- Email: [email protected]
Assessment of hepatic indicators of subchronic carbon tetrachloride injury and recovery in rats
FIJNDAMENTALANDAPPLIEDTOXICOLOGY
(1990)
l&558-570
Assessment of Hepatic Indicators of Subchronic Carbon Tetrachloride Injury and Recovery in Rats’ JOHNW.ALLIS,THOMASR.WARD,JOHN
C.SEELY,*ANDJANEELLENSIMMONS
Health Effects Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 2771 I: and*PATHCO. Inc., Research Triangle Park, North Carolina 27709
Received November 2, 1989; accepted May 14, 1990 Assessment of Hepatic Indicators of Subchronic Carbon Tetrachloride Injury and Recovery in Rats. ALLIS, J. W., WARD, T. R.. SEELY, J. C., AND SIMMONS, J. E. (1990). Amdam. Appl. Toxicol. 15, 558-570. To determine the course of hepatic recovery from subchronic oral administration of carbon tetrachloride (CC&). male F-344 rats were gavaged with 0,20, or 40 mg CC&/kg. 5 days/week, for 12 weeks. Exposure to Ccl, caused dosage-dependent increases in relative liver weight and the serum levels of aspartate aminotransferase, alanine aminotransferase, lactate dehydrogenase, alkaline phosphatase, and cholesterol as well as a dosage-dependent decrease in hepatic cytochrome P450. Centrilobular hepatocellular vacuolar degeneration, necrosis, and cirrhosis occurred at both 20 and 40 mg/kg, with dosage-dependent severity. Reversibility of these reported effects varied with parameter. By Day 8 postexposure, necrosis had disappeared and all serum indicators and cytochrome P450 had returned to control levels. By Day 15 postexposure, the severity of the vacuolar degeneration had decreased. Reversibility of cirrhosis was dosage dependent; complete recovery occurred in the low- but not the high-dose group by Day 15. The disappearance of the increase in relative liver weight was also dependent on dosage; the low- but not the high-dose group had returned to the control level by Day 22. In an attempt to measure persistent hepatic damage, liver uptake relative to the spleen was determined for a sulfur colloid labeled with technetium-99m and for tritiated 2-deoxyglucose. Neither method consistently measured hepatic damage in cirrhotic livers due, in part, to the high degree of variability in the tracer uptake data. o 1990 society of Toxicology.
The hepatotoxic effects of carbon tetrachloride (CCL,) have been widely studied and mechanisms have been proposed for the production of necrosis in the liver (for a review see Kalf et al., 1987). Most of the information on CC4 hepatotoxicity has been developed from acute or short-term studies and relatively little is known about subchronic exposure and recovery after cessation of exposure.
’ The research described in this article has been reviewed by the Health Effects Research Laboratory and approved for publication. Approval does not signify that the contents necessarily reflect the views of the Agency nor does mention of trade names or commercial products constitute endorsement or recommendation for use. 0272-0590/90 $3.00 Copyright 0 1990 by the Society of Toxicology. All rights of reproduction in any form reserved.
558
Bruckner et al. (1986) conducted a comprehensive subchronic study in which rats were gavaged five times weekly for 12 weeks with 1, 10, and 33 mg CC&/kg body wt. Beginning at Week 2, rats dosed with 33 mg/ kg exhibited increased serum activities for the enzymes alanine aminotransferase (ALT), sorbital dehydrogenase (SDH), and ornithine carbamoyltransferase (OCT). After 12 weeks of exposure at this dosage, periportal fibrosis and other degenerative changes associated with cirrhosis were also found. For the 10 mg/ kg dosage level, SDH and ALT were elevated and centrilobular vacuolization was found after 12 weeks of exposure. After a 2-week recovery period, all serum enzymes (with the
SUBCHRONIC
HEPATIC
possible exception of ALT) had returned to normal, but the cirrhotic changes remained. In another study, Hayes et al. (1986) treated mice with 12 to 1200 mg CCWkg daily for 90 days, finding evidence of hepatotoxicity in all groups. Several serum indicators of hepatotoxicity plus liver-to-body weight ratio were elevated, with males more susceptible than females. Necrosis and other histopathological changes were also found in each dose group. A similar study (Condie et al., 1986) used dosages of 1.2, 12, and 120 mg/kg 5 days per week for 12 weeks. Again, necrosis and increased serum enzymes were found at both 12 and 120 mg/kg, but no changes were noted at the lowest dosage. Earlier subchronic studies with rats (Rubin et al., 1963; Cameron and Karunaratne, 1936) utilized a twice weekly dosing regimen at dosages substantially higher than the lowest effective dosages in the reports cited above. In this earlier work, dosages of more than 1000 mg/kg twice a week produced fibrosis and cirrhosis at various times between 5 and 20 weeks. Cameron and Karunaratne (1936) also reported that dosages of approximately 40 mg/kg delivered on a similar schedule did not produce structural alterations in the liver. Except for the single time point investigated by Bruckner et al. ( 1986), recovery from exposure and permanent damage to the liver from subchronic dosing has been investigated at high doses given infrequently (more than 1000 mg/kg two or three times weekly), and with one exception (Sellers et al., 1948) by routes other than oral (for a review see Perez-Tamayo, 1979). In these studies, cirrhosis and presence of collagen were the principal endpoints studied. The data generally support the finding of Cameron and Karunaratne ( 1936) that experimental cirrhosis is reversible if dose and duration of exposure are sufficiently low, but eventually it appears to be irreversible. The work described below had the twofold objective of determining the time course of recovery from subchronic oral CC4 administration for a number of experimental end-
559
INJURY AND RECOVERY
points and attempting to develop a measure for persistent impairment of the liver’s functional capacity. The doses employed are much lower and the experimental endpoints more varied than in previous studies investigating recovery. Treatment with CCL, produces experimental cirrhosis in rats and should provide a suitable test for the sensitivity and appropriateness of the functional measures employed. Doses were chosen to provide differing degrees of cirrhosis in order to test the sensitivity of the functional measures. The assay for functional capacity was based on a clinical procedure used in humans but its validity to measure functional decrement in animals has not yet been established. Harper et al. ( 1964) developed a technique using a sulfur colloid labeled with technetium-99m (99mTc) as a liver imaging tool employing a gamma camera. By measuring the relative uptake of the colloid in the liver compared to that in the spleen, the relative efficiency of the liver can be determined; this liver-to-spleen ratio is used as a diagnostic test for establishing the degree of human cirrhosis (Prakash et al., 1977). This ratio may also be useful as a functional test in animals. However, the colloid is taken up by Kupffer cells rather than by hepatocytes; thus the technique is an indirect measure of hepatocyte function. To compare the indirect sulfur colloid assay with a more direct method, we measured hepatic uptake of tritiated 2-deoxyglucose as an indicator of hepatic glucose utilization. Uptake of 2-deoxylglucose has been widely used as an indicator of cerebral metabolic rate (Sokoloff, 198 1) and its analogue ([2-‘8F]fluoro-2-deoxy-D-glucose) has been suggested as a diagnostic tool in hepatic studies (Yonekura et al., 1982). MATERIALS
AND
METHODS
Male 60-day-old Fischer 344 rats (Charles River Labo-
ratories, Raleigh, NC) were housed three per cage in polycarbonate cages with bedding of heat-treated pine shavings, fed Purina Rodent Lab Chow (No. 5001, Rals-
ALLIS ET AL. ton Purina, Checkerboard Square. MO) and given tap water ad libitum. Animals were acclimatized for 5 days before the experiment was begun. Feed was removed from the cages the afternoon prior to termination. A 12hour light-to-dark cycle with light beginning at 6 AM was maintained during the experiment. Sentinel animals were housed in the animal room as part of an ongoing surveillance program for parasitic, bacterial, and viral infections. Treatment. One hundred forty-four rats were gavaged 5 days per week for 12 weeks with 0,20, or 40 mg/kg/day carbon tetrachloride (Sigma Chemical Co.; Calbiochem) in corn oil (Sigma Chemical Co.) (0, 10, or 20 mg CCL,/ ml corn oil). Dosing was begun on a Monday and ended on a Monday to avoid terminating animals after 2 days of recovery (i.e., after the weekend). Rats were weighed on Days 1, 3, 5, and 8 and thereafter on a weekly basis. A new gavage solution was used daily for each dosage; solutions were made weekly. The gavage vehicle was corn oil; all rats including controls received 2 ml/kg with dosing volumes adjusted for each rat on the basis of its most recent weight. Food consumption was measured by cage on a regular basis throughout dosing. The rats were divided into two groups of 72 animals each (24 animals per dose level). One group of animals was used for measuring hepatic cytochrome P450, serum chemistries, and light microscopic histopathology. Six animals from each dose level were terminated on Days I, 3.8, and 15 postexposure. The other group was used to measure radiolabeled sulfur colloid and 2-deoxyglucose (2DG) uptake and six animals from each dose level were sacrificed on Days 1.8, 15, and 22 postexposure. At termination, the first group was anesthetized with 50 mg/kg sodium pentobarbital (Abbott Lab., Chicago, IL) injected ip and exsanguinated from the abdominal aorta into serum-separation tubes (Becton-Dickinson, Lincoln Park, NJ). The blood was held on ice for a minimum of 30 min and then centrifuged at 1OOOgfor 30 min. Serum was collected and frozen at -40°C until analyzed. Serum concentrations of total bilirubin (Bib), triglycerides, and cholesterol and the activities of aspartate aminotransferase (AST), ALT. alkaline phosphatase (Alk Phos), and lactate dehydrogenase (LDH) were assayed commercially (Vet-Path, Teterboro, NJ). Each liver was quickly excised, rinsed in saline, blotted, and weighed. Slices of the left lobe were removed for open light-microscopic histopathological examination (Society of Toxicologic Pathologists. 1986: Prasse et al.. 1986) and placed in 10% phosphate-buffered formalin. Two grams of liver was taken to assay for cytochrome P450 by the method of Omura and Sato (1964). Protein determinations were performed on the microsomal samples by the method of Lowry et al. ( 195 1). Histological evaluation of livers was conducted on hematoxylin- and eosin-stained tissue sections. In addition, all livers in the 20 mg/kg/day and selected livers in the 40 mg/kg/day groups were stained with Masson’s trichrome to confirm the presence or absence ofcollagen.
