Detection of endogenous malondialdehyde-deoxyguanosine adducts in human liver.

Endogenous DNA adducts may contribute to the etiology of human genetic disease and cancer. One potential source of endogenous DNA adducts is lipid peroxidation, which generates mutagenic carbonyl compounds such as malondialdehyde. A sensitive mass spectrometric method permitted detection and quantitation of the major malondialdehyde-DNA adduct, a pyrimidopurinone derived from deoxyguanosine. DNA from disease-free human liver was found to contain 5400 adducts per cell, a frequency comparable to that of adducts formed by exogenous carcinogens.

Identification of protein-bound riboflavin in rat hepatocyte plasma membrane as a source of autofluorescence.

The presence of flavin compound(s) giving a yellowish-green autofluorescence in rat hepatocyte plasma membrane has recently been reported (Nokubo, M. et al. (1988) Biochim. Biophys. Acta 939, 441-448). The fluorophore can quantitatively be extracted with water at 80 degrees C from isolated plasma membranes. Gel filtration of the extract eluted with water showed two peaks, the fluorescence of which closely resembled that of riboflavin. The major peak comigrated with proteins and the minor one displayed a position identical to authentic riboflavin. When the components of the major peak were rechromatographed after acetic acid treatment and eluted with 20 mM of acetic acid, the fluorescent compound separated from the proteins and eluted at the same position as riboflavin. In paper chromatography and HPLC, the behavior of the fluorescent compound (separated by acid treatment from the proteins) was identical to that of riboflavin. SDS gel filtration of subcellular fractions of rat liver revealed that riboflavin was the dominant flavin, whereas FAD and FMN were not detectable in the plasma membrane. Microsomes and mitochondria contain predominantly FAD and FMN, and only minor quantities of riboflavin. The presence of riboflavin in the plasma membrane is a novel finding, the functional significance of which is still unclear; however, a hypothesis can be forwarded on the basis of the ability of flavins to generate superoxide anion radicals during their autoxidation.

Characterization of the autofluorescence of rat liver plasma membranes.

The autofluorescence of isolated rat liver cell plasma membranes was characterized in vitro in relation to the autofluorescence used previously for fluorescence recovery after photobleaching (FRAP) studies. The fluorescence of membrane preparations displayed an emission pattern with a maximum at around 525 nm when excited with a 468 nm blue light. The excitation spectrum monitored at 525 nm closely resembled that of flavin compounds (riboflavin, FAD, FMN). The chloroform extract of the membrane fraction showed practically no fluorescence, whereas, both the water-soluble and water-insoluble protein fractions remaining after chloroform extraction were strongly fluorescent. The fluorescence disappeared almost completely under the effect of sodium hydrosulfite, and recovered after oxidation either by shaking in air or by adding buffered hydrogen peroxide solution. The fluorescence of the acid extract of the plasma membranes photolyzed in an alkaline medium was quite similar to that of lumiflavin obtained from the photolysis of riboflavin in an alkaline medium. The plasma membranes prepared from isolated hepatocytes (which were completely devoid of endothelial cell contamination) exhibited the same autofluorescence in the liver cell plasma membranes. The results suggest that the autofluorescence of the liver cell plasma membranes is most likely of a character similar to that of flavin, bound to hepatocyte plasma membrane proteins. This fluorescence is suitable for measuring the average lateral diffusion constant of proteins by means of FRAP methods.

Age-related alterations of enzyme activities and subunits of hepatic glutathione S-transferases in male and female Fischer-344 rats.

