The inducible isoform of the nitric oxide synthase (iNOS) is considered as an important part of the host response to infection1. iNOS produces NO that may limit bacterial growth and even prove lethal to invading pathogens2. In addition, NO has been implicated in both acute and chronic models of inflammation including septic shock3, inflammatory arthritis4 and ulcerative colitis5. It is well established that murine macrophages express iNOS when stimulated with cytokines and bacterial cell wall components such as lipopolysaccharide (LPS)1, but it has been difficult to demonstrate NO production in isolated human leukocytes, including monocytes/macrophages6 and neutrophils7. However, other human cells such as hepatocytes8 and epithelial cells9,10 have been shown to express iNOS in vitro.
Urinary tract infections (UTIs) are among the most common bacterial infections in humans and the majority is caused by Escherichia coli. Neutrophils increase their iNOS activity during UTI in humans11, and elevated urinary nitrite concentration, iNOS activity12,13 and increased gaseous NO concentrations14 have been demonstrated in the bladder of patients with UTI. High levels of NO can have detrimental consequences and damage the host tissue, and therefore inhibition of iNOS may suppress tissue destruction resulting from inflammation4,15. Recent studies on uroepithelial cells suggest that NO may participate in urothelial damage by interfering with cellular differentiation and growth16,17. The mechanisms by which NO mediates toxicity include generation of reactive derivatives such as peroxynitrite (ONOO-). The cellular targets of reactive nitrogen derivatives are multiple and involve DNA, proteins and lipids2. Thus, iNOS expression appears to be increased in UTI, but the cellular origin of iNOS remains unclear.
Uroepithelial cells are the first to encounter bacteria, and have been proposed to express iNOS. iNOS immunoreactivity was observed in rat bladder urothelial cells six to nine hours after intraperitoneal challenge with isolated LPS18,19, and human isolated urothelial cells demonstrated iNOS activity after cytokine treatment17. Kidney epithelial cells also have been shown to express iNOS, in animals20,21,22,23, and in humans24,25 when stimulated with cytokines or LPS.
Epithelial cells respond directly to bacterial challenge by producing various pro-inflammatory cytokines, such as IL-6 and IL-826. The NO response might be part of the same cascade, or may occur only after the uroepithelial cells have been stimulated by other pro-inflammatory mediators. The iNOS response to direct challenge of human uroepithelial cells with a uropathogenic strain of E. coli has not yet been tested and few studies, overall, have examined the human epithelial iNOS response to bacteria.
The present study investigated the NO response of isolated primary human renal tubular epithelial cells (HRTEC) to bacterial infection and cytokines, and compared this response to that of monocytes. The effect of NO and oxidative stress on the epithelial cell viability was examined in parallel. The results suggest that epithelial NO production results from the stimulation by inflammatory mediators but not from a direct effect of bacteria on these cells.
METHODS
Reagents
The following cytokines were used: human interferon gamma (IFN-
), recombinant human tumor necrosis factor-
(TNF-), recombinant human interleukin-1
(IL-1), mouse recombinant IFN-, mouse recombinant TNF-, mouse recombinant IL-1 and lipopolysaccharide (LPS) from E. coli serotype 0127:B8 (all from Sigma).
Bacteria
Escherichia coli Hu734 is the lac- mutant of the wild-type pyelonephritis strain GR12, serotype 075:K5:H-27. It is phenotypically positive for type 1 and P fimbriae. E. coli Hu734 was maintained on tryptic soy agar (TSA; Difco, Detroit, MI, USA) plates. For experiments, bacterial colonies from the TSA plate were inoculated in Luria broth, incubated overnight at 37°C, harvested by centrifugation at 4000 rpm for 10 minutes and finally diluted in fresh Dulbecco's modified Eagle's medium (DMEM; 108 CFU/mL). Bacterial multiplication was limited by gentamicin (50 
g/mL). Separate experiments showed that addition of gentamicin did not affect iNOS expression.
Primary cultures of human kidney epithelial cells
Human renal tubular epithelial cells (HRTEC) were isolated from the kidneys of four children who underwent surgery due to hydronephrosis, dysfunction, or reflux nephropathy. The removal of tissue for research purposes was approved by the ethics committee of the Faculty of Medicine, Lund University. Cortical epithelial cells were isolated and cultured as described previously28. At confluence, cells were trypsinized and passaged in Primaria flasks (75 cm2) no more than five times.
The cells were identified as epithelial cells by cytokeratine staining with MNF116 (Dakopatts AB, Stockholm, Sweden) and CAM 5.2 (Becton Dickinson, San Jose, CA, USA) using the alkaline phosphatase anti-alkaline phosphatase technique. The presence of leukocytes was ruled out by lack of reactivity with a monoclonal antibody directed to human leukocyte common antigen. Furthermore, the cells were negative for endothelial cell markers like von Willebrand factor, CD31 and CD34 (all antibodies from Dakopatts AB).
Primary cultures of human renal pelvis epithelial cells
Renal pelvis epithelial cells were isolated from kidneys of children undergoing surgery according to the method described for HRTEC (see above). As the isolated renal pelvis cells grow slowly and cannot be generated in as large quantities as HRTEC, we selected to use them for immunohistochemistry.
Mouse macrophage cell line, RAW 264.7
The mouse macrophage cell line RAW 264.7 (ATCC TIB-71) was obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA). Cells were grown in phenol red-free DMEM (Sigma) supplemented with 10% fetal calf serum (FCS), 2 mmol/L L-glutamine, 1 mmol/L sodium pyruvate, 1 mmol/L non-essential amino acids, 100 U/mL penicillin and 100 g/mL streptomycin (all from Sigma).
Cell stimulation procedure
HRTEC.
The medium from confluent cell cultures was aspired and replaced with fresh medium or medium containing bacteria (108 CFU/mL) and IFN- (400 U/mL), LPS (1g/mL) and IFN-, individual cytokines IL-1 (1 ng/mL), TNF- (25 ng/mL) or IFN-, or a cytokine mixture combined of IL-1, TNF- and IFN-. HRTEC were incubated at 37°C for 24 hours when used for reverse transcription-polymerase chain reaction (RT-PCR) and 48 to 72 hours when used for Western blot analysis, immunohistochemistry and nitrite assay. All cytokines used were of human origin, and the time points were chosen based on highest iNOS expression observed in initial time-course studies.
