1000, or 5000 mg/kg body weight once daily by intubation did not differ from control animals (Stubblefield et al. 1995).1 These concentrations were much higher than the concentrations of fuel oil fed to the mink in this study (3, 33, and 58 mg fuel oil/kg body weight/day for the 48-, 520-, and 908-ppm groups). The effects of petroleum hydrocarbons on adrenal size and function have been studied extensively in waterfowl and seabirds. Nestling herring gulls fed a single daily dose of 0.2–2 g/kg body weight Prudhoe Bay crude oil or fractions of crude oil enriched in heavy aromatic hydrocarbons for 8 days developed adrenal hypertrophy (Peakall et al. 1982). In contrast, mallard ducks (Anas platyrhynchos) fed 10–100 g/kg (i.e., parts per thousand [ppt]) of diet of naturally weathered Exxon Valdez crude oil for 14 days showed no change in adrenal weights or adrenal relative weights (Stubblefield et al. 1995). These concentrations of petroleum products are higher than the concentrations of fuel oil fed to the mink in this exposure. Plasma concentrations of corticosterone were measured in petroleum oil-fed mallard ducks in two studies. Ducks were fed South Louisiana crude oil or one of its distillates 1
Fuel oil as parts per million is equivalent to milligrams per kilogram body weight. Based on an approximate consumption of 0.125 kg feed per day and an average body weight of 1.95 kg for the mink in this exposure, the daily consumption of fuel oil is 3, 33, and 58 mg/kg body weight/day for the mink fed 48, 520, and 908 ppm fuel oil daily, respectively (cf. Stubblefield et al. [1995] for ferrets). The density of bunker C fuel oil is approximately 1 g/ml (Irwin 1998). The concentrations of fuel oil in the diet expressed as a percentage (weight/volume) are 0.0048, 0.0520, and 0.0908% (v/w) for the 48-, 520-, and 908-ppm groups, respectively (cf. Gorsline and Holmes [1981, 1982a] and Harvey et al. [1981] for mallard ducks).
Arch Environ Contam Toxicol (2008) 54:337–347
(0.5 to 3%, v/w) in the food for 10 days (Gorsline and Holmes 1982a) or North Sea crude oil (5%, v/w) daily for 20 weeks (Harvey et al. 1981). These concentrations of petroleum product are higher than the concentrations of fuel oil fed to the mink in this study (0.0048, 0.0520, and 0.0908%, v/w, for the 48-, 520-, and 908-ppm groups, respectively). In both studies plasma corticosterone concentrations were decreased in the petroleum oil-fed ducks compared to the control-fed ducks. In another study mallard ducks fed 3% (v/w) South Louisiana crude oil showed declines in plasma corticosterone concentrations after acute exposure (3–9 days); however, with chronic exposure (50 and 200 days) the corticosterone concentrations were similar to those in the controls (Gorsline and Holmes 1981). Interestingly after 500 days of oral exposure to the crude oil, plasma corticosterone concentrations in crude oil-fed birds were significantly increased over those in controls. From all of these studies in mammals and birds it is clear that the type of petroleum product used, the length of the exposure, the species involved, and also the form of dosing (oil mixed into food or given separately) will influence the development of adrenal hypertrophy and the secretory activity of the adrenal glands. The adrenal hypertrophy seen in this and in a previous fuel oil exposure study with mink (Schwartz et al. 2004a) is likely an adaptive response to maintain the blood glucocorticoid concentration within its normal range. This suggests that fuel oil is causing an adrenal insufficiency that is triggering the adaptive response. A mechanism that may be responsible is the disruption of steroid biosynthesis in adrenal cortical cells from inhibition of signal transduction or cholesterol transport pathways or direct inhibition of enzymes responsible for adrenosteroid biosynthesis. Because serum glucocorticoid and progesterone concentrations were not different between fuel oil-exposed and control animals, fuel oil may be inhibiting both 17hydroxylase and 3-b-dehydrogenase d-4,5-isomerase or an earlier step such as cytochrome P450 side-chain cleavage or proteins involved in cholesterol transport, e.g., steroidogenic acute regulatory (StAR) protein (Clark et al. 1994). In rainbow trout (Oncorhynchus mykiss), stimulation of the aryl hydrocarbon receptor inhibited steroidogenic enzymes and attenuated stress responses (Aluru and Vijayan 2006). Benz[a]anthracene inhibited steroidogenic enzymes in cultures of bovine adrenal cortical cells (Dibartolomeis and Jefcoate 1984). Inhibition of steroidogenic enzymes by chemical inhibitors of steroidogenesis has been shown to cause adrenal enlargement (Akana et al. 1983). The histological appearance of enlarged, heavily vacuolated adrenocortical cells found predominantly in fuel oil-exposed animals supports the concept of decreased glucocorticoid synthesis and secretion because the number of vacuoles in these cells
345