Degenerative and necrotic lesions were graded according to the following criteria: none, minimal (one to several centrilobular hepatocytes affected), mild (no more than one-fourth of the hepatocytes of the centrilobular region affected), moderate (expansion of the affected hepatocytes up to one-half of the centrilobular region), or marked (over one-half of the lobular architecture involved). These criteria have been described more fully in Simmons et al. (1988). The presence of cirrhosis was based on the following criteria: disorganization of the liver by interconnecting fibrous bands or scarring that tended to be either portalcentral or portal-portal; formation of parenchymal nodules that varied in size and were due to regenerative activity within the network of scars; and within regenerative nodules, the lobular architecture of the liver appeared disorganized, i.e., hepatocytes varied in size, mitotic figures were present, and there was a loss of central veins. Accordingly, cirrhosis was graded as follows: none, minimal (fibrous scarring to a thickness of one hepatocyte or less), mild (fibrous tissue extended to a thickness of one or two hepatocytes, with some micronodular formation), or moderate (fibrous tissue extended to a thickness of more than two hepatocytes with micronodular formation pronounced). The second group of animals was used to evaluate the sulfur colloid and 2DG uptake assays. The uptake of these two radiolabeled tracers was measured and the liver-to-spleen ratios were calculated. Because of technical problems in handling I8 animals in I day, this group was split into two parts. Half of the animals were dosed for I2 weeks and 1 day. At termination on Days 1,8, 15, and 22 postexposure, rats were anesthetized with 50 mg/kg pentobarbital injected ip. When the animal was not responsive to a foot pinch (approximately 10 min), the left femoral vein was surgically exposed. The animal was then injected iv with 0. I ml of a mixture of sulfur colloid (donated by Duke University Medical Center, Durham, NC) labeled with technetium-99m ( 10 &i/rat, measured at 8 AM each day) and 2-deoxy+ghtcose (New England Nuclear) labeled with tritium (1 &/rat). Ten minutes after injection of the radioactive tracers. the animal was terminated by introducing perfusate (0. IS M NaCl) into the left ventricle and cutting the inferior vena cava anterior to the liver. After 5 min of perfusion (approximately 50 ml), the liver and spleen were removed and weighed. Tissue samples of about 0.2 g were placed in liquid scintillation vials and counted in a gamma counter (Packard Instruments, Auto/Gamma) immediately after termination of all rats each day. To reduce variability due to the short half-life of 99mTc (6 hr), hepatic and splenic samples were counted consecutively for each rat and the data analyzed as the liver-tospleen ratio for each rat. Following gamma counting, 2 ml of tissue solubilizer (Protosol, New England Nuclear) was added to each sample. After incubation for 2 days at 37°C the solubilizer
SUBCHRONIC
HEPATIC
was neutralized with 0.5 ml 3 M HCl and 10 ml ofscintillation cocktail (Liquiscint, National Diagnostics) was added. The samples were counted in a liquid scintillation counter (Packard Instruments, Model 2660) after the 99mT~ had decayed for at least a week. Scintillation data were adjusted for quenching and reported as dpm. Ratios for each rat were calculated and analyzed in the same manner as the 99mT~data. Statistical analysis. Rat weight gain during the course of dosing was analyzed by obtaining the slope and intercept for each rat by linear regression analysis (SAS, 1985). The slopes and intercepts were analyzed by analysis of variance (ANOVA) (SAS, 1985). Weighted average daily food consumption was calculated (Weeks I to 11) for each cage of rats. A quadratic equation was fitted to the data for each cage; these coefficients were analyzed by ANOVA. All other numerical data were also analyzed by ANOVA. A p value 5 0.05 was used for all ANOVAs, and when that criterion was met, the Sidak method of comparison of means (for multiple comparisons) was used, with p 5 0.05 required for significance. Homogeneity of variance was checked with Bartlett’s test (Sokal and Rolf, 198 1) and when necessary a log transformation and reanalysis were performed. The criterion of significance for Bartlett’s test wasp G 0.00 1 (Anderson and McLean, 1974).
RESULTS Hepatotoxicity and recovery. One day after the cessation of the 12-week exposure, there was considerable evidence of hepatotoxicity. The serum indicators ALT, AST, and LDH were elevated for both the 20 and 40 mg/kg/ day groups (Table 1). Alk Phos and cholesterol were also increased for the higher dose group only (Table l), while total bilirubin (data not shown) and triglycerides were unaffected. Cytochrome P450 levels (Fig. 1) were significantly reduced for both dose groups. Histopathological evaluation confirmed cellular damage from Ccl4 administration (Table 2 and Fig. 2). The primary lesions were in the centrilobular region. At 20 mg/ kg, two of the six livers were cirrhotic, and all showed vacuolar degeneration and hepatocellular necrosis. There was fibrous tissue in the noncirrhotic livers of this group, but it was not interconnecting. At the higher dose, all livers were cirrhotic and showed evidence for degeneration and necrosis. Staining with
INJURY AND RECOVERY
561
Masson’s trichrome showed the presence of collagen, confirming the finding of cirrhosis in both treated groups. By Day 3 postexposure, the serum indicators of hepatotoxicity (Table 1) were decreasing. With the exception of cholesterol, magnitudes of all indicators significant on Day 1 declined by Day 3. Significant elevations remained in the higher dose group for the five measures ALT, AST, LDH, Alk Phos, and cholesterol and in the lower group only for ALT. Similarly, cytochrome P450 levels were returning to normal and only the high-dose group showed a significant decrement. The histopathological results (Table 2) were similar to those on Day 1. At Days 8 and 15 postexposure, all serum indicators and cytochrome P450 had returned to control levels for both dose groups (the small but statistically significant decrements for ALT on Day 8 and AST on Day 15 were not considered to be biologically significant). Cirrhosis persisted in the high-dose group (Table 2 and Fig. 2d), although there was some decrease in its severity. The decrease in severity was confirmed by the Masson’s trichrome data showing that collagen and bridging decreased with time postexposure. In the low-dose group, cirrhosis was not evident by Day 8. For both groups, centrilobular hepatocellular necrosis disappeared by Day 8 and vacuolar degeneration decreased in severity at Days 8 and 15 but was still present. Liver weight relative to body weight at termination is shown for all 144 animals in Fig. 3. At Days 1, 3, 8, and 15 postexposure, significant differences were found between both dose groups and controls. In contrast to most of the other measures described above, relative liver weight for the exposed animals in the high-dose group did not return to the control level, even by Day 22. Radiolabeled tracer uptake as a measure of persistent hepatic damage. Two techniques of ascertaining persistent hepatic damage and functional decrement were assessed. The first, the uptake of sulfur colloid labeled with
562
ALLIS ET AL. TABLE 1 SERUM CHEMISTRY VALUES FOLLOWING SUBCHRONICEXPOSURE TO CARBON TETRACHLORIDE
Day post exposure
Dose km/k/day)
Day I
ALT (W/liter)
0
7s+1sb 387i. l52+ 3046 f 1299*
20 40
Day 3
0
0 20 40
Day 15
106+20 299+ 121+ 2593,1142*
0 20 40
’ CHOL, cholesterol. b Data are reported as means + Significantly different from * Significantly different from a Significantly different from
CHOL” bw/dl)
45+ 10
49k 3 47* 4 88+25*
90,17 108,19 365 t62*
l65k26 180+39 361+77*
36+ 4 4lF 3 63+13*
100 -+ 28*
64k5 54*4+ 61+5
9ort17 85k 86+6
2OOk72 234+41 217?26
34* 39+ 42+
5 4 6
432 4lk 47f
4 6 3
67k5 58k5 75k9'
89~5 80+4+ 95,3"
194k45 166t40 222*43
33+ 33k 38+
7 5 8
48+43+ 52k
3 6 8
JO
44f 52kII
5
+ SD. control, p i 0.05. control and from 20 mg/kg, p < 0.05. 20 mg/kg but not from control. p < 0.05.
Do6e(m@k@day) 00 I3 20 q 40
3
Alk Phos (W/liter)
55-c 8 84-c21*
99mT~, is reported as the liver-to-spleen ratio in two ways: In counts per gram of tissue and as counts per organ. When reported as
1
LDH (W/liter) 208k62 6052413' 2370+720*
67k9 112-t-30+ 5lOk 85*
20 40
Day 8
AST (W/liter)
8
15
Day Post-Exposure
FIG. 1. The effect of subchronic exposure to 20 and 40 mg CClJkg/day on the concentration of hepatic cytochrome P450. N = 6 per group. Significantly different from the concurrent control group, p < 0.05. *Significantly different from the concurrent control group and the 20 mg/kg group, p < 0.05.
counts/gram (Fig. 4a), on Day 1 postexposure the 40 mg/kg group was significantly lower than controls. This effect was not seen at 20 mg/kg, or on later termination days. When reported as the ratio of counts/organ (Fig. 4b), there was neither a dose nor a day effect. These apparently conflicting results may be explained by examining liver weights: The livers are enlarged on Day 1 postexposure (Fig. 3) giving rise to a decrease in counts/gram although the total uptake by the liver is unchanged. The controls for both methods when analyzed separately showed no day effect. The uptake was also analyzed as counts/gram for the liver and the spleen individually (results not shown), and no effect was seen. The second technique of ascertaining persistent damage, that of deoxyglucose uptake, is reported in the same manner, counts/gram and counts/organ (Figs. 5a and 5b). The liverto-spleen ratio of deoxyglucose uptake reported as counts/gram showed no effect of
SUBCHRONIC
HEPATIC
563
INJURY AND RECOVERY
TABLE 2
Day postexposure I
8
15
0
20
40
0
20
40
0
20
40
0
20
40
5
-
-
(j
-
-
6
-
-
5
-
-
lb -
-
6
_--
6 -
2 4
-
1 5 -
-
2------5--3 4 2--3------
CCl,dosage(mg/kg): Normal Vacuolar degeneration Minimal Mild Moderate Marked Hepatocellular necrosis Minimal Mild Moderate Marked Cirrhosis Minimal Mild Moderate
3
1
I-------5 5 6 6 -
-
6
(j
_--
-
5 1
-
-----_
-
1*
-
_
6
-
-
-
-
-
-
-
_
3
-----
1
-
3
a Lesions were graded on hematoxylin- and eosin-stained tissue sections. Observed lesions were centrilobular in location and were graded according to the criteria described under Materials and Methods. Each group contained six animals. b Lesion not considered associated with experimental procedures.
dose. When reported as counts/organ the analysis showed a significant effect of dose which appears to be primarily based on the statistical significance of the 20 mg/kg point at Day 8 (Fig. 5b). However, both methods of analysis show a significant day effect as hepatic uptake relative to the spleen decreased with time postexposure. The controls for the two methods, analyzed separately, also demonstrate the day effect. In sum, the 2 DG data indicate a decrease in uptake of deoxyglucose after the cessation of dosing. Animal weight gain. Figure 6 shows animal weights for each dose group during the initial 78 days of exposure. Soon after the initiation of dosing, both the 20 and 40 mg/kg groups began to exhibit apparent growth retardation. The animal weight curves showed an overall dose effect with weight decreasing with increased dose; the slopes of the weight curve in
both the 20 and 40 mg/kg groups were significantly lower than control. The decrement in weight gain in the exposed groups could not be accounted for by food consumption. The average food consumption curves for the controls and the two exposed groups (data not shown) were statistically indistinguishable from one another. At Day 22 postexposure, animal weights at termination for all dose groups (data not shown) were equivalent, indicating recovery from the weight decrement developed during dosing. DISCUSSION Hepatotoxicity and recovery. The hepatotoxicity data presented in this paper show that low doses of carbon tetrachloride, 20 and 40 mg/kg administered subchronically, pro-
564
ALLIS ET AL.
