The development of an in vitro model for studying mechanisms of nephrotoxicity as an alternative for animal experiments
SUMMARYPresently in our society animal tests still form the main starting point for the assessment of the possible risks of chemicals with regard to human and animal health. For scientific. economic, and ethical reasons. attempts are undertaken continuously to develop cell models as alternatives to animal testing. However, the predictive value of in vitro test systems is often limited due to the unawareness about the mechanisms of toxicity and the complexity of organisms. As a consequence, a strategy increasingly used is one of studying mechanisms of toxicity evolving in rational cell models. Following this strategy, this thesis aims at contributing to the development of cell models as an alternative to animal testing.A usefull cell model for studying mechanisms of toxicity requires the presence of characteristics which may play a role in this toxicity in vivo. and the possibility to demonstrate their involvement. The organ of our choice was the kidney. Since very often the cortex. more in particular the proximal tubule. is affected by chemicals. cortical cells of different origin have been used: A renal cell line derived from a Hamshire pig (LLC-PK1) and primary cultures of rat renal cortical cells (Chapter 2). The applicability of these cells in a cell model was tested considering two major aspects of importance for nephrotoxicity: Firstly. the functional polarity of the proximal tubular cell, and secondly the presence of specific biotransformation processes and transport systems (Chapter 1).In vivo luminal (apical) as well as serosal (basolateral) exposure of the proximal tubular cell can occur. However. when monolayers are cultured in a conventional way, i.e. on a solid support, only the apical side is accessible for the compounds to be investigated. To overcome this problem a new system for nephrotoxicological in vitro studies has been developed. Cells were cultured on porous substrates, which enables exposure of monolayers on either side (Chapter 2).The nephrotoxic effects of haloalkenes and halobenzenes seem to be mediated via the formation of glutathione conjugates (Chapter 1). In part II of this thesis a glutathione conjugate of hexachlorobutadiene and its derivatives are used as model compounds to investigate the possibilities of the cell systems described in chapter 2. LLC-PK1monolayers possess the most important characteristics necessary for cytotoxicity of S-(1.2,3,4,4-pentachlorobutadienyl)glutathione (PCBD- GSH) to occur (Chapter 3). Via the use of acivicin, an irreversible inhibitor of γ-glutamyltranspeptidase (γGT). and aminooxyacetic acid. an inhibitor of β-lyase, the importance of these enzymes in the activation of PCBD-GSH was demonstrated. After metabolism by γGT and dipeptidase S-(1,2,3,4,4-pentachlorobutadienyl)-L-cysteine (PCBD-CYS) is formed extracellularly. This is taken up and converted by β-lyase into a reactive thiol resulting in toxicity. In vivo and in the LLC-PK1cell line γGT is predominantly present on the luminal and apical membrane respectively. In agreement with this. apical exposure of LLC-PK1monolayers to PCBD-GSH caused a stronger toxicity than basolateral exposure . In the nephrotoxicity of the corresponding mercapturate N-acetyl-S-(1,2,3,4,4-pentachlorobutadlenyl)-L-cysteine (PCBD-NAC), which is deacetylated intracellularly into PCBD-CYS, a serosally located probenecid sensitive organic anion transporter has been suggested to be of importance (Chapter 1). Apical exposure of LLC-PK1monolayers to PCBD-NAC caused hardly any toxicity; only at relatively high concentrations a toxic effect was detected. Surprisingly, basolateral exposure and consequently administration of the compound to the possible transporter, did not cause substantial toxicity either. Further investigations revealed that this is probably caused by the absence of an organic anion transporter in the LLC-PK1cell line: The model organic anion para-aminohippurate (PAH) was not transported by LLC-PK1cells. In contrast, LLC-PK1monolayers are in the possession of an organic cation transporter. A quinine and ouabalne inhibitable transcellular transport of the cation tetraethylammonium (TEA) was demonstrated (Chapter 3).The organic anion transporter was also postulated (Lash and Anders, 1986) to play a role in the cytotoxicity of cysteine conjugates . However, in spite of the demonstrated absence of an organic anion transporter , PCBD-CYS causes toxicity in the LLC-PK1monolayer. In contrast to PCBD-GSH, basolateral exposure to PCBD-CYS resulted in a stronger toxic effect than apical exposure (Chapter 4). A possible candidate for the transport of PCBD-CYS in LLC-PK1cells is an amino acid transporter. The inhibitory effect of D-tryptophan and histidine on the toxicity of PCBD-CYS strongly suggests the involvement of the amino acid transport System T (Chapter 4).Culturing primary cultures of rat renal cortical cells resulted in confluent monolayers (Chapter 2 and 5). However, investigating transcellular transport of organic ions in a similar