Hydroquinone

Specific investigations: other studies

Description of key information

Nephrotoxic effects of HQ were found after acute and subchronic oral exposure of F344 rats with LOAELs of 200 and 25 mg/kg bw/d, respectively. The nephrotoxic action is dependent on the formation of metabolites that require processing by GGT, that are glutathionyl conjugates of HQ. 2,3,5-(tris-glutathion-S-yl)HQ displayed the highest nephrotoxic potency (about 600-fold that of HQ). Renal lesions are characterized by tubular cell necrosis followed by cell regeneration. Thus, HQ-induced renal cell proliferation, presumably is an attempt to compensate for proximal tubular cell loss rather than representing a direct mitogenic effect of HQ. This is further supported by the absence of DNA damage in the kidney from transgenic mice treated by oral gavage for 28 days in an in vivo TGR assay (Matsumoto et al., 2014), as well as in an in vivo Comet assay in F344 rats administered oral doses up to the maximum tolerated dose, despite initial indication of renal tubular necrosis after 2 administrations (Beevers, 2016).
In summary, based on a weight of evidence approach of the strain- and species-specific susceptibility of male F344 rats to the nephrotoxic action of HQ after acute, subacute and subchronic oral application, the significance of the reported kidney tumour response in F344 rats (NTP, 1989) for evaluating human risk is unclear. Mechanistically, nephrotoxicity and hyperplasia in the kidney induced by HQ administration could predispose the male F344 rats to kidney tumour formation by an interaction of regenerative cell proliferation associated with spontaneous chronic progressive nephropathy. This mode of tumour formation is not expected to operate in the absence of the renal disease process or at subtoxic exposure levels (Boatman et al., 1992, 1993, 1996).
Subacute to subchronic dermal exposure to HQ may lead to reduced pigmentation on both animal and human skin at concentrations of ca. 1% and higher.

Additional information

STUDIES ON NEPHROTOXIC EFFECTS

Single application studies

For examination of sex-, strain- and species-differences of acute nephrotoxic effects, HQ was administered to both male and female F344 rats, Sprague-Dawley rats and B6C3F1 mice at doses of 0, 200 and 400 mg/kg in rats and 0 and 350 mg/kg in mice (near lethal dose) by single gavage application. Blood and urine were collected for up to 96 h for analysis of blood urea nitrogen (BUN) and urinary glucose, creatinine, osmolality, specific gravity, volume, activity assays of enzymes (alanine aminopeptidase (AAP), N-acetyl-ß-D-glucosaminidase (NAG), alkaline phosphatase (ALP), gamma-glutamyl transpeptidase (GGT)), and microscopic examination of urine. A quantitative evaluation of tubular changes was made by microscopic examination of kidney sections.

Both male and female F344 rats, at 400 and 200 mg/kg bw, respectively, displayed enzymuria and glucosuria, and increased levels of BUN with maximum values up to 48 h after administration. Significant increases of epithelial cells in urine corresponded with similar increases of urinary enzymes of proximal tubular cell origin as well with the severity of histopathological findings in renal tubuli, suggesting an increased rate of tubular cell necrosis in this strain of rats and the existence of susceptible and resistant subpopulations, especially within the male dose groups. There was no indication of alpha-2µ-globulin nephropathy. Female F344 rats were significantly more sensitive to the acute nephrotoxic effects of HQ than male F344 rats. This relationship is reversed to findings after repeated administration of HQ for longer exposure periods or at lower dose levels. Sex differences in enzyme induction may explain the diametric findings obtained after single versus repeated exposure to HQ. Also, spontaneous nephropathy, which progresses more rapidly in male rats than in female rats, may be a factor of greater susceptibility of males to HQ after chronic dosing. In contrast, the examined endpoints of nephrotoxicity were generally unaffected in Sprague-Dawley rats (same dose groups), demonstrating a strain-specific effect. B6C3F1 mice were distinctly less susceptible than F344 rats as only minor nephrotoxic effects occurred following a high dose of 350 mg/kg, just under a lethal dose (Boatman et al., 1992, 1993, 1996).