SUBCHRONIC
HEPATIC
565
INJURY AND RECOVERY DOSS (mglk@W
1
3
a
15
22
Day Post-Exposure
FIG. 3. The effect of subchronic exposure to 20 and 40 mg CCld/kg/day on relative liver weight. N = 12 per group. ‘Significantly different from the concurrent control group, p G 0.05. *Significantly different from the concurrent control group and the 20 mg/kg group, p d 0.05.
q
2s
40
20 1s 10 5
duce hepatotoxicity and that over the 12 weeks of exposure, persistent damage arises in the form of cirrhosis. The recovery from the ongoing toxicity was rapid, with serum indicators of hepatotoxicity and cytochrome P450 levels returning to normal at parallel rates and within 8 days of cessation of exposure. These data are consistent with the disappearance of necrosis over the same time span. However, vacuolar degeneration persisted longer and was still evident although decreased in severity at Day 15 postexposure. The times to recovery for the parameters in this study are summarized in Table 3. The elevation of ALT, AST, and LDH on Days 1 and 3 postexposure are indicative of ongoing parenchymal cell damage, with ALT being specific for the liver (Guzelian, 1983). The presence of vacuolar degeneration and
0 1
8
Day
1s
22
Post-Exposure
FIG. 4. The effect of subchronic exposure to 20 and 40 mg CC&/kg/day on the liver-to-spleen ratio of uptake of sulfur colloid 99mTc expressed as cpm/g tissue (a) and cpm/organ (b). N = 6 per group. *Significantly different from the concurrent control group and the 20 mg/kg group, p G 0.05. “Significantly different from 20 mg/kg but not from control, p G 0.05.
necrosis in the liver is consistent with these findings, indicating hepatocyte damage and death. Elevated levels of Alk Phos and cholesterol are consistent with malfunction of the hepatobiliary system, but this result was not confirmed by changes in total bilirubin levels which were unaffected during the experiment. The putative hepatobiliary involve-
FIG. 2. Histopathology. (a) Control liver. Normal centrilobular area is shown. H&E, 40X. (b) Day I postexposure, 20 mgjkg. Moderate centrilobular vacuolar degeneration and mild centrilobular necrosis is present. H&E, 40X. (c) Day 1 postexposure, 40 mg/kg. Hepatic cirrhosis is characterized by fibrous bands surrounding regenerative hepatocellular parenchyma. Moderate vacuolar degeneration and mild hepatocellular necrosis are present along the periphery of the lobule. H&E, 16X. (d) Day 15 postexposure, 40 mg/ kg. Cirrhosis is still evident; however, the amount of vacuolar degeneration is reduced. H&E, 16X.
566 a
ALLIS ET AL. 3.0
Dose (mgkgday,
T
1
T
b
50
c
I
1
8
15
22
Day Post-Exposure
FIG. 5. The effect of subchronic exposure to 20 and 40 mg CC&/kg/day on the liver-to-spleen ratio of uptake of tritiated 2-deoxylgucose expressed as dpm/g tissue (a) and dpm/organ (b). N = 6 per group except N = 5 for Day 15 controls. +Significantly different from the concurrent control group, p < 0.05.
two experiments provide a consistent picture of hepatic toxicity and recovery for subchronic exposure of rats to carbon tetrachloride. These studies indicate that the lowest effective subchronic dosage for hepatotoxicity in rats is near 10 mg/kg. They also indicate that the threshold for cirrhosis persisting for at least 2 weeks following cessation of exposure is between 20 and 33 mg/kg for this dosing schedule. The work of Hayes et al. (1986) and Condie et al. (1986) with mice shows that subchronic exposure produces effects similar to those in rats. While neither paper provides recovery data, a subchronic dosing schedule similar to that used for rats gave evidence for hepatotoxicity at 12 mg/kg but not at 1.2 mg/ kg (Condie et al., 1986). It would appear that the lowest effective level of CC& in mice is similar to that in rats. All four of these studies show that daily exposure to CC& is hepatotoxic at much lower doses than twice weekly dosing over a protracted period of time. For example, Rubin et al. ( 1963) used more than 1500 mg/kg twice weekly (actually 0.15 ml per animal, delivered by subcutaneous injection, in female Sprague-Dawley rats) to produce necrosis in 10 days, fibrosis between 20 and 65 days, and cirrhosis beginning 95 days
350l’ ment (Widman, 1979) either is in an early stage of development or may be secondary to general inflammation of the liver, as indicated by the increased relative liver weight. These data extend the work of Bruckner et al. ( 1986) by providing the time course of recovery for ongoing parenchymal toxicity, the decrease of cytochrome P450 due to treatment and subsequent recovery, and the minimal involvement of the hepatobiliary system at this stage of liver injury. Their result that ALT levels had returned to normal but cirrhosis remained at 2 weeks postexposure is consistent with our findings, notwithstanding the different strain of rat and somewhat different doses used. Taken together, these
15ol0
I 15
30
Days
of Exposure
45
60
75
to CC14
FIG. 6. The effect of subchronic exposure to 20 and 40 mg CC&/kg/day on body weight during the course of dosing; each dose group consisted of 48 animals. Both treatment groups were significantly different from control (see text). Error bars are not shown for clarity.
SUBCHRONIC
HEPATIC
INJURY TABLE
AND
567
RECOVERY
3
SUMMARY OF RECOVERY TIME FOLLOWING SUBCHRONICEXPOSURE TO 40 MG/KG CC& Parameters
Day 1
Day 3
Day 8
Day 15
Day 22
Serum chemistries Cytochrome P450 Liver wt/body wt Body wt Degeneration Necrosis Cirrhosis 99mT~uptake 2DG uptake
+
+ + 0 + + + ND ND
0 0 + 0 + 0 + 0 0
0 0 + + 0 + 0 0
ND ND + 0 ND ND ND 0 0
+ + + + 0 0
Note. 0, not different from controls for quantitative data; absent for histopathology. +, significantly increased compared to controls for quantitative data; present for histopathology. -, significantly decreased compared to controls. ND, no data; data not taken.
after starting exposures. Cameron and Karunaratne ( 1936) found necrosis in albino rats (Graham, sex not reported) after about 10 days, fibrosis after 3 weeks, and cirrhosis after 14 weeks with twice weekly dosing of lOOO2600 mg/kg (0.10-0.25 ml delivered subcutaneously). They also found only slight, reversible degeneration after 4 months of twice weekly dosing with 40 mg/kg (0.0037 ml), the higher dose used in our study. Sellers et al. (1948) used 250 mg/kg (by gavage) every third day for 4 months to produce cirrhosis. Our study demonstrated recovery from cirrhosis. Recovery was complete for the 20 mg/ kg group and the severity of cirrhosis decreased in the 40 mg/kg group during the post-treatment period. While cirrhosis is generally regarded as irreversible, in fact, recovery from CC&-induced cirrhosis was first noted by Cameron and Karunaratne (1936) who found reversible and nonreversible components depending on the duration of exposure. The components apparently are indistinguishable histologically. A number of other workers have also demonstrated recovery from CCkinduced cirrhosis (see a review by Perez-Tamayo, 1979). In this work, cirrhosis has been produced by injection at dosages of at least 1000 mg/kg, usually given twice weekly, or by inhalation
at unspecified doses. In the one exception, Sellers et al. ( 1948) exposed one group by inhalation at 1000 ppm and another group to 250 mg/kg by gavage 2-3 days per week. In general, these studies conclude that cirrhosis is produced after several weeks of exposure, and a reversible stage is followed by an irreversible stage as dosing continues. Our results are clearly within the reversible stage based on the low doses delivered, the relatively mild cirrhosis found, and the demonstrated decline in severity in the 40 mg/kg group within 15 days postexposure.
Tracer uptake methods for persistent hepatic damage. Because cirrhosis cannot be measured quantitatively and is not a direct measure of liver function, we investigated another measure in the interest of determining persistent damage by a method other than histopathology. Cirrhosis was used as the indicator of permanent damage with the presumption of a functional decrement. For the functional assay, we chose the uptake of radioactive tracer-labeled substances by the liver and spleen. The rate of uptake should indicate the liver’s functional capacity. Comparison to the rate of uptake by the spleen should provide an internal control and reduce variability (based on the premise that the spleen was not affected by CC& treatment).
568
ALLIS
Two different tracer methods were employed. The first was a sulfur colloid labeled with technetium-99m that is used clinically to assess the extent of hepatic cirrhosis by gamma-camera imaging. Clinicians have long used this technique as the method of choice (Oppenheim et al., 1976) to determine liver size and pattern of hepatic uptake. The liver-to-spleen uptake ratio is also measured clinically as a qualitative measure of human cirrhosis; a decreased ratio indicates impaired function. Although the colloid is a particulate and as such is phagocytized by the Kupffer cells and not by hepatocytes, the decreased uptake of colloid has been found to be correlated with a decrease in number of Kupffer cells and increased injury in the livers of rats with CC&induced cirrhosis (Lough et al., 1987). More recently, another chemical, [2‘8F]fluoro-2-deoxy-D-glucose (2FDG), has been suggested as an alternative liver imaging agent (Yonekura et al., 1982). The short halflife (2 hr) gamma-emitting isotope of fluorine is attached to 2-deoxy-D-glucose allowing the uptake and distribution of this compound to be monitored with a gamma camera. Both 2FDG and its parent compound 2DG are glucose analogues which have been used extensively to study brain metabolism. Because of uncertainties in their metabolism, the use of glucose analogs in the liver has come under criticism (Welsh, 1982). 2DG is transported across membranes and phosphorylated to 2-deoxyglucose 6-phosphate in the same manner as is glucose. In the liver, the deoxyglucose compounds are not trapped as efficiently as they are in the brain. However, Jenkins et al. ( 1986) who have shown that the metabolic products of 2DG are different in the brain and liver, conclude “provided these products are not lost from the tissues to a substantial extent (as appears to be the case), they should not invalidate the use of 2-deoxyglucose as a monitor of glucose metabolism” in the liver. To the authors’ knowledge, the 2DG technique has been tried only once in the liver, by Yonekura et al.