RAW 264.7.
The medium from confluent cell cultures was aspired and replaced with fresh medium or medium containing bacteria (108 CFU/mL) and IFN- (10 ng/mL), LPS (1g/mL) and IFN-, individual cytokines IL-1 (1 ng/mL), TNF- (25 ng/mL) or IFN-, or a cytokine mixture combined of IL-1, TNF- and IFN-. RAW 264.7 were incubated at 37°C for six hours when used for RT-PCR, and 24 hours when used for Western blot analysis, immunohistochemistry and nitrite assay. All cytokines used were mouse recombinants and the time points were chosen based on the highest iNOS expression observed in the initial time-course studies.
Nitrite assay
Nitric oxide is rapidly converted into the stable end products nitrite and nitrate, and may be used as indirect measures of the amount of NO produced. Nitrite accumulation in culture supernatants was analyzed in duplicate by the Griess assay. Briefly, 50 L of the culture supernatants were mixed with 20 L of water and 100 L of Griess reagent [one part 0.1% N-(1-naphtyl) ethylene-diamine dihydrochloride in water and one part 1% sulfanilamide in 5% concentrated H3PO4; both purchased from Sigma]. The mixture was incubated for five minutes at room temperature and the absorbance was measured at 540 nm (Labsystems Multiscan PLUS; Labsystems AB, Lund, Sweden). The readings were compared to a standard curve for sodium nitrite with a lower detection limit of 1 mol/L nitrite. In order to confirm that the formed nitrite was derived from NOS, cells were incubated with the NOS-inhibitor, NG-monomethyl-L-arginine (L-NMMA), for one hour prior to stimulation with the cytokine mixture.
E. coli is a member of the family Enterobacteriaceae, known to reduce nitrate to nitrite. E. coli Hu734 did not cause nitrite accumulation for up to 72 hours, demonstrating that the observed nitrite accumulation in infected cells was not produced by the bacteria. Furthermore, DMEM was used for the cellular assays because this medium contains lower amounts of nitrate (<1 mol/L) than many other media.
RT-PCR
Total cellular RNA was prepared from HRTEC and RAW 264.7 following the TRIzol® reagent RNA protocol (Life Technologies AB, Täby, Sweden). RT-PCR was performed according to the Perkin Elmer PCR-kit (GeneAmp® RNA PCR kit; Perkin Elmer, Foster City, CA, USA), using 2 g total RNA and with oligo-dT as the first strand primer and MuLV reverse transcriptase according to the manufacturer's instructions. Primers for mouse and human iNOS were obtained from DNA Technology Aps (Aarhus C, Denmark), and were as follows; mouse sense, 5'-CTT CCG AAG TTT CTG GCA GCA GCG-3'; mouse antisense, 5'-GAG CCT CGT GGC TTT GGG CTC CTC-3';29 human sense, 5'-AGA CAT CAA CAA CAA TGT G-3'; and human antisense, 5'-GAC CTG ATG TTG CCA TTG TTG-3',30 amplifying a 487 bp and 658 bp product, respectively. The mouse and human GAPDH primers (also obtained from DNA Technology) were as follows; mouse sense, 5'-GAC GTG CCG CCT GGA GAA AC-3'; mouse antisense, 5'-GGG TCT GGG ATG GAA ATT GTG AG-3' (mouse GAPDH sequence found through GenBank database search); human sense, 5'-ATT CCA TGG CAC CGT CAA GGC T-3'; and human antisense, 5'-TCA GGT CCA CCA CTG ACA CGT T-3'30, amplifying a 389 bp and 571 bp product, respectively. PCR was performed in an automated thermal cycler (OmiGene, Hybaid, Middlesex, UK) one initial step at 95°C for two minutes, followed by 35 cycles at 95°C for 60 seconds, at 58°C for 60 seconds and at 72°C for 60 seconds. Negative controls were performed PCR without template or MuLV reverse transcriptase. PCR products were separated by 2% agarose gel electrophoresis and bands were visualized by ethidium bromide staining.
Western blot analysis
Cells were washed in sterile PBS (pH 7.4), lysed in Laemmli sample buffer and boiled. Protein concentrations were determined with Bio-Rad DC Protein assay (Bio-Rad Laboratories, Hercules, CA, USA) using BSA as a standard. Equal amounts of protein (100 g/lane) were subjected to 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE; Bio-Rad Laboratories) and transferred to a polyvinylidene difluoride (PVDF) – Plus transfer membrane. Unspecific sites were blocked by incubating the membrane in 5% non-fat milk overnight at 4°C. iNOS protein was detected using a rabbit polyclonal antibody raised to murine iNOS (1/1000) or a rabbit polyclonal antibody raised to human iNOS (1/500; both antibodies from Santa Cruz Biotechnology, Santa Cruz, CA, USA) followed by donkey anti-rabbit IgG (1/5000) linked to horseradish peroxide (HRP; Santa Cruz Biotechnology). Blots were developed using enhanced chemiluminescence Western blotting detection reagent (ECL+; Amersham Life Science, Arlington Heights, IL, USA) and exposed to X-ray film (Hyperfilm ECL; Amersham Life Science).
Immunohistochemistry
Inducible nitric oxide synthase-expressing cells were visualized using conventional immunohistochemistry. HRTEC, pelvic epithelial cells and RAW 264.7 were grown in 8-well culture slides (Biocoat Collagen 1 and Falcon glass culture slides, respectively; Becton Dickinson Labware, Bedford, MA). Cells were washed three times in sterile PBS (pH 7.4) and fixed for 15 minutes in cold 4% paraformaldehyde in 0.1 mol/L PBS. Following rinsing, cells were incubated with 0.2% BSA and 0.05% Triton X-100 in PBS for 30 minutes at 37°C, and with the primary antibody for one hour at 37°C.
Immunoreactive products were visualized by incubation for one hour with fluorescein isothiacyanate (FITC)-conjugated donkey anti-rabbit IgG (1/80) (Jackson Immunoresearch Laboratories Inc., West Grove, PA, USA) diluted in PBS containing 0.2% BSA and 0.05% Triton X-100, in the dark at 37°C. The cells were washed and mounted in glycerol with p-phenylenediamine to prevent fluorescence fading. Control experiments showed no immunoreactivity in cells incubated with the secondary antibody.
