genotoxicity of azidoalanine in mammalian cells

Environmental and Molecular Mutagenesis 13:263-270 (1 989)

Genotoxicity of Azidoalanine in Mammalian Cells

P. Arenaz and L. Hallberg Department of Biological Sciences, University of Texas at El Paso

Sodium azide mutogenesis is mediated through a metabolic intermediate in bacteria and plant species. However, very little is known about the interaction of this intermediate with nucleic acids, its genotoxic potential, or its mechanism of action, especially in mammalian cells. Chinese hamster cells and normal human skin fibroblasts were treated with extracts from Salmonella typhimurium or Hordeum vulgare (barley) containing a crude mutagenic metabolite, as well as with synthetically produced azidoalanine. The cells were evaluated for the induction of sister chromatid exchanges and the ability to perform

unscheduled DNA synthesis. With the purified azidoalanine and the azide-treated extracts from Hordeurn vulgare, there was a statistically significant increase in the frequency of sister chromatid exchanges observed in both Chinese hamster cells and human fibroblasts. This increase was about twofold, as compared with the control. On the other hand, there was no detectable genotoxic response when cells were exposed to azide-treated extract from Salmonella fyphirnu- rium. The results imply that azidoalanine and the crude mutagenic metabolite from Hordeurn vulgare are weakly genotoxic in mammalian cells.

Key words: sodium azide, sister chromatid exchange, UDS, CHO cells, human fibroblasts

INTRODUCTION DNA synthesis in human EUE cells [Slamenova and Gabe- Sodium azide (NaN3) has wide application in agriculture

as an herbicide, nematocide, and fungicide as well as a nitrification inhibitor. In industry, azides, including sodium azide, are used as contributors of nitrogen atoms for many chemical reactions. This property of sodium azide has prompted the federal government to examine the possibility of using azide as a nitrogen gas generator for automobile passive restraints and escape chutes for airplanes. In addition, sodium azide is used in clinical and analytical labora- tories as an antibacterial agent because of its properties as a respiratory poison and growth inhibitor [see Kleinhofs et al., 1978 for review]. The widespread use of azide, along with its demonstrated mutagenicity in bacterial and plant species, suggests that this compound possesses the potential for becoming a major environmental mutagen [see Klein- hofs et al., 19781.

Although sodium azide is a potent mutagen in many or- ganisms, its genotoxic (and mutagenic) potential in mammalian cells is suspect. Azide did not induce chromosome aberrations or sister chromatid exchanges (SCE) in human lymphocytes [Sanders et al., 1978; Arenaz and Nilan, 19811, and in Chinese hamster cells [Arenaz and Nilan, 19811 or single strand breaks in Chinese hamster DNA [Arenaz et al., 19831. Azide has been shown to elicit a weak to nonmutagenic response in Chinese hamster cells [Jones et al., 1980; Slamenova and Gabelova, 19801, mouse L cells [Clive et al., 19791, and human EUE cells [Slamenova and Gabelova, 19801. Azide was not able to induce unscheduled

lova, 19801. In addition, azide has been shown to be nonmutagenic in Drosophila [Karnra and Gollapudi, 19791 and Arabidopsis [Gichner and Veleminsky, 19771 as well as Neurospora [Landers, 19711.

Sodium azide is metabolized by both Hordeum vulgare [Owais et al., 19781 and Salmonella ryphimurium [Owais et al., 19791 to a stable mutagenic intermediate. This mutagenic metabolite has been isolated from azide-treated Hor- deum vulgare embryos and from Salmonella typhimurium [Owais et al., 1978, 19791 as well as synthesized from cell-free extracts of both Salmonella typhimurium and Hor- deum vulgare [Owais et al., 1981b, 1983; Rosichan et al., 19831. The putative mutagenic intermediate has been iden- tified as azidoalanine [Owais et al., 19831 and involves the condensation of azide and 0-acetylserine by the enzyme 0-acetylserine(thi0)-lyase (EC 4.2.99.8) [Owais et al., 1981a, 1983; Rosichan et al., 19831. Crude metabolite, extracted from azide treated Hordeum vulgare seed, was shown to be mutagenic in repair deficient Saccharomyces Rad 2-5 [Veleminsky et al., 19791. However, the crude

Received January 18, 1988; revised and accepted December 6, 1988.

Address reprint requests to Dr. P. Arenaz, University of Texas at El Paso, El Paso, TX 79968.

0 1989 Alan R. Liss, Inc.