ET AL.
(1982) who concluded that it may be useful. Because deoxyglucose is taken up by all cells including hepatocytes, it should give a good indication of the total metabolic activity of this organ. Use of this compound, if measured shortly after injection, should assessthe hepatic metabolic rate which in turn should indicate hepatic function. One potential confounding factor is the affect of insulin which influences the nonmetabolic uptake and trapping of glucose (Horn et al., 1984). Insulin was not measured in this study. Neither of the two tracer techniques detected a decreased functional capacity in cirrhotic livers. Based on the histopathological finding of cirrhosis in the liver that persisted through at least Day 15 postexposure in the 40 mg/kg group, the lack of corroboration by either or both measures of liver functional capacity was surprising. Because of the extensive swelling of the liver due to Ccl, exposure, the counts per organ method of measurement was probably the more meaningful for either assay. Indeed, the correlation between decreased number of Kupffer cells and colloid uptake has been determined on an organ basis (Lough et al., 1987). As mentioned earlier, the decrease in the liver-to-spleen ratio for the sulfur colloid when calculated by the counts/gram method but not by counts/ organ may be an artifact of hepatic hypertrophy for the high-dose group on Day 1 postexposure. The reasons for the failure of the tracer techniques to detect a decreased functional capacity are not clear. The cirrhosis found in the high-dose group was moderate (results at Days 8 and 15 postexposure showed fibrosis extended only to a thickness of two layers of hepatocytes from the centrilobular area). Perhaps the techniques were not sufficiently sensitive to measure the limited amount of damage found or they were masked by the liver’s reserve functional capacity. A factor contributing to the relative insensitivity was the high degree of variability found among the samples of both tracers; as seen in Figs. 4 and 5, the standard deviations of the uptake ratios
SUBCHRONIC
HEPATIC
are large. Inspecting the data of Lough et al. ( 1987), one also sees high variability and may infer that the decrease in colloid uptake may be difficult to distinguish from control when liver damage is not extensive. Others have also reported high variability and low sensitivity (Klingensmith et al., 1983; Fischer et al., 1980) that have not been satisfactorily explained. Dynamic measurements of sulfur colloid in normal versus diseased livers have been made (Biersack et al., 198 1; Houston and Macleod, 1980) that show promise as a better technique for establishing the functional capacity of the liver. The basis for the decrease in uptake of 2DG with time postexposure is not clear. We speculate that it could be due to the cessation of corn oil in the diet after termination of the exposures. The change in type of caloric intake from high- to low-oil may affect the uptake of 2DG, which is presumably handled by the cell as a nutrient. No such effect was found in the 99mTc uptake studies. Since the 99mT~ is in the form of a colloid and is phagocytized, it should not be affected by changes in nutrient type and is consistent with our speculation. However, the experiment was not designed to explore vehicle effects and no definitive statement can be made. Summary. We produced hepatotoxicity and persistent hepatic cirrhosis by subchronic oral administration of low doses of CC& and monitored recovery after termination of exposure. By Day 8 postexposure, cytochrome P450 and the serum indicators of hepatotoxicity had returned to control levels and necrosis had disappeared. By Days 15 and 22, hepatocellular degeneration and increased relative liver weight still persisted. The severity and apparent reversibility of cirrhosis were dependent on dose: 20 mg CC&/kg produced cirrhosis that disappeared within 8 days, but 40 mg Ccl&g produced more severe cirrhosis that, while decreasing in severity, persisted through Day 15 postexposure. ACKNOWLEDGMENTS The authors thank Duke University Medical Center, Radiology Department, Nuclear Medicine, for the gener-
INJURY AND RECOVERY
569
ous donation of the technetium-99m sulfur colloid. Statistical consultations were provided by Jerry W. Highfill and Dr. David J. Svendsgaard. The authors gratefully acknowledge the assistance of Fred A. Talley for production of the histopathological prints and the technical assistance of B. L. Robinson, B. Terrill, W. Toler, and H. Smith. We thank Drs. William H. Briner, Kirk T. Kitchin, and B. Alex Merrick for reviewing this manuscript.
REFERENCES ANDERSON, V. L., AND MCLEAN, R. A. (1974). Design ofExperiments: A RealisticApproach. pp. 16-22. Dekker, New York. BIERSACK,H. J., TORRES, J., THELEN, M., MONZON, O., AND WINKLER, C. (198 1). Determination of liver and spleen perfusion by quantitative sequential scintigraphy: Results in normal subjects and in patients with portal hypertension. Clin. Nucl. Med. 6,2 1g-220. BRUCKNER, J. V., MACKENZIE, W. F., MURALIDHARA, S., LUTHRA. R., KYLE, G. M., AND ACOSTA, D. (1986). Oral toxicity of carbon tetrachloride: Acute, subacute, and subchronic studies in rats. Fundam. Appl. Toxicol. 6, 16-34. CAMERON, G. R., AND KARUNARATNE, W. A. E. (1936). Carbon tetrachloride cirrhosis in relation to liver regeneration. J. Pathol. Bacterial. 92, l-20. CONDIE, L. W., LAURIE, R. D., MILLS, T., ROBINSON, M., AND BERCZ, J. P. (1986). Effect of gavage vehicle on hepatotoxicity of carbon tetrachloride in CD-l mice: Corn oil versus Tween-60 aqueous emulsion. Fundam. Appl. Toxicol. 7, 199-206. FISCHER,G. J., STAAB, E. V.. LESESNE,H. R.. AND McCARTNEY, W. H. (1980). Liver-spleen imaging in patients with subacute hepatic necrosis and chronic hepatitis. Clin. Nucl. Med. 5, I3- 18. GUZELIAN, P. S. (1983). Research needs for hepatic injury due to environmental agents. Environ. Health Perspect. 48,65-7 1. HARPER, P. V., LATHROP, K. A., AND RICHARDS, P. ( 1964). Tc-99m as a radiocolloid. J. Nucl. Med. 5,382. HAYES, J. R., CONDIE, L. W., AND BORZELLECA, J. F. (1986). Acute, 14-day repeated dosing, and 90-day subchronic studies of carbon tetrachloride in CD-l mice. Fund. Appl. Toxicol. 7,454-463. HOM, F. G., GOODNER, C. J., AND BERRIE, M. A. ( 1984). A (3H)2-deoxyglucose method for comparing rates of glucose metabolism and insulin responses among rat tissues in vivo. Diabetes 33, I4 I- 152. HOUSTON, A. S., AND MACLEOD, M. A. (I 980). Processing of liver dynamic studies with technetium-labelled sulfur colloid. Brit. J. Radiol. 53,87-92. JENKINS, A. B., FURLER, S. M., AND KRAEGEN, E. W. (1986). 2-Deoxy-D-glucose metabolism in individual tissues of the rat in vivo. Int. J. Biochem. l&3 1 l-3 18.
ALLIS
KALF, G. F., POST, G. B., AND SNYDER, R. (1987). Solvent toxicology: Recent advances in the toxicology of benzene, the glycol ethers, and carbon tetrachloride. Annu. Rev. Pharmacol. Toxicol. 27,399-427. KLINGENSMITH, W. C.. SPITZER, V. M., FRITZBERG, A. R., AND KUNI, C. C. (1983). Normal appearance and reproducibility of liver-spleen studies with Tc99m sulfur colloid and Tc-99m microalbumin colloid. J. Nucl. Med. 24,8- 13. LOUGH, J., ROSENTHALL, L., ARZOUMANIAN, A., AND GORESKY, C. A. (1987). Kupffer cell depletion associated with capillarization of liver sinusoids in carbon tetrachloride-induced rat liver cirrhosis. J. Hepazol. 5, 190-198. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDALL, R. J. (195 1). Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193,265-275. OMURA, T., AND SATO, R. (1964). The carbon monoxide-binding pigment of liver microsomes. J. Biol. Chem. 239,2370-2378. OPPENHEIM, B. E., GOTTSCHALK, A., AND HOFFER, P. B. (1976). Gastrointestinal nuclear medicine. In Diagnostic Nuclear Medicine (A. Gottschalk, and E. J. Potchen, Eds.), Chap. 33, pp. 399-455. Williams & Wilkins, Baltimore. PEREZ-TAMAYO, R. (1979). Cirrhosis of the liver: A reversible disease? In Pathology Annual (S. C. Sommers, and P. P. Rosen, Eds.), Vol. 14, Part 2. pp. 183-2 13. Appleton-Century-Crofts, London. PRAKASH, V.. LIN, M. S.. AND KRISS, J. P. (1977). Liver scintigraphy in alcoholic liver disease. Clin. Nucl. Med. 2,308-309.
PRASSE, K.. HILDEBRANDT, P., AND DODD, D. (1986). Letter to the Editor. Vet. Pathol. 23,540-54 1.
ET AL.
RUBIN, E., HUTTERER, F., AND POPPER,H. (1963). Cell proliferation and fiber formation in chronic carbon tetrachloride intoxication. Amer. J. Pathol. 42, 7 15728.
SAS Institute, Inc. (1985). SAS Users Guide: Statistics, 5 Edition. SAS Institute, Gary, NC. SELLERS, E. A., LUCAS, C. C.. AND BEST, C. H. (1948). The lipotropic factors in experimental cirrhosis. Brit. Med. J. 1,1061-1065. SIMMONS, J. E., DEMARINI, D. M., AND BERMAN, E. (1988). Lethality and hepatotoxicity of complex waste mixtures. Environ. Res. 46,74-85. Society of Toxicologic Pathologists (1986). Society of Toxicologic Pathologists’ position paper on blinded slide reading. Toxicol. Pathol. 14,493-494. SOKAL, R. R., AND ROHLF, F. J. (1981). Biometry: The Principles and Practices of Statistics in Biological Research, 2nd ed., pp. 179-453. Freeman, San Francisco. SOKOLOFF, L. (I 98 1). Localization of functional activity in the nervous system by measurement of glucose utilization with radioactive deoxyglucose. J. Cereb. Blood FlowMetab. 1,7-36. WELSH, M. J. (1982). New radiopharmaceuticals for studying liver function. J. Nucl. Med. 23, 1138-l 139. WIDMAN, F. K. (1979). Clinical Interpretation of Laboratory Tests. 8th ed., pp. 350-386, Davis, Philadelphia. YONEKURA, Y., BENUA, R. S., BRILL, A. B., SOM, P.. YEH, S. D. J., KEMENY, N. E., FOWLER, J. S., MACGREGOR, R. R., STAMM, R., CHRISTMAN, D. R., AND WOLF, A. P. (1982). Increased accumulation of 2deoxy-2-[‘*F]fluoro-D-glucose in liver metastases from colon carcinoma. J. Nucl. Med. 23, 1133-I 137.