Micrographs of the immunolabeled cells were obtained using a digital camera system (Nikon E400 microscope and the Optronix DEI-750 camera), and the pictures were captured using appropriate filter settings for FITC. Adobe® Photoshop™ was used for image handling, and the three-color channels were handled separately. Only the background level, contrast and brightness of the entire image were changed in the final picture.
Cell viability
The cell viability was assayed by the mitochondrial dependent reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma) to formazan31. HRTEC and RAW 264.7, cultured in sterile 96-well plates, were stimulated in duplicate as previously described. After stimulation, 50 L of the culture medium was saved for nitrite determination and the remaining culture medium removed. The cells were incubated with 20 L of MTT solution (5 mg/mL) for one hour at 37°C. The MTT solution was removed and the cells were solubilized in 100 L of dimethylsulphoxide (DMSO; Sigma) with shaking for five minutes. The extent of MTT reduction to formazan was measured as the absorbance at 540 nm. Results are expressed as a ratio of stimulated compared to control cells.
In addition, the cell viability in response to exogenously applied NO (DETA NONOate; 10 to 500 mol/L; Alexis Biochemicals, Lausen, Switzerland), peroxynitrite (SIN-1; 10 to 500 mol/L; Casella AG, Frankfurt am Main, Germany) and H2O2 (10 to 500 mol/L; Sigma) was examined. The dependence of NOS and the second messenger cGMP on cell viability was examined by pretreating the cells for one hour with the NOS-inhibitor L-NMMA (1 to 2 mmol/L; Calbiochem, La Jolla, CA, USA) or the guanylate cyclase inhibitor ODQ (5 mol/L; Tocris Cookson Inc., St Louis, MO, USA), respectively.
Analysis of data
Data are presented as means 
SEM. The Student unpaired t test was used to compare two means and ANOVA followed by the Bonferroni-Dunn test was used for multiple comparisons. P < 0.05 was considered statistically significant.
RESULTS
E. coli Hu734 and LPS do not induce nitrite production, iNOS mRNA or protein expression in HRTEC
The iNOS response of HRTEC to uropathogenic E. coli or LPS was analyzed using Western blot, RT-PCR and nitrite measurements. Nitrite production was not increased in HRTEC exposed for 72 hours to E. coli Hu734 or LPS alone or in combination with IFN- (Hu734/IFN-; 4.6 0.64 mol/L; N = 5; LPS/IFN-; 2.8 0.23 mol/L; N = 5), when compared to unstimulated control cells (2.9 0.3 mol/L; N = 5; Figure 1). Stimulation of HRTEC with E. coli Hu734/IFN- or LPS/IFN- did not increase iNOS mRNA expression Figure 2a or iNOS protein expression Figure 3a. Furthermore, no iNOS immunoreactivity was observed by morphological studies in unstimulated cells Figure 4a or HRTEC exposed to LPS/IFN- and E. coli Hu734/IFN- Figure 4 b, c. Epithelial cells isolated from the renal pelvis area showed no iNOS immunoreactivity before Figure 5a or after stimulation with E. coli Hu734/IFN-Figure 5b.
Figure 1.
Nitrite concentration in cell culture supernatants of human renal tubular epithelial cells (HRTEC). The cells were stimulated for 72 hours with various cytokines, lipopolysaccharide (LPS) or E. coli Hu734 as indicated. The Cytokine mixture (CM) was comprised of interleukin-1 (IL-1)/tumor necrosis factor- (TNF-)/interferon- (IFN-). Data are expressed as mean SEM. Statistical comparison of control vs. treated cells ***P < 0.001 (N = 3 to 7).
Figure 2.
Reverse transcription-polymerase chain reaction (RT-PCR) analysis of inducible nitric oxide synthase (iNOS) mRNA expression in (A, B) HRTEC and (C) RAW 264.7 stimulated for 6 and 24 hours as indicated. Abbreviations are: C, control; CM (cytokine mixture) comprised of IL-1/TNF-/IFN-. GAPDH mRNA expression was similar in control and stimulated cells.
Figure 3.
Western blot analysis of iNOS protein expression in (A) HRTEC stimulated for 72 hours and (B) RAW 264.7 stimulated for 24 hours with cytokines, LPS orE. coli Hu734. Abbreviations are: C, control; CM (cytokine mixture) comprised of IL-1/TNF-/IFN-; MW, molecular weight.
Figure 4.
Immunohistochemical demonstration of iNOS expression in HRTEC stimulated for 48 to 72 hours. (A) Unstimulated cells showed no iNOS expression. (B) Cells stimulated with LPS/IFN- showed no iNOS expression. (C) Cells stimulated with E. coli Hu734/IFN- showed no iNOS expression. (D) Cells stimulated with IL-1/IFN- were iNOS positive. (E and F) Cells stimulated with a cytokine mixture of IL-1/TNF-/IFN- showed iNOS expression. Note the altered cellular morphology of iNOS positive cells characterized by long extensions. Scale bars (A, B, C, E) = 15 m, (D, F) = 10 m.
Figure 5.
Immunohistochemical demonstration of iNOS in pelvis epithelial cells stimulated for 72 hours. (A) Unstimulated cells showed no iNOS expression. (B) Cells stimulated with E. coli Hu734/IFN- showed no iNOS expression. (C) Cells stimulated with a cytokine mixture of IL-1/TNF-/IFN- were iNOS positive. Scale bars = 15 m.
Stimulation of the macrophage cell line RAW 264.7 with LPS/IFN- or E. coli Hu734/IFN- caused a marked increase in iNOS mRNA Figure 2c and protein expression Figure 3a after 6 and 24 hours, respectively. These experiments confirmed that LPS and E. coli Hu734 delivered a signal to the macrophages, but that HRTEC were refractory to E. coli Hu734 and LPS in combination with IFN-.