negatively corresponds to the amount of glucocorticoid secretion (Sharawy et al. 1979). The large variability in concentrations of glucocorticoid or progesterone in the 0ppm and, to a lesser degree, in the 48-ppm groups may be explained as individual variation among animals. The smaller variability in serum hormone concentration in the two highest fuel oil groups may be attributed to the effects of fuel oil-induced inhibition of adrenal steroid synthesis because inhibition of steroidogenesis would possibly tend to reduce the variability between samples. All of these findings suggest that adrenal glands from fuel oil-exposed mink would have a decreased ability to respond appropriately to ACTH stimulation, a finding demonstrated in ducks exposed to crude oil (Gorsline and Holmes 1982b). It is also possible that fuel oil-induced adrenal hypertrophy is not being mediated by direct toxic effects to the adrenal gland. Instead, fuel oil may be acting as a nonspecific chemical stressor to the HPA axis. It has been suggested that chronic stress is not always characterized by elevated serum glucocorticoid levels (Rich and Romero 2005) Chronic stress from various stressors (immune stimulation, social disruption, physical restraint) was not associated with elevation in serum glucocorticoid concentration (Pignatelli et al. 2000; Bauer et al. 2001; Stavisky et al. 2001; Veissier et al. 2001; Rich and Romero 2005). In one study there was an increase in adrenal weight associated with decreased glucocorticoid responses (Grinevich et al. 2001). In some forms of chronic stress, the HPA axis may become hypoactive and not respond to acute novel stressors even for days after the chronic stressor is removed (Ostrander et al. 2006). Organ weights, hematologic parameters, and serum chemistries were also measured to document any fuel oilor stress-associated changes. Liver relative weights were elevated in the 520- and 908-ppm fuel oil groups, consistent with our previous findings (Schwartz et al. 2004a). The hematological parameters measured in this exposure were all within published reference ranges for ranch mink (Fletch and Karstad 1972; Weiss et al. 1994). However, there were significant differences between the two highest fuel oil-exposed groups and the control and low fuel oil-exposed groups. Chronic exposure to fuel oil caused a small decrease in hematocrit, erythrocyte count, and hemoglobin concentration, and this is similar to our previous findings (Schwartz et al. 2004a). The associated increase in MCV in fuel oil-exposed mink is consistent with a mild increase in erythrocyte turnover. The slight decrease in erythrocyte numbers in the fuel oil-exposed mink is not sufficient to be regarded as anemia, and this decrease is considered to be attributed to fuel oil exposure and not adrenal status, as glucocorticoids have been shown to stimulate erythropoiesis (Greco 2004).
123
346
Changes in leukocyte parameters were not significant but showed some trends consistent with our previous observations (Schwartz et al. 2004a). The total leukocyte count was elevated over the control in the 48- and 908-ppm fuel oil groups, and that increase was due to an increase in neutrophil count and, in the 48-ppm group, also in the lymphocyte count. None of these changes was of sufficient magnitude or direction to suggest a stress response mediated by glucocorticoids. All serum chemistry values were within normal reference ranges for ranch mink except for the ALT concentration, which was elevated in all groups (normal range: 71.6± 56.9) (Weiss et al. 1994; Aulerich et al. 1999); therefore, significant fuel oil-related changes in serum chemistry values must be interpreted as having no biological effects on the health of the animals. Mononuclear cell infiltrates in hepatic portal areas were positively correlated with fuel oil-independent changes in ALT. This suggests that the elevation in serum concentrations of this enzyme was more likely due to mild inflammation than exposure to fuel oil. Although alterations in some serum chemistry parameters could be related to the effects of fuel oil on the liver, e.g., ALP and bilirubin, other changes could not be explained by either fuel oil exposure or changes in adrenal or liver physiology, e.g., sodium and calcium. The liver-related changes in serum chemistries were often decreased in fuel oil-exposed animals compared to control animals and were not considered to be evidence of hepatic toxicity. We have demonstrated in vivo that chronic exposure to fuel oil in mink is associated with adrenal hypertrophy with no increase in glucocorticoid output. This finding does not fit with the paradigm of chemical stressors initiating a general adaptive response that leads to adrenal enlargement and increased output of glucocorticoid into the plasma. The mechanism for the fuel oil alteration of adrenal activity in mink is subject to ongoing investigation. The ramification of our finding for marine foraging mammals such as mink, sea otters and river otters, which serve as important environmental sentinels, is that endocrine disruption from chronic petroleum oil contamination is likely to interfere with the capacity of these animals to adapt to other stressors and to changing environmental conditions. Acknowledgments This project was supported by the California Department of Fish and Game’s Oil Spill Response Trust Fund through the Oiled Wildlife Care Network at the Wildlife Health Center, School of Veterinary Medicine, University of California, Davis. The sponsor had no involvement in any aspect of this work. The authors would like to thank Dr. K. Kannan, School of Public Health, SUNY, New York State Department of Health, for evaluation of organochlorine contaminants and determination of fuel oil concentrations in the experimental diets.
