This article is mentioned in:

Introduction

At low levels, nitric oxide (NO•) is a signaling molecule required for many physiological functions. However, when produced in excessive amounts, for example, by inflammatory cells, it can cause cell death and mutagenicity, as demonstrated by numerous experiments in cultured cells and experimental animal models (1,2). The concentration, source and cumulative dosage of NO•, as well as the composition of biological milieu, can affect cellular responses. NO• acts on cells and tissues in three ways: By diffusing into cells and undergoing intracellular consumption, autoxidating to form nitrous anhydride (N2O3), and reacting with superoxide (O2•−) to form peroxynitrite (ONOO−) (1,2). The reactive oxygen species (ROS) formed by these reactions may then react further: ONOO− reacts with carbon dioxide to form nitrosoperoxycarbonate (ONOOCOO−); N2O3, ONOO− and ONOOCOO− breakdown to form reactive species such as hydroxyl radicals (OH•), nitrogen dioxide radicals (NO2•) and carbonate radical anions (CO3•−); and all of these species may react with cellular molecules (1,3–9). Overproduction of these reactive species under pathological conditions such as inflammation induces oxidative and nitrosative stress and damages cellular macromolecules, which increases levels of mutation and carcinogenesis (3–9).

Adverse responses to NO• reflect qualitative and quantitative variation among different types of cells. Apoptosis is the main mechanism by which cell death occurs at low doses of NO•, whereas necrosis occurs more readily at higher doses. However, there is a wide variation of responses to NO• among different cell types. For example, sub-millimolar concentrations of NO•-donor drugs rapidly induce cell death in macrophages, however, they have no obvious effect on cultured hepatocytes. Furthermore, some cell types, such as human lymphoblastoid cells, are sensitive to apoptosis, whereas others, including HCT116 human colon cancer cells, exhibit resistance (2). These differences may be down to the actions of other factors acting on the cell. Antioxidants (for example, glutathione and flavonoids) can alter the apoptosis-inducing potency of different NO• donors. Other cellular factors, such as enzymatic scavengers of free radicals (including superoxide dismutase and catalase) may also contribute to the differences in sensitivity of cell to NO•-induced cell death.

Factors affecting the mutagenic potency induced by NO• have not yet been extensively characterized. Previous studies by our group have demonstrated that NO• and the ROS derived from it are strongly mutagenic in the supF shuttle vector pSP189 exposed to large single bolus doses of ONOO−, NO• gas or unstable donors (10–13). While providing useful mechanistic information, exposure under these conditions limits extrapolation of the data to in vivo pathological states in which these radicals may play a role. To address this issue, studies aimed at developing well-defined experimental models that could be exposed to more pathophysiologically relevant conditions, and defining the mechanisms responsible for NO• and ROS-associated genotoxicity have been performed by our group (14–20). The effects of DNA damage, mutational frequencies and spectra induced by varying bolus doses of ONOO− were compared with those induced by slow infusion of ONOO− and by SIN-1, which slowly decomposes to release NO• and O2•−, thus producing ONOO− (14,15). A co-culture system (17,18) was also employed in which target cells were co-cultivated with mouse macrophages (RAW 264.7) and stimulated to produce NO• by interferon-γ (IFN-γ) and lipopolysaccharide (LPS), to characterize genotoxic responses in endogenous hypoxanthine-guanine phosphoribosyltransferase (HPRT) and thymidine kinase (TK1) genes (17) as well as the transfected supF gene (18). NO•-induced mutagenesis of the TK1 gene in TK6 (wild-type p53) and NH32 (p53 null) cells using an NO• reactor specifically designed to provide tightly controlled steady state concentrations of NO• and molecular oxygen (O2) have also been examined (16).

The present study sought to mimic the conditions in inflamed tissues to assess how NO• and the reactive species derived from it increase carcinogenic risk. The modulating influences of additional factors on cytotoxicity and mutagenesis induced by NO• and associated ROS were assessed using two delivery methods (reactor and co-culture systems). For co-culture experiments, a modification of the Costar Transwell™ system was introduced that places target and generator cells into close proximity (1 µm) whilst facilitating their separation following treatment. The target cells used for these experiments were human lymphoblastoid TK6 cells, which have been used extensively in mutagenicity studies (14,16,21–23). The relative contribution of NO• to mutagenesis was also investigated by employing an NO• synthase inhibitor.