264 Arenaz and Hallberg

metabolite was not mutagenic in repair proficient rad +

cells. Although the chemical structure of azidoalanine has been

determined [Mangold and LaVelle, 1986; Owais et al., 19861, very little is known about its interaction with nucleic acids, its genotoxic potential, or its mechanism of action, especially in mammalian cells. In Salmonella typhimurium, it has been suggested that azidoalanine acts as a base substitution mutagen as it increases the reversion rate only in the putative base substitution strains TA 1530 and TA 100 [Owais et a]., 1983, 1986; Mangold and LaVelle, 1986; LaVelle and Mangold, 19871, suggesting that the mechanism involved in the induction of mutations by azidoalanine may be similar to that of azide.

Studies were performed to more thoroughly understand the mechanism of action of sodium azide and the mutagenic intermediate, as well as the interaction of the putative mutagenic intermediate with mammalian cells. Two standard assays, were employed to evaluate the genotoxic potential of azidoalanine and the putative mutagenic intermediate. The data support the conclusion that azidoalanine and the crude mutagenic metabolite, although a potent mutagen in Salmonella typhimurium, are mildly genotoxic in mammalian cells. A possible explanation for the poor response of mammalian cells to azidoalanine and/or the mutagenic intermediate is discussed.

MATERIALS AND METHODS

Cell Lines

Chinese hamster V79 cells (gift from R. Snyder, Stauffer Chemical Co.) and normal human skin fibroblasts (CRL 1222, American Type Culture Collection) were used throughout these experiments. The Chinese hamster cells and human fibroblasts were maintained in Dulbecco's mod- ified Eagles medium (DMEM) supplemented with 10% fe- tal calf serum (Gibco), 2 mM L- glutamine and either 0.1 pg/ml gentamycin (Sigma) or 100 U penicillin/100 pg streptomycin (Gibco) at 37°C in an atmosphere containing 95% air/5% COZ.

Metabolite Production

Salmonella typhimurium, in early log phase, was treated with sodium azide (5 x M) and grown for 22 hr at 37°C with vigorous shaking according to the procedure of Owais et al. [1979]. The cells were harvested by centrifugation at 15,OOOg for 20 min and the supernatant was saved. The supernatant, designated as growth medium, was lyoph- ilized and reconstituted in 25 ml of KHP04 buffer (pH 7.2), acidified to pH 2.0 with conc. HCI to remove excess azide, and evaporated to dryness. The residue was dissolved in 20

ml of 0.1 M KHP04 (pH 7.2) and applied to a DOWEX 50 cation exchange column (0.7 X 25 cm). The metabolite was eluted with a step gradient of 0.01, 0.2, and 0.75 N HCl [Owais et al., 198 1 b]. The fractions were evaporated to dryness, redissolved in phosphate-buffered saline (PBS) , adjusted to pH 7.2 with I N NaOH, filter sterilized, and tested for mutagenicity with the plate incorporation method of Maron and Ames [ 19831 with strain TA 1530. Fraction(s) that demonstrated the highest mutagenicity were pooled and used for experiments on mammalian cells. Cell-free extracts from Salmonella typhimurium were produced according to the procedures outlined by Owais et al. [ 19811. Cells from exponentially growing cultures were harvested by centrifugation at 15,OOOg for 20 min at 4°C. The cell pellets were suspended in 0.1 M KHP04 (pH 7.2) and washed twice. The pellets were resuspended in 0.1 M KHP04 (pH 7.2) and sonicated 3 x 45 seconds at 60 watts with a Bransor. sonicator and centrifuged at 15,000g for 20 min. The supernatant was saved and stored at -70°C. The reaction mixture for metabolite production consisted of 2 ml of the crude extract (4-5 mg proteidml) plus 1 mM sodium azide and 1 mM o-acetyl-L-serine (OAS) and 0.1 M KHP04 (pH 7.2) for a total of 5 ml. The reaction mix was incubated for 10 min at 37"C, acidified with conc. HCl, centrifuged at 10,OOOg for 20 min to remove precipitated protein, and evaporated to dryness. The residue was redissolved in 0.1 M KHP04 and partially purified through DOWEX 50 as described above.

