Characterization of Nitric Oxide Production following Isolation of Rat Hepatocytes

M. A. Tirmenstein1, F. A. Nicholls-Grzemski, T. D. Schmittgen, B. A. Zakrajsek and M. W. Fariss

Department of Pharmaceutical Sciences, College of Pharmacy, Washington State University, Pullman, Washington 99164–6510

Received June 10, 1999; accepted August 10, 1999


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Freshly isolated suspensions of rat parenchymal liver cells (hepatocytes) produce large amounts of nitrite following isolation. Nitrite production was inhibited by the inducible nitric oxide synthase (iNOS) inhibitor aminoguanidine, as well as the transcription inhibitor actinomycin D. Increases in iNOS mRNA, protein, and activity levels correlated with the formation of nitrite. iNOS mRNA was first detectable 2 h after the onset of hepatocyte incubations and peaked at 4 h. These results indicate that nitrite formation is a result of the large scale production of nitric oxide (NO) by hepatocytes in response to the time-dependent induction of iNOS. NO production by hepatocytes was attenuated by pretreatment of rats with the Kupffer cell inhibitor, gadolinium chloride. Also, the addition of the endotoxin neutralizing agent, polymyxin B; the protein kinase inhibitor, staurosporine, and antioxidants to perfusion buffers and hepatocyte suspensions also decreased nitrite formation. Collectively, our results suggest that iNOS is induced in hepatocytes in response to the stresses generated during collagenase isolation procedures. The response appears to be triggered by a complex interaction between several different factors including Kupffer cell activation, reactive oxygen species generation, and endotoxin contamination of collagenase preparations.

Key Words: nitric oxide; hepatocytes; nitric oxide synthase; endotoxin; nitrite; iNOS; NOS II.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The nitric oxide radical (NO) has been recognized as an important physiologic mediator that induces a large number of biological actions. NO is the active component of endothelium-derived relaxing factor (Ignarro et al., 1987Go; Palmer et al., 1987Go) and as such, plays an important role in regulating blood flow. NO can also modify normal cellular function by inhibiting enzymes and altering signal transduction events. In the liver, NO has been demonstrated to decrease protein synthesis, mitochondrial respiration and cytochrome P450 activity while increasing intracellular levels of cGMP (Harbrecht and Billiar, 1995Go). NO has also been reported to function as a suppressor of apoptosis by inhibiting caspase activity (Li et al., 1997Go; Rossig et al., 1999Go). These NO-induced biochemical alterations may have important implications in toxicological studies.

There is increasing evidence that NO can function as a cellular antioxidant and can, under certain conditions, protect against reactive oxygen species-induced toxicity (Wink et al., 1995Go). Sergent et al. (1997) reported that NO protected against iron-mediated oxidative stress in primary rat hepatocyte cultures and hypothesized that NO-iron binding prevented the iron catalyzed formation of reactive oxygen species. Although NO can function as an antioxidant, NO can also react with superoxide anion to form the reactive oxidant species peroxynitrite. The dual nature of NO actions is also seen in studies conducted with hepatotoxicants. Muriel (1998) demonstrated that NO formation was protective against carbon tetrachloride-induced liver damage while Gardner et al. (1998) and Hinson et al. (1998) have proposed that NO is an important mediator of acetaminophen hepatotoxicity. Therefore, since NO can modulate toxicity and affect cellular functions, it is important to recognize when NO production is occurring in in vitro cell models.

The enzyme nitric oxide synthase (NOS) converts L-arginine to NO. At least three isoforms of NOS are known to exist in tissues. The inducible form of the nitric oxide synthase or type II nitric oxide synthase (iNOS) is expressed in several cell types including parenchymal liver cells (hepatocytes), smooth muscle cells, and macrophages (Harbrecht and Billiar, 1995Go). Kupffer cells are generally recognized as a source of NO in the liver, but hepatocytes can also contribute to liver NO production. Once iNOS is expressed in hepatocytes, extremely high levels of NO can be produced. For example, Curran et al. (1989) measured extracellular nitrite/nitrate (stable end products of NO formation) levels of over 500 µM in cultured rat hepatocytes after 18 h. Several treatments have been shown to induce iNOS in rat hepatocytes. Endotoxin and cytokines (tumor necrosis factor, interferon {gamma} and interleukin 1) are potent inducers of iNOS and act synergistically to evoke maximal NO production (Nussler et al., 1993Go). Oxidative stress has also been proposed to induce iNOS expression in rat hepatocytes (Duval et al., 1995Go).