Male Fischer 344 rats were exposed to a single dose of 0, 1.8 and 4.5 mmol HQ/kg bw corresponding to ca. 0, 200, and 500 mg/kg bw by gavage (vehicle water) and indicators of nephrotoxicity were examined up to 72 h post application. These included urinary parameters (glucose, activities of glutathion-S-transferase (GST), alkaline phosphatase (ALP), gamma-glutamyl transpeptidase (GGT), lactate dehydrogenase (LDH)), blood urea nitrogen (BUN), renal cell proliferation by BrdU incorporation assay and histopathological changes. To investigate the possible role of biotransformation on the nephrotoxic effects of HQ, additional groups were pretreated with acivicin, an inhibitor of gamma-glutamyl transpeptidase (GGT), 1 hour prior to treatment with 4.5 mmol HQ/kg, or were exposed to a metabolite of HQ, the glutathione conjugate 2,3,5-(tris-glutathion-S-yl)HQ (TGSH), at a dose of 7.5 µmol/kg bw in PBS by intravenous injection.

During 10 to 20 min after dosing, general toxicity was indicated by tremors. At the 200 mg/kg dose level, there was no significant change. Effects were only observed in single animals, showing both renal tubular necrosis and elevated urinary enzymes (GST, GGT and ALP) and labeling indices, but not in others. At 400 mg/kg, significant increases of GST, GGT, ALP, LDH and glucose compared to vehicle control were observed together with a significant correlation between labeling index and urinary GGT excretion and microscopic renal lesions.

HQ was found to be acutely toxic to the kidney of Fischer344 rats in a manner requiring the activity of gamma-glutamyl transpeptidase (GGT) (no nephrotoxic effect after pre-treatment with GGT inhibitor acivicin). This suggests that nephrotoxicity of HQ is dependent on the formation of metabolites that require processing by GGT. Consistent with this interpretation, 2,3,5-(tris-glutathion-S-yl)hydroquinone (TGSH), an in vivo metabolite of HQ, causes a similar pattern of renal toxicity. The severity of nephrotoxicity following i.v. administration of 7.5 µmol TGSH/kg is comparable to oral application of 4.5 mmol HQ/kg by gavage. Consequently, TSGH is about 600-fold more potent than HQ, indicating that only a small fraction of HQ requires metabolism to TSGH in order to produce renal toxicity. Renal lesions are characterized by tubular cell necrosis followed by cell regeneration without any evidence of interstitial infiltration or inflammation. Both the cell proliferation assay (BrdU incorporation) and the microscopic findings show that cell regeneration is confined to areas of cellular toxicity in the renal tubules, presumably as an attempt to compensate for proximal tubular cell loss. The absence of cell proliferation distal to the site of toxicity suggests that cell proliferation is a response to cytolethality and not a mitotic stimulus per se. The site-selectivity of nephrotoxicity at the outer stripe of the outer medulla (OSOM, corresponding to S₃M segment of the nephron or P3 region) may be a consequence of the susceptibility of this area to oxidative stress, and to the high concentrations of GGT in the brush border membrane of proximal tubular cells (Peters et al., 1997).

Investigations to identify possible critical metabolites for the acute nephrotoxic action of HQ were performed with acute intravenous administration of glutathione conjugates of different degree of substitution (mono-, di-, tri- and tetra-conjugates) (Lau et al., 1988, 1995).

Nephrotoxic and hepatotoxic effects of glutathione conjugates of HQ (2-GSylHQ, 2,3-diGSylHQ, 2,5-diGSylHQ, 2,6-diGSylHQ, 2,3,5-TriGSylHQ or 2,3,5,6-TetraGSylHQ, 5 to 250 µmol/kg bw i.v.) were examined in male Sprague-Dawley rats by measuring BUN and SGPT, and by macroscopic and microscopic examination of kidneys and livers. Blood and tissues were sampled 24 h after application. Additionally, using specific inhibitors, roles of renal transport and of renal metabolism of the conjugates by gamma-glutamyltranspeptidase (GGT) and cysteine conjugate ß-lyase was investigated.