(1990)
l&558-570
Assessment of Hepatic Indicators of Subchronic Carbon Tetrachloride Injury and Recovery in Rats’ JOHNW.ALLIS,THOMASR.WARD,JOHN
C.SEELY,*ANDJANEELLENSIMMONS
Health Effects Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 2771 I: and*PATHCO. Inc., Research Triangle Park, North Carolina 27709
Received November 2, 1989; accepted May 14, 1990 Assessment of Hepatic Indicators of Subchronic Carbon Tetrachloride Injury and Recovery in Rats. ALLIS, J. W., WARD, T. R.. SEELY, J. C., AND SIMMONS, J. E. (1990). Amdam. Appl. Toxicol. 15, 558-570. To determine the course of hepatic recovery from subchronic oral administration of carbon tetrachloride (CC&). male F-344 rats were gavaged with 0,20, or 40 mg CC&/kg. 5 days/week, for 12 weeks. Exposure to Ccl, caused dosage-dependent increases in relative liver weight and the serum levels of aspartate aminotransferase, alanine aminotransferase, lactate dehydrogenase, alkaline phosphatase, and cholesterol as well as a dosage-dependent decrease in hepatic cytochrome P450. Centrilobular hepatocellular vacuolar degeneration, necrosis, and cirrhosis occurred at both 20 and 40 mg/kg, with dosage-dependent severity. Reversibility of these reported effects varied with parameter. By Day 8 postexposure, necrosis had disappeared and all serum indicators and cytochrome P450 had returned to control levels. By Day 15 postexposure, the severity of the vacuolar degeneration had decreased. Reversibility of cirrhosis was dosage dependent; complete recovery occurred in the low- but not the high-dose group by Day 15. The disappearance of the increase in relative liver weight was also dependent on dosage; the low- but not the high-dose group had returned to the control level by Day 22. In an attempt to measure persistent hepatic damage, liver uptake relative to the spleen was determined for a sulfur colloid labeled with technetium-99m and for tritiated 2-deoxyglucose. Neither method consistently measured hepatic damage in cirrhotic livers due, in part, to the high degree of variability in the tracer uptake data. o 1990 society of Toxicology.
The hepatotoxic effects of carbon tetrachloride (CCL,) have been widely studied and mechanisms have been proposed for the production of necrosis in the liver (for a review see Kalf et al., 1987). Most of the information on CC4 hepatotoxicity has been developed from acute or short-term studies and relatively little is known about subchronic exposure and recovery after cessation of exposure.
’ The research described in this article has been reviewed by the Health Effects Research Laboratory and approved for publication. Approval does not signify that the contents necessarily reflect the views of the Agency nor does mention of trade names or commercial products constitute endorsement or recommendation for use. 0272-0590/90 $3.00 Copyright 0 1990 by the Society of Toxicology. All rights of reproduction in any form reserved.
558
Bruckner et al. (1986) conducted a comprehensive subchronic study in which rats were gavaged five times weekly for 12 weeks with 1, 10, and 33 mg CC&/kg body wt. Beginning at Week 2, rats dosed with 33 mg/ kg exhibited increased serum activities for the enzymes alanine aminotransferase (ALT), sorbital dehydrogenase (SDH), and ornithine carbamoyltransferase (OCT). After 12 weeks of exposure at this dosage, periportal fibrosis and other degenerative changes associated with cirrhosis were also found. For the 10 mg/ kg dosage level, SDH and ALT were elevated and centrilobular vacuolization was found after 12 weeks of exposure. After a 2-week recovery period, all serum enzymes (with the
SUBCHRONIC
HEPATIC
possible exception of ALT) had returned to normal, but the cirrhotic changes remained. In another study, Hayes et al. (1986) treated mice with 12 to 1200 mg CCWkg daily for 90 days, finding evidence of hepatotoxicity in all groups. Several serum indicators of hepatotoxicity plus liver-to-body weight ratio were elevated, with males more susceptible than females. Necrosis and other histopathological changes were also found in each dose group. A similar study (Condie et al., 1986) used dosages of 1.2, 12, and 120 mg/kg 5 days per week for 12 weeks. Again, necrosis and increased serum enzymes were found at both 12 and 120 mg/kg, but no changes were noted at the lowest dosage. Earlier subchronic studies with rats (Rubin et al., 1963; Cameron and Karunaratne, 1936) utilized a twice weekly dosing regimen at dosages substantially higher than the lowest effective dosages in the reports cited above. In this earlier work, dosages of more than 1000 mg/kg twice a week produced fibrosis and cirrhosis at various times between 5 and 20 weeks. Cameron and Karunaratne (1936) also reported that dosages of approximately 40 mg/kg delivered on a similar schedule did not produce structural alterations in the liver. Except for the single time point investigated by Bruckner et al. ( 1986), recovery from exposure and permanent damage to the liver from subchronic dosing has been investigated at high doses given infrequently (more than 1000 mg/kg two or three times weekly), and with one exception (Sellers et al., 1948) by routes other than oral (for a review see Perez-Tamayo, 1979). In these studies, cirrhosis and presence of collagen were the principal endpoints studied. The data generally support the finding of Cameron and Karunaratne ( 1936) that experimental cirrhosis is reversible if dose and duration of exposure are sufficiently low, but eventually it appears to be irreversible. The work described below had the twofold objective of determining the time course of recovery from subchronic oral CC4 administration for a number of experimental end-
559
INJURY AND RECOVERY
points and attempting to develop a measure for persistent impairment of the liver’s functional capacity. The doses employed are much lower and the experimental endpoints more varied than in previous studies investigating recovery. Treatment with CCL, produces experimental cirrhosis in rats and should provide a suitable test for the sensitivity and appropriateness of the functional measures employed. Doses were chosen to provide differing degrees of cirrhosis in order to test the sensitivity of the functional measures. The assay for functional capacity was based on a clinical procedure used in humans but its validity to measure functional decrement in animals has not yet been established. Harper et al. ( 1964) developed a technique using a sulfur colloid labeled with technetium-99m (99mTc) as a liver imaging tool employing a gamma camera. By measuring the relative uptake of the colloid in the liver compared to that in the spleen, the relative efficiency of the liver can be determined; this liver-to-spleen ratio is used as a diagnostic test for establishing the degree of human cirrhosis (Prakash et al., 1977). This ratio may also be useful as a functional test in animals. However, the colloid is taken up by Kupffer cells rather than by hepatocytes; thus the technique is an indirect measure of hepatocyte function. To compare the indirect sulfur colloid assay with a more direct method, we measured hepatic uptake of tritiated 2-deoxyglucose as an indicator of hepatic glucose utilization. Uptake of 2-deoxylglucose has been widely used as an indicator of cerebral metabolic rate (Sokoloff, 198 1) and its analogue ([2-‘8F]fluoro-2-deoxy-D-glucose) has been suggested as a diagnostic tool in hepatic studies (Yonekura et al., 1982). MATERIALS
AND
METHODS
Male 60-day-old Fischer 344 rats (Charles River Labo-
ratories, Raleigh, NC) were housed three per cage in polycarbonate cages with bedding of heat-treated pine shavings, fed Purina Rodent Lab Chow (No. 5001, Rals-
ALLIS ET AL. ton Purina, Checkerboard Square. MO) and given tap water ad libitum. Animals were acclimatized for 5 days before the experiment was begun. Feed was removed from the cages the afternoon prior to termination. A 12hour light-to-dark cycle with light beginning at 6 AM was maintained during the experiment. Sentinel animals were housed in the animal room as part of an ongoing surveillance program for parasitic, bacterial, and viral infections. Treatment. One hundred forty-four rats were gavaged 5 days per week for 12 weeks with 0,20, or 40 mg/kg/day carbon tetrachloride (Sigma Chemical Co.; Calbiochem) in corn oil (Sigma Chemical Co.) (0, 10, or 20 mg CCL,/ ml corn oil). Dosing was begun on a Monday and ended on a Monday to avoid terminating animals after 2 days of recovery (i.e., after the weekend). Rats were weighed on Days 1, 3, 5, and 8 and thereafter on a weekly basis. A new gavage solution was used daily for each dosage; solutions were made weekly. The gavage vehicle was corn oil; all rats including controls received 2 ml/kg with dosing volumes adjusted for each rat on the basis of its most recent weight. Food consumption was measured by cage on a regular basis throughout dosing. The rats were divided into two groups of 72 animals each (24 animals per dose level). One group of animals was used for measuring hepatic cytochrome P450, serum chemistries, and light microscopic histopathology. Six animals from each dose level were terminated on Days I, 3.8, and 15 postexposure. The other group was used to measure radiolabeled sulfur colloid and 2-deoxyglucose (2DG) uptake and six animals from each dose level were sacrificed on Days 1.8, 15, and 22 postexposure. At termination, the first group was anesthetized with 50 mg/kg sodium pentobarbital (Abbott Lab., Chicago, IL) injected ip and exsanguinated from the abdominal aorta into serum-separation tubes (Becton-Dickinson, Lincoln Park, NJ). The blood was held on ice for a minimum of 30 min and then centrifuged at 1OOOgfor 30 min. Serum was collected and frozen at -40°C until analyzed. Serum concentrations of total bilirubin (Bib), triglycerides, and cholesterol and the activities of aspartate aminotransferase (AST), ALT. alkaline phosphatase (Alk Phos), and lactate dehydrogenase (LDH) were assayed commercially (Vet-Path, Teterboro, NJ). Each liver was quickly excised, rinsed in saline, blotted, and weighed. Slices of the left lobe were removed for open light-microscopic histopathological examination (Society of Toxicologic Pathologists. 1986: Prasse et al.. 1986) and placed in 10% phosphate-buffered formalin. Two grams of liver was taken to assay for cytochrome P450 by the method of Omura and Sato (1964). Protein determinations were performed on the microsomal samples by the method of Lowry et al. ( 195 1). Histological evaluation of livers was conducted on hematoxylin- and eosin-stained tissue sections. In addition, all livers in the 20 mg/kg/day and selected livers in the 40 mg/kg/day groups were stained with Masson’s trichrome to confirm the presence or absence ofcollagen.