Enzyme activities of glutathione S-transferases (GSTs) toward five different substrates (benzalacetone (PBO), styrene oxide (STOX), sulfobromophthalein (BSP), 1,2-dichloro-4-nitrobenzene (DCNB) and 1-chloro-2,4-dinitrobenzene (CDNB)) as well as concentrations of four subunits of GST isozymes (1, 2, 3 and 4) were determined using cytosol fractions obtained from livers of young (6 months) and old (26 months) Fischer-344 rats of both sexes. Values for enzyme activities for three substrates (DCNB, BSP and PBO) in young male rats were significantly higher than the corresponding values in female rats. In old male rats, values were generally lower than the corresponding values in young male rats, becoming close to corresponding values in young female rats. Old female rats, however, exhibited values close to those in young female rats, except for DCNB and STOX values, which were slightly lower in old female rats. GST subunits 3 and 4, as determined by high-performance liquid chromatography after purification by affinity chromatography using S-hexyl-glutathione, were predominant in young males, whereas concentrations of subunits 1 and 2 were higher in females than in males. In male rat livers, concentrations of subunits 3 and 4 decreased considerably with age while those of subunits 1 and 2 increased, so that the subunit pattern in old male rats tended to be similar to that of young female rats. In old females, a decrease in the concentration of subunits 3 and 4 and an increase in the concentration of subunit 1 were also observed as in old male rats, while the subunit 2 concentration tended to decline. Furthermore, the elution pattern of affinity chromatography changed with age, yielding an earlier elution of most subunits in old male rats and of subunit 1 in old female rats. The results suggest that age-related changes that occur with GSTs in livers of male rats are essentially a feminization of the isozyme pattern. However, despite rather unremarkable changes in enzyme activities with age in females, considerable changes of subunit pattern (a general decrease in concentration of subunits 2, 3 and 4 and an increase in the concentration of subunit 1) were also observed in female rats, and these were much greater than could be predicted from enzyme activity changes with age in this sex.

Difference in response of hepatic glutathione S-transferase activities to protein-free diet between young and old C57/BL male mice.

Responses of hepatic glutathione S-transferase (GST) activities to protein-free diet (PFD) and normal diet (ND) refeeding were compared for young (6-month-old) and old (22-month-old) C57/BL male mice. Enzyme activities toward 1-chloro-2,4-dinitrobenzene (CDNB) were not significantly different between young and old rat livers in the basal condition without diet manipulation. When animals were fed PFD for 1 week, GST activities toward CDNB significantly declined in both age groups in comparison to respective basal values, but there was no significant difference in activities between the two age groups after a 7-day PFD. When they were refed with ND for 2 days (on day 2 of ND), the activities in young mice rose to a level significantly higher than the corresponding basal value. In contrast, in old animal livers, the activity slightly but further tended to decline on day 2 of ND. Activities in old rat livers returned to the basal level on day 5 of ND, while activities in young animal livers that increased to levels higher than basal levels due to the overshoot returned to the basal level on day 7 of ND. Enzyme activities toward 1,2-dichloro-4-nitrobenzene (DCNB) were significantly higher in young rat livers than in old ones at the basal period. However, enzyme activities also overshot the basal level on day 2 of ND after 7-day PFD in young mouse livers, while in old mouse livers the activities were lowest on this day. Activities returned to the basal level on day 7 of ND in both age groups. Thus, the greatest difference in enzyme activities between young and old mouse livers for both substrates was observed on day 2 of ND after 7-day PFD, rather than at either the basal period or immediately after 7-day PFD. The results essentially agree with our previous findings on female C57/BL mice as well as female Fischer-344 rats, suggesting that the age-induced changes in the GST system become clearly manifest after diet manipulation of PFD followed by ND refeeding, rather than in values during a basal period without diet manipulation, regardless of sex or species of animal.

Effect of protein-free diet on activities and subunits of glutathione S-transferase in livers of young and aged female rats.

Female Fischer-344 rats of different ages (8 and 25 months old) were fed a protein-free diet (PFD) for 7 days and refed a normal diet (ND) (23% protein) thereafter. Rats were killed immediately after the PFD was stopped (day 0) and at different time intervals during refeeding of a ND. Four subunits (1,2,3 and 4) and activities of glutathione S-transferases (GSTs) toward five different substrates, [styrene oxide (STOX), 1,2-dichloro-4-nitrobenzene (DCNB), 1-chloro-2,4-dinitro benzene (CDNB), sulfobromophthalein (BSP) and benzalacetone (PBO)] were determined. There were no significant differences between young and old rats in the liver enzyme activities before the PFD. The PFD caused significant decreases in activities for three substrates (DCNB, BSP and STOX) in both age groups, with no significant differences between young and old rats a day 0. During recovery from the PFD, activities for the three substrates exceeded basal levels in young rats but at different time intervals (STOX, day 2; BSP, day 5; DCNB, day 9), while enzyme values in old rats tended to return slowly to basal values with no "overshoot." Concentrations of subunits 3 and 4 in young rat livers that were depressed by the PFD did not recover until day 9 of the ND, while subunits 2 and especially 1 increased during the ND refeeding, overshooting the basal levels. In contrast, in old rat livers the only change was a reduction of subunit 1 by the PFD and its gradual recovery during ND refeeding. These results demonstrate that our previous observation of overshooting of enzyme activities in mice is reproducible in rats but with certain substrate specificities. Furthermore, changes in subunit concentrations caused by aging and a PFD are more complex than what was predicted from changes in enzyme activities of GSTs.