Materials and methods

Cell cultures and chemicals

Mouse macrophage-like RAW264.7 cells (American Type Culture Collection, Manassas, VA, USA) were cultured in high glucose Dulbecco's modified Eagle's medium (DMEM) containing L-glutamine supplemented with 100 U/ml penicillin, 100 µg/ml streptomycin and 10% heat-inactivated fetal bovine serum (Atlanta Biologicals, Lawrenceville, GA, USA). Cells from the human lymphoblastoid TK6 cell line were provided by Dr Gerald N. Wogan (Massachusetts Institute of Technology, Cambridge, MA, USA), and maintained in RPMI-1640 medium supplemented with antibiotics and 10% heat-inactivated horse serum (Lonza Group Ltd., Basel, Switzerland). Before each experiment, cell cultures were treated with CHAT (10 µM 2′-deoxycytidine, 20 µM hypoxanthine, 0.1 µM aminopterin, and 17.5 µM thymidine) according to a standard protocol to remove pre-existent mutant cells (21).

Reagents and materials were obtained as follows: Gases from Air Gas (Edison, NJ, USA); Silastic™ tubing (0.058 in. i.d., 0.077 in. o.d.) from Dow Corning (Auburn, MI, USA); NO• synthase and •OH/ONOO− detection kit from Cell Technology, Inc. (Fremont, CA, USA); total NO• immunoassay kit and recombinant mouse IFN-γ from R&D Systems, Inc. (Minneapolis, MN, USA); Escherichia coli LPS (serotype 0127:B8), 4,5-dihydroxy-1,3-benzene-disulfonic acid (tiron), uric acid, 4-nitroquinoline 1-oxide (4-), 6-thioguanine (6-TG), and trifluorothymidine (TFT) all from Sigma-Aldrich (St. Louis, MO, USA); and N-methyl-L-arginine monoacetate (NMA) from CalBiochem Research (EMD Millipore, Billerica, MA, USA).

Exposure of target cells to NO• in a reactor system

Cells were exposed to NO• by diffusion through permeable Silastic™ tubing utilized for specially designed reactors, with which NO• dose and dose-rate is tightly controlled at steady state concentrations as described previously (16,24). TK6 cells, which grow in suspension, were cultivated with a density of 5×105 cells/ml in 110 ml of culture medium and exposed to NO• at a steady state concentration of 0.6 µM. The total NO• dose delivered into the medium was controlled by varying the exposure time. Cells exposed to argon gas under the same conditions served as negative controls.

Intracellular NO• and ROS synthesis in RAW 264.7 cells

Generation of NO• and ROS by RAW 264.7 macrophages was determined using fluorescent probes (Cell Technology, Inc.). Macrophages were plated in a black, clear bottom 96-well microplate (1×105 cells/well) and incubated for 6 h to allow cells to adhere. Cells were then washed twice with HBSS and preloaded with dye before incubation at 37°C for 60 min with either 10 µM diaminofluorescein-2 diacetate (DAF-2DA) for NO• detection, or 10 µM hydroxyphenyl fluorescein (HPF) for detection of ROS. After washing twice with HBSS to remove excess dye, cells were incubated for 8 h in 100 µl of fresh medium in the presence or absence of 20 U/ml IFN-γ and 20 ng/ml LPS. In some experiments, NMA, tiron, uric acid or combinations thereof were also added. After 8 h incubation, fluorescence intensity was measured in a multi plate Spectra Max Gemini fluorescence reader (Molecular Devices LLC, Sunnyvale, CA, USA) in triplicate and averaged every 100 sec for 1 h at 37°C. Excitation and emission wavelengths were set at 488 and 515 nm, and medium blanks corrected for auto fluorescence.