Degenerative and necrotic lesions were graded according to the following criteria: none, minimal (one to several centrilobular hepatocytes affected), mild (no more than one-fourth of the hepatocytes of the centrilobular region affected), moderate (expansion of the affected hepatocytes up to one-half of the centrilobular region), or marked (over one-half of the lobular architecture involved). These criteria have been described more fully in Simmons et al. (1988). The presence of cirrhosis was based on the following criteria: disorganization of the liver by interconnecting fibrous bands or scarring that tended to be either portalcentral or portal-portal; formation of parenchymal nodules that varied in size and were due to regenerative activity within the network of scars; and within regenerative nodules, the lobular architecture of the liver appeared disorganized, i.e., hepatocytes varied in size, mitotic figures were present, and there was a loss of central veins. Accordingly, cirrhosis was graded as follows: none, minimal (fibrous scarring to a thickness of one hepatocyte or less), mild (fibrous tissue extended to a thickness of one or two hepatocytes, with some micronodular formation), or moderate (fibrous tissue extended to a thickness of more than two hepatocytes with micronodular formation pronounced). The second group of animals was used to evaluate the sulfur colloid and 2DG uptake assays. The uptake of these two radiolabeled tracers was measured and the liver-to-spleen ratios were calculated. Because of technical problems in handling I8 animals in I day, this group was split into two parts. Half of the animals were dosed for I2 weeks and 1 day. At termination on Days 1,8, 15, and 22 postexposure, rats were anesthetized with 50 mg/kg pentobarbital injected ip. When the animal was not responsive to a foot pinch (approximately 10 min), the left femoral vein was surgically exposed. The animal was then injected iv with 0. I ml of a mixture of sulfur colloid (donated by Duke University Medical Center, Durham, NC) labeled with technetium-99m ( 10 &i/rat, measured at 8 AM each day) and 2-deoxy+ghtcose (New England Nuclear) labeled with tritium (1 &/rat). Ten minutes after injection of the radioactive tracers. the animal was terminated by introducing perfusate (0. IS M NaCl) into the left ventricle and cutting the inferior vena cava anterior to the liver. After 5 min of perfusion (approximately 50 ml), the liver and spleen were removed and weighed. Tissue samples of about 0.2 g were placed in liquid scintillation vials and counted in a gamma counter (Packard Instruments, Auto/Gamma) immediately after termination of all rats each day. To reduce variability due to the short half-life of 99mTc (6 hr), hepatic and splenic samples were counted consecutively for each rat and the data analyzed as the liver-tospleen ratio for each rat. Following gamma counting, 2 ml of tissue solubilizer (Protosol, New England Nuclear) was added to each sample. After incubation for 2 days at 37°C the solubilizer
SUBCHRONIC
HEPATIC
was neutralized with 0.5 ml 3 M HCl and 10 ml ofscintillation cocktail (Liquiscint, National Diagnostics) was added. The samples were counted in a liquid scintillation counter (Packard Instruments, Model 2660) after the 99mT~ had decayed for at least a week. Scintillation data were adjusted for quenching and reported as dpm. Ratios for each rat were calculated and analyzed in the same manner as the 99mT~data. Statistical analysis. Rat weight gain during the course of dosing was analyzed by obtaining the slope and intercept for each rat by linear regression analysis (SAS, 1985). The slopes and intercepts were analyzed by analysis of variance (ANOVA) (SAS, 1985). Weighted average daily food consumption was calculated (Weeks I to 11) for each cage of rats. A quadratic equation was fitted to the data for each cage; these coefficients were analyzed by ANOVA. All other numerical data were also analyzed by ANOVA. A p value 5 0.05 was used for all ANOVAs, and when that criterion was met, the Sidak method of comparison of means (for multiple comparisons) was used, with p 5 0.05 required for significance. Homogeneity of variance was checked with Bartlett’s test (Sokal and Rolf, 198 1) and when necessary a log transformation and reanalysis were performed. The criterion of significance for Bartlett’s test wasp G 0.00 1 (Anderson and McLean, 1974).
RESULTS Hepatotoxicity and recovery. One day after the cessation of the 12-week exposure, there was considerable evidence of hepatotoxicity. The serum indicators ALT, AST, and LDH were elevated for both the 20 and 40 mg/kg/ day groups (Table 1). Alk Phos and cholesterol were also increased for the higher dose group only (Table l), while total bilirubin (data not shown) and triglycerides were unaffected. Cytochrome P450 levels (Fig. 1) were significantly reduced for both dose groups. Histopathological evaluation confirmed cellular damage from Ccl4 administration (Table 2 and Fig. 2). The primary lesions were in the centrilobular region. At 20 mg/ kg, two of the six livers were cirrhotic, and all showed vacuolar degeneration and hepatocellular necrosis. There was fibrous tissue in the noncirrhotic livers of this group, but it was not interconnecting. At the higher dose, all livers were cirrhotic and showed evidence for degeneration and necrosis. Staining with
INJURY AND RECOVERY
561
Masson’s trichrome showed the presence of collagen, confirming the finding of cirrhosis in both treated groups. By Day 3 postexposure, the serum indicators of hepatotoxicity (Table 1) were decreasing. With the exception of cholesterol, magnitudes of all indicators significant on Day 1 declined by Day 3. Significant elevations remained in the higher dose group for the five measures ALT, AST, LDH, Alk Phos, and cholesterol and in the lower group only for ALT. Similarly, cytochrome P450 levels were returning to normal and only the high-dose group showed a significant decrement. The histopathological results (Table 2) were similar to those on Day 1. At Days 8 and 15 postexposure, all serum indicators and cytochrome P450 had returned to control levels for both dose groups (the small but statistically significant decrements for ALT on Day 8 and AST on Day 15 were not considered to be biologically significant). Cirrhosis persisted in the high-dose group (Table 2 and Fig. 2d), although there was some decrease in its severity. The decrease in severity was confirmed by the Masson’s trichrome data showing that collagen and bridging decreased with time postexposure. In the low-dose group, cirrhosis was not evident by Day 8. For both groups, centrilobular hepatocellular necrosis disappeared by Day 8 and vacuolar degeneration decreased in severity at Days 8 and 15 but was still present. Liver weight relative to body weight at termination is shown for all 144 animals in Fig. 3. At Days 1, 3, 8, and 15 postexposure, significant differences were found between both dose groups and controls. In contrast to most of the other measures described above, relative liver weight for the exposed animals in the high-dose group did not return to the control level, even by Day 22. Radiolabeled tracer uptake as a measure of persistent hepatic damage. Two techniques of ascertaining persistent hepatic damage and functional decrement were assessed. The first, the uptake of sulfur colloid labeled with
562
ALLIS ET AL. TABLE 1 SERUM CHEMISTRY VALUES FOLLOWING SUBCHRONICEXPOSURE TO CARBON TETRACHLORIDE
Day post exposure
Dose km/k/day)
Day I
ALT (W/liter)
0
7s+1sb 387i. l52+ 3046 f 1299*
20 40
Day 3
0
0 20 40
Day 15
106+20 299+ 121+ 2593,1142*
0 20 40
’ CHOL, cholesterol. b Data are reported as means + Significantly different from * Significantly different from a Significantly different from
CHOL” bw/dl)
45+ 10
49k 3 47* 4 88+25*
90,17 108,19 365 t62*
l65k26 180+39 361+77*
36+ 4 4lF 3 63+13*
100 -+ 28*
64k5 54*4+ 61+5
9ort17 85k 86+6
2OOk72 234+41 217?26
34* 39+ 42+
5 4 6
432 4lk 47f
4 6 3
67k5 58k5 75k9'
89~5 80+4+ 95,3"
194k45 166t40 222*43
33+ 33k 38+
7 5 8
48+43+ 52k
3 6 8
JO
44f 52kII
5
+ SD. control, p i 0.05. control and from 20 mg/kg, p < 0.05. 20 mg/kg but not from control. p < 0.05.
Do6e(m@k@day) 00 I3 20 q 40
3
Alk Phos (W/liter)
55-c 8 84-c21*
99mT~, is reported as the liver-to-spleen ratio in two ways: In counts per gram of tissue and as counts per organ. When reported as
1
LDH (W/liter) 208k62 6052413' 2370+720*
67k9 112-t-30+ 5lOk 85*
20 40
Day 8
AST (W/liter)
8
15
Day Post-Exposure
FIG. 1. The effect of subchronic exposure to 20 and 40 mg CClJkg/day on the concentration of hepatic cytochrome P450. N = 6 per group. Significantly different from the concurrent control group, p < 0.05. *Significantly different from the concurrent control group and the 20 mg/kg group, p < 0.05.
counts/gram (Fig. 4a), on Day 1 postexposure the 40 mg/kg group was significantly lower than controls. This effect was not seen at 20 mg/kg, or on later termination days. When reported as the ratio of counts/organ (Fig. 4b), there was neither a dose nor a day effect. These apparently conflicting results may be explained by examining liver weights: The livers are enlarged on Day 1 postexposure (Fig. 3) giving rise to a decrease in counts/gram although the total uptake by the liver is unchanged. The controls for both methods when analyzed separately showed no day effect. The uptake was also analyzed as counts/gram for the liver and the spleen individually (results not shown), and no effect was seen. The second technique of ascertaining persistent damage, that of deoxyglucose uptake, is reported in the same manner, counts/gram and counts/organ (Figs. 5a and 5b). The liverto-spleen ratio of deoxyglucose uptake reported as counts/gram showed no effect of
SUBCHRONIC
HEPATIC
563
INJURY AND RECOVERY
TABLE 2
Day postexposure I
8
15
0
20
40
0
20
40
0
20
40
0
20
40
5
-
-
(j
-
-
6
-
-
5
-
-
lb -
-
6
_--
6 -
2 4
-
1 5 -
-
2------5--3 4 2--3------
CCl,dosage(mg/kg): Normal Vacuolar degeneration Minimal Mild Moderate Marked Hepatocellular necrosis Minimal Mild Moderate Marked Cirrhosis Minimal Mild Moderate
3
1
I-------5 5 6 6 -
-
6
(j
_--
-
5 1
-
-----_
-
1*
-
_
6
-
-
-
-
-
-
-
_
3
-----
1
-
3
a Lesions were graded on hematoxylin- and eosin-stained tissue sections. Observed lesions were centrilobular in location and were graded according to the criteria described under Materials and Methods. Each group contained six animals. b Lesion not considered associated with experimental procedures.
dose. When reported as counts/organ the analysis showed a significant effect of dose which appears to be primarily based on the statistical significance of the 20 mg/kg point at Day 8 (Fig. 5b). However, both methods of analysis show a significant day effect as hepatic uptake relative to the spleen decreased with time postexposure. The controls for the two methods, analyzed separately, also demonstrate the day effect. In sum, the 2 DG data indicate a decrease in uptake of deoxyglucose after the cessation of dosing. Animal weight gain. Figure 6 shows animal weights for each dose group during the initial 78 days of exposure. Soon after the initiation of dosing, both the 20 and 40 mg/kg groups began to exhibit apparent growth retardation. The animal weight curves showed an overall dose effect with weight decreasing with increased dose; the slopes of the weight curve in
both the 20 and 40 mg/kg groups were significantly lower than control. The decrement in weight gain in the exposed groups could not be accounted for by food consumption. The average food consumption curves for the controls and the two exposed groups (data not shown) were statistically indistinguishable from one another. At Day 22 postexposure, animal weights at termination for all dose groups (data not shown) were equivalent, indicating recovery from the weight decrement developed during dosing. DISCUSSION Hepatotoxicity and recovery. The hepatotoxicity data presented in this paper show that low doses of carbon tetrachloride, 20 and 40 mg/kg administered subchronically, pro-
564
ALLIS ET AL.