Differences in the influence of diet on hepatic glutathione S-transferase activity and glutathione content between young and old C57 black female mice.

The effect of a protein-free diet (PFD) on hepatic activity of glutathione S-transferase (GST) and hepatic content of total glutathione (GSH) was examined in young (9-month-old), middle-aged (17-month-old) and old (27-month-old) C57BL/6CrS1c female mice. There were no significant differences in the control values of GSH or of enzyme activity for four of five substrates among young, middle-aged, and old animals fed normal diet (ND) only. Both GSH and GST activity were significantly decreased by the 7-day PFD in both young and old groups but the decrement was generally greater in old mice. After a 2-3-day refeeding of ND, young mouse enzyme activities were significantly higher than control (basal) values for all five substrates, whereas old mouse values were still significantly lower than corresponding control values. There was no overshooting of GSH levels after refeeding of ND in either young or old animals. This study indicates that an age difference in this detoxification system can be clearly demonstrated in the hepatic response to PFD feeding and especially to ND refeeding, despite the enzymes' stable basal activities with aging.

Biliary excretion of digitoxin and its metabolites in young and old male Wistar rats.

The biliary excretion of digitoxin (Dt3) and its metabolites were compared between young (3-month-old) and old (25-month-old) male Wistar rats after an iv injection of [3H]Dt3 (0.03 mg/100g body weight) for 2 hrs. The 2-hr. total biliary recovery of iv injected radioactivity (percent of the dose) was two times lower in old rats (7.40 +/- 1.36% mean +/- SD) compared with young rats (14.74 +/- 4.10%). This difference was primarily due to the decrease in the excretion of Dt3 metabolites in the bile, while the excretion of the parent drug, Dt3 was 1.3 times higher in old rats. Among various Dt3 metabolites in the bile, digitoxigenin bis-digitoxoside (Dt2), digoxigen bis-digitoxoside (Dg2), and polar (conjugated) metabolites were major components, which all decreased with age. In accord with the decreased excretion of the radioactivity in the bile of old rats, the plasma disappearance of radioactivity was generally slower in old animals compared with young ones, yielding significantly higher plasma levels at different times of observation. Despite the increase in plasma radioactivity, the radioactivity concentration in the liver 2 hrs. after the injection was almost equal between the two age groups. It is suggested that at least in this rat strain and sex the biliary excretion of Dt3 metabolites was markedly age-dependent, presumably due to the decreased capacity of the liver to biotransform Dt3 with age. Furthermore, the lower liver plasma radioactivity ratio in old animals suggested the possibility that the distribution of Dt3 in the liver may also decrease with age.

Age-dependent decrease in the lethal threshold of pentylenetetrazole in mice.

We previously reported that the minimal effective concentrations in plasma and brain for inducing the maximal seizure after intraperitoneal injection of pentylenetetrazole (PTZ) significantly increased with age in mice. We also observed that some old mice died without the maximal seizure during the observation period of 15 min. To examine how the lethal dose changes with age in the mouse, in the present study we employed a continuous intravenous infusion of PTZ that enabled us to control the dose, either stopping the infusion after the maximal seizure or continuing it until the lethal dose was reached. This study has demonstrated that the lethal threshold of PTZ in the oldest mice (27 months for males and 30 months for females) was significantly lower than that for the respective younger groups. We conclude that our observation in old mice of death occurring before the appearance of maximal seizure is due to a decrease in the lethal threshold combined with an increase in the threshold for maximal seizure.

Ursodeoxycholic acid reduces the systemic toxicity of 1,2-dichloro,4-nitrobenzene by stimulating hepatic glutathione S-transferase in mice.