way as with the LLC- PK1 monolayers was not possible, because a non-leaking monolayer on a porous support was not obtained. This may be due to the diversity of the cell population in the culture, and the concomitant differences in junctional complexes necessary for complete sealing of the monolayer. The presence of both PAH and TEA transporters was demonstrated by measuring intracellular accumulation, which could be inhibited by probenecid and quinine respectively. Inhibition of γGT and/or β-lyase decreased the toxic effect of PCBD-GSH, PCBD-CYS, and PCBD-NAC. In contrast to the experiments with LLC- PK1 monolayers, no effect of amino acid transporter substrates or probenecid on the toxicity was observed. Apparently uptake of PCBD-CYS or PCBD-NAC in primary cultures of rat renal cortical cells is not the rate-limiting step in the toxicity. The probenecid sensitivity of toxicity observed by others thus must be caused by other mechanisms.In part II, glutathione conjugated halohydroquinones are the subject of investigation. These conjugates resemble the bromohydroquinone glutathione conjugates which are supposed to be responsible for the nephrotoxic effects of bromobenzene. The exact mechanism of nephrotoxicity is not known. In vivo experiments indicate that γGT is an important factor, as in the case of PCBD-GSH. However, β-lyase mediated metabolism appears to be of lesser importance than oxidative metabolism resulting in quinone conjugates (Chapter 1). In a first attempt to validate the in vitro test system and to see whether the mechanism of nephrotoxicity of glutathione conjugated bromohydroquinones is more generally valid, dichloro(glutathion-S-yl)hydroquinone (DC-(GSyl)HQ) and trichloro(glutathion-S-yl)hydroquinone (TC-(GSyl)HQ) were studied in vivo (Chapter 6). Administration (i.v.) of these compounds in the quinone form caused nephrotoxicity evidenced by elevations in blood urea nitrogen (BUN) , an increase in the urinary excretion of glucose, lactate dehydrogenase and γGT, and by the pathological changes in the kidney slices. Reducing the glutathione conjugated quinones with ascorbic acid caused a drastic increase in nephrotoxicity . This protocol may result in an increased delivery of DC-(GSyl)HQ and TC-(GSyl)HQ to the kidney , by preventing their interaction with nucleophilic sites on plasma proteins and/or with other extra-renal macromolecules. Surprisingly, inhibition of γGT increased the nephrotoxicity.Apparently γGT has a dual role in the nephrotoxicity of halogenated hydroquinone glutathione conjugates. This was further investigated in vitro in chapter 7. The LLC-PK1cell line was chosen, since the apical presence of γGT makes this cell line a very suitable model for this purpose. However, in contrast to the experiments with PCBD-GSH and its derivatives, the in vitro experiments did not completely clarify the mechanism of activation of halogenated hydroquinone-glutathione conjugates. In a postulated scheme (Fig.7.7) the reactions that can occur with halogenated GSyl-HQ conjugates are summarized , and an explanation for the results obtained in vitro as well as in vivo is offered. The results indicate that γGT is not the rate-limiting step in the toxicity. They suggest that γGT on the one hand initiates a detoxication by 1.4-benzothiazine formation and/or polymerization, and on the other hand activates by targetting the oxidation that causes the toxicity. In addition, without metabolism by γGT these compounds are not stable and can be detoxified via air-oxidation and presumably polymerization. Further studies will be necessary to elucidate the exact site and mechanism of oxidation.CONCLUDING REMARKSThe studies described in this thesis demonstrate that exposing of monolayers on porous substrates closely resembles the in vivo luminal and serosal exposure of a renal proximal tubular cell. In the future this system offers the possibility to investigate and compare the relative importance of polarized functions of epithelial cells of different organs and origin.By using model compounds it could be demonstrated that the LLC-PK1cell may be a relatively simple model for determining a specific nephrotoxic potential of glutathione and cysteine conjugates. For this purpose, the absence of an organic anion transporter in the LLC-PK1cell line proved not to be a handicap. However, for investigating the nephrotoxic potential of other chemicals, the presence of an organic anion transporter may be a prerequisite. Consequently, further characterization of other cell lines as the Opussum Kidney cell fine (Koyama et al, 1978) or the recently established human renal cell line (KRC/Y) (Yano et al. 1988) should focus on the presence of this transporter.Finally. the experiments on the role of γGT in the nephrotoxicity of glutathione conjugated halogenated hydroquinones underline once more the importance of in vivo validation. Differences of a more physical natures between in vivo and in vitro, as e.g. oxygen pressure and pH, may cause different or additional effects of compounds in vivo. which may be missed in vitro when these differences are not taken into account.