Previously we reported that isolated rat hepatocytes in suspension (free of Kupffer cells) produce high levels of NO shortly after isolation (Nicholls-Grzemski et al., 1999Go). Hepatocyte NO production occurred in a variety of incubation media, and was not affected by adding antibiotics to incubations suggesting that bacterial contamination was not involved in nitrite synthesis. However, the addition of the NOS inhibitors aminoguanidine or NG-nitro-L-arginine methyl ester (L-NAME) completely blocked NO production, implicating iNOS involvement. Since isolated hepatocyte suspensions are a commonly used in vitro model system in metabolism and toxicology studies, a better understanding of NO production in this cell system is needed. In the present manuscript, we further characterize the biochemical events involved in NO production by freshly isolated rat hepatocyte suspensions and explore possible mechanisms that could initiate NO production. We propose that hepatocyte suspensions offer a convenient model system to investigate the cellular signaling events involved in iNOS induction.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Chemicals.
2,3-Diaminonaphthalene was obtained from ICN Pharmaceuticals, Inc. (Costa Mesa, CA). Liberase RH enzyme was a generous gift from Roche Molecular Biochemicals (Indianapolis, IN). Trolox was purchased from Aldrich (Milwaukee, WI). Collagenase (type L), the components of modified Waymouth's medium, actinomycin D, gadolinium chloride, staurosporine, aminoguanidine, polymyxin B sulfate, deferoxamine mesylate, and all other chemicals were obtained from Sigma (St. Louis, MO). Collagenase (type L) is referred to as the standard collagenase throughout this manuscript to distinguish it from Liberase RH enzyme preparations.

Preparation of isolated hepatocyte suspensions.
Adult male Sprague-Dawley rats (180–220 g) were obtained from Simonsen Laboratories, Inc. (Gilroy, CA). They were housed in small groups and given food and water ab libutum in a 12 h light/dark cycle for at least 1 week prior to use. Rats pretreated with gadolinium chloride (10 mg/kg) dissolved in phosphate-buffered saline (10 mg/ml) were iv-injected into the tail vein 24 h prior to hepatocyte isolation. Rat hepatocytes were prepared by collagenase perfusion as previously described (Fariss et al., 1985Go). Briefly, the liver was cannulated via the portal vein, surgically removed, and perfused in a recirculating system at 30 ml/min for 3 min with Hanks'-bicarbonate buffer containing 0.6 mM EGTA. The liver was then perfused in a recirculating system with Hanks'-bicarbonate buffer containing 3 mM calcium chloride and 60–100 mg standard collagenase (Sigma L blend) for up to 10 min. To obtain the same degree of separation, 2 mg of Liberase was perfused for 4 min. After dispersion and washing, cells were assessed for viability by Trypan blue exclusion. A yield of 4.5 to 5.5 x 108 was routinely obtained with over 93% viability. Hepatocyte suspensions were prepared as described in Fariss et al., (1997). Briefly, hepatocyte suspensions (2 x 106 cells/ml, 12 ml total) were prepared in serum-free modified Waymouth's medium as previously described (Fariss et al., 1997Go) with the addition of 0.12 mM ornithine but without cystine. Cells were placed in 125-ml boiling flasks and rotated at 37°C under ambient air. After a 15-min equilibration time, an aliquot of cells was taken as the 0 time point. Following collection of the 0 time point, the chemical treatments (prepared in media or as indicated) were immediately added. At hourly intervals thereafter, cells were sampled and pelleted, and supernatants were collected for nitrite and lactate dehydrogenase (LDH) leakage analysis. Cell pellets were stored at –80°C for iNOS Western blot analysis and quantitative RT-PCR.

Enzyme assays.
Hepatocyte calcium-independent NOS activity was determined by measuring the formation of nitrite as previously described (Nicholls-Grzemski et al., 1999Go). LDH activity was determined by monitoring the enzymatic formation of NADH from NAD+ in the presence of L-lactic acid. Post-centrifugation supernatants were diluted 1:80 with phosphate buffered saline, pH 7.4. A 100 µl aliquot was mixed with 100 µl of reagent to give a final concentration of 3.75 mM NAD+ and 25 mM L-lactic acid in 125 mM Tris–HCl buffer, pH 8.9 in a 96-well plate. The increase in fluorescence (360 nm ex, 460 nm em) due to the formation of NADH was immediately monitored at room temperature (gain 70) using a CytoFluor 4000, Perseptive Biosystems (Framingham, MA). The percent LDH leakage was calculated by comparing values to total LDH activity. Total LDH was measured from a sample of hepatocytes collected at zero time and lysed with a final concentration of 0.2% Triton X-100.

Nitrite determinations.
Nitrite levels in media were determined according to the procedures of Misko et al. (1993) with slight modifications as previously described (Nicholls-Grzemski et al., 1999Go). Nitrite standards were prepared fresh and standard curves were constructed in the same media as samples. Samples and standards were read at 360 nm ex and 460 nm em with a gain setting of 90 on a CytoFluor 4000 fluorescence plate reader, Perseptive Biosystems (Framingham, MA).

Quantitative RT-PCR.
Total cellular RNA was isolated from the hepatocytes using the RNeasy Mini, RNA isolation kit from Qiagen (Valencia, CA) per the manufacturer's protocol. Total RNA was eluted from the matix with 35 µl of RNase-free water. Residual genomic DNA was removed by incubating the RNA solution with 15 units of RNase-free DNase I in 2 mM MgCl2 for 10 min at 37°C, followed by 5 min at 90°C to inactivate the DNase. DNase-treated RNA solution (25 µl) was reverse-transcribed to cDNA as previously described (Horikoshi, et al., 1992Go).