The in vivo crude metabolite isolated from Hordeum vulgare (cv Himalaya, kindly provided by Robert Nilan, Washington State University) was prepared according to the procedures Owais et al. [1978]. Seed from Hordeurn vulgare was soaked for 8 hr, then germinated in aerated water for 16 hr at ambient temperature. The germinated seeds were treated with 5 x M sodium azide in 0.1 M KHP04 buffer (pH 3.2) for 3-4 hr and washed for 1 hr with tap water. The embryos from 100 g of seed were excised from the endosperm and homogenized in 0.1 M KHP04 (pH 7.2) with a mortar and pestle. The homogenate was centrifuged at 15,000g and the supernatant was saved. The supernatant was acidified to pH 2 with conc. HCI and evaporated to dryness. The residue was dissolved in 0.1 M KHP04 and applied to a DOWEX 50 column as described above. To generate cell-free extracts of Hordeum vulgare, seeds were germinated as described above but without treatment with azide. The extracts were reacted with sodium azide and OAS as described above and the metabolite was purified through a DOWEX 50 column. Chemically synthesized and pure azidoalanine was a gift from W. Owais (Yarmouk University, Irbid, Jordan). Mutagenicity was determined with the plate incorporation method of Maron and Ames [1983] with strains TA 1530. Protein determination was done according to the method of Lowry [1951].

Azidoalanine in Mammalian Cells 265

TABLE 1. Induction of Sister Chromatid Exchange in Normal Human Skin Fibroblasts by a Putative Mutagenic Metabolite of Sodium Azide Extracted From Salmonella hwhimurium*

Treatment of Cells and SCE Induction

Chinese hamster cells or human fibroblasts were seeded onto 100 mm culture dishes at 1-2 X lo5 cells/ml and incubated for 48 or 72 hr at 37°C. 5-Bromodeoxyuridine (BrdU), at a final concentration of 30 pM, was added for the final 24 or 48 hr. All cultures were kept in the dark after BrdU treatment.

The crude mutagenic metabolite or azidoalanine, dissolved in PBS, was added directly to the culture media to give the final concentrations described in Results. Treat- ment was initiated 24 hr post seeding and lasted for two rounds of replication (24 hr for Chinese hamster cells or 48 hr for human fibroblasts). Alternatively, cells were treated with the crude metabolites or azidoalanine for 6 hr. The plates were then washed twice with PBS and fresh medium containing 30 p M BrdU was added. The cells were incubated for two rounds of replication as described above. Colcemid (0.1 pg/ml) was added 2 hr before harvest for the Chinese hamster cells and 4 hr before harvest for the human fibroblasts. Controls for the Salmonella typhimurium growth media-derived metabolite and the Hordeum vulgare in vivo metabolite consisted of equivalent amounts (based on protein concentration) of extract that had not been treated with azide. For the bacterial and barley cell free extract- derived metabolite, the controls consisted of equal amounts of extract alone or extract that had been interacted with azide or OAS. In all cases, the total amount of protein was equal. In addition, a control with the equivalent amount of BSA in PBS was used.

The Chinese hamster cells and human fibroblasts were collected by scraping the dish with a rubber policeman and centrifugation at 200g for 10 min. The cell pellets were resuspended in 0.075 M KCl, fixed in 3:l methano1:acetic acid, dropped onto cold, wet slides, and flame-dried. The chromatids were differentially stained by the fluorescence plus Giemsa method described by Perry and Wolfe [ 19741. Fifty well spread and differentially stained metaphases containing 20-25 chromosomes for the Chinese hamster cells or 42-47 chromosomes for the human cells, were scored for each treatment and experiment. Each experiment was repeated at least twice. Data were analyzed with a one-way analysis of variance (ANOVA) to determine whether treatment with the crude metabolite or azidoalanine had any effect on the induction of SCE (at the .05 significance level) when compared to controls.

Unscheduled DNA Synthesis

The ability of Chinese hamster cells to perform DNA repair in response to insult with azidoalanine was measured as unscheduled DNA synthesis (UDS) in the presence of 10 mM hydroxyurea (HU). In this study, 1 x lo6 cells were exposed to azidoalanine or MMS (2 mM) dissolved in PBS. Freshly prepared hydroxyurea was added to a final concen-

Source of metabolite Concentrationa SCE/metaphase (mean 2 SE)

Growth media 10,Ooo 10.66 ? .81 Growth media 5 ,ooo 8.33 2 .52 Growth media 2,000 8.70 f .52 Growth media 1 ,Ooo 7.76 ? .41 Growth media 500 8.13 ? .46 Control 9.48 f .41

Cell extract 10,Ooo 8.19 5 .34 Cell extract 5 ,Ooo 1.41 5 .35 Cell extract 2.000 1.64 2 .35 Cell extract 1 ,Ooo 8.31 f .35 Cell extract 500 1.32 f .40 Control 1.83 f .34