Co-culture of TK6 with RAW264.7 cells in a modified Transwell™ system

Transwell™ permeable supports (Corning Inc., NY, USA) consist of 100 mm culture dishes each containing a 75 mm diameter insert carrying a polycarbonate membrane with 0.4 µm pores that separates two chambers. This design allows the free exchange of fluids but no direct contact between generator and target cells, even when both are adherent. Co-culture of TK6 cells with macrophages was performed by the following modified procedure: i) The membrane-supporting insert was inverted and macrophages (1×107 cells) were seeded onto the bottom surface of the membrane after pretreatment with medium for 1 h at 37°C, ii) After 6 h incubation to allow macrophage attachment, the insert was restored and the two-chamber unit was assembled in the usual configuration, iii) 12 ml of culture medium, containing 20 units/ml IFN-γ and 20 ng/ml LPS, with or without various combinations of 2 mM NMA, 1.25 mM tiron, and 1.25 mM uric acid, was inserted into the lower chamber of the Transwell™ device, iv) TK6 cells suspended in 9 ml medium were placed into the upper compartment, v) The concentration of NMA and a nonspecific NOS inhibitor, which were effective in reducing cytotoxicity (MTT assay, data not shown) was established using modifications of conditions outlined in Li et al (14).

Two preliminary experiments were performed to establish optimal co-culture conditions suitable for assessment of mutagenic responses. First, it was confirmed that TK6 target doses do not produce detectable amounts of NO• when cultured with IFN-γ/LPS for 24 h. Following this, the TK6 cell to macrophage ratio was established based on NO•-induced cytotoxicity. In order to evaluate mutagenic responses at equivalent levels of cell survival, TK6 cells were co-cultured with macrophages at a ratio of 1:2 (5×106:1×107) in DMEM for 24 h with shaking to prevent settling of suspended cells. Following co-culture, TK6 cells were collected, washed twice and re-suspended in 10 ml of culture medium prior to analysis.

Following each period of co-culture, NO2− as well as total NO• (NO3− plus NO2−) content of cell supernatants were measured using an immunoassay kit (R&D Systems, Inc.). NO2− was measured by allowing 50 µl of culture supernatant to react with 100 µl of Griess reagent at room temperature for 10–30 min. To measure total NO• production, NADH and NO3− reductase were added before reaction with the Griess reagent, and absorbance was measured at 540 nm using a micro plate reader. Total NO• and NO2− concentrations were calculated from standard curves derived from standard solutions provided, with fresh culture media serving as the control.

Cell viability

Cell viability 24 h after treatment was determined by trypan blue exclusion, which had previously produced results comparable to those determined by plating efficiency and MTT assay (18,21).

HPRT mutation assay

Following treatment, TK6 cells were maintained for 7 days to allow phenotypic expression. A total of 24×106 cells were placed into ten 96-well micro titer plates at densities of 4×104 cells/well in medium containing 2 µg/ml of 6-TG and TFT to select HPRT and TK1 mutants, respectively. For plating efficiency, cells from each culture were plated into 96-well dishes at 1 cell/100 µl/well in the absence of selective agents. After 2 weeks of incubation, colonies were counted and mutation fractions (MFs) were calculated as described by Li et al (16). A single HPRT mutant colony was then transferred to 24 well plates to propagate mutant cells. Approximately 2×106 mutant cells were collected for molecular analysis. The spontaneous MF was estimated from the argon-treated cells for NO• treatment or untreated cells for co-culture. Cells treated with 4-NQO (140 ng/ml for 1.5 h) served as positive controls.

Total RNA isolation, RT-PCR for determining point mutations and intragenic deletions in HPRT mutants and sequencing