The amount of iNOS mRNA relative to the ß actin endogenous control was determined using real-time, quantitative PCR. The sequence of the ß actin primers was 5' ACCAACTGGGACGATATGGAGAAGA 3' bases 1557–1581, forward primer and 5' TACGACCAGAGGCATACAGGGACAA 3' bases 2210–2234 reverse primer of the genomic ß actin sequence (Nudel et al., 1983Go). The sequence of the iNOS primers was 5' CACGACACCCTTCACCACAAG 3' bases 301–321, forward primer and 5' TTGAGGCAGAAGCTCCTCCA 3' bases 419–438, reverse primer of the iNOS cDNA sequence (Kosuga et al., 1994Go). The PCR was performed in the Perkin-Elmer Applied Biosystems GeneAmp 5700 Sequence Detection System (Foster City, CA) using the SYBR green PCR kit as recommended by the manufacturer (Foster City, CA). Briefly 2.5 µl of the 10x SYBR green buffer, 1 mM dA, dG, dC and dUTP, 2 mM MgCl2, 0.25 units of uracil N-glycosylase, 0.625 units of Amplitaq Gold DNA polymerase, 250 nM of the forward and reverse primer, 5 µl of a 1:10 dilution of the cDNA and water to 25 µl. The reactions were performed in the MicroAmp 96-well plate capped with MicroAmp optical caps Perkin-Elmer Applied Biosystems (Foster City, CA). The reactions were incubated at 50°C for 2 min to activate the uracil N'-glycosylase and then for 10 min at 95°C to inactivate the uracil N'-glycosylase and activate the Amplitaq Gold polymerase. The reactions were performed for 40 cycles of 15 sec at 95°C, 30 sec at 55°C and 30 sec at 72°C.

The data generated by real-time PCR were plotted as the {Delta}Rn fluorescence signal versus the cycle number. The Applied Biosystems, Inc. (ABI) 5700 sequence detection system software from Perkin Elmer calculates the {Delta}Rn using the equation {Delta}Rn = (Rn+) – (Rn-), where Rn+ is the fluorescence signal of the product at any given time and Rn- is the fluorescence signal of the baseline emission during cycles 3 to15. An arbitrary threshold was set at the midpoint of the log {Delta}Rn versus cycle number plot. The Ct value is defined as the cycle number at which the {Delta}Rn crosses this arbitrary threshold. The amount of iNOS cDNA relative to the ß actin endogenous control was determined using a modification of the 2-{Delta}{Delta}Ct method as described in the ABI user bulletin number 2. The amount of iNOS mRNA relative to ß actin was calculated equal to 2-{Delta}Ct where {Delta}Ct = CtiNOS – CtActin.

Western blot analysis of iNOS.
Hepatocyte pellets were resuspended in phosphate-buffered saline, pH 7.4, and protein levels were determined by the BCA protein assay (Pierce Chemical Co., Rockford, IL). Aliquots of resuspended pellets were heat-denatured and equal amounts of total protein (20 µg) were electrophoresed under reducing conditions on a 7.5% polyacrylamide-SDS gel. Proteins were transferred to nitrocellulose blots and blocked overnight at 4°C in 5% dried milk powder dissolved in 150 mM sodium chloride, 0.1% Tween-20 in 50 mM Tris–HCl, pH 7.4. The blots were washed and expression of iNOS protein was detected with rabbit anti-mouse antibody Transduction Laboratories (Lexington, KY). This polyclonal antibody has been shown by the manufacturer to recognize 2 protein bands in mouse macrophage lysates. Specific immunocomplexes were detected using peroxidase-labeled anti-rabbit antibody with the ECL Western Blot Analysis System Amersham Pharmacia (Piscataway, NJ). Developed blots were immediately exposed to Kodak X-Omat scientific imaging film and both bands were quantified together using a Hewlett-Packard ScanJet 4c (Palo Alto, CA) equipped with Bioscan Analysis software, Biosoft (Ferguson, MO).

Statistics.
Results are presented as mean ± SD. Both the Student's t-test and ANOVA, followed by Dunnett's post hoc test were performed with the InStat 2.03 GraphPad Software Inc. (San Diego, CA) statistical package.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Rat hepatocytes were isolated from rats by established collagenase perfusion techniques. Following isolation, nitrite levels were measured in the extracellular media as a measure of NO production (Fig. 1Go). Nitrite represents a stable end product of NO production and is commonly used to assess NO formation. In control hepatocyte preparations, there was typically a 3-h delay prior to the onset of nitrite production. The addition of the iNOS inhibitor, aminoguanidine, blocked nitrite production (less than 0.8 µM) over the entire 8-h-incubation period, supporting the conclusion that the measured extracellular nitrite was derived from NO. As Figure 1Go indicates, the addition of the transcription inhibitor, actinomycin D, at zero time also prevented nitrite formation. Neither actinomycin D nor aminoguanidine significantly altered cell death, and control hepatocytes remained at least 70% viable after 8 h, as assessed by LDH analysis (data not shown). These data suggest that the production of NO does not significantly decrease hepatocyte viability during the 8-h incubation.