There was no indication of hepatotoxic effects. With the exception of the fully substituted GSyl-HQ conjugate, nephrotoxicity correlated with the extent of glutathione substitution in the following order 2,3,5-TriGSylHQ > 2,6-DiGSylHQ > 2,3- or 2,5-diGSylHQ > 2-GSylHQ. Nephrotoxicity was found to be dependent on renal metabolism of GSyl-HQ conjugates by GGT. For the most effective conjugate 2,3,5-TriGSylHQ, BUN showed a steep dose response, with a significant increase from 20 µmol adduct/kg. Pathological changes were dose-related and specifically localized and consisted of extensive coagulative necrosis of the proximal tubular cells in the cortico-medullary junction. Severe necrosis was localized in the S3 segments of the proximal tubules, which contained eosinophilic cells with pyknotic nuclei(Lau et al., 1988).

For the most nephrotoxic metabolite, 2,3,5(triGSyl)HQ, a species and strain comparison of the potency was performed in F344 rats (20 µmol/kg intravenous application), B6C3F1, Balb/C and C57B1 mice (200 µmol/kg bw intravenous application), Golden Syrian hamsters and Albino guinea pigs (200 µmol/kg bw intracardiac application). F344 rats were found to be highly sensitive to the nephrotoxic action (excretion of urinary enzymes and BUN). Susceptibility was also indicated in guinea pigs by an increase of BUN, whereas no indications of a nephrotoxic effect were observed in different strains of mice or hamsters. The species differences in the metabolism and nephrotoxicity of 2,3,5(triGSyl)HQ were consistent with reported differences in HQ-mediated nephrotoxicity and neoplastic effects in kidneys (Lau et al., 1995).

Repeated dosing studies

Nephrotoxic effects and renal cell proliferation were compared in male and female F344 rats and male Sprague-Dawley rats after oral exposure to HQ by gavage (vehicle water). F344 rats were dosed with 0, 2.5, 25 and 50 mg/kg bw/d, and Sprague-Dawley rats with 0 or 50 mg/kg bw/d for 1, 3 or 6 wk, on 5 d/wk. Kidney weights and histopathology, and activities of urinary enzymes were investigated as indicators of nephrotoxicity. Further, renal cell proliferation in different proximal tubule segments (P1, P2, P3 and distal cells) was investigated via a BrdU incorporation assay.

HQ induced cell proliferation in renal proximal tubules occurred in male F344 rats in P1 and P2 segments but not in F344 female rats or male Sprague-Dawley rats, which correlates with the sex- and strain-specific findings of renal toxicity and benign kidney tumors observed in other studies. The delay in cell proliferation, occurring between 3 and 6 weeks, suggests that this is in response to nephrotoxicity, rather than representing a direct mitogenic effect of HQ. A similar time course of induction of nephrotoxic effects in male F344 rats was suggested by a modest significant elevation of urinary enzymes (ALP, AAP, GGT, NAG, and glucose), indicative of proximal tubular damage, and by histopathological changes as increased incidences of degenerative and regenerative renal tubule foci and interstitial inflammation. Presumably, the cell proliferation seen in male F344 rats reflects a regenerative response to replace damaged and dead proximal tubular epithelial cells after relatively short-term repeated exposure (about 6 wk) at doses of 50 mg HQ/kg and possibly 25 mg HQ/kg (English et al., 1994).

In a study with examination of similar endpoints after subchronic dermal exposure (5 d/w, for 6 h/d during 13 w)ofmale and female F344 rats to doses of up to 74 and 110 mg/kg bw, respectively,no signs of treatment-related nephrotoxicity or renal cell proliferation were observed, which is consistent with the known reduced dermal absorption of HQ (David et al., 1998; for details see Section 7.5.2).