IL-1
in combination with IFN-
induce nitrite production, iNOS mRNA and protein expression in HRTEC
The iNOS response of HRTEC to different cytokines and cytokine mixtures was investigated. Supernatants of HRTEC exposed to a cytokine mixture (IL-1/TNF-/IFN-) for 72 hours showed increased nitrite accumulation (16 2.4 mol/L; N = 5) compared to unstimulated cells (2.9 0.3 mol/L; N = 5; P < 0.001; Figure 1). The combination of IL-1 and IFN- (10 1.4 mol/L; N = 5) stimulated nitrite production, but the individual cytokines and the combination of TNF- and IFN- (3.0 0.16 mol/L; N = 5) did not. The results suggest that the IL-1–induced nitrite production was dependent on the presence of IFN-Figure 1. Unstimulated cells or cells stimulated with IL-1 alone showed no iNOS mRNA as revealed by RT-PCR. Stimulation of HRTEC with the combination of IL-1/IFN- or cytokine mixture, but not TNF-/IFN-, caused an increase in iNOS mRNA expression after 24 but not 6 hours Figure 2 a, b. To confirm iNOS induction at the protein level a Western blot analysis was performed. No iNOS protein expression was detected in unstimulated cells or in cells stimulated with IL-1 alone or TNF-/IFN-Figure 3a. HRTEC stimulated with IL-1/IFN- or the cytokine mixture gave a positive band at 
130 kD, corresponding to the size of iNOS protein Figure 3a. Immunohistochemistry was performed to visualize iNOS expressing cells. Unstimulated cells did not show any iNOS labeling Figure 4a, whereas iNOS immunoreactivity was demonstrated in HRTEC stimulated with IL-1/IFN-Figure 4d and the cytokine mixture Figure 4 e, f. In addition, iNOS immunoreactivity was found in epithelial cells isolated from the renal pelvis area when stimulated with cytokines Figure 5c. Notably, iNOS positive cells showed changes in morphology characterized by thin podial-like extensions Figures 4e and 5c. RAW 264.7 cells stimulated with the cytokine mixture showed distinct iNOS mRNA Figure 2c and protein expression Figure 3b.
The cell viability in iNOS expressing cultures is decreased independent of NO
The effects of iNOS expression/NO production on the cell viability were examined by a viability assay based on formazan formation from MTT. Nitrite concentrations were determined in the same wells by Griess assay. HRTEC stimulated for 72 hours with a cytokine mixture (IL-1/TNF-/IFN-) showed a significant (P < 0.001) increase in nitrite production, and cell viability was reduced by 28 6.3%; N = 9 (P < 0.001) compared to unstimulated control cells Figure 6a. In contrast, stimulation with LPS/IFN- and E. coli Hu734/IFN- caused no significant increase in nitrite accumulation or reduction in cell viability Figure 6a.
Figure 6.
Relationship between nitrite levels and cell viability demonstrated in (A, B) HRTEC and (C, D) RAW 264.7. The cells were stimulated for 72 and 24 hours, respectively, with CM, LPS/IFN- or E. coli Hu734/IFN- as indicated. Cytokine mixture (CM) was comprised of IL-1/TNF-/IFN-. The bar graphs show the nitrite levels and the line graphs the cell viability. The cell viability in control cells is set to 100%. In B and D, the involvement of NOS activity for cell viability was studied after treatment with the NOS-inhibitor L-NMMA (1-2 mmol/L). The cell viability in L-NMMA treated control cells is set to 100%. Data are expressed as mean SEM (N = 5 to 9).
To elucidate whether the decreased cell viability involved NOS activation, the NOS-inhibitor L-NMMA (1-2 mmol/L) was added before stimulation with cytokines. L-NMMA prevented cytokine-induced increases in nitrite production, which confirmed that NOS was inhibited Figure 6b, but the cell viability was still reduced to the same extent (25 and 30%, respectively) as in the absence of L-NMMA Figure 6b. Thus, the cytokine-induced decrease in cell viability in HRTEC was independent of NOS.
RAW 264.7 macrophages showed a pronounced increase in nitrite production (P < 0.001) and a significant (P < 0.001) decrease in cell viability by 83 0.4%, 85 2.1%, 78 1.8% (N = 6) after stimulation with the cytokine mixture, LPS/IFN- and E. coli Hu734/IFN-, respectively Figure 6c. The LPS-induced decrease in cell viability was less pronounced in the presence of L-NMMA (1 to 2 mmol/L; Figure 6d), suggesting that the decrease in RAW 264.7 cell viability involved NOS. The production of nitrite was not completely inhibited by L-NMMA, probably because of the high concentration of LPS used.
Exogenous NO reduces cell viability more in RAW 264.7 than in HRTEC
Since the maximal concentration of nitrite produced by HRTEC (15 to 20 mol/L) is three- to fourfold lower than the concentration produced by RAW 264.7 (60 to 70 mol/L), it is possible that the lower viability of RAW 264.7 compared with HRTEC was related to the NO concentration. Therefore, similar concentrations of the NO-donor DETA/NO (10 to 500 mol/L) were applied to HRTEC and RAW 264.7 for 24 hours and the viability was compared by the MTT assay. DETA/NO spontaneously releases NO and provides a constant NO supply over hours32. Both HRTEC and RAW 264.7 were insensitive to low concentrations (
100 mol/L) of DETA/NO, and decreasing viability was only seen at the highest concentrations Table 1. RAW 264.7 was more sensitive to DETA/NO than HRTEC (P < 0.001). These experiments demonstrated that approximately 120 mol/L nitrite derived from DETA/NO was needed to obtain a similar decrease in HRTEC viability as the decrease produced by 15 to 20 mol/L endogenously produced nitrite. Thus, although exogenous NO has some effect on viability in HRTEC, the mediator responsible for the cytokine-induced decreased in cell viability is not likely to be NO.
Table 1 - Effect of DETA/NO stimulation for 24 hours on cell viability and nitrite accumulation.
Full table The involvement of the second messenger cGMP on cell viability was tested by inhibition of guanylate cyclase by ODQ (5 mol/L). ODQ had no effect on the medium nitrite levels. Pretreatment with ODQ did not affect the DETA/NO-induced decrease in cell viability in HRTEC or RAW 264.7 (data not shown).
H2O2, but not SIN-1, affects the cell viability in HRTEC
Since NO did not explain the decrease in HRTEC viability we investigated SIN-1, a peroxynitrite donor, and H2O2. SIN-1 caused an increase in nitrite accumulation, but had no significant effect on cell viability in HRTEC and RAW 264.7 Table 2. In contrast, oxidative stress induced by H2O2 decreased the cell viability in HRTEC, with a maximum decrease of 58 7.1% (N = 6) at 500 mol/L Table 3, but RAW 264.7 was not significantly affected by H2O2 (P < 0.05; Table 3). H2O2 did not increase nitrite levels in the culture medium. Thus, oxidative stress induced by H2O2 was found to have more detrimental effects on cell viability in uroepithelial cells than NO and nitrogen species.