View larger version (18K):
[in this window]
[in a new window]
 
FIG. 1. Effects of aminoguanidine and actinomycin D on nitrite production by hepatocyte suspensions. Aminoguanidine (1 mM) or actinomycin D (1 µM) were added at zero time and extracellular nitrite levels were monitored over time. Values represent mean ± SD (n = 3 for all determinations). *Significantly different from controls (p < 0.05) as determined by ANOVA followed by Dunnett's post hoc test.

 
The time course of iNOS mRNA, protein levels and activity were assayed following hepatocyte isolations. After hepatocytes were isolated and washed, hepatocytes were resuspended in Waymouth's medium and incubated at 37°C. The beginning of the incubation was designated as 0 time. In these experiments, the time required for hepatocyte isolation and equilibration was approximately 80 min. The first detectable increase in iNOS mRNA levels were measured 2 h after the beginning of the incubation, with levels peaking at about 4 h (Fig. 2aGo). Western blots indicated that there was no iNOS protein expression at zero time. The protein was first detected at 3 h after the beginning of the incubations and continued to increase for up to 7 h after the incubations were initiated (Figs. 2b and 2cGoGo). In addition, calcium-independent NOS activity was measured. In the liver, calcium-independent NOS activity corresponds to the iNOS isoform of the enzyme (Harbrecht and Billiar, 1995Go). Activity measurements agreed well with the Western blot analysis. iNOS activity levels were significantly increased at 4 and 8 h relative to those measured at zero time (Fig. 2cGo).



View larger version (31K):
[in this window]
[in a new window]
 
FIG. 2. Time-dependent induction of iNOS mRNA, protein and activity levels in hepatocyte suspensions. (A) The amount of iNOS mRNA relative to ß actin was determined at various time intervals following the beginning of hepatocyte incubations, by using real-time quantitative PCR analysis as described in Materials and Methods. Values represent the mean ± SD (n = 3). (B) Western blots of iNOS protein levels (molecular mass 130 kDa). (C) iNOS protein levels were determined by densitometric analysis of Western blots (B); iNOS activity levels were measured and are expressed as nmol nitrite formed per mg protein. Values represent mean ± SD (n = 3). *Significantly different from zero time (p < 0.05) as determined by ANOVA followed by Dunnett's post hoc test.

 
The control of iNOS expression in hepatocytes is complex, with many agents inducing iNOS expression (Forstermann and Kleinert, 1995Go). Since Kupffer cell-derived cytokines have been shown to induce iNOS expression in hepatocytes (Nussler et al., 1993Go), experiments were conducted to examine whether alterations in Kupffer cell function would have an effect on iNOS expression in hepatocytes. Previous studies (Hoglen et al., 1998Go), have shown that gadolinium chloride pretreatment inhibits Kupffer cell function. In our experiments, pretreatment of rats with gadolinium chloride significantly decreased nitrite production after 6 h, as compared to controls (Fig. 3Go). Hepatocytes from gadolinium chloride pretreated rats had a 57% reduction in nitrite production after 6 h. The decrease was not due to loss of cell viability. Less than 30% of hepatocytes were nonviable after 6 h following gadolinium chloride pretreatment as judged by LDH release (data not shown).



View larger version (15K):
[in this window]
[in a new window]
 
FIG. 3. Effects of gadolinium chloride pretreatment on nitrite production by hepatocytes suspensions. Rats were administered gadolinium chloride (10 mg/kg) dissolved in phosphate buffered saline (10 mg/ml) by iv tail vein injection 24 h prior to hepatocyte isolations. Following isolations, extracellular nitrite levels were monitored over time. Values represent mean ± SD (n >= 3). *Significantly different from controls (p < 0.05) as determined by Student's t test.

 
Endotoxin has been shown to induce iNOS in cells (Forstermann and Kleinert, 1995Go). Since most collagenase preparations are derived from bacterial sources, endotoxin contamination of collagenase preparations may be involved in iNOS induction in isolated hepatocytes. To test this hypothesis, livers were perfused with Liberase. Liberase is a specially purified blend of collagenase with very low levels of endotoxin (Less than 1 EU per mg protein). This represents a 6000-fold reduction in endotoxin levels found in crude preparations (Roche Molecular Biochemicals, personal communication). Hepatocytes isolated by the Liberase perfusion still produced high levels of nitrite, but the extent of nitrite production was more variable and the onset was somewhat delayed in comparison with hepatocytes prepared from standard collagenase perfusions.