123
Arch Environ Contam Toxicol (2008) 54:337–347
References Akana SF, Shinsako J, Dallman MF (1983) Relationships among adrenal weight, corticosterone, and stimulated adrenocorticotropin levels in rats. Endocrinology 113:2226–2231 Aluru N, Vijayan MM (2006) Aryl hydrocarbon receptor activation impairs cortisol response to stress in rainbow trout by disrupting the rate-limiting steps in steroidogenesis. Endocrinology 147:1895–1903 Anderson J W, Neff JM, Cox BA, Tatem HF, Hightower GM (1974) Characteristics of dispersions and water-soluble extracts of crude oil and refined oils and their toxicity to estuarine crustaceans and fish. Mar Biol 27:75–88 Aulerich RJ, Powell DC, Bursian SJ (1999) Handbook of biological data for mink. Michigan State University, East Lansing Bauer M E, Perks P, Lightman SL, Shanks N (2001) Restraint stress is associated with changes in glucocorticoid immunoregulation. Physiol Behav 73:525–532 Beckett KJ, Aulerich RJ, Duffy LK, Patterson J S, Bursian S J (2002) Effects of dietary exposure to environmentally relevant concentrations of weathered Prudhoe Bay crude oil in ranch-raised mink (Mustela vison). Bull Environ Contam Toxicol 69:593–600 Bodkin JL, Ballachey BE, Dean TA, Fukuyama AK, Jewett SC, McDonald LM, Monson DH, O’Clair CE, Van Blaricom GR (2002) Sea otter population status and the process of recovery from the Exxon Valdez oil spill. Mar Ecol Prog Series 241:237–253 Clark B J, Wells J, King S R, Stocco DM (1994) The purification, cloning, and expression of a novel luteinizing hormone-induced mitochondria protein in MA-10 mouse Leydig tumor cells. Characterization of the Steroidogenic Acute Regulatory Protein (StAR). J Biol Chem 269:28314–28322 Dibartolomeis MJ, Jefcoate CR (1984) The interrelationship of polycyclic hydrocarbon metabolism and steroidogenesis in primary cultures of bovine adrenal cortical cells. Mol Pharmacol 25:476–486 Fletch SM, Karstad LH (1972) Blood parameters of healthy mink. Can J Comp Med 36:275–281 Gorsline J, Holmes WN (1981) Effects of petroleum on adrenocortical activity and on hepatic naphthalene-metabolizing activity in mallard ducks. Arch Environ Contam Toxicol 10:765–777 Gorsline J, Holmes WN (1982a) Adrenocortical function and hepatic naphthalene metabolism in mallard ducks (Anas platyrhynchos) consuming petroleum distillates. Environ Res 28:139–146 Gorsline J, Holmes WN (1982b) Suppression of adrenocortical activity in mallard ducks exposed to petroleum-contaminated food. Arch Environ Contam Toxicol 11:497–502 Greco DS (2004) Laboratory evaluation of thyroid, parathyroid, adrenal, and pituitary glands. In: Troy DB (ed) Veterinary hematology and clinical chemistry. Lippincott Williams & Wilkins, Philadelphia, pp 445–459 Grinevich V, Ma X-M, Herman JP, Jezova D, Akmayev I, Aguilera G (2001) Effect of repeated lipopolysaccharide administration on tissue cytokine expression and hypothalamic-pituitary-adrenal axis activity in rats. J Neuroendocrinol 13:711–723 Hartung R, Hunt GS (1966) Toxicity of some oils to waterfowl. J Wild Manage 30:564–570 Harvey PW, Everett DJ (2003) The adrenal cortex and steroidogenesis as cellular and molecular targets for toxicity: Critical omissions from regulatory endocrine disrupter screening strategies for human health? J Appl Toxicol 23:81–87 Harvey S, Klandorf H, Phillips JG (1981) Reproductive performance and endocrine responses to ingested petroleum in domestic ducks (Anas platyrhynchos). Gen Comp Endocrinol 45:372–380 Irwin R J, Van Mouwerik M, Stevens L, Seese MD, Basham W (1998). Environmental contaminants encyclopedia. National