In a further study, with gavage administration of25 or 50 mg HQ/kg bw/d (vehicle water) for 6 weeks on 5 d/w to male and female F344 rats and male Sprague-Dawley rats there was no indication of any nephrotoxic effect based on the measured urinary markers (AAP, GGT, NAG). However, no further endpoints of nephrotoxicity were investigated (Boatman et al., 2000).

In vitro studies

Freshly isolated peripheral tubular cells from male Fischer 344 or Sprague-Dawley rats were incubated in vitro either with HQ or metabolites (the mono- or thriglutathione conjugate of HQ, or the cysteine or N-acetylcysteine conjugate) at concentrations of 0.05-1.0 mM for 0.5 to 4 hrs either in the presence of low- or high-oxygen atmosphere (5% oyxgen / 5% carbon dioxide /90% nitrogen or 95% oxygen / 5% carbon dioxide, respectively).

Under conditions of low oxygen, which are representative of physiologically relevant conditions, both 1.0 mM HQ and 0.5 – 1.0 mM Cys-HQ displayed cytotoxicity simultaneous to inhibition of mitochondrial function. Among the other HQ metabolites studied, only the monoGSH-conjugate was active, possibly due to production of Cys-HQ in the cell media. No significant differences between strains of rats were observed at low oxygen tension.In the presence of 95% oxygen, the greatest levels of cytotoxicity were observed. Significant decreases of cell viability and ATP/ADP ratio were observed at 0.15 mM HQ and higher. Male F344 rats were found to be more sensitive than Sprague-Dawley rats under the high oxygen atmospheres, thus correlating with in vivo results (results of Boatman et al., 1996). In addition, under this condition Cys-HQ and NAcCys-HQ produced slight but detectable amounts of cytotoxicity at a concentration of 0.15 mM.Presumably, oxidative stress may be the primary determinant of cytotoxicity at elevated oxygen levels.The strain difference in cytotoxicity was only observed in the presence of high oxygen and the absence of BSA, conditions favouring oxidative stress. Biochemical characteristics of the isolated proximal tubular cells from untreated control rats show the SD strain to have a significantly higher capacity to respond to oxidative stress. Thus, the F344 rats kidney may possess an inferior capacity for coping with toxic insult from HQ. No strain differences in sensitivity under physiological conditions of low oxygen tension were detected, which supports a role for extra renal factors in the known in vivo strain differences (Boatman et al., 1999, 2001, 2004).

Conclusions

HQ was found to be acutely toxic to the kidney of Fischer344 rats in a manner requiring the activity of gamma-glutamyl transpeptidase (GGT), suggesting that the nephrotoxic action is dependent on the formation of metabolites that require processing by GGT. Consistent with this interpretation, 2,3,5-(tris-glutathion-S-yl)hydroquinone (TGSH), an in vivo metabolite of HQ, causes a similar pattern of renal toxicity and was found to be about 600-fold more potent than HQ itself. Consequently, only a small fraction of HQ requires metabolism to TSGH in order to produce renal toxicity (Peters et al., 1997). After intravenous application to male Sprague-Dawley rats, glutathione conjugates of HQ were found to induce nephrotoxic effects being dependent on metabolism by GGT and on the degree of substitution (2,3,5-TriGSylHQ > 2,6-DiGSylHQ > 2,3- or 2,5-diGSylHQ > 2-GSylHQ) (Lau et al., 1988).The species differences in the metabolism and nephrotoxicity of 2,3,5(triGSyl)HQ were consistent with reported differences in HQ-mediated nephrotoxicity and neoplastic effects in kidneys (Lau et al., 1995). An in vitro study showed that oxidative stress may contribute to the cytotoxic action of HQ and its metabolites on kidney cells but is not the determining factor under physiological conditions ((Boatman et al., 1999, 2001, 2004).