Polymyxin B binds and neutralizes endotoxin and should therefore diminish any endotoxin-mediated induction of iNOS. The addition of 100 U/ml Polymyxin B to both the perfusion and incubation media significantly decreased nitrite production by hepatocyte suspensions (Fig. 4Go). Polymyxin B decreased nitrite production by about 50% in hepatocytes prepared from standard collagenase perfusions and completely blocked nitrite production (less than 0.8 µM) in hepatocytes isolated by the Liberase perfusion. Polymyxin B had no effect on cell viability (data not shown). In addition to binding endotoxin, polymyxin B also inhibits protein kinase activity (Rando, 1988Go). To assess whether the inhibitory effects of polymyxin B on nitrite production are due to protein kinase inhibition, another protein kinase inhibitor was tested. The protein kinase inhibitor staurosporine was tested for its effects on nitrite production. In the results reported in Table 1Go, polymyxin B and staurosporine were not present in the perfusion buffer, in contrast to Figure 4Go, but instead were added at zero time. Staurosporine added at the onset of the hepatocyte incubations was as effective as polymyxin B in inhibiting nitrite production (Table 1Go). Polymyxin B and staurosporine did not induce hepatocyte cell death during the 6 h incubation (data not shown).



View larger version (20K):
[in this window]
[in a new window]
 
FIG. 4. Effects of polymyxin B on nitrite production by hepatocytes suspensions isolated by Liberase and standard collagenase perfusions. Polymyxin B (100 U/ml) was added to hepatocyte perfusion buffers and added again to suspensions at zero time. Following isolation, extracellular nitrite levels were monitored over time. Controls contained no additions. Values represent mean ± SD (n = 3). *Significantly different from corresponding collagenase or Liberase controls (p < 0.05) as determined by Student's t test.

 

View this table:
[in this window]
[in a new window]
 
TABLE 1 Six Hour Extracellular Nitrite Levels in Hepatocyte Suspensions
 
Since reactive oxygen species have been shown to induce iNOS (Duval et al., 1995Go), the effects of the antioxidants deferoxamine (DFO) and Trolox on nitrite production were examined (Table 1Go). The addition of DFO and Trolox to the both the perfusion buffers and hepatocyte suspensions significantly decreased nitrite production in hepatocytes isolated with Liberase but not with standard collagenase. The DFO/Trolox addition did not affect cell viability as determined by LDH analysis (data not shown).


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Hepatocytes isolated from rat liver release large amounts of nitrite into the medium after isolation. Our data strongly supports the conclusion that nitrite is derived from NO following the induction of iNOS. The addition of the iNOS inhibitor, aminoguanidine, and the transcription inhibitor, actinomycin D, blocked nitrite formation throughout the incubation period. In addition, the formation of nitrite was preceded by increases in iNOS mRNA and coincided with increases in iNOS protein and activity levels. Previously, we demonstrated that the removal of Kupffer cells following the isolation of hepatocytes with a collagenase perfusion method had no effect on the subsequent production of nitrite by hepatocyte suspensions (Nicholls-Grzemski et al., 1999Go). Therefore, although Kupffer cells may be important for the initiation of NO formation, these cells are not directly involved in the large scale production of NO observed in hepatocyte suspensions.

In our studies, increases in extracellular nitrite levels first occurred at about 4 h after the beginning of the hepatocyte incubations (Fig. 1Go). Since the collagenase isolation procedures require about 80 min to complete, the total time we observed between increases in nitrite levels and the isolation of hepatocytes was about 5 to 6 h. This time period is consistent with the time required to induce nitrite production in cultured hepatocytes following the addition of cytokines/endotoxin (4–6 h) reported by Geller et al. (1993). The time frame for nitrite production with hepatocyte suspensions suggests that the stimulus for nitrite production occurs during the isolation of hepatocytes. In addition, a comparison of the time-dependent formation of iNOS mRNA and protein (Fig. 2Go) indicates that there is approximately a 1-h delay between mRNA expression and the subsequent formation of detectable levels of the iNOS protein. Once the iNOS protein is expressed, nitrite rapidly increases in the extracellular medium.

Curran et al. (1989) demonstrated that activated Kupffer cells release products which can induce NO formation in hepatocytes. Activated Kupffer cells produce cytokines and reactive oxygen species, both of which can induce iNOS expression in hepatocytes (Duval et al., 1995Go; Geller et al., 1993Go). Studies have shown that gentle manipulation of rat livers can lead to the activation of Kupffer cells (Schemmer et al., 1998Go). Since collagenase isolation techniques involve liver manipulations, this mechanism may be relevant to our studies. Our results with the Kupffer cell inhibitor, gadolinium chloride, also support a role for Kupffer cells in the sequence of events leading to the production of NO by hepatocyte suspensions. Pretreatment of rats with gadolinium chloride attenuated nitrite formation by freshly isolated hepatocyte suspensions (Fig. 3Go).