Male C57BL/6 mice were fed with normal diet (ND) or diets containing 0.3 or 0.5% ursodeoxycholic acid (UDCA) for 3 weeks. Glutathione S-transferase (GST) activities in the liver cytosolic fraction of these animals toward 1,2-dichloro,4-nitrobenzene (DCNB) as well as to 1-chloro,2,4-dinitrobenzene (CDNB) were significantly increased in a dose dependent manner in UDCA-treated groups compared with the control (ND-fed) animal group (one-way ANOVA). Reduced glutathione (GSH) levels tended to slightly decrease with UDCA diets but the difference did not attain a statistical significance (P > 0.05, one-way ANOVA). Twenty four hr survival rates after an oral challenge of 3.5 mg/kg of DCNB were significantly higher (P < 0.05, Chi-square test) in the two UDCA fed groups (10/10 for 0.5% group, 8/11 for 0.3% group) compared with the control group (3/11). Thus, UDCA appears to reduce the systemic toxicity of DCNB which is detoxified by the hepatic GST system. Although UDCA has been shown to exert hepatoprotective effects in experimental animals and humans in the past, to the best of our knowledge, the present study is the first report that UDCA reduces the systemic toxicity of a toxicant which is detoxified by the hepatic GST system. Although a direct proof is not available, it is most likely that the reduction of the systemic toxicity of DCNB was achieved by the increase in GST activity caused by UDCA feeding. This finding may open a new research field with regard to the unique biological properties of this bile salt in modulating hepatic detoxifying enzymes.

(-) deprenyl induces activities of both superoxide dismutase and catalase but not of glutathione peroxidase in the striatum of young male rats.

Daily s.c. injection of (-)deprenyl (2.0 mg/kg/day) for three weeks in young male rats caused a threefold increase in superoxide dismutase (SOD) activity in the striatum of the brain compared with the value in saline-injected control rats. Furthermore, the activity of catalase (but not of glutathione peroxidase) was also increased significantly by deprenyl treatment. The results confirmed the previous findings of Knoll on SOD activity and furthermore provided evidence that the activity of catalase is also significantly induced by the drug, which was not found in the previous study.

Age related increased threshold for electroshock seizure in BDF1 mice.

The thresholds for inducing the minimal and maximal electroshock seizures were examined in relation to age in BDF1 mice of both sexes. The 50 percent effective intensities for the maximal electroshock seizure (tonic hindlimb extensor component) were lowest in the youngest age groups (6-month-old) for both male (10.68 mA) and female (9.18 mA) animals. The threshold increased with age and became significantly higher at 24 months (14.00 mA, 12.70 mA for male and female mice respectively). There was also a further increase in threshold at 30 months for both sexes. Similarly, the threshold for inducing the minimal seizure also increased with age but the differences in mean threshold levels between the youngest and oldest groups were much smaller in comparison to the maximal seizure. It was concluded that the threshold for inducing electroshock seizures significantly increases with age in mice of both sexes.

Increasing anticonvulsant effect of AD-810 (zonisamide) in aging BDF1 mice.

The anticonvulsant efficacy of a newly developed anticonvulsant, AD-810 (zonisamide, 3-sulfamoylmethyl-1,2-benzisoxazole) was examined in relation to mouse age in three different age groups of female BDF1 mice (7-, 25- and 29-month-old). The minimal effective concentration (MEC) of AD-810 in both plasma and brain for abolishing the electroshock-induced maximal seizure steadily decreased with age, the 25- and 29-month values being 50 and 30% of respective 7-month values. The observation in the present study was almost identical to previous observations by the authors on phenytoin, phenobarbital and oxazepam. The present results support our previous contention that the dose and plasma concentration of anticonvulsants can (and probably should) be reduced in the elderly regardless of the drug. Since the anticonvulsant mechanism of AD-810 has been reported to differ from those of previously examined drugs (phenobarbital and oxazepam), the results also suggest that the apparent increase in the pharmacological effect of these anticonvulsants may be due to old animals' lowered response capability for seizures rather than a specific age effect on the pharmacological reaction sites for individual anticonvulsants.

Increased anticonvulsant effect of phenobarbital with age in mice--a possible pharmacological index for brain aging.

We have recently reported that the anticonvulsant effect of phenytoin increases with age in mice (1). Since some of the mechanisms of anticonvulsant action of phenytoin and phenobarbital may be different, the present study sought to determine whether a similar increase with age in the anticonvulsant effect of phenobarbital could also be observed. The anticonvulsant effect of phenobarbital was examined in BDF1 female mice of different ages (6, 12, 24 and 30 months old) using the abolition of the tonic hindlimb extensor component of maximal electro-shock seizure as the index. The minimal effective concentration (MEC) values of phenobarbital in plasma and brain were significantly lower in aged (24 and 30 month old) mice compared with the respective values in the youngest animal group (6 month old). Series using nearly two-fold different intensities of electroshock (30 and 55 mA) showed almost identical MEC values in 24 month-old mice. It was concluded that the brain of aged mice is more sensitive to phenobarbital, as it is to phenytoin.

The optimal dosage of (-)deprenyl for increasing superoxide dismutase activities in several brain regions decreases with age in male Fischer 344 rats.