* All treatments were for 48 hr (approximately two cell cycles). a Concentration is expressed as his revertants per milliliter as determined by reversion at the histidine locus in Salmonella typhimurium strain TA 1530.

tration of 10 mM in the growth media, and cells were incubated for 30 min at 37°C. The medium was removed and cultures were washed twice with PBS. Azidoalanine was added to each dish, and the cells were incubated for 30 min at 37°C. The azidoalanine was removed, and the cells were washed twice with PBS and once with Hanks’ balanced salt solution (HBSS) before incubation in 10 mM HU. In control experiments, the equivalent amount of PBS was added. Both damaged and control cells received [3H] thymidine (60 Ci/mmol, 10 pCi/culture) for 2 hr in the presence of 10 mM HU. Unscheduled DNA synthesis was determined by the incorporation of [3H] thymidine into acid-precipitable ma- terial, as quantified by liquid scintillation spectrophometry. Repair activity is expressed as the difference between exposure to azidoalanine or MMS and the control.

RESULTS

Normal human skin fibroblasts (NHSF) were exposed for 48 hr (two rounds of replication) to crude azide mutagenic metabolite isolated from the growth media of azide treated Salmonella typhimurium or crude metabolite produced from azide treated cell-free extracts of Salmonella typhimurium. Neither the metabolite isolated from growth media or produced from cell free extracts significantly increased the incidence of SCE above control (Table I). At concentrations greater than the equivalent of 12,000 his revertants/milliliter there was little cell growth. In addition, at 5,000 and 10,000 his revertants/milliliter, there was a delay in cell cycle progression, as noted by a decrease in the frequency of cells in the second cell cycle (Table 11). When normal human fibroblasts were treated with either crude mutagenic metabolite for 6 hr and subsequently allowed to pass through two


TABLE 11. Effect of the Crude Mutagenic Metabolite(s) From Salmonella typhimurium and Hordeum vulgare as well as Azidoalanine on Cell Cycle Progression in Normal Human Fibroblasts and Chinese Hamster Cells Cell type Metabolite source 1st" 2nd 3rd

v79 Control 20 51 29 v79 Salmonella in vivob 50 40 10 V79 Salmonella in vitro 51 42 7 v79 Hordeum in vivo 48 42 10 V7Y Hordeum in vitro 52 37 1 1 V7Y Azidoalanine 25 55 20

TABLE 111. Induction of Sister Chromatid Exchange in Chinese Hamster VT9 Cells by a Putative Mutagenic Metabolite of Sodium Azide Extracted From Salmonella typhimurium* Source of metabolite Concentration" SCElmetaphase (mean f SEI

Growth media 10.000 14.12 f .59 Growth media 5 ,000 12.91 2 .60 Growth media 2,000 12.26 f .59 Growth media 1 ,000 12.94 f .51 Growth media 500 13.02 f .57 Control 13.52 2 .46

Fibroblast Control 30 51 19 Cell extract 10,Ooo 10.50 ? .64 Fibroblast Salmonella in vivo 55 40 5 Cell extract 5 ,m 9.38 ? .65 Fibroblast Salmonella in vitro 57 40 3 Cell extract 2,000 9.80 ? .64 Fibroblast Hordeum in vivo 60 32 8 Cell extract 1 ,Ooo 11.01 f .44 Fibroblast Hordeum in vitro 58 35 7 Cell extract 500 10.20 f .40 Fibroblast Azidoalanine 35 50 15 Control 10.02 * .37

"Represents the percentage of cells in Ist, 2nd, or 3rd metaphase, based on three replications of I 0 0 randomly selected metaphases. bConcentrations used were the highest concentration that elicited an increase in SCE frequency.

* All treatments were for 24 hr (approximately two cell cycles). a Concentration is expressed as his revertants per milliliter, as determined by reversion at the histidine locus in Salmonella typhimurium strain T A 1530.

rounds of replication, no change in SCE frequency was observed (data not shown). This finding is similar to what was observed above and suggests that the mutagenic metabolite derived from Salmonella typhimurium is not genotoxic in human fibroblasts. Because of the unknown com- position of the crude metabolite mixture, quantification of the dosage was based on the number of his revertants per milliliter as determined by the Ames assay at each dilution. This was a crude measure but provided a good estimate of mutagenic potential. It should be noted that there is a difference in the spontaneous SCE frequency between the values from the growth media-treated cells and the cells treated with cell-free extracts (Table I). However, this difference is not statistically significant. The elevated SCE level observed in growth media treated cells may represent nonspecific effects of differences in protein concentration as the BSA control also showed a roughly equivalent increase in SCE when compared to the cell-free extract. This increase was also observed for the Chinese hamster VW cells (Table 111).