Table 2 - Effect of peroxynitrite (SIN-1) stimulation for 24 hours on cell viability and nitrite accumulation.
Full table Table 3 - Effect of H2O2 stimulation for 24 hours on cell viability and nitrite accumulation.
Full table DISCUSSION
The present study examined the iNOS response of human kidney uroepithelial cells to E. coli Hu734, a human wild-type pyelonephritis strain, LPS and inflammatory cytokines. The bacteria did not trigger a response as no iNOS induction was found for up to 72 hours, no increase in the mRNA was observed, and no nitrite was produced. Similarly, the cells did not respond to LPS, even though both of these agonists caused a response in macrophages. The epithelial cells were able to increase their iNOS expression and nitrite production, however, if stimulated with the cytokine IL-1 in combination with IFN-. Our results suggest that NO may be produced by epithelial cells in response to cytokine stimulation but that bacteria failed to elicit such a response. Furthermore, our results emphasize the difference between epithelial cells and macrophages in this regard.
Studies using murine inner medullary collecting duct cells21 or other kidney cells22,33 have demonstrated increased iNOS mRNA and/or protein expression after LPS treatment, but we found no iNOS expression in HRTEC when stimulated by LPS. This may be another example of species-specific differences in LPS-induced iNOS expression, as rat and hamster alveolar macrophages express iNOS, but not human cells34. This hypo-responsiveness to LPS in human cells has been explained at the transcriptional level by a lack of a LPS-inducible NF-
B complex in the human iNOS promoter35. In addition, the cytokine response to LPS is known to be poor in most non-immune cells, including human uroepithelial cells36, due to the lack of specific LPS receptors like CD14.
Most studies have reported iNOS induction in human epithelial cells when stimulated with LPS or cytokines, and only a few studies using intestinal epithelial cells have examined iNOS expression in response to bacteria. Invasive bacteria were found to activate iNOS expression in human intestinal epithelial cells37,38,39, but the intestinal and uroepithelial cells are known to differ greatly in response to bacteria. Bacterial invasion is needed to trigger a cytokine response in intestinal epithelial cells,40 but not for the uroepithelial cytokine response36. Uropathogenic E. coli express several virulence factors that influence their ability to trigger a host response. Under the growth conditions used in this study, they attach to urothelial cells through surface fimbriae, like P and type 1 fimbriae26. Both P and type 1 fimbriated E. coli are able to enhance the human uroepithelial cell cytokine response in vitro41,42,43. E. coli Hu734 has been shown to induce IL-8 production in HRTEC (unpublished observations; Godaly) under similar conditions as those in the present study, with a peak after 6 to 12 hours. The IL-8 and iNOS responses to bacterial activation in uroepithelial cells differ from the response in intestinal epithelium. In human intestinal epithelial cells, both IL-840 and iNOS37 were induced by bacterial activation, whereas bacterial activation induced IL-842, but not iNOS in kidney uroepithelial cells. The lack of iNOS induction in HRTEC by E. coli Hu734 suggest that the signaling pathways triggered by the bacterial fimbriae do not cause iNOS transcription. Indeed, additional data from studies using a human kidney epithelial cell line, A498, confirm that other strains of E. coli such as recombinant strains expressing either the P or type 1 fimbriae do not activate iNOS expression (unpublished observations).
Our recent in vivo study examined iNOS expression in a mouse UTI model44. Uropathogenic E. coli strain caused iNOS up-regulation in polymorphonuclear (PMN) cells within four to six hours after bacterial challenge. iNOS was detected in the transitional and columnar epithelial cells lining the renal pelvis 12 hours after bacterial challenge, and in the tubular epithelial cells in the outermost parts of cortex after 72 hours44. The delay suggests that iNOS expression in uroepithelial cells was triggered by inflammatory mediators released during mucosal infection, rather than a direct result of bacterial/epithelial interactions. Indeed, this hypothesis was supported by findings in the present study using isolated HRTEC. Stimulation of HRTEC with inflammatory cytokines increased nitrite production and iNOS mRNA and protein expression, but direct contact with bacteria did not.
When the individual cytokines were examined separately, it was found that IL-1, but not TNF-, increased iNOS expression in HRTEC. In animal studies, TNF- caused iNOS up-regulation in cultures of rat proximal tubules and inner medullary collecting duct cells20, and in mouse inner medullary collecting duct cells21. In agreement with our results, iNOS up-regulation in human alveolar epithelium-like cancer cells was found when stimulated with IL-1, but not TNF-45. Again, these differences are likely to reflect species variations in regulation of the iNOS gene, since different localization of the cytokine responsive elements have been described for the human and mouse iNOS promotor45,46. Moreover, iNOS expression in HRTEC was dependent on IFN-, suggesting that IFN--responsive transcription factors are necessary for induction of the human iNOS gene.
Patients with UTI have elevated iNOS activity and expression in neutrophil-enriched fractions of urine compared with noninfected controls11. Neutrophils are important for the antibacterial defense of the urinary tract, and are the main cells involved in the initial stages of the inflammatory response26. However, the contribution of neutrophil-derived NO to bacterial clearance from the urinary tract is not known. Many UTIs resolve spontaneously, without antibiotic treatment, which suggests that an endogenous factor, like NO, may play a role. Early studies have suggested that uroepithelial cells produce a bactericidal factor for E. coli47. It is not known whether NO produced from the uroepithelial cells during UTI is bactericidal or if other functions are attributed to urothelial NO production. Given the delayed iNOS expression in HRTEC, most bacteria may already be cleared by neutrophil mediated phagocytos before the uroepithelial cells express iNOS. Thus, NO production by uroepithelial cells and by inflammatory cells are likely to play different roles in response to bacterial infection. The mechanisms of NO-related antimicrobial activity have been shown to involve formation of bactericidal peroxynitrite, a reaction product of NO and phagocyte-derived superoxide anion48. Lundberg and co-workers suggested that nitrite-producing bacteria induce their own death in urine by supplying substrate for generation of bacteriostatic NO49. In an experimental model of pyelonephritis, inhibition of NO production induced a greater renal infection in LPS-nonresponder C3H/HeJ mice compared to LPS-responder C3H/HeN mice50. These results suggest that adequate NO production and LPS responsiveness work synergistically to provide a mechanism of renal resistance to bacterial infection50.