In addition, Kupffer cells may become activated by exposure to endotoxin. Geller et al. (1993) found endotoxin ineffective in inducing iNOS in cultured hepatocytes, but endotoxin did potentiate the induction of iNOS by cytokines. In our experiments, hepatocytes isolated by Liberase, which contains very low levels of endotoxin, still produced nitrite. The addition of the endotoxin neutralizing agent, polymyxin B, inhibited NO production by rat hepatocyte suspensions, but polymyxin B was much more effective in blocking NO production in hepatocytes isolated by Liberase than by the standard collagenase preparations. There may be two explanations for this observation. Either the addition of polymyxin B was sufficient to bind and neutralize any residual endotoxin present in the Liberase preparation and block Kupffer cell activation, or the addition of polymyxin B inhibits NO production by a mechanism unrelated to endotoxin neutralization. Polymyxin B is known to be a protein kinase inhibitor and has been shown to affect protein kinase C activity (Rando, 1988Go). Reports have implicated protein kinase C as being involved in the signaling events triggering iNOS expression in hepatocytes (Hortelano et al., 1992Go). Interestingly, Hortelano and colleagues also demonstrated that protein kinase C activation antagonized iNOS induction by endotoxin in rat hepatocytes, indicating that there is considerable complexity in the regulation of iNOS in hepatocytes. In our experiments, adding polymyxin B at zero time (Table 1Go) was almost as effective in inhibiting NO production as including polymyxin B in the collagenase perfusion buffers and the hepatocyte suspensions (Fig. 4Go). If polymyxin B is acting by binding endotoxin from collagenase preparations, then it should be much more effective if it is added during the collagenase perfusion. However, we did not find this to be true with either Liberase or the standard collagenase. In addition, the effects of the protein kinase inhibitor, staurosporine, on nitrite production were almost identical to that of polymyxin B (Table 1Go). These results suggest that in addition to neutralizing endotoxin, polymyxin B may be affecting NO production by inhibiting protein kinases.

Adding the antioxidants DFO/Trolox to perfusion buffers and hepatocyte suspensions significantly inhibited nitrite production by hepatocytes isolated from Liberase preparations, but not from hepatocytes isolated from standard collagenase preparations. Duval et al. (1995) concluded that reactive oxygen species induced iNOS in rat hepatocytes and that DFO and Trolox inhibited this induction. Our results agree with this conclusion and suggest that reactive oxygen species may contribute to the induction of iNOS in hepatocytes isolated by Liberase. The sources of the reactive oxygen species are unknown, but may involve Kupffer cell activation, ischemia, the presence of trace amounts of iron, alterations in extracellular calcium, or other events occurring during isolation. Trace amounts of iron may be present as a contaminant in perfusion and incubation buffers or may be released from damaged cells. Several groups have suggested that protein kinase C is activated by reactive oxygen species (Bagchi et al., 1997Go; Gopalakrishna et al., 1991; Kass et al., 1989Go). A reasonable hypothesis would be that reactive oxygen species are generated during collagenase perfusions and isolation procedures and these species may induce iNOS by activating the protein kinase C signaling pathway in hepatocytes. However, further work is required to specifically identify the role of protein kinase C in inducing iNOS in hepatocyte suspensions.

As Table 1Go indicates, there are clear differences in the ability of hepatocytes isolated from standard collagenase and Liberase preparations to generate NO in the presence of polymyxin B, staurosporine, and DFO/Trolox. Our findings suggest that endotoxin may be involved in inducing iNOS in standard collagenase preparations, but other signaling pathways such as protein kinase C and reactive oxygen species may take a more prominent role in the absence of endotoxin. Our work suggests that endotoxin is not solely responsible for iNOS induction in isolated hepatocyte suspensions, but its presence may modify existing signaling pathways. Further work is necessary to characterize the differences between hepatocytes isolated with standard collagenase and Liberase preparations.

Recently, Wang et al. (1998) reported that collagenase perfusions induced the expression of iNOS mRNA in cultured mouse hepatocytes and speculated that the procedures involved in hepatocyte isolation may be responsible for these events. Although these researchers detected iNOS mRNA expression, they did not detect increases in extracellular nitrite. In agreement with our results, Lopez-Garcia (1998) recently reported that cultured rat hepatocytes produce large amounts of nitrite shortly after isolation, and they concluded that isolation procedures are involved in initiating this process. Jones and Czaja (1998) suggested that NO is generated in liver in response to injury. Hepatotoxicants, ischemia-reperfusion (Jones and Czaja, 1998Go), and partial hepatectomy (Hortelano et al., 1995Go) have been shown to increase NO production by the liver. Our data are consistent with this viewpoint, and suggest that hepatocytes as well as Kupffer cells may contribute to NO production in the liver. Therefore, the induction of iNOS in isolated hepatocytes may represent a cellular response to the stresses imposed on the liver during the isolation procedures.