Chinese hamster cells were treated for 24 hr (two rounds of replication) with crude metabolite isolated from Salmo- nella typhimurium as described above. Exposure to the metabolite isolated from growth media produced no statistically significant increase in the frequency of SCEs, compared to control (Table III), even at concentrations giv- ing as high as lo4 his revertantdm1 (as measured in the Ames assay). This trend was also observed when the Chi- nese hamster cells were treated with the metabolite produced from cell-free extracts derived from Salmonella typhimurium. In addition, there was an observed delay in the cell cycle progression of the treated cells, as demonstrated by a decrease in the frequency of cells in the second meta-

phase (Table 11). Treatment of the Chinese hamster cells with crude metabolites for 6 hr induced no significant increase in SCE frequency (data not shown). These data, along with the data on the lack of induction of SCEs in human fibroblasts suggest strongly that the crude mutagenic metabolite isolated from Salmonella typhimurium is not genotoxic in mammalian cells.

When normal human skin fibroblasts were exposed for 48 hr to crude metabolite synthesized from Hordeum vulgare, there was a statistically significant increase in the incidence of SCEs (Table IV). Treatment with in vivo synthesized crude metabolite at 2.5 x lo3 his revertantdm1 produced an increase 175% above control and a 228% increase over the untreated extract. It is interesting to note that there is a slight decrease in the SCE frequency at 5 X lo3 his+ revertantdm1 (162% of control and 211% of extract). This may reflect the observed cytotoxic effects at this concentration. A 175% increase in the frequency of SCEs was noted for the in vitro produced crude azide metabolite at 2.5 x lo3 his revertantdml, an increase roughly similar to that observed for the in vivo produced metabolite at the same concentration. However, when compared to the extract control, the increase in SCE was only 190% as compared to a 228% increase for the in vivo metabolite. This difference may reflect differences in the total amount of the crude metabolite present in the two sources, even though the total protein concentration and number of his revertants were the same. At treatment with 5 x lo3 his revertantslml a twofold increase in SCE frequency was observed, compared with the BSA control, and a 2.15-fold increase, when compared to the extract control. Controls, with similar amounts of crude extract plus azide or OAS, did not increase the incidence of SCE above BSA control. In fact, these treatments slightly decreased the SCE frequency. These


TABLE V. Induction of Sister Chromatid Exchange in Chinese Hamster V79 Cells by a Putative Mutagenic Metabolite of Sodium Azide Extracted From Hordeum vulgare*

TABLE IV. Induction of Sister Chromatid Exchange in Normal Human Skin Fibroblasts by a Putative Mutagenic Metabolite of Sodium Azide Extracted from Hordeum vulgare* Source of metabolite Concentration" SCWmetaphase (mean ? SE)

In vivo 5.000 16.24 -C .8Ib In vivo 2,500 17.56 2 .93b In vivo 1 ,OOo 13.08 2 .35b In vivo 500 13.37 2 .70b In vivo 200 9.24 f .82 Untreated extract 7.70 2 .58 Control 10.02 2 .37

In vitro 5 ,OOo 16.58 -C .73b In vitro 2,500 14.64 C .5gb In vitro 1 ,OOo 13.41 2 .80b In vitro 500 10.57 * .8Ib In vitro 200 9.40 4 .80 Untreated extract 7.70 f .58 Extract plus azide 6.86 2 .43 Extract plus OAS 7.22 ? .38 Control 8.30 2 .73

* All treatments were for 48 hr (approximately two cell cycles). a Concentration is expressed as his revertants per milliliter, as determined by reversion at the histidine locus in Salmonella typhimurium strain TA 1530.

Significantly different from the controls at the .05 level. See Mate- rials and Methods for specifics of statistical analysis.

data suggest that the increase in SCE frequency observed for both the in vivo and in vitro crude metabolite from Hordeum vulgare was due to the mutagenic intermediate and not to another component of the crude mixture. Treat- ment of cells for 6 hr induced no significant increase in SCE frequency above control (data not shown).