Bacterial colonization in vivo causes shedding of infected and damaged urothelial cells51,52. A delayed iNOS induction may play a role in the later stage of inflammation by removing infected and damaged urothelial cells. We compared the effect of microbial products and cytokines on cell viability. Stimulation with E. coli Hu734 and LPS had no effect on cell viability in HRTEC, but cytokines caused a decrease in cell viability. We speculated that NO might be the endogenous factor responsible for the decreased viability, since an effect on viability was only observed in iNOS-expressing cell cultures. Experiments with a NOS-inhibitor showed, however, that the cytokine-induced decrease in cell viability was not associated with NO. Previously, NO or peroxynitrite have been found to modulate cell growth and differentiation16,17 and to impair the cell-matrix adhesion properties of uroepithelial cell lines without affecting cell death53. Thus, NO/peroxynitrite may preferentially affect the cell adhesion properties of uroepithelial cells and exert cytostatic rather than cytotoxic effects.
Nitric oxide was more harmful to RAW 264.7 than to HRTEC. Experiments with a NOS-inhibitor showed that the decrease in RAW 264.7 viability was associated with an increase in NO production, while the decrease in HRTEC viability was unrelated to NO. It may be speculated that uroepithelial cells have a protective mechanism, like reported in some other cells54, which enables them to resist the damaging effects of NO. The consequences of NO exposure on cell viability are related to the non-heme iron content of the cells. Cells with low non-heme iron levels, such as RAW 264.7, are more sensitive to NO than cells with high non-heme iron levels, such as hepatocytes54. In addition, both protective and toxic effects of NO may involve activation of soluble guanylate cyclase and cGMP formation55. When investigating the involvement of guanylate cyclase for the DETA/NO-induced decrease in cell viability, we found that the cytotoxic capacity of NO was independent of guanylate cyclase in both HRTEC and RAW 264.7.
Since the generation of NO could not explain the decrease in HRTEC viability other possible factors were investigated. SIN-1, known to release ONOO- during breakdown56, had practically no effect on cell viability in HRTEC; however, the actual amount of ONOO- generated from SIN-1 in our cell system is unknown. On the other hand, since treatment with a NOS-inhibitor was without effect on cell viability it is unlikely that NO-derived ONOO- is involved in the cytokine-triggered decrease in HRTEC viability. Another candidate may be H2O2, produced as a reaction between 2O2- and 2H+. Several cell types, including macrophages, neutrophils and epithelial cells, generate large amounts of O2- during inflammation from enzymes such as xanthine oxidase and NADPH oxidase57. Oxidative stress induced by H2O2 was demonstrated to have more profound effects on cell viability in uroepithelial cells than NO and nitrogen species. More studies are needed to identify the endogenous factor(s) associated with the cytokine-triggered decrease in HRTEC viability.
In conclusion, we showed iNOS induction in HRTEC when stimulated with inflammatory cytokines but not when stimulated with uropathogenic bacteria or LPS. This suggests that the epithelial iNOS expression in the human kidney is not induced by bacterial activation factors, but rather by inflammatory mediators. Epithelial NO thus may play a role in the later stages of inflammation.
References
| 1. | Nussler AK & Billiar TR. Inflammation, immunoregulation and inducible nitric oxide synthase. J Leukocyte Biol 1993; 54: 171–178. | PubMed | ISI | ChemPort | |
| 2. | Fang FC. Mechanisms of nitric oxide-related antimicrobial activity. J Clin Invest 1997; 99: 2818–2825. | PubMed | ISI | ChemPort | |
| 3. | Petros A, Bennett D & Vallance P. Effect of nitric oxide synthase inhibitors on hypertension in patients with septic chock. Lancet 1991; 338: 1557–1558. | Article | PubMed | ISI | ChemPort | |
| 4. | McCartney-Francis N, Allen JB & Mizel DE et al. Suppression of arthritis by an inhibitor of nitric oxide synthase. J Exp Med 1993; 178: 749–454. | Article | PubMed | ChemPort | |
| 5. | Middleton SJ, Shorthouse M & Hunter J. Increased nitric oxide synthesis in ulcerative colitis. Lancet 1993; 341: 465–466. | Article | PubMed | ISI | ChemPort | |
| 6. | Denis M. Human monocytes/macrophages: NO or no NO. J Leukocyte Biol 1994; 55: 682–684. | PubMed | ISI | ChemPort | |
| 7. | Yan L, Vandivier RW & Suffredini AF et al. Human polymorphonuclear leukocytes lack detectable nitric oxide synthase activity. J Immunol 1994; 153: 1825–1834. | PubMed | ISI | ChemPort | |
| 8. | Nussler AK, Di Silvio M & Billiar TR et al. Stimulation of the nitric oxide synthase pathway in human hepatocytes by cytokines and endotoxin. J Exp Med 1992; 176: 261–264. | Article | PubMed | ISI | ChemPort | |
| 9. | Guo FH, Uetani K & Haque SJ et al. Interferon gamma and interleukin 4 stimulate prolonged expression of inducible nitric oxide synthase in human airway epithelium through synthesis of soluble mediators. J Clin Invest 1997; 100: 829–838. | PubMed | ISI | ChemPort | |
| 10. | Linn SC, Morelli PJ & Edry I et al. Transcriptional regulation of human inducible nitric oxide synthase gene in an intestinal epithelial cell line. Am J Physiol 1997; 272: G1499–G1508. | PubMed | ISI | ChemPort | |