In summary, freshly isolated hepatocyte suspensions produce large amounts of nitrite as a result of NO production. Our data suggests that iNOS induction occurs during the isolation of hepatocytes as a result of several factors, including Kupffer cell activation, reactive oxygen species production, and to a certain extent endotoxin exposure. The induction of iNOS may be triggered in response to the stress of hepatocyte isolations. In our system, the large production of NO had no effect on cell viability; however, other cellular functions were not examined. Further research is necessary to determine if NO is also produced in cultured hepatocytes and to examine the toxicological implications of NO production in these cells.


    ACKNOWLEDGMENTS
 
This work was supported by NIEHS/NIH grant #R01ES05452. The authors would like to acknowledge the helpful advice and assistance of Dr. Margaret Black in performing this study. The technical assistance of Christopher Wuestefeld is also gratefully acknowledged.


    NOTES
 
1 To whom correspondence should be addressed. Fax: (509) 335-5902. Email: tirmen{at}mail.wsu.edu. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Bagchi, D., Bagchi, M., Tang, L., Stohs, S. J. (1997). Comparative in vitro and in vivo protein kinase C activation by selected pesticides and transition metal salts. Toxicol. Lett. 91, 31–37.[ISI][Medline]

Curran, R. D., Billiar, T. R., Stuehr, D. J., Hofmann, K., and Simmons, R. L. (1989). Hepatocytes produce nitrogen oxides from L-arginine in response to inflammatory products of Kupffer cells. J. Exp. Med. 170, 1769–1774.[Abstract]

Duval, D. L., Sieg, D. J., and Billings, R. E. (1995). Regulation of hepatic nitric oxide synthase by reactive oxygen intermediates and glutathione. Arch. Biochem. Biophys. 316, 699–706.[ISI][Medline]

Fariss, M. W., Brown, M. K., Schmitz, J. A., and Reed, D. J. (1985). Mechanisms of chemical-induced toxicity: I. Use of a rapid centrifugation technique for the separation of viable and nonviable hepatocytes. Toxicol. Appl. Pharmacol. 79, 283–295.[ISI][Medline]

Fariss, M. W., Bryson, K. F., and Tirmenstein, M. A. (1997). Role of cellular thiol status in tocopheryl hemisuccinate cytoprotection against ethyl methanesulfonante-induced toxicity. Biochem. Pharmacol. 53, 651–661.[ISI][Medline]

Forstermann, U., and Kleinert, H. (1995). Nitric oxide synthase: expression and expressional control of the three isoforms. Naunyn Schmiedebergs Arch. Pharmacol. 352, 351–64.[ISI][Medline]

Gardner, C. R., Heck, D. E., Yang, C. S., Thomas, P. E., Zhang, X. J., DeGeorge, G. L., Laskin, J. D., and Laskin, D. L. (1998). Role of nitric oxide in acetaminophen-induced hepatotoxicity in the rat. Hepatology 27, 748–754.[ISI][Medline]

Geller, D. A., Nussler, A. K., Di Silvio, M., Lowenstein, C. J., Shapiro, R. A., Wang, S. C., Simmons, R. L., and Billiar, T. R. (1993). Cytokines, endotoxin, and glucocorticoids regulate the expression of inducible nitric oxide synthase in hepatocytes. Proc. Natl. Acad. Sci. U S A 90, 522–526.[Abstract]

Gopalakrishna, R., and Anderson, W. B. (1991). Reversible oxidative activation and inactivation of protein kinase C by the mitogen/tumor promoter periodate. Arch. Biochem. Biophys. 285, 382–387.[ISI][Medline]

Harbrecht, B. G., and Billiar, T. R. (1995). The role of nitric oxide in Kupffer cell-hepatocyte interactions. Shock 3, 79–87.[ISI][Medline]

Hinson, J. A., Pike, S. L., Pumford, N. R., and Mayeux, P. R. (1998). Nitrotyrosine-protein adducts in hepatic centrilobular areas following toxic doses of acetaminophen in mice. Chem. Res. Toxicol. 11, 604–607.[ISI][Medline]

Hoglen, N. C., Younis, H. S., Hartley, D. P., Gunawardhana, L., Lantz, R. C., and Sipes, I. G. (1998). 1,2-Dichlorobenzene-induced lipid peroxidation in male Fischer 344 rats is Kupffer cell dependent. Toxicol. Sci. 46, 376–385.[Abstract]

Horikoshi, T., Danenberg, K. D., Stadlbauer, T.H., Volkenandt, M., Shea, L. C., Aigner, K., Gustavsson, B., Leichman, L., Frosing, R., Ray, M., Gibson, N. W., Spears, P., and Danenberg, P. V. (1992). Quantitation of thymidylate synthase, dihydrofolate reductase, and DT-diaphorase gene expression in human tumors using the polymerase chain reaction. Cancer Res. 52, 108–116.[Abstract]

Hortelano, S., Genaro, A. M., and Bosca, L. (1992). Phorbol esters induce nitric oxide synthase activity in rat hepatocytes. Antagonism with the induction elicited by lipopolysaccharide. J. Biol. Chem. 267, 24937–24940.[Abstract/Free Full Text]