We previously reported that the optimal dosage of (-)deprenyl to increase superoxide dismutase (SOD) activities in striatum in rats differs 10 fold between young male and female rats (1). Furthermore, in female rats the optimal dosage increased with age (1). In the present study in order to clarify how the optimal dosage of this effect changes with age in male rats, we examined the effects of four different dosages of deprenyl on SOD enzyme activities in striatum and several other tissues in old (28-29-month-old) male Fischer 344 (F-344) rats. Continuous s.c. infusion of deprenyl for 3 wks increased activities of SOD and catalase (CAT) in striatum, substantia nigra and cortical regions but not in hippocampus, cerebellum or the liver. The dose of 0.5 mg/kg/day was found to be most effective, while higher (1.0, 2.0 mg/kg/day) or lower (0.1 mg/kg/day) dosages were less effective. This value of 0.5 mg/kg/day was 4 fold lower than the dosage of 2.0 mg/kg/day which was most effective in increasing SOD and CAT activities in young (5-7 month old) male rats of the same strain (1,2). The decline of the optimal dosage with age found in male rats is best explained by a possible decline with age in the hepatic microsomal monooxygenase enzyme activities that are involved with the metabolism of deprenyl. In view of the large differences in the optimal dosages shown among different sexes and ages of rats, future studies regarding the unique effect of this drug in prolonging the life span of rats must be carefully investigated with the caution in mind that the optimal dosage for the life prolonging effect may well differ depending on sex, age and possibly strain and species of animal model used.

The effect of age on the adaptation of the brain to the anticonvulsant effect of phenobarbital in mice.

The anticonvulsant effect of phenobarbital was examined in young (6 month old) and old (24 month old) BDF1 female mice consisting of three groups each (one control and two chronically dosed phenobarbital groups), using the abolition of the tonic hindlimb extensor component of maximal electroshock seizure as the index. The minimal effective concentrations (MEC) of phenobarbital in plasma and brain in old control mice that were given a vehicle (tragacanth) for one week were significantly lower in comparison to the respective values in young adult control mice with the same treatment, confirming our previous findings. In young mice chronically treated with phenobarbital for one week (20 mg/kg daily for two days followed by daily dose of 50 mg/kg for 5 days), the MECs in both plasma and brain were significantly higher compared with respective control values. The 3 week treatment also produced an increase in MEc comparable to the one-week treatment. The same one-week treatment with phenobarbital in old mice similarly caused significantly higher plasma and brain MEC values but 3-week-treatment values were not significantly different from corresponding control values. It is concluded that the development of brain adaptation to phenobarbital is almost equal for young and old mice, so that the reduction in MEC with age indicates the need for lowered dosages for the aged, even when the age effect on brain adaptation developed to chronic dosing is taken into consideration.

Age-dependent increase in the threshold for pentylenetetrazole induced maximal seizure in mice.

The thresholds for inducing the maximal seizure by pentylenetetrazole (PTZ) were compared for BDF1 mice of both sexes with varying ages after intraperitoneal administration of various doses of PTZ. The minimal effective PTZ concentrations (MECs) in the brain for inducing the maximal seizure were significantly higher in 24-month or older mice than in 6-month-old animals of both sexes. Some mice of 30 months did not demonstrate the maximal seizure but died within the 15-min observation period, a phenomenon never observed in mice of 24 months or younger. The relationship between plasma and brain concentrations of PTZ changed little during aging. It was concluded that the brain becomes less sensitive to PTZ with age in regard to its convulsant activity, as was previously shown for electroshock by the authors. This observation, coupled with our earlier observations on anticonvulsants, appears to support the classical hypothesis that age has a dual effect on drug sensitivity i.e. a decrease for stimulants but an increase for sedative (or depressant) drugs.

Effect of lecithin on the release of 5'-nucleotidase from liver plasma membrane of rat by bile acids.