Arch Environ Contam Toxicol (2008) 54:337–347 Parks Service, Water Resources Division, Fort Collins, CO (distributed on the Internet via the Nature Net portion of the Park Service home page; www.nps.gov) Lipscomb TP, Harris RK, Rebar AH, Ballachey BE, Haebler RJ (1994) Pathology of sea otters. In: Loughlin TR (ed) Marine mammals and the Exxon Valdez. Academic Press, San Diego, pp 265–279 Mazet JAK, Gardner IA, Jessup D A, Lowenstine L J (2001) Effects of petroleum on mink applied as a model for reproductive success in sea otters. J Wild Dis 37:686–692 Mazet JK, Gardner IA, Jessup D A, Lowenstine L J, Boyce W M (2000) Evaluation of changes in hematologic and clinical biochemical values after exposure to petroleum products in mink (Mustela vison) as a model for assessment of sea otters (Enhydra lutris). Am J Vet Res 61:1197–1203 Ostrander MM, Ulrich-Lai YM, Choi DC, Richtand NM, Herman JP (2006) Hypoactivity of the hypothalamo-pituitary-adrenocortical axis during recovery from chronic variable stress. Endocrinology 147:2008–2017 Peakall DB, Hallett DJ, Bend JR, Foureman GL, Miller DS (1982) Toxicity of Prudhoe bay crude oil and its aromatic fractions to nestling herring gulls. Environ Res 27:206–215 Pignatelli D, Maia M, Castro AR, da Conceicao Magalhaes M, Vivier J, Defaye G (2000) Chronic stress effects on the rat adrenal cortex. Endocr Res 26:537–544 Rich EL, Romero M (2005) Exposure to chronic stress downregulates corticosterone responses to acute stressors. Am J Physiol Regul Integr Comp Physiol 288:1628–1636 Schwartz JA, Aldridge BM, Lasley BL, Snyder PW, Stott J L, Mohr FC (2004a) Chronic fuel oil toxicity in American mink (Mustela vison). Systemic and hematological effects of ingestion of a low
347 concentration of bunker C fuel oil. Toxicol Appl Pharmacol 200:146–158 Schwartz JA, Aldridge BM, Stott JL, Mohr FC (2004b) Immunophenotypic and functional effects of bunker C fuel oil on the immune system of American mink (Mustela vison). Vet Immunol Immunopathol 101:179–190 Sharawy M, Dirksen T, Chaffin J (1979) Increase in free cholesterol content of the adrenal cortex after stress: radiographic and biochemical study. Am J Anat 156:567–576 Stavisky RC, Adams MR, Watson S L, Kaplan JR (2001) Dominance, cortisol, and behavior in small groups of female cynomolgus monkeys (Macaca fascicularis). Horm Behav 39:232–238 Stubblefield WA, Hancock GA, Ford WH, Ringer RK (1995) Acute and subchronic toxicity of naturally weathered Exxon Valdez crude oil in mallards and ferrets. Environ Toxicol Chem 14:1941–1950 Vandermeulen J H, Singh J G (1994) Arrow oil spill, 1970–90: persistence of 20-yr weathered bunker C fuel oil. Can J Fish Aquatic Sci 51:845–855 Veissier I, Boissy A, dePassille AM, Rushen J, van Reenen C G, Roussel S, Andanson S, Pradel P (2001) Calves’ responses to repeated social regrouping and relocation. J Anim Sci 79:2580–2593 Weiss DJ, Wustenberg W, Bucci TJ, Perman V (1994) Hematologic and serum chemistry reference values for adult brown mink. J Wild Dis 30:599–602 Wiliams TM, O’Conner DJ, Nielsen SW (1995) The effects of oil on sea otters: histopathology, toxicology, and clinical history. In: Williams TM, Davis RW (eds) Emergency care and rehabilitation of oiled sea otters: a guide for oil spills involving furbearing marine mammals. University of Alaska Press, Fairbanks, pp 3–22
123