| 11. | Wheeler MA, Smith SD & Garcia-Gardena G et al. Bacterial infection induces nitric oxide synthase in human neutrophils. J Clin Invest 1997; 99: 110–116. | PubMed | ISI | ChemPort | |
| 12. | Smith SD, Wheeler MA & Weiss RM. Nitric oxide synthase: An endogenous source of elevated nitrite in infected urine. Kidney Int 1994; 45: 586–591. | PubMed | ISI | ChemPort | |
| 13. | Smith SD, Wheeler MA & Foster HE et al. Urinary nitric oxide synthase activity and cyclic GMP levels are decreased with interstitial cystitis and increased with urinary tract infections. J Urol 1996; 155: 1432–1435. | PubMed | ISI | ChemPort | |
| 14. | Lundberg JON, Ehren I & Jansson O et al. Elevated nitric oxide in the urinary bladder in infectious and noninfectious cystitis. Urology 1996; 48: 700–702 10.1016/s0090-4295(96)00423-2. | Article | PubMed | ISI | ChemPort | |
| 15. | Hierholzer C, Harbrecht B & Menezes JM et al. Essential role of induced nitric oxide in the initiation of the inflammatory response after hemorrhagic shock. J Exp Med 1998; 187: 917–928. | Article | PubMed | ISI | ChemPort | |
| 16. | Elgavish A, Robert B & Lloyd K et al. Nitric oxide mediates the action of lipotechoic acid on the function of human urothelial cells. J Cell Physiol 1996; 169: 66–77 10.1002/(sici)1097-4652(199610)169:1<66::aid-jcp7>3.3.co;2-7. | Article | PubMed | ISI | ChemPort | |
| 17. | Morcos E, Jansson OT & Adolfsson J et al. Endogenously formed nitric oxide modulates cell growth in bladder cancer cell lines. Urology 1999; 53: 1252–1257 10.1016/s0090-4295(99)00033-3. | Article | PubMed | ISI | ChemPort | |
| 18. | Cook HT, Bune AJ & Jansen AS et al. Cellular localization of inducible nitric oxide synthase in experimental endotoxic shock in the rat. Clin Sci 1994; 87: 179–186. | PubMed | ISI | ChemPort | |
| 19. | Persson K, Poljakovic M & Johansson K et al. Morphological and biochemical investigation of nitric oxide synthase and related enzymes in the rat and pig urothelium. J Histochem Cytochem 1999; 47: 739–750. | PubMed | ISI | ChemPort | |
| 20. | Markowitz BA, Michael JR & Kohan DE. Cytokine-induced expression of a nitric oxide synthase in rat renal tubule cells. J Clin Invest 1993; 91: 2138–2143. | PubMed | ChemPort | |
| 21. | Mohaupt MG, Schwöbel J & Elzie JL et al. Cytokines activate inducible nitric oxide synthase gene transcription in inner medullary collecting duct cells. Am J Physiol (Renal Fluid Electrolyte Physiol) 1995; 268: F770–F777. | ChemPort | |
| 22. | Amoah-Apraku B, Chandler LJ & Harrison JK et al. NF- B and transcriptional control of renal epithelial-inducible nitric oxide synthase. Kidney Int 1995; 48: 674–682. | PubMed | ChemPort | |
| 23. | Kone B, Schwöbel J & Turner P et al. Role of NF-B in the regulation of inducible nitric oxide synthase in an MTAL cell line. Am J Physiol 1995; 269 Renal Fluid Electrolyte Physiol 38: F718–F729. | PubMed | ISI | ChemPort | |
| 24. | Jansson OT, Morcos E & Brundin L et al. Nitric oxide synthase activity in human renal cell carcinoma. J Urol 1998; 160: 556–560. | Article | PubMed | ISI | ChemPort | |
| 25. | Chatterjee PK, Hawksworth GM & McLay JS. Cytokine-stimulated nitric oxide production in the human renal proximal tubule and its modulation by natriuretic peptides: A novel immunomodulatory mechanism. Exp Nephrol 1999; 7: 438–448. | Article | PubMed | ISI | ChemPort | |
| 26. | Svanborg C, Godaly G & Hedlund M. Cytokine responses during mucosal infections: Role in disease pathogenesis and host defense. Curr Opin Microbiol 1999; 2: 99–105. | Article | PubMed | ISI | ChemPort | |
| 27. | Svanborg-Edén C, Freter R & Hagberg L et al. Inhibition of experimental ascending urinary tract infection by receptor analogue. Nature 1982; 298: 560–562. | Article | PubMed | ISI | |
| 28. | Karpman D, Hakansson A & Perez MT et al. Apoptosis of renal cortical cells in the hemolytic-uremic syndrome: in vivo and in vitro studies. Infect Immun 1998; 66: 636–644. | PubMed | ISI | ChemPort | |
| 29. | Baskin H, Ellermann-Eriksen S & Lovmand J et al. Herpes simplex virus type 2 synergizes with interferon-gamma in the induction of nitric oxide production in mouse macrophages through autocrine secretion of tumour necrosis factor-alpha. J Gen Virol 1997; 78: 195–203. | PubMed | ISI | ChemPort | |
| 30. | Adcock IM, Brown CR & Kwon O et al. Oxidative stress induces NF-B DNA binding and inducible NOS mRNA in human epithelial cells. Biochem Biophys Res Commun 1994; 199: 1518–1524 10.1006/bbrc.1994.1403. | Article | PubMed | ISI | ChemPort | |
| 31. | Mosmann T. Rapid colorimetric assay for cellular growth and survival: Application to proliferation and cytotoxicity assays. J Immunol Methods 1983; 65: 55–63. | Article | PubMed | ISI | ChemPort | |
| 32. | Maragos CM, Wang JM & Hrabie JA et al. Nitric oxide/nucleophile complexes inhibit the in vitro proliferation of A375 melanoma cells via nitric oxide release. Cancer Res 1993; 53: 564–568. | PubMed | ISI | ChemPort | |
| 33. | Trachtman H, Futterweit S & Singhal PC et al. Renal tubular epithelial cell–E. coli interaction products stimulate nitric oxide production in cultured rat renal medullary interstitial and mesangial cells. Res Commun Mol Pathol Pharmacol 1996; 94: 227–238. | PubMed | ISI | ChemPort | |
| 34. | Jesch NK, Dorger M & Enders G et al. Expression of inducible nitric oxide synthase and formation of nitric oxide by alveolar macrophages: An interspecies comparison. Environ Health Perspect 1997; 105: 1297–1300. | PubMed | ISI | ChemPort | |