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Doctoral thesis biblioteca |
| Language: | English |
| Published: |
Landbouwuniversiteit Wageningen
|
| Subjects: | animal testing alternatives, in vitro, nephrotoxicity, organs, tissues, alternatieven voor dierproeven, nefrotoxiciteit, organen, weefsels, |
| Online Access: | https://research.wur.nl/en/publications/the-development-of-an-in-vitro-model-for-studying-mechanisms-of-n |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | SUMMARYPresently in our society animal tests still form the main starting point for the assessment of the possible risks of chemicals with regard to human and animal health. For scientific. economic, and ethical reasons. attempts are undertaken continuously to develop cell models as alternatives to animal testing. However, the predictive value of in vitro test systems is often limited due to the unawareness about the mechanisms of toxicity and the complexity of organisms. As a consequence, a strategy increasingly used is one of studying mechanisms of toxicity evolving in rational cell models. Following this strategy, this thesis aims at contributing to the development of cell models as an alternative to animal testing.A usefull cell model for studying mechanisms of toxicity requires the presence of characteristics which may play a role in this toxicity in vivo. and the possibility to demonstrate their involvement. The organ of our choice was the kidney. Since very often the cortex. more in particular the proximal tubule. is affected by chemicals. cortical cells of different origin have been used: A renal cell line derived from a Hamshire pig (LLC-PK1) and primary cultures of rat renal cortical cells (Chapter 2). The applicability of these cells in a cell model was tested considering two major aspects of importance for nephrotoxicity: Firstly. the functional polarity of the proximal tubular cell, and secondly the presence of specific biotransformation processes and transport systems (Chapter 1).In vivo luminal (apical) as well as serosal (basolateral) exposure of the proximal tubular cell can occur. However. when monolayers are cultured in a conventional way, i.e. on a solid support, only the apical side is accessible for the compounds to be investigated. To overcome this problem a new system for nephrotoxicological in vitro studies has been developed. Cells were cultured on porous substrates, which enables exposure of monolayers on either side (Chapter 2).The nephrotoxic effects of haloalkenes and halobenzenes seem to be mediated via the formation of glutathione conjugates (Chapter 1). In part II of this thesis a glutathione conjugate of hexachlorobutadiene and its derivatives are used as model compounds to investigate the possibilities of the cell systems described in chapter 2. LLC-PK1monolayers possess the most important characteristics necessary for cytotoxicity of S-(1.2,3,4,4-pentachlorobutadienyl)glutathione (PCBD- GSH) to occur (Chapter 3). Via the use of acivicin, an irreversible inhibitor of γ-glutamyltranspeptidase (γGT). and aminooxyacetic acid. an inhibitor of β-lyase, the importance of these enzymes in the activation of PCBD-GSH was demonstrated. After metabolism by γGT and dipeptidase S-(1,2,3,4,4-pentachlorobutadienyl)-L-cysteine (PCBD-CYS) is formed extracellularly. This is taken up and converted by β-lyase into a reactive thiol resulting in toxicity. In vivo and in the LLC-PK1cell line γGT is predominantly present on the luminal and apical membrane respectively. In agreement with this. apical exposure of LLC-PK1monolayers to PCBD-GSH caused a stronger toxicity than basolateral exposure . In the nephrotoxicity of the corresponding mercapturate N-acetyl-S-(1,2,3,4,4-pentachlorobutadlenyl)-L-cysteine (PCBD-NAC), which is deacetylated intracellularly into PCBD-CYS, a serosally located probenecid sensitive organic anion transporter has been suggested to be of importance (Chapter 1). Apical exposure of LLC-PK1monolayers to PCBD-NAC caused hardly any toxicity; only at relatively high concentrations a toxic effect was detected. Surprisingly, basolateral exposure and consequently administration of the compound to the possible transporter, did not cause substantial toxicity either. Further investigations revealed that this is probably caused by the absence of an organic anion transporter in the LLC-PK1cell line: The model organic anion para-aminohippurate (PAH) was not transported by LLC-PK1cells. In contrast, LLC-PK1monolayers are in the possession of an organic cation transporter. A quinine and ouabalne inhibitable transcellular transport of the cation tetraethylammonium (TEA) was demonstrated (Chapter 3).The organic anion transporter was also postulated (Lash and Anders, 1986) to play a role in the cytotoxicity of cysteine conjugates . However, in spite of the demonstrated absence of an organic anion transporter , PCBD-CYS causes toxicity in the LLC-PK1monolayer. In contrast to PCBD-GSH, basolateral exposure to PCBD-CYS resulted in a stronger toxic effect than apical exposure (Chapter 4). A possible candidate for the transport of PCBD-CYS in LLC-PK1cells is an amino acid transporter. The inhibitory effect of D-tryptophan and histidine on the toxicity of PCBD-CYS strongly suggests the involvement of the amino acid transport System T (Chapter 