Bile acids solubilize proteins from liver plasma membrane both in vivo and in vitro. The ability to solubilize the proteins is dependent on the species of bile acid. In this paper, the effect of phospholipid on the solubilization of a membrane-bound enzyme by bile acids was investigated in vitro. Taurocholate (TC) and tauroursodeoxycholate (TUDC) solubilized the enzyme, 5'-nucleotidase, from the liver plasma membrane of the rat in a concentration-dependent manner, although there was a great difference in their effect; at 40 mM, TC solubilized 55.4% of the original 5'-nucleotidase activity of the membrane, but TUDC only 5.7%. While lecithin alone had no solubilizing effect, its addition to the bile acids provoked a 10-fold increase in the solubilizing effect of TUDC, but virtually no change for TC, essentially equalizing the solubilizing effect of the two. Both TC-rich and TUDC-rich bile were obtained from rats infused with the respective bile acids via the jugular vein after their endogenous bile acid pool had been depleted. The solubilization effect of these biles was quite similar to the bile acid-lecithin mixtures. These findings demonstrate that lecithin enhances the ability of the bile acids to solubilize the membrane protein and eliminates the difference in the two bile acid species in their solubilizing ability.

Hepatic biotransformation and choleretic effect of 7-ketolithocholic acid in the rat.

Hepatic biotransformation and the effect on bile flow of 7-ketolithocholic acid (7-oxo-3 alpha-hydroxy-5 beta-cholan-24-oic acid), in comparison to ursodeoxycholic acid, were examined in rats under conditions of continuous infusion of solutions of sodium salts of these bile acids (1.2 mumol/min/100 g body wt) for 2 hr. Both bile salts elevated the bile flow rate as well as the bile bicarbonate concentration to a similar degree. The minor difference observed was a transient (10-20 min) and subtle drop of bile flow during the first hour in rats given 7-ketolithocholate. In ursodeoxycholate infused rats, the major bile salt in the bile was its taurine conjugate, although excretion of tauroursodeoxycholate dropped considerably during the second hour. In 7-ketolithocholate infused rats, the major bile salt in the bile was again its taurine conjugate, but ursodeoxycholate and chenodeoxycholate and their conjugates were also excreted. In contrast to ursodeoxycholate infused rats, the drop in excretion of taurine conjugates and the increase of glycine conjugates in rats infused with 7-ketolithocholate were more rapid. In rats infused with 7-ketolithocholate, excretion of ursodeoxycholate and its conjugates was significantly higher than the corresponding values for chenodeoxycholate, suggesting that 7-ketolithocholate is reduced predominantly to the 7 beta-epimer in this species. However, the concentration of ursodeoxycholate and its conjugates excreted into the bile in rats infused with 7-ketolithocholate was only 10% of that of rats infused with ursodeoxycholate, yet the magnitude of choleresis and the rise in bile bicarbonate concentration were similar in both experiments.(ABSTRACT TRUNCATED AT 250 WORDS)

Morphological, physiological and biochemical alterations in livers of rodents induced by protease inhibitors: a comparison with old livers.

A "Protease inhibitor model of aging" has been proposed primarily based on observations on brain tissues exposed to a thiol protease inhibitor, leupeptin (Ivy et al., 1984a). In order to validate this model in terms of a mechanism of cellular aging, as well as of lipofuscin formation in particular, attempts have been made to induce lipofuscin in hepatocytes in young rodent (rat and mouse) livers by continuous i.p. infusion of two different thiol protease inhibitors, leupeptin and E-64C. With doses of leupeptin higher than 1.0 mg/100g/day for 2 wks, a fine granular lipofuscin-like deposition with distinct yellowish-green fluorescence was induced in young rat hepatocytes. The deposition became greater in degree with increasing leupeptin doses. In Kupffer cells and other endothelial cells, fluorescent granules were also induced. In contrast to rat livers, lipofuscin-like pigments induced in hepatocytes in mice were much less, even with a higher dose (20 mg/100 g/day). E-64C also induced the accumulation of lipofuscin-like pigments at a dose of 5 mg/100 g/day, their characteristics being very similar to those induced by leupeptin, but the accumulation being smaller in degree. The fluorescence of leupeptin induced lipopigments was yellowish-green having a peak around 520 nm in emission profile, closely resembling that observed in old rat livers. The hepatobiliary transport functions such as biliary transport maximum (Tm) for sulfobromophthalain and the biliary recovery of iv injected ouabain which are known to decline with age tended to decline in young (6-wk-old) rats administered with leupeptin at a dose of 5 mg/100 g/day for 2 wks. On the other hand, dolichol concentration in leupeptin treated livers was not increased in comparison to control livers, whereas in old rat livers, the dolichol concentration was more than 2 times greater than in young livers. A clear-dose-dependent deposition of ceroid-lipofuscin induced in young rodent livers by protease inhibitors strongly suggests that the "Protease inhibitor model" is generally valid not only for the brain but for other tissues such as the liver, and for two different thiol protease inhibitors.