SUBCHRONIC
HEPATIC
565
INJURY AND RECOVERY DOSS (mglk@W
1
3
a
15
22
Day Post-Exposure
FIG. 3. The effect of subchronic exposure to 20 and 40 mg CCld/kg/day on relative liver weight. N = 12 per group. ‘Significantly different from the concurrent control group, p G 0.05. *Significantly different from the concurrent control group and the 20 mg/kg group, p d 0.05.
q
2s
40
20 1s 10 5
duce hepatotoxicity and that over the 12 weeks of exposure, persistent damage arises in the form of cirrhosis. The recovery from the ongoing toxicity was rapid, with serum indicators of hepatotoxicity and cytochrome P450 levels returning to normal at parallel rates and within 8 days of cessation of exposure. These data are consistent with the disappearance of necrosis over the same time span. However, vacuolar degeneration persisted longer and was still evident although decreased in severity at Day 15 postexposure. The times to recovery for the parameters in this study are summarized in Table 3. The elevation of ALT, AST, and LDH on Days 1 and 3 postexposure are indicative of ongoing parenchymal cell damage, with ALT being specific for the liver (Guzelian, 1983). The presence of vacuolar degeneration and
0 1
8
Day
1s
22
Post-Exposure
FIG. 4. The effect of subchronic exposure to 20 and 40 mg CC&/kg/day on the liver-to-spleen ratio of uptake of sulfur colloid 99mTc expressed as cpm/g tissue (a) and cpm/organ (b). N = 6 per group. *Significantly different from the concurrent control group and the 20 mg/kg group, p G 0.05. “Significantly different from 20 mg/kg but not from control, p G 0.05.
necrosis in the liver is consistent with these findings, indicating hepatocyte damage and death. Elevated levels of Alk Phos and cholesterol are consistent with malfunction of the hepatobiliary system, but this result was not confirmed by changes in total bilirubin levels which were unaffected during the experiment. The putative hepatobiliary involve-
FIG. 2. Histopathology. (a) Control liver. Normal centrilobular area is shown. H&E, 40X. (b) Day I postexposure, 20 mgjkg. Moderate centrilobular vacuolar degeneration and mild centrilobular necrosis is present. H&E, 40X. (c) Day 1 postexposure, 40 mg/kg. Hepatic cirrhosis is characterized by fibrous bands surrounding regenerative hepatocellular parenchyma. Moderate vacuolar degeneration and mild hepatocellular necrosis are present along the periphery of the lobule. H&E, 16X. (d) Day 15 postexposure, 40 mg/ kg. Cirrhosis is still evident; however, the amount of vacuolar degeneration is reduced. H&E, 16X.
566 a
ALLIS ET AL. 3.0
Dose (mgkgday,
T
1
T
b
50
c
I
1
8
15
22
Day Post-Exposure
FIG. 5. The effect of subchronic exposure to 20 and 40 mg CC&/kg/day on the liver-to-spleen ratio of uptake of tritiated 2-deoxylgucose expressed as dpm/g tissue (a) and dpm/organ (b). N = 6 per group except N = 5 for Day 15 controls. +Significantly different from the concurrent control group, p < 0.05.
two experiments provide a consistent picture of hepatic toxicity and recovery for subchronic exposure of rats to carbon tetrachloride. These studies indicate that the lowest effective subchronic dosage for hepatotoxicity in rats is near 10 mg/kg. They also indicate that the threshold for cirrhosis persisting for at least 2 weeks following cessation of exposure is between 20 and 33 mg/kg for this dosing schedule. The work of Hayes et al. (1986) and Condie et al. (1986) with mice shows that subchronic exposure produces effects similar to those in rats. While neither paper provides recovery data, a subchronic dosing schedule similar to that used for rats gave evidence for hepatotoxicity at 12 mg/kg but not at 1.2 mg/ kg (Condie et al., 1986). It would appear that the lowest effective level of CC& in mice is similar to that in rats. All four of these studies show that daily exposure to CC& is hepatotoxic at much lower doses than twice weekly dosing over a protracted period of time. For example, Rubin et al. ( 1963) used more than 1500 mg/kg twice weekly (actually 0.15 ml per animal, delivered by subcutaneous injection, in female Sprague-Dawley rats) to produce necrosis in 10 days, fibrosis between 20 and 65 days, and cirrhosis beginning 95 days
350l’ ment (Widman, 1979) either is in an early stage of development or may be secondary to general inflammation of the liver, as indicated by the increased relative liver weight. These data extend the work of Bruckner et al. ( 1986) by providing the time course of recovery for ongoing parenchymal toxicity, the decrease of cytochrome P450 due to treatment and subsequent recovery, and the minimal involvement of the hepatobiliary system at this stage of liver injury. Their result that ALT levels had returned to normal but cirrhosis remained at 2 weeks postexposure is consistent with our findings, notwithstanding the different strain of rat and somewhat different doses used. Taken together, these
15ol0
I 15
30
Days
of Exposure
45
60
75
to CC14
FIG. 6. The effect of subchronic exposure to 20 and 40 mg CC&/kg/day on body weight during the course of dosing; each dose group consisted of 48 animals. Both treatment groups were significantly different from control (see text). Error bars are not shown for clarity.
SUBCHRONIC
HEPATIC
INJURY TABLE
AND
567
RECOVERY
3
SUMMARY OF RECOVERY TIME FOLLOWING SUBCHRONICEXPOSURE TO 40 MG/KG CC& Parameters
Day 1
Day 3
Day 8
Day 15
Day 22
Serum chemistries Cytochrome P450 Liver wt/body wt Body wt Degeneration Necrosis Cirrhosis 99mT~uptake 2DG uptake
+
+ + 0 + + + ND ND
0 0 + 0 + 0 + 0 0
0 0 + + 0 + 0 0
ND ND + 0 ND ND ND 0 0
+ + + + 0 0
Note. 0, not different from controls for quantitative data; absent for histopathology. +, significantly increased compared to controls for quantitative data; present for histopathology. -, significantly decreased compared to controls. ND, no data; data not taken.
after starting exposures. Cameron and Karunaratne ( 1936) found necrosis in albino rats (Graham, sex not reported) after about 10 days, fibrosis after 3 weeks, and cirrhosis after 14 weeks with twice weekly dosing of lOOO2600 mg/kg (0.10-0.25 ml delivered subcutaneously). They also found only slight, reversible degeneration after 4 months of twice weekly dosing with 40 mg/kg (0.0037 ml), the higher dose used in our study. Sellers et al. (1948) used 250 mg/kg (by gavage) every third day for 4 months to produce cirrhosis. Our study demonstrated recovery from cirrhosis. Recovery was complete for the 20 mg/ kg group and the severity of cirrhosis decreased in the 40 mg/kg group during the post-treatment period. While cirrhosis is generally regarded as irreversible, in fact, recovery from CC&-induced cirrhosis was first noted by Cameron and Karunaratne (1936) who found reversible and nonreversible components depending on the duration of exposure. The components apparently are indistinguishable histologically. A number of other workers have also demonstrated recovery from CCkinduced cirrhosis (see a review by Perez-Tamayo, 1979). In this work, cirrhosis has been produced by injection at dosages of at least 1000 mg/kg, usually given twice weekly, or by inhalation
at unspecified doses. In the one exception, Sellers et al. ( 1948) exposed one group by inhalation at 1000 ppm and another group to 250 mg/kg by gavage 2-3 days per week. In general, these studies conclude that cirrhosis is produced after several weeks of exposure, and a reversible stage is followed by an irreversible stage as dosing continues. Our results are clearly within the reversible stage based on the low doses delivered, the relatively mild cirrhosis found, and the demonstrated decline in severity in the 40 mg/kg group within 15 days postexposure.
Tracer uptake methods for persistent hepatic damage. Because cirrhosis cannot be measured quantitatively and is not a direct measure of liver function, we investigated another measure in the interest of determining persistent damage by a method other than histopathology. Cirrhosis was used as the indicator of permanent damage with the presumption of a functional decrement. For the functional assay, we chose the uptake of radioactive tracer-labeled substances by the liver and spleen. The rate of uptake should indicate the liver’s functional capacity. Comparison to the rate of uptake by the spleen should provide an internal control and reduce variability (based on the premise that the spleen was not affected by CC& treatment).
568
ALLIS
Two different tracer methods were employed. The first was a sulfur colloid labeled with technetium-99m that is used clinically to assess the extent of hepatic cirrhosis by gamma-camera imaging. Clinicians have long used this technique as the method of choice (Oppenheim et al., 1976) to determine liver size and pattern of hepatic uptake. The liver-to-spleen uptake ratio is also measured clinically as a qualitative measure of human cirrhosis; a decreased ratio indicates impaired function. Although the colloid is a particulate and as such is phagocytized by the Kupffer cells and not by hepatocytes, the decreased uptake of colloid has been found to be correlated with a decrease in number of Kupffer cells and increased injury in the livers of rats with CC&induced cirrhosis (Lough et al., 1987). More recently, another chemical, [2‘8F]fluoro-2-deoxy-D-glucose (2FDG), has been suggested as an alternative liver imaging agent (Yonekura et al., 1982). The short halflife (2 hr) gamma-emitting isotope of fluorine is attached to 2-deoxy-D-glucose allowing the uptake and distribution of this compound to be monitored with a gamma camera. Both 2FDG and its parent compound 2DG are glucose analogues which have been used extensively to study brain metabolism. Because of uncertainties in their metabolism, the use of glucose analogs in the liver has come under criticism (Welsh, 1982). 2DG is transported across membranes and phosphorylated to 2-deoxyglucose 6-phosphate in the same manner as is glucose. In the liver, the deoxyglucose compounds are not trapped as efficiently as they are in the brain. However, Jenkins et al. ( 1986) who have shown that the metabolic products of 2DG are different in the brain and liver, conclude “provided these products are not lost from the tissues to a substantial extent (as appears to be the case), they should not invalidate the use of 2-deoxyglucose as a monitor of glucose metabolism” in the liver. To the authors’ knowledge, the 2DG technique has been tried only once in the liver, by Yonekura et al.
ET AL.