4).Culturing primary cultures of rat renal cortical cells resulted in confluent monolayers (Chapter 2 and 5). However, investigating transcellular transport of organic ions in a similar way as with the LLC- PK1 monolayers was not possible, because a non-leaking monolayer on a porous support was not obtained. This may be due to the diversity of the cell population in the culture, and the concomitant differences in junctional complexes necessary for complete sealing of the monolayer. The presence of both PAH and TEA transporters was demonstrated by measuring intracellular accumulation, which could be inhibited by probenecid and quinine respectively. Inhibition of γGT and/or β-lyase decreased the toxic effect of PCBD-GSH, PCBD-CYS, and PCBD-NAC. In contrast to the experiments with LLC- PK1 monolayers, no effect of amino acid transporter substrates or probenecid on the toxicity was observed. Apparently uptake of PCBD-CYS or PCBD-NAC in primary cultures of rat renal cortical cells is not the rate-limiting step in the toxicity. The probenecid sensitivity of toxicity observed by others thus must be caused by other mechanisms.In part II, glutathione conjugated halohydroquinones are the subject of investigation. These conjugates resemble the bromohydroquinone glutathione conjugates which are supposed to be responsible for the nephrotoxic effects of bromobenzene. The exact mechanism of nephrotoxicity is not known. In vivo experiments indicate that γGT is an important factor, as in the case of PCBD-GSH. However, β-lyase mediated metabolism appears to be of lesser importance than oxidative metabolism resulting in quinone conjugates (Chapter 1). In a first attempt to validate the in vitro test system and to see whether the mechanism of nephrotoxicity of glutathione conjugated bromohydroquinones is more generally valid, dichloro(glutathion-S-yl)hydroquinone (DC-(GSyl)HQ) and trichloro(glutathion-S-yl)hydroquinone (TC-(GSyl)HQ) were studied in vivo (Chapter 6). Administration (i.v.) of these compounds in the quinone form caused nephrotoxicity evidenced by elevations in blood urea nitrogen (BUN) , an increase in the urinary excretion of glucose, lactate dehydrogenase and γGT, and by the pathological changes in the kidney slices. Reducing the glutathione conjugated quinones with ascorbic acid caused a drastic increase in nephrotoxicity . This protocol may result in an increased delivery of DC-(GSyl)HQ and TC-(GSyl)HQ to the kidney , by preventing their interaction with nucleophilic sites on plasma proteins and/or with other extra-renal macromolecules. Surprisingly, inhibition of γGT increased the nephrotoxicity.Apparently γGT has a dual role in the nephrotoxicity of halogenated hydroquinone glutathione conjugates. This was further investigated in vitro in chapter 7. The LLC-PK1cell line was chosen, since the apical presence of γGT makes this cell line a very suitable model for this purpose. However, in contrast to the experiments with PCBD-GSH and its derivatives, the in vitro experiments did not completely clarify the mechanism of activation of halogenated hydroquinone-glutathione conjugates. In a postulated scheme (Fig.7.7) the reactions that can occur with halogenated GSyl-HQ conjugates are summarized , and an explanation for the results obtained in vitro as well as in vivo is offered. The results indicate that γGT is not the rate-limiting step in the toxicity. They suggest that γGT on the one hand initiates a detoxication by 1.4-benzothiazine formation and/or polymerization, and on the other hand activates by targetting the oxidation that causes the toxicity. In addition, without metabolism by γGT these compounds are not stable and can be detoxified via air-oxidation and presumably polymerization. Further studies will be necessary to elucidate the exact site and mechanism of oxidation.CONCLUDING REMARKSThe studies described in this thesis demonstrate that exposing of monolayers on porous substrates closely resembles the in vivo luminal and serosal exposure of a renal proximal tubular cell. In the future this system offers the possibility to investigate and compare the relative importance of polarized functions of epithelial cells of different organs and origin.By using model compounds it could be demonstrated that the LLC-PK1cell may be a relatively simple model for determining a specific nephrotoxic potential of glutathione and cysteine conjugates. For this purpose, the absence of an organic anion transporter in the LLC-PK1cell line proved not to be a handicap. However, for investigating the nephrotoxic potential of other chemicals, the presence of an organic anion transporter may be a prerequisite. Consequently, further characterization of other cell lines as the Opussum Kidney cell fine (Koyama et al, 1978) or the recently established human renal cell line (KRC/Y) (Yano et al. 1988) should focus on the presence of this transporter.Finally. the experiments on the role of γGT in the nephrotoxicity of glutathione conjugated halogenated hydroquinones underline once more the importance of in vivo validation. Differences of a more physical natures between in vivo and in vitro, as e.g. oxygen pressure and pH, may cause different or additional effects of compounds in vivo. which may be missed in vitro when these differences are not taken into account. |
|---|