| 35. | Zhang X, Laubach VE & Alley EW et al. Transcriptional basis for hyporesponsiveness of the human inducible nitric oxide synthase gene to lipopolysaccharide/interferon-gamma. J Leukocyte Biol 1996; 59: 575–585. | PubMed | ISI | ChemPort | |
| 36. | Hedlund M, Svensson M-L & Hang L et al. Bacterial adherence, transmembrane signaling and epithelial cytokine responses,. inRecent Advances in the Pathogenesis of Gastrointestinal Bacterial Infections 1998; edited by Rampal P, Boquet P Paris, John Libbey Eurotext pp 65–79. |
| 37. | Witthöft T, Eckmann L & Kim JM et al. Enteroinvasive bacteria directly activate expression of iNOS and NO production in human colon epithelial cells. Am J Physiol 1998; 275: G565–G571. |
| 38. | Li CK, Seth R & Gray T et al. Production of proinflammatory cytokines and inflammatory mediators in human intestinal epithelial cells after invasion by Trichinella spiralis. Infect Immun 1998; 66: 2200–2206. | PubMed | ISI | ChemPort | |
| 39. | Kim YM, Kim TH & Seol DW et al. Nitric oxide suppression of apoptosis occurs in association with an inhibition of Bcl-2 cleavage and cytochrome c release. J Biol Chem 1998; 273: 31437–31441. | Article | PubMed | ISI | ChemPort | |
| 40. | Eckmann L, Kagnoff MF & Fierer J. Epithelial cells secrete the chemokine interleukin-8 in response to bacterial entry. Infect Immun 1993; 61: 4569–4574. | PubMed | ISI | ChemPort | |
| 41. | Hedges S, Svensson M & Svanborg C. Interleukin-6 response of epithelial cell lines to bacterial stimulation in vitro. Infect Immun 1992; 60: 1295–1301. | PubMed | ISI | ChemPort | |
| 42. | Hedges S, Agace W & Svensson M et al. Uroepithelial cells are part of a mucosal cytokine network. Infect Immun 1994; 62: 2315–2321. | PubMed | ISI | ChemPort | |
| 43. | Svensson M, Linstedt R & Radin NS et al. Epithelial glucosphingolipid expression as a determinant of bacterial adherence and cytokine production. Infect Immun 1994; 62: 4404–4410. | PubMed | ISI | ChemPort | |
| 44. | Poljakovic M, Svensson ML & Svanborg C et al. Escherichia coli-induced inducible nitric oxide synthase and cyclooxygenase expression in the mouse bladder and kidney. Kidney Int 2001; 59: 893–904 10.1046/j.1523-1755.2001.059003893.x. | Article | PubMed | ISI | ChemPort | |
| 45. | Chu SC, Marks-Konczalik J & Wu H-P et al. Analysis of cytokine-stimulated human inducible nitric oxide synthase (iNOS) gene. Characterization of differences between human and mouse iNOS promotors. Biochem Biophys Res Commm 1998; 248: 871–878. | ChemPort | |
| 46. | de Vera ME, Shapiro RA & Nussler AK et al. Transcriptional regulation of human inducible nitric oxide synthase (NOS2) gene by cytokines: Initial analysis of the human NOS2 promoter. Proc Natl Acad Sci USA 1996; 93: 1054–1059. | Article | PubMed | ChemPort | |
| 47. | Schulte-Wissermann H, Mannhardt W & Schwarz J et al. Comparison of the antibacterial effect of uroepithelial cells from healthy donors and children with asymptomatic bacteriuria. Eur J Pediatr 1985; 144: 230–233. | Article | PubMed | ChemPort | |
| 48. | Evans TJ, Buttery LDK & Carpenter A et al. Cytokine-treated human neutrophils contain inducible nitric oxide synthase that produces nitration of ingested bacteria. Proc Natl Acad Sci USA 1996; 93: 9553–9558. | Article | PubMed | ChemPort | |
| 49. | Lundberg JO, Carlsson S & Engstrand L et al. Urinary nitrite: More than a marker of infection. Urology 1997; 50: 189–191 10.1016/s0090-4295(97)00257-4. | Article | PubMed | ISI | ChemPort | |
| 50. | Nowicki B, Singhal J & Fang L et al. Inverse relationship between severity of experimental pyelonephritis and nitric oxide production in C3H/HeJ mice. Infect Immun 1999; 67: 2421–2427. | PubMed | ISI | ChemPort | |
| 51. | Aronson M, Medalia O & Amichay D et al. Endotoxin-induced shedding of viable uroepithelial cells is an antimicrobial defense mechanism. Infect Immun 1988; 56: 1615–1617. | PubMed | ISI | ChemPort | |
| 52. | Mulvey MA, Lopez-Boado YS & Wilson CL et al. Induction and evasion of host defense by type 1-piliated uropathogenic Escherichia coli. Science 1998; 282: 1494–1497. | Article | PubMed | ISI | ChemPort | |
| 53. | Wangsiripaisan A, Gengaro PE & Nemenoff RA et al. Effect of nitric oxide donors on renal tubular epithelial cell-matrix adhesion. Kidney Int 1999; 55: 2281–2288 10.1046/j.1523-1755.1999.00484.x. | Article | PubMed | ISI | ChemPort | |
| 54. | Kim YM, Chung HT & Simmons RL et al. Cellular non-heme iron content is a determinant of nitric oxide-mediated apoptosis, necrosis, and caspase inhibition. J Biol Chem 2000; 275: 10954–10961. | Article | PubMed | ISI | ChemPort | |
| 55. | Brune B, Messmer UK & Sandau K. The role of nitric oxide in cell injury. Toxicol Lett 1995; 82–83: 233–237. |
| 56. | Beckman JS. Oxidative damage and tyrosine nitration from peroxynitrite. Chem Res Toxicol 1996; 9: 836–844. | Article | PubMed | ISI | ChemPort | |
| 57. | Maeda H & Akaike T. Nitric oxide and oxygen radicals in infection, inflammation and cancer. Biochemistry 1998; 63: 854–865. | PubMed | ISI | ChemPort | |
Acknowledgments
This project was supported by the Swedish Medical Research Council (12601, 7934), the Royal Physiographic Society, the National Board of Health and Welfare, the Swedish Society of Medicine, the Foundations of Crafoord, Magnus Bergwall and Memorial Lars Hierta. A preliminary report has previously been published in abstract form (Acta Physiologica Scandinavia 167(Suppl 645):48, 1999). Dr Kjell Johansson is acknowledged for help with the image handling.