(1982) who concluded that it may be useful. Because deoxyglucose is taken up by all cells including hepatocytes, it should give a good indication of the total metabolic activity of this organ. Use of this compound, if measured shortly after injection, should assessthe hepatic metabolic rate which in turn should indicate hepatic function. One potential confounding factor is the affect of insulin which influences the nonmetabolic uptake and trapping of glucose (Horn et al., 1984). Insulin was not measured in this study. Neither of the two tracer techniques detected a decreased functional capacity in cirrhotic livers. Based on the histopathological finding of cirrhosis in the liver that persisted through at least Day 15 postexposure in the 40 mg/kg group, the lack of corroboration by either or both measures of liver functional capacity was surprising. Because of the extensive swelling of the liver due to Ccl, exposure, the counts per organ method of measurement was probably the more meaningful for either assay. Indeed, the correlation between decreased number of Kupffer cells and colloid uptake has been determined on an organ basis (Lough et al., 1987). As mentioned earlier, the decrease in the liver-to-spleen ratio for the sulfur colloid when calculated by the counts/gram method but not by counts/ organ may be an artifact of hepatic hypertrophy for the high-dose group on Day 1 postexposure. The reasons for the failure of the tracer techniques to detect a decreased functional capacity are not clear. The cirrhosis found in the high-dose group was moderate (results at Days 8 and 15 postexposure showed fibrosis extended only to a thickness of two layers of hepatocytes from the centrilobular area). Perhaps the techniques were not sufficiently sensitive to measure the limited amount of damage found or they were masked by the liver’s reserve functional capacity. A factor contributing to the relative insensitivity was the high degree of variability found among the samples of both tracers; as seen in Figs. 4 and 5, the standard deviations of the uptake ratios
SUBCHRONIC
HEPATIC
are large. Inspecting the data of Lough et al. ( 1987), one also sees high variability and may infer that the decrease in colloid uptake may be difficult to distinguish from control when liver damage is not extensive. Others have also reported high variability and low sensitivity (Klingensmith et al., 1983; Fischer et al., 1980) that have not been satisfactorily explained. Dynamic measurements of sulfur colloid in normal versus diseased livers have been made (Biersack et al., 198 1; Houston and Macleod, 1980) that show promise as a better technique for establishing the functional capacity of the liver. The basis for the decrease in uptake of 2DG with time postexposure is not clear. We speculate that it could be due to the cessation of corn oil in the diet after termination of the exposures. The change in type of caloric intake from high- to low-oil may affect the uptake of 2DG, which is presumably handled by the cell as a nutrient. No such effect was found in the 99mTc uptake studies. Since the 99mT~ is in the form of a colloid and is phagocytized, it should not be affected by changes in nutrient type and is consistent with our speculation. However, the experiment was not designed to explore vehicle effects and no definitive statement can be made. Summary. We produced hepatotoxicity and persistent hepatic cirrhosis by subchronic oral administration of low doses of CC& and monitored recovery after termination of exposure. By Day 8 postexposure, cytochrome P450 and the serum indicators of hepatotoxicity had returned to control levels and necrosis had disappeared. By Days 15 and 22, hepatocellular degeneration and increased relative liver weight still persisted. The severity and apparent reversibility of cirrhosis were dependent on dose: 20 mg CC&/kg produced cirrhosis that disappeared within 8 days, but 40 mg Ccl&g produced more severe cirrhosis that, while decreasing in severity, persisted through Day 15 postexposure. ACKNOWLEDGMENTS The authors thank Duke University Medical Center, Radiology Department, Nuclear Medicine, for the gener-
INJURY AND RECOVERY
569
ous donation of the technetium-99m sulfur colloid. Statistical consultations were provided by Jerry W. Highfill and Dr. David J. Svendsgaard. The authors gratefully acknowledge the assistance of Fred A. Talley for production of the histopathological prints and the technical assistance of B. L. Robinson, B. Terrill, W. Toler, and H. Smith. We thank Drs. William H. Briner, Kirk T. Kitchin, and B. Alex Merrick for reviewing this manuscript.
REFERENCES ANDERSON, V. L., AND MCLEAN, R. A. (1974). Design ofExperiments: A RealisticApproach. pp. 16-22. Dekker, New York. BIERSACK,H. J., TORRES, J., THELEN, M., MONZON, O., AND WINKLER, C. (198 1). Determination of liver and spleen perfusion by quantitative sequential scintigraphy: Results in normal subjects and in patients with portal hypertension. Clin. Nucl. Med. 6,2 1g-220. BRUCKNER, J. V., MACKENZIE, W. F., MURALIDHARA, S., LUTHRA. R., KYLE, G. M., AND ACOSTA, D. (1986). Oral toxicity of carbon tetrachloride: Acute, subacute, and subchronic studies in rats. Fundam. Appl. Toxicol. 6, 16-34. CAMERON, G. R., AND KARUNARATNE, W. A. E. (1936). Carbon tetrachloride cirrhosis in relation to liver regeneration. J. Pathol. Bacterial. 92, l-20. CONDIE, L. W., LAURIE, R. D., MILLS, T., ROBINSON, M., AND BERCZ, J. P. (1986). Effect of gavage vehicle on hepatotoxicity of carbon tetrachloride in CD-l mice: Corn oil versus Tween-60 aqueous emulsion. Fundam. Appl. Toxicol. 7, 199-206. FISCHER,G. J., STAAB, E. V.. LESESNE,H. R.. AND McCARTNEY, W. H. (1980). Liver-spleen imaging in patients with subacute hepatic necrosis and chronic hepatitis. Clin. Nucl. Med. 5, I3- 18. GUZELIAN, P. S. (1983). Research needs for hepatic injury due to environmental agents. Environ. Health Perspect. 48,65-7 1. HARPER, P. V., LATHROP, K. A., AND RICHARDS, P. ( 1964). Tc-99m as a radiocolloid. J. Nucl. Med. 5,382. HAYES, J. R., CONDIE, L. W., AND BORZELLECA, J. F. (1986). Acute, 14-day repeated dosing, and 90-day subchronic studies of carbon tetrachloride in CD-l mice. Fund. Appl. Toxicol. 7,454-463. HOM, F. G., GOODNER, C. J., AND BERRIE, M. A. ( 1984). A (3H)2-deoxyglucose method for comparing rates of glucose metabolism and insulin responses among rat tissues in vivo. Diabetes 33, I4 I- 152. HOUSTON, A. S., AND MACLEOD, M. A. (I 980). Processing of liver dynamic studies with technetium-labelled sulfur colloid. Brit. J. Radiol. 53,87-92. JENKINS, A. B., FURLER, S. M., AND KRAEGEN, E. W. (1986). 2-Deoxy-D-glucose metabolism in individual tissues of the rat in vivo. Int. J. Biochem. l&3 1 l-3 18.
ALLIS
KALF, G. F., POST, G. B., AND SNYDER, R. (1987). Solvent toxicology: Recent advances in the toxicology of benzene, the glycol ethers, and carbon tetrachloride. Annu. Rev. Pharmacol. Toxicol. 27,399-427. KLINGENSMITH, W. C.. SPITZER, V. M., FRITZBERG, A. R., AND KUNI, C. C. (1983). Normal appearance and reproducibility of liver-spleen studies with Tc99m sulfur colloid and Tc-99m microalbumin colloid. J. Nucl. Med. 24,8- 13. LOUGH, J., ROSENTHALL, L., ARZOUMANIAN, A., AND GORESKY, C. A. (1987). Kupffer cell depletion associated with capillarization of liver sinusoids in carbon tetrachloride-induced rat liver cirrhosis. J. Hepazol. 5, 190-198. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDALL, R. J. (195 1). Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193,265-275. OMURA, T., AND SATO, R. (1964). The carbon monoxide-binding pigment of liver microsomes. J. Biol. Chem. 239,2370-2378. OPPENHEIM, B. E., GOTTSCHALK, A., AND HOFFER, P. B. (1976). Gastrointestinal nuclear medicine. In Diagnostic Nuclear Medicine (A. Gottschalk, and E. J. Potchen, Eds.), Chap. 33, pp. 399-455. Williams & Wilkins, Baltimore. PEREZ-TAMAYO, R. (1979). Cirrhosis of the liver: A reversible disease? In Pathology Annual (S. C. Sommers, and P. P. Rosen, Eds.), Vol. 14, Part 2. pp. 183-2 13. Appleton-Century-Crofts, London. PRAKASH, V.. LIN, M. S.. AND KRISS, J. P. (1977). Liver scintigraphy in alcoholic liver disease. Clin. Nucl. Med. 2,308-309.
PRASSE, K.. HILDEBRANDT, P., AND DODD, D. (1986). Letter to the Editor. Vet. Pathol. 23,540-54 1.
ET AL.
RUBIN, E., HUTTERER, F., AND POPPER,H. (1963). Cell proliferation and fiber formation in chronic carbon tetrachloride intoxication. Amer. J. Pathol. 42, 7 15728.
SAS Institute, Inc. (1985). SAS Users Guide: Statistics, 5 Edition. SAS Institute, Gary, NC. SELLERS, E. A., LUCAS, C. C.. AND BEST, C. H. (1948). The lipotropic factors in experimental cirrhosis. Brit. Med. J. 1,1061-1065. SIMMONS, J. E., DEMARINI, D. M., AND BERMAN, E. (1988). Lethality and hepatotoxicity of complex waste mixtures. Environ. Res. 46,74-85. Society of Toxicologic Pathologists (1986). Society of Toxicologic Pathologists’ position paper on blinded slide reading. Toxicol. Pathol. 14,493-494. SOKAL, R. R., AND ROHLF, F. J. (1981). Biometry: The Principles and Practices of Statistics in Biological Research, 2nd ed., pp. 179-453. Freeman, San Francisco. SOKOLOFF, L. (I 98 1). Localization of functional activity in the nervous system by measurement of glucose utilization with radioactive deoxyglucose. J. Cereb. Blood FlowMetab. 1,7-36. WELSH, M. J. (1982). New radiopharmaceuticals for studying liver function. J. Nucl. Med. 23, 1138-l 139. WIDMAN, F. K. (1979). Clinical Interpretation of Laboratory Tests. 8th ed., pp. 350-386, Davis, Philadelphia. YONEKURA, Y., BENUA, R. S., BRILL, A. B., SOM, P.. YEH, S. D. J., KEMENY, N. E., FOWLER, J. S., MACGREGOR, R. R., STAMM, R., CHRISTMAN, D. R., AND WOLF, A. P. (1982). Increased accumulation of 2deoxy-2-[‘*F]fluoro-D-glucose in liver metastases from colon carcinoma. J. Nucl. Med. 23, 1133-I 137.