Exposure of Chinese hamster V79 cells to the in vivo produced crude metabolite of Hordeum vulgare induced an increase in SCE frequency similar to that observed for normal human fibroblasts (Table V). At 2.5 X lo3 his revertantslml, the increase is slightly less than that observed in human fibroblasts (for V79 cells 138% of control and 175% for fibroblasts). On the other hand, at 5 X lo3 his revertants/ml, the increase observed was slightly higher in the V79 cells than in the fibroblasts (198% vs. 162% of control) and probably reflects the inherent variability within the cell types rather than a difference in response to the metabolite. If the treatments with the crude metabolite at either concentration are compared with their respective untreated controls, the difference disappears (210% for V79 cells compared to 215% for fibroblasts). Chinese hamster cells exposed to the in vitro derived Hordeum vulgare mutagenic metabolite also exhibited a statistically significant increase in SCE frequency at concentrations above 500 his revertants/ml (Table V). It should be noted that as observed above, there was no significant increase in the SCE frequency after a 6 hr exposure to either the in vivo or in vitro synthesized metabolite from Hordeum vulgare.

Source of metabolite Concentration" SCWmetaphase (mean ? SE)

In vivo 5 ,m 19.90 2 .45 In vivo 2,500 13.95 * .39

13.17 2 .40 In vivo 1 ,OOo 11.44 * .36 In vivo 500

In vivo 200 10.01 2 .35 Untreated extract 9.47 2 .30 Control 10.05 2 .33

In vitro 5 ,OOo 14.34 * .35 13.48 ? .41 In vitro 2,500

In vitro 1 ,OOO 13.08 f .35 In vitro 500 13.37 * .70 In vitro 200 10.01 * .37 Untreated extract 9.47 2 .30 Extract plus azide 9.48 2 .51 Extract plus OAS 9.62 2 .59 Control 9.50 2 .35

* All treatments were for 24 hr (approximately two cell cycles). a Concentration is expressed as his revertants/ml as determined by reversion at the histidine locus in Salmonella ryphirnuriurn strain TA 1530.

Significantly different from the controls at the .05 level. See Mate- rials and Methods for specifics of statistical analysis.

The administration of chemically synthesized azidoalanine to normal human skin fibroblasts produced a significant increase in the frequency of SCEs over control values (Table VI). At 1 mM azidoalanine, the increase was almost twice the control and was 2.5 times control at 5 mM. The increase observed with azidoalanine roughly parallels that observed after exposure to the crude metabolite(s) synthesized from Hordeum vulgare, suggesting that there is a similarity in the interaction of both the crude metabolite and azidoalanine with mammalian cells. On the other hand, a short exposure (6 hr) to azidoalanine increased the frequency of SCEs in the fibroblasts (Table VI). This increase was roughly equal to that observed for the 48 hr exposure. In addition, azidoalanine appeared less disruptive to the fibroblasts as cell cycle progression appeared normal after either a 6 hr or 48 hr exposure (Table 11).

The effects of chemically synthesized azidoalanine on the induction of SCEs in Chinese hamster cells are presented in Table VII. As can be noted, there was a significant increase in the frequency of SCE in cells treated with azidoalanine for 24 hr, which is similar to those observed for human fibroblasts. Azidoalanine treatment of the Chinese hamster cells for 6 hr induced a significant increase in SCE frequency similar to that observed at 48 hr exposure (Table VII). These data taken as a whole suggest that azidoalanine and the crude mutagenic metabolite from Hordeum vulgare are capable of increasing the SCE frequency. The ability of azidoalanine to induce unscheduled DNA synthesis in Chi-


TABLE VI. Induction of Sister Chromatid Exchange in Normal Human Skin Fibroblasts by Azidoalanine, the Putative Azide Mutagenic Metabolite

Concentration (mM/plate) Time (hr) SCEhetaphase (mean 2 SE)

5.00 (17,000)” 48 18.59 2 .44 1 .OO (3,400) 48 16.80 2 .40 0.50 (1,700) 48 12.18 f .61 0.10 (340) 48 9.38 2 .56 0.05 (170) 48 8.58 5 .41 Control 8.62 2 .37

5.00 (17,000) 6 16.23 2 .62 1 .oo (3,400) 6 15.46 2 .62 0.50 (1,700) 6 11.72 -t .51 0.10 (340) 6 9.76 2 .61 0.05 (170) 6 9.90 2 .46 Control 8.92 t .51

a Concentration as expressed as his revertants per milliliter, as determined by reversion at the histidine locus in Salmonella typhimurium strain TA 1530. ’ Significantly different from the controls at the .05 level. See Mate- rials and Methods for specifics of statistical analysis.