Acute renal lesions induced by the metabolite 2,3,5-TriGSylHQ are characterized by tubular cell necrosis followed by cell regeneration without any evidence of interstitial infiltration or inflammation. Both the cell proliferation assay (BrdU incorporation) and the microscopic findings show that cell regeneration is confined to areas of cellular toxicity in the renal tubules, presumably as an attempt to compensate for proximal tubular cell loss. The absence of cell proliferation distal to the site of toxicity suggests that cell proliferation is a response to acute cytolethality and not a mitotic stimulus per se. The site-selectivity of nephrotoxicity at the outer stripe of the outer medulla (OSOM, corresponding to S₃M segment of the nephron or P3 region) may be a consequence of the susceptibility of this area to oxidative stress, and to the high concentrations of GGT in the brush border membrane of proximal tubular cells (Peters et al., 1997). Principally, this was also the site of renal toxicity and renal tubular adenomas in the 13-week gavage study and in the 2 year bioassay with Fischer 344 rats (NTP, 1989; Kari, personal communication, cited by Peters et al., 1997).

HQ-induced cell proliferation in renal proximal tubules in P1 and P2 segments of male F344 rats correlated with the sex- and strain-specific findings of renal toxicity and benign kidney tumors observed in the NTP study. The increase in the P3 segment was not significant. The delay in cell proliferation, occurring between 3 and 6 weeks, suggested that this is in response to simultaneously appearing nephrotoxic effects, rather than representing a direct mitogenic effect of HQ. Presumably, the cell proliferation seen in male F344 rats reflects a regenerative response to replace damaged and dead proximal tubular epithelial cells after relatively short-term repeated exposure (about 6 wk at doses of 25-50 mg HQ/kg bw/d by gavage) (English et al., 1994).

In summary, based on a weight of evidence approach of the strain- and species-specific susceptibility of male F344 rats to the nephrotoxic action of HQ after acute, subacute and subchronic oral application, the significance of the reported kidney tumour response in F344 rats (NTP, 1989) for evaluating human risk is unclear. Mechanistically, nephrotoxicity and hyperplasia in the kidney induced by HQ administration could predispose the male F344 rats to kidney tumour formation by an interaction of regenerative cell proliferation associated with spontaneous chronic progressive nephropathy. This mode of tumour formation is not expected to operate in the absence of the renal disease process or at subtoxic exposure levels (Boatman et al., 1992, 1993, 1996).

 

 

 

STUDIES ON SKIN DEPIGMENTATION

Subchronic exposure to HQ may lead to reduced pigmentation on both animal and human skin.

In black guinea pigs skin pigmentation was reduced following daily application of 1 to 10% HQ in an ointment formulation for up to 1 month (Bleehen et al., 1968; Jimbow et al., 1974; cited both from DeCaprio, 1999). Hydroquinone (1 or 5% in a hydrophilic ointment) was found to induce depigmentation in the skin of black guinea pigs after 63 weeks of topical treatment (5 d/w) (Maibach, 1989).

Early controlled clinical studies suggested that dermal applications of creams containing at least 1.5% HQ were effective in producing at least mild cosmetic depigmentation in human skin after approximately one month of use (cited from DeCaprio, 1999).

Studies on the mechanism of the depigmentating action of HQ showed that HQ application caused a reduction in melanin content of melanosomes and a reduction in the formation of melanosomes, a marked alteration in the internal structure of melanosomes, and eventually produced a degeneration of melanocytes and a destruction of the membranous organelles of the melanosomes (Jimbow et al., 1974). These findings are consistent with the hypothesis that HQ is an inhibitor of tyrosinase to L-dopa and then to dopaquinone, the initial steps in melanin formation. Investigations on the participation of tyrosinase inhibition in cytotoxicity yielded conflicting results. There are indications that inhibition of GSH synthesis results in potentiation of the depigmenting action of HQ on black guinea pig skin, presumably by increasing the level of oxidative species that would normally be trapped by the intracellular thiol (cited from DeCaprio, 1999).