Total RNA from each mutant clone was isolated using TRI-reagent (Sigma Aldrich) according to the manufacturer's protocol, and stored at −80°C. 2 µg RNA was used as a template for the reverse transcription polymerase chain reaction (RT-PCR) with Omniscript Reverse Transcription kits (Qiagen GimbH, Hilden, Germany) and the oligo(dT)15 primer (Promega Corporation, Madison, WI, USA) provided for first-strand cDNA synthesis. The reaction was performed at 37°C for 1 h to allow the lysis of cell membrane and the synthesis of first-strand cDNA from polyadenylated mRNA. Amplification of the cDNA was performed in two rounds of nested PCR in a PTC-200 DNA Engine Thermal Cycler (Bio-Rad Laboratories, Inc., Hercules, CA, USA). A 10 µl aliquot of cDNA solution was transferred into the first round mix of 10 µl 10x PCR buffer, 2 µl dNTP mix, 0.5 µl taq polymerase, 73.5 µl high powered liquid chromatography-grade water, 0.2 µl each 25 mM forward (bases −60 to −41; 5′-CTGCTCCGCCACCGGCTTCC-3′) and reverse (bases −721 to −702; 5′-GATAATTTTACTGGCGATGT-3′) primer and amplified with a PCR profile of 94°C: 1 min, 30 cycles of 94°C; 1 min, 61°C; 1 min, 72°C; 1 min and a final extension of 72°C for 7 min. The product from this reaction was filtered using a Centricon 50 concentrator (Amicon; EMD Millipore) and resuspended in 100 µl sterile water to avoid unspecific binding with remaining primers. 10 µl aliquot was used as template in the second round of PCR using nested primers (bases −36 to −17; 5′-CCTGAGCAGTCAGCCCGCGC-3′ and bases 701–682; 5′-CAATAGGACTCCAGATGTTT-3′). The PCR conditions were the same in the second round reaction as those in the first round. The final product was run on a 1% agarose gel, stained with etidium bromide and observed under ultraviolet light. For direct sequencing of HPRT PCR products, the nested PCR products were purified by a QIAquick PCR Purification Kit (Qiagen GimbH) and aliquots of these PCR products were then sequenced by the Dana-Farber/Harvard Cancer Center DNA Resource Core (Boston, MA, USA) using three primers with the following sequences: Bases 264–252; 5′-ATTTCTATTCAGT-3′, bases 169–181; 5′-ATGGGAGGCCATC-3′ and bases 405–417; 5′-TATAATTGACACT-3′ (Integrated DNA Technologies, Coralville, IA, USA), using an adapted method (25).

Statistical analysis

All experiments were repeated 2–4 times. The two-tailed Student's t-test was used for the comparison of test and control group, and P<0.05 was considered to indicate a statistically significant difference.

Results

Cytotoxicity and mutagenicity of NO• in target cells by reactor system

TK6 cells were exposed for 2–24 h to NO• at a steady state concentration of 0.6 µM, resulting in cumulative total NO• doses ranging from 90–918 µM/min, which induced cell death in a dose-dependent manner (Fig. 1A). The cell viability of TK6 cells was 36% 24 h after a dose of 540 µM/min NO• (Fig. 1A). The mutagenicity of NO• in the HPRT and TK1 genes of TK6 cells was also investigated (Fig. 1B). At a total dose of 540 e/min of NO•, induced MFs in the HPRT and TK1 genes were 7.7×10−6 (P<0.05) and 24.8×10−6 (P<0.01), 2.7- and 3.7-fold higher than background (2.9×10−6 and 6.7×10−6), respectively (Fig. 1B). MFs at the HPRT and TK1 loci in 4-NQO-treated positive controls were 14.6×10−6 and 28.1×10−6, respectively.

Figure 1. Cell survival and mutagenesis following exposure of TK6 cells to nitric oxide (NO•) by a reactor system. (A) Survival was determined by a trypan blue assay 24 h after NO• treatment. (B) Mutation fraction in the thymidine kinase (TK1) and hypoxanthine-guanine phosphoribosyltransferase (HPRT) genes of TK6 cells exposed to 540 µM • min NO•. Cells treated with Argon acted as negative controls. Data represent mean ± standard deviation for three measurements. *P<0.05 and **P<0.01 vs. the untreated control.

Intracellular production of NO• and ROS in activated macrophages

To quantify intracellular NO• and ROS production by macrophages, DAF-2T and HPF fluorescence intensities were measured for 1 h, beginning 8 h after activation with IFN-γ/LPS (Fig. 2). NO• converts the non-fluorescent dye, DAF-2, to its fluorescent triazole derivative, DAF-2T (26). The fluorescence of DAF-2T increased continuously (Fig. 2A), confirming increased NO• production by macrophages stimulated with IFN-γ/LPS. The probe HPF was used as an index of the production of ROS, primarily ONOO− and •OH, which exhibited a modest increase up to ~30 min, after which production tended to plateau (Fig. 2B). These experiments indicate that NO• and ROS were intracellularly generated by activated macrophages (Fig. 2).