TABLE VII. Induction of Sister Chromatid Exchange in Chinese Hamster V79 Cells by Azidoalanine, the Putative Azide Mutagenic Metabolite

Concentration (mM/plate) Time (hr) SCEImetaphase (mean f SE) ~~

5.00 (17,000) 24 17.23 2 .62 I .oo (3,400) 24 15.46 f .62 0.50 (1,700) 24 13.72 * .51 0.10 (340) 24 13.76 2 .61 0.05 (170) 24 9.90 2 .46 Control 8.92 * .51

5.00 (17,000) 6 16.11 2 .52 1 .oo (3,400) 6 14.87 & .63 0.50 (1,700) 6 13.15 5 .53 0.10 (340) 6 12.34 * .67 0.05 (170) 6 10.03 t .46 Control 9.01 t .51

a Concentration expressed as his revertants per milliliter, as determined by reversion at the histidine locus in Salmonella ryphimurium strain TA 1530. ’Significantly different from the controls at the .05 level. See Mate- rials and Methods for specifics of statistical analysis.

nese hamster V79 cells is shown in Table VIII. Treatment of the cells with 10 mM azidoalanine induced a response equivalent to the control suggesting that azidoalanine was unable to induce UDS. On the other hand, treatment of these cells with 2 mM methyl methanesulfonate induced an increase in r3H] thymidine uptake of over twofold. Thus, it appears that although azidoalanine is capable of inducing an increase in SCE frequency, it is unable to induce DNA repair, as measured by UDS.

DISCUSSION

Previous reports on the genotoxic and mutagenic effects of sodium azide in mammalian cells have suggested that this

TABLE VIII. Induction of Unscheduled DNA Synthesis (UDS) by Azidoalanine in Chinese Hamster Vpo Cells

Concentration CPM/ lo6 cells Percent of controla

10 rnM 3,836 1 mM 3,607 0.1 mM 4,036 2 mM MMS 10,200 PBS control 4.405

87 82 92

232

~ ~ ~ _ _ _ _ _

a Average of two experiments with two replicates/experiment.

compound is at best a weak genotoxin [Clive et al., 1979; Jones et al., 19801 and possibly nongenotoxic [Arenaz and Nilan, 1981; Arenaz et al., 1983; Sanders et al., 19791, despite its demonstrated mutagenic efficiency in bacteria and several plant species. The data presented here are the first to examine the effects of azidoalanine, the putative mutagenic intermediate of sodium azide, on a genotoxic endpoint in higher eukaryotes. From the results of this study it can be suggested that azidoalanine and the crude mutagenic intermediate from Hordeum vulgare are weakly genotoxic in two different mammalian cell lines, as they induced an increase in the frequency of sister chromatid exchanges approximately twofold. This finding is similar to the work of Veleminsky and coworkers [ 19791, who demonstrated that a cell lysate from azide-treated Hordeum vulgare seed increased the mutation rate in Saccharomyces. In contrast, treatment of mammalian cells with a crude mutagenic metabolite did not induce single-strand breaks or proteinase K-sensitive sites in Chinese hamster cells, as measured by alkaline elution [Arenaz et al., 19831. However, the fact that these metabolites did not induce single-strand breaks does not preclude their being mutagenic or genotoxic. In addition, this observation may reflect a unique mechanism of action of azidoalanine and/or the “ultimate” mutagenic intermediate.

There are apparent differences in the ability of the crude metabolite from Salmonella typhirnurium and Hordeum vulgare to elicit a genotoxic response in mammalian cells. That is, when human and Chinese hamster cells were treated with metabolite derived from Hordeum, there was an increase in the frequency of SCE (Tables IV, V). However, when cells were treated with metabolite derived from Salmonella, no increase in SCE frequency was observed (Tables I, 111). These data suggest that the two crude metabolites may be different and that the “ultimate genotoxin” may possess distinct chemical structures, although eliciting similar genotoxic responses in tester strain TA 1530. However, Ro- sichan et al. [I9831 demonstrated that Hordeum vulgare produces azidoalanine by a mechanism similar to that of Salmonella typhimurium, suggesting that at least initially the intermediate is the same. This finding would not preclude further metabolism of this compound in either Sal- monella typhimurium or Hordeum vulgare to produce the