Hortelano, S., Dewez, B., Genaro, A. M., Diaz-Guerra, M. J., and Bosca, L. (1995). Nitric oxide is released in regenerating liver after partial hepatectomy. Hepatology 21, 776–786.[ISI][Medline]

Ignarro, L. J., Buga, G. M., Wood, K. S., Byrns, R. E., and Chaudhuri, G. (1987). Endothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc. Natl. Acad. Sci. U S A 84, 9265–9269.[Abstract]

Jones, B. E., and Czaja, M. J. (1998). Mechanisms of hepatic injury: III. Intracellular signaling in response to toxic liver injury. Am. J. Physiol. 275, G874–878.[Abstract/Free Full Text]

Kass, G. E., Duddy, S. K., and Orrenius, S. (1989). Activation of hepatocyte protein kinase C by redox-cycling quinones. Biochem. J. 260, 499–507.[ISI][Medline]

Kosuga, K., Yui, Y., Hattori, R., Sase, K., Eizawa, H., Aoyama, T., Inoue, R., and Sasayama, S. (1994). Cloning of an inducible nitric oxide synthase from rat polymorphonuclear neutrophils. Endothelium 2, 217–221.

Li, J., Billiar, T. R., Talanian, R. V., and Kim, Y. M. (1997). Nitric oxide reversibly inhibits seven members of the caspase family via S-nitrosylation. Biochem. Biophys. Res. Commun. 240, 419–424.[ISI][Medline]

Lopez-Garcia, M. P. (1998). Endogenous nitric oxide is responsible for the early loss of P450 in cultured rat hepatocytes. FEBS Lett. 438, 145–149.[ISI][Medline]

Misko, T. P., Schilling, R. J., Salvemini, D., Moore, W. M., and Currie, M. G. (1993). A fluorometric assay for the measurement of nitrite in biological samples. Anal. Biochem. 214, 11–16.[ISI][Medline]

Muriel, P. (1998). Nitric oxide protection of rat liver from lipid peroxidation, collagen accumulation, and liver damage induced by carbon tetrachloride. Biochem. Pharmacol. 56, 773–779.[ISI][Medline]

Nicholls-Grzemski, F. A., Tirmenstein, M. A., and Fariss, M. W. (1999). Time-dependent production of nitric oxide by rat hepatocyte suspensions. Biochem. Pharmacol. 57, 1223–1226.[ISI][Medline]

Nudel, U., Zakut, R., Shani, M., Neuman, S., Levy, Z., and Yaffe, D. (1983). The nucleotide sequence of the rat cytoplasmic beta-actin gene. Nucleic Acids Res. 11, 1759–1771.[Abstract]

Nussler, A. K., Geller, D. A., Sweetland, M. A., Di Silvio, M., Billiar, T. R., Madariaga, J. B., Simmons, R. L., Lancaster, J. R., Jr. (1993). Induction of nitric oxide synthesis and its reactions in cultured human and rat hepatocytes stimulated with cytokines plus LPS. Biochem. Biophys. Res. Commun. 194, 826–835.[ISI][Medline]

Palmer, R. M., Ferrige, A. G., and Moncada, S. (1987). Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor. Nature 327, 524–526.[ISI][Medline]

Rando, R. R. (1988). Regulation of protein kinase C activity by lipids. FASEB J. 2, 2348–2355.[Abstract/Free Full Text]

Rossig, L., Fichtlscherer, B., Breitschopf, K., Haendeler, J., Zeiher, A. M., Mulsch A, and Dimmeler, S. (1999). Nitric oxide inhibits caspase-3 by S-nitrosation in vivo. J. Biol. Chem. 274, 6823–6826.[Abstract/Free Full Text]

Schemmer, P., Schoonhoven, R., Swenberg, J. A., Bunzendahl, H., and Thurman, R. G. (1998). Gentle in situ liver manipulation during organ harvest decreases survival after rat liver transplantation: Role of Kupffer cells. Transplantation 65, 1015–1020.[ISI][Medline]

Sergent, O., Griffon, B., Morel, I., Chevanne, M., Dubos, M. P., Cillard, P., Cillard, J. (1997). Effect of nitric oxide on iron-mediated oxidative stress in primary rat hepatocyte culture. Hepatology 25, 122–127.[ISI][Medline]

Wang, H., Gao, X., Fukumoto, S., Tademoto, S., Sato, K., and Hirai, K. (1998). Post-isolation inducible nitric oxide-synthase gene expression due to collagenase buffer perfusion and characterization of the gene regulation in primary cultured murine hepatocytes. J. Biochem. (Tokyo) 124, 892–899.[Abstract]

Wink, D. A., Cook, J. A., Pacelli, R., Liebmann, J., Krishna, M. C., and Mitchell, J. B. (1995). Nitric oxide (NO) protects against cellular damage by reactive oxygen species. Toxicol. Lett. 82/83, 221–226.