Figure 2. Increase in fluorescence intensity according to nitric oxide (NO•) and reactive oxygen species (ROS) production from RAW 264.7 cells. DAF-2T (A) and HPF (B) fluorescence intensity induced by macrophages stimulated with interferon-γ/lipopolysaccharide (IFN-γ/LPS) in the absence or presence of N-methyl-L-arginine monoacetate (NMA) and uric acid. Fluorescence is expressed as the % of control values obtained in the absence of IFN-γ/LPS. Data represent mean ± standard deviation for three measurements.

Various treatment regimens were used to provide additional evidence regarding the reactive species responsible for these effects, by introducing an inhibitor and scavengers, including NMA (a NO• synthase inhibitor), tiron (an O2•− scavenger) and uric acid (O2•− and ONOO− scavengers), alone or in combination (NMA+ tiron or NMA+ tiron +uric acid). Intracellular fluorescence of DAF-2T, i.e., NO•-derived fluorescence, was almost completely suppressed by NMA, and was partially suppressed by uric acid (Fig. 2A). By contrast, HPF fluorescence intensity, the ROS probe, was effectively blocked by uric acid but was not significantly affected by NMA (Fig. 2B). Results obtained in the presence of the both probes demonstrated that tiron and/or uric acid in combination with NMA reduced total fluorescence almost completely (data not shown), confirming that NO• and ROS were of central importance as a reactive species produced by activated macrophages. Additionally, MTT assays performed at the beginning and end of each experiment confirmed that the observed changes in fluorescence intensity were not attributable to changes in cell survival (data not shown).

Viability, NO• production and mutagenesis in TK6 cells co-cultivated with activated macrophages

The relative survival, concentration of NO• and mutation in the HPRT and TK1 genes of TK6 cells co-cultivated with macrophages for 24 h was determined (Fig. 3). Activated macrophages caused reduced (38%) cell survival rates compared to controls (100%; P<0.01), and 18.3-fold higher total NO• production compared with untreated controls (P<0.01; Figs. 3A and B). The MFs induced in HPRT and TK1 genes (8.8×10−6 and 13.9×10−6, respectively) were 2.8- and 1.9-fold higher than the spontaneous MFs (3.2×10−6 and 7.3×10−6, respectively; P<0.05; Fig. 3C). Co-treatment with both IFN-γ/LPS and NMA restored cell survival to 95.5%, completely blocked total NO• and NO2− production and strongly suppressed the increases in MF by 95 and 89%, respectively, indicating that NO• was largely responsible for the cytotoxicity and induced mutations in TK6 cells (Fig. 3; Table I).

Figure 3. The effects nitric oxide (NO•) on cell viability and mutation rates. Cell survival rates (A), total NO• and nitrogen dioxide (NO2−) concentrations (B), and mutagenesis rates (C) in the hypoxanthine-guanine phosphoribosyltransferase (HPRT) and thymidine kinase 1 (TK1) genes of TK6 cells co-cultivated with interferon-γ/lipopolysaccharide-stimulated RAW264.7 cells in the absence or presence of methyl-L-arginine monoacetate (NMA). TK6 cells (0.5×107) were co-cultivated with macrophages (1×107) for 24 h. Data represent mean ± standard deviation for three measurements. *P<0.05 and **P<0.01 vs. the untreated control.

Table I. Suppressive effects of NMA on total NO• and NO2− production, mutagenesis, and cytotoxicity in TK6 cells co-cultured with IFN-γ/LPS-stimulated RAW 264.7 cells.

Table I.

Suppressive effects of NMA on total NO• and NO2− production, mutagenesis, and cytotoxicity in TK6 cells co-cultured with IFN-γ/LPS-stimulated RAW 264.7 cells.

	Inhibition (%)
Treatment	Nitrite+Nitrate	Nitrite	Mutation	Cytotoxicity, %
IFN-γ/LPS	–		–	62.0
IFN-γ/LPS+NMA	100	99.7±1.44	94.5±9.32 (HPRT)	4.5
			89.2±5.69 (TK1)

[i] NO^•, nitrogen oxide; NO₂⁻, nitrogen dioxide radical; NMA, N-methyl-L-arginine monoacetate; IFN-γ/LPS, interferon-γ/lipopolysaccharide; HPRT, hypoxanthine-guanine phosphoribosyltransferase; TK1, thymidine kinase.