‘‘ultimate genotoxin” with distinct chemical properties. Alternatively, the observed increase in SCE frequency may have been due to either azidoalanine or metabolite from Hordeum vulgare that escaped detoxification in mammalian cells, whereas the metabolite from Salmonella typhimurium was efficiently detoxified. This point is further supported by the observation that there was no concentration-dependent increase in SCE frequency upon azidoalanine treatment with either mammalian cell line and that there appeared to be a point at which there was no substantive increase in SCE frequency in cells treated with crude mutagenic metabolite. The idea that sodium azide may be detoxified has also been proposed by de Flora and coworkers [ 19791 and Dierickx [ 19791, who observed that azide mutagenesis in Salmonella typhimurium was decreased after preincubation of azide with rat liver S9 or gastric juices. There appears to be a species difference in response to azidoalanine or the crude metabolites, as noted by the slightly lower induction of SCEs in Chinese hamster cells compared with human cells on a SCE/metaphase basis. This finding, again, may reflect a difference in the ability of the cell lines to detoxify or activate these compounds. Although the present data cannot distinguish between these possibilities, we are investigating the metabolic fate of azidoalanine in mammalian cells.

It is interesting to note the lack of induction of unscheduled DNA synthesis in Chinese hamster cells after exposure to azidoalanine, despite the increase in SCE observed. This was true even for 10 mM azidoalanine. This result is similar to that observed by Arenaz [1981] for crude metabolite from Salmonella typhimurium. In addition, there was a significant increase in the length of cell cycle, as noted by the decrease in cells in second cell cycle after exposure to crude metabolite from both Salmonella typhimurium and Hor- deum vulgare. These data would suggest that azidoalanine or the “ultimate mutagen” may possess a mechanism of action that is unique. That is, the mutagenic intermediate may not produce a lesion that is recognized by the mammalian excision repair system. Alternatively, azidoalanine or the crude metabolite may interfere with the DNA repli- cative or repair machinery and therefore may be considered a comutagen. Preliminary evidence suggests that the crude mutagenic intermediate may inhibit DNA polymerase (un- published data). Such inhibition could account for the increase in SCE frequency observed as well as the lack of induction of UDS.

ACKNOWLEDGMENTS

The authors wish to thank Drs. E. Rae1 and R. Elizondo for their critical evaluation of this manuscript. This work was supported in part by research grant ES04154 from the National Institute of Environmental Health Sciences and Grant RR08012 from the National Institute of Health.

REFERENCES Arenaz P (1981): Doctoral Dissertation, Washington State University,

Pullman, WA. Arenaz P, Nilan RA (1981): Effect of sodium azide on sister chromatid

exchanges in human lymphocytes and Chinese hamster cells. Mutat Res 88:217-221.

Arenaz P, Nilan RA, Kleinhofs A (1983): Induction of single strand breaks by sodium azide and its proximal mutagen. Mutat Res 116:423- 429.

Clive D, Johnson KO, Spector JFS, Batson AG, Brown MM (1979): Validation and characterization of the L 5 178 TK + 1- mouse lym- phoma mutagen assay system. Mutat Res 59:61-108.

Dierickx PJ (1979): Deactivation of sodium azide in the Salmonella micro- some test. Bull Environ Contam Toxic01 22660-665.

de Flora S, Coppola R, Zanacchi P, Bennicelli C (1979): Reversal of sodium-azide mutagenicity by liver preparations and by gastric juice. Mutat Res 61:387-392.

Gichner T, Veleminsky J (1977): The very low mutagenic activity of sodium azide in Arabidopsis rhalianu. Biol Planta 19: 153-155.

Jones JA, Starkey JR, Kleinhofs A (1980): Toxicity and mutagenicity of sodium a i d e in mammalian cell cultures. Mutat Res 77:293-299.

Kamra OP, Gollapudi B (1979): Mutagenic effect of sodium azide in Drosophila melanogaster. Mutat Res 66:381-384.

Kleinhofs A, Owais W, Nilan RA (1978): Azide. Mutat Res 55:165-195. Lander L (1971): Genetic recombination in Neurospora crassa affected by

changes in the supply of cellular energy. Hereditas 69:315-319. La Velle JM, Mangold JB (1987): Structure-activity relationships of the

azide metabolite, azidoalanine in S. typhimurium. Mutat Res 177:

Lowry OH, Rosebrough NJ, Farr AL, Randall RS (1951): Protein mea- surement with the fohn phenol reagent. J Biol Chem 193:265-275.

Mangold JB, La Velle JM (1986): Synthesis and enantioselective mutagenicity of azidoalanine in Salmonella typhimun‘um. Chem Biol In- teract 60:183-190.

Maron DM, Ames BM (1983): Revised methods for the Salmonella mutagenicity test. Mutat Res 113:173-215.

Owais WM, Kleinhofs A, Nilan RA (1979): In vivo conversion of sodium azide to a stable