Molecular analysis of HPRT mutants

For the molecular analysis of the HPRT mutations, RT-mediated production of HPRT cDNA, PCR amplification and cDNA sequencing were used to define small alterations sin the coding sequence. sRNA was extracted and analyzed from 22 and 48 mutants in argon and 540 µM min of NO• treatment groups in the reactor system, and 20 and 51 mutants in untreated and IFN-γ/LPS-treatment groups in the co-culture system, respectively. Table II highlights the altered DNA sequences of 15 and 30 cDNA products synthesized from 22 argon and 48 NO• treated HPRT mutants, respectively. Single base pair substitutions (60%) were the major type of argon control mutations, with G:C to T:A transversions (13%) and G:C to T:A (13%) and A:T to G:C (13%) transitions. 20% (3/15) of mutants had base deletions and 13% (2/15) had one complete exon exclusion. Similar mutation profiles were observed in NO•-induced mutants (Table II). Base substitutions were the predominant type of mutation (61%), followed by single or two consecutive exon exclusions (23%) and deletions (13%). Of the 18 mutants carrying single base pair substitutions, an equal number occurred at G:C (9/18) or A:T (9/18) base pairs with G:C to T:A (13%) and A:T to T:A (13%) transversions and G:C to A:T (13%) and A:T to G:C (13%) transitions. Among these, a clustering of base changes on exons 2 and 4 was observed in comparison to the spontaneous distribution. Four frame shift mutations, consisting of 1 or 2-bp deletions and 1-bp insertions occurred in exon 3.

Table II. A summary of results from the molecular analysis of human HPRT mutant TK6 cells treated with NO• exposure by reactor or co-culture systems.

Table II.

A summary of results from the molecular analysis of human HPRT mutant TK6 cells treated with NO• exposure by reactor or co-culture systems.

	Proportion of mutants (%)
	Reactor		Co-culture
	Spontaneousa	NO•-inducedb	Spontaneousc	NO•-inducedd
Transversion	5 (34)	10 (35)	4 (26)	9 (30)
G:C to T:A	2 (13)	4 (13)	1 (7)	6 (20)
G:C to C:G	1 (7)	1 (3)	0 (0)	1 (3)
A:T to T:A	1 (7)	4 (13)	2 (13)	1 (3)
A:T to C:G	1 (7)	1 (3)	1 (7)	1 (3)
Transition	4 (26)	8 (26)	3 (20)	7 (24)
G:C to A:T	2 (13)	4 (13)	2 (13)	2 (7)
A:T to G:C	2 (13)	4 (13)	1 (7)	5 (17)
Insertions	1 (7)	1 (3)	0 (0)	0 (0)
Deletions	3 (20)	4 (13)	1 (7)	6 (20)
Multiple	0 (0)	0 (0)	1 (6)	1 (3)
Exon exclusions	2 (13)	7 (23)	6 (40)	7 (23)
One exon	2 (13)	1 (3)	0 (0)	7 (23)
Two exon	0 (0)	6 (20)	6 (40)	0 (0)
Total	15 (100)	30 (100)	15 (100)	30 (100)

a Cells exposed to argon gas under the same conditions served as negative controls

b Cells exposed to 540 µM min of NO^•

c Untreated cells served as a negative control

d Cells co-cultivated with IFN-γ/LPS-stimulated RAW264.7 cells for 24 h. NO^•, Nitric oxide; HPRT, hypoxanthine-guanine phosphoribosyltransferase; IFN-γ/LPS, interferon-γ/lipopolysaccharide.

In total, 30 of 51 HPRT mutants from TK6 cells co-cultivated with activated macrophages produced HPRT cDNA. These mutations included 54% (16/30) base substitutions, 23% (7/30) single exon exclusions and 20% (6/30) deletions (Table II). The proportion of point mutations at the G:C and A:T base pairs was 56% (9/16) and 44% (7/16), respectively. The major base substitutions were G:C to T:A transversions (20%) and A:T to G:C transitions (17%). NO• derived from co-culture system induced base substitutions that occurred most frequently in exons 2 and 3 of the HPRT gene (Table II). 8- to 16-bp deletions at positions between 387 and 402 occurred mostly frequently in exon 5 and were observed in the co-culture system. Analysis of spontaneous mutations that formed in the HPRT gene of TK6 cells co-cultivated with untreated macrophages revealed that 46% were base substitutions, 7% 1-bp deletions and 40% two exon exclusions (Table II).

Discussion

It has been previously demonstrated that the delivery method of NO• and ROS affects cellular responses, sometimes producing conflicting findings regarding cytotoxicity, apoptosis and mutation (27,28). The present study has characterized the mutagenic effects in cells exposed to NO• and ROS associated with inflammation through two delivery modes: A reactor system to deliver NO• by diffusion into cell culture media at tightly controlled steady state concentrations (16,24); and a co-culture system to expose target cells to NO• and ROS generated by activated macrophages. In the first approach, cytotoxicity and mutagenesis were compared in TK6 cells exposed via a reactor system for delivering NO• at predictable, reproducible rates, and induced DNA damage in CHO cells comparable to those observed in the nuclear DNA of NO•-producing macrophages (2,29). The results highlighted that NO• treatment reduced the percentage of viable cells dose-dependently (Fig. 1A), and mutagenic responses were characterized at total NO• dose allowing ~30% cell survival (540 µM/min NO•, closed square in Fig. 1A). Statistically significant mutagenic effects of NO• have been observed previously, however the current study demonstrated that NO• had a different mutagenic potency at the two endogenous HPRT and TK1 loci (Fig. 1B), consistent with earlier findings (16,17,19,21). Continuous exposure of TK6 cells for 2 h, at a rate of 533 nM/s NO•, caused a 2-fold increase in HPRT and 4.2-fold in TK1 gene expression as compared with argon-treated controls (17,21). These differences in susceptibility to mutations at the two loci may be explained by the fact that deletions >1.3 Mbp at the HPRT locus decreases cell viability (30–32), however, cells at the TK1 locus can tolerate much larger deletions (33). Furthermore, allelic recombination is possible at the autosomal TK1 locus but not at the X-linked hemizygous HPRT locus (34).

The present study also characterized the mutagenic potency of NO• and ROS produced by IFN-γ/LPS activated macrophages using a modified Transwell™ co-culture system. Before examining mutagenic events in co-cultivated target cells, NO• and ROS production were assessed by macrophages activated with IFN-γ/LPS using fluorescent probes specific for these factors (Fig. 2). Upon activation, macrophages produced NO• and ROS, detected at the intracellular level by specific probes and at the extracellular level through the formation of NO3− and NO2−, the metabolites of NO• in cell supernatants (Fig. 3B; Table I). These data expand on previous results (35,36), demonstrating that NO• concentration after 8 h activation approached a plateau of 0.6 µM at ~34 min as well as a substantial increase in O2•− production by activated macrophages. These findings are further substantiated by decreases in fluorescence intensity of probes specific for NO• and ROS caused by NMA, an NO• synthase inhibitor, and O2•− and/or ONOO− scavengers (Fig. 2).

Previous co-culture studies indicated that NO• and ROS produced by macrophages were lethal and strongly mutagenic in both target cells and generator cells (17,18,20). Others have demonstrated that products of IFN-γ/LPS-stimulated RAW264.7 or TPA-stimulated HL-60 cells induced mutagenesis in AS52 cells exposed in mixed co-culture systems (37). As well as increased MF, high levels of 8-hydroxy-2′-deoxyguanosine, reflecting oxidative DNA damage have been detected in AS52 co-cultivated with both types of generator cells (37,38). Suppressors of NO• and ROS strongly inhibited MF (38–41), supporting the conclusion that NO• and ROS are major causes of induced mutations.

In the present study, a Transwell™ system was used that permits cell-cell communication through diffusible soluble factors independent of cell-cell contact, in contrast to the co-culture systems used in previous studies. These allow direct generator-target cell contact, potentially maximizing the transfer of reactive species from generator to target cells, which could amplify estimates of target cell exposure in inflamed tissues. Additionally, macrophage and target cells share the same medium without cell-cell interactions, owing to the physical separation of cells by a polycarbonate membrane. Using the Transwell™ co-culture system, activation of macrophages increased production of NO• and the MF of TK6 cells as compared to unstimulated controls (Fig. 3; Table I). By employing an NO• synthase inhibitor, the relative contribution of NO• to mutagenesis in this system was assessed. Inhibition of NO• production through addition of NMA to the culture medium abrogated much of the cytotoxicity and genotoxicity in TK6 cells, confirming the role of NO• in inducing these effects (Fig. 3; Table I). A contributing factor may have been distances between NO• generator and target cells, due to TK6 cells settling out of suspension were less uniformly distributed than adherent cells growing in close proximity relative to the monolayer of macrophages. O2•− and ONOO−-derived ROS have much shorter half-lives (<50 ms) than NO• and shorter diffusion radii, thus they may not have reached effective concentrations in proximity to TK6 cells above the monolayer of adherent macrophages (17,18).

Mutagenesis induced by NO• has previously been examined in a number of experimental systems. Exposure of S. typhimurium to NO• gas induced G:C to T:A, T:A to A:T, and C:G to A:T mutations (19), and also mutagenized TK6 cells, causing A:T to T:A and A:T to G:C mutations in the HPRT gene (19). Peroxynitrite, the reaction product of NO• and superoxide, also induced pSP189 supF mutations, predominantly G:C to T:A (10,15). G:C to T:A mutations occurred in the supF gene of pSP189 replicating in AD293 cells co-cultivated with activated macrophages (18). Treatment of the same plasmid with NO• donor drugs (diethylamine/NO• or spermine/NO• complexes) induced predominantly G:C to A:T and A:T to G:C mutations (42). In each of these experimental systems, base substitutions were the main type of mutations induced, the specific type depended on the form of NO• used.

In the present study, the majority of spontaneous and NO• induced mutations were single base pair substitutions, with transversions (30–35%) outnumbering transitions (24–26%). Other mutations were deletions and an insertion in both reactor and co-culture systems (Table II). 29–30 and 23–29% of mutations occurred at G:C and A:T base pairs, respectively, and G:C to T:A transversions (13–20%) and A:T to G:C transitions (13–17%) were the single base pair substitutions most often induced by NO• treatment in both delivery systems (Table II). In human cancer, a significant proportion of mutations in oncogenes and tumor suppressor genes are G:C to T:A transversions, particularly in lung cancer (43,44). Previous studies have demonstrated that these mutations may be caused by a variety of DNA lesions, such as apurinic sites, 8-oxodeoxyguanosine (8-oxo-dG), or 8-nitroguanine (8-nitro-G; (45–47). ROS-associated G:C to T:A transversions may arise from oxidative damage, leading to dA rather than dC being incorporated opposite dG. A:T to G:C transitions, generally induced by oxidative damage (19), were also frequently observed (Table II). Deamination of adenine to hypoxanthine, which pairs with cytosine rather than thymine in DNA, could account for the high proportion of A:T to G:C transitions observed (19). The predominance of G:C to T:A transversions is consistent with previously reported mutation types induced by peroxynitrite (ONOO−) and its derivatives generated by reaction of NO• with O2•− (10,15,18) and NO• gas (42), suggesting that NO• may be a major contributor to the observed mutagenesis.

To conclude, the present study provides evidence that exposure to NO• induces cytotoxicity and mutagenicity in target cells and that inhibition of NO• production by activated macrophages is effective at abrogating these properties. Moreover, both the delivery method of NO• and target genes at which the TK6 cells are exposed was revealed to strongly influence the cytotoxicity, the mutagenic potency, types and distribution of mutations. The systems used to introduce NO• in these experiments were designed to approximate conditions of exposure physiologically relevant to chronic inflammation states. Further studies will be required to elucidate precise mechanisms underlying these effects and their potential relevance to NO• induced genotoxicity in vivo.

Acknowledgements

This work was supported by Basic Science Research Program (no. 2016R1A6A1A03012862 and no. 2014R1A1A2056292) through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology, Republic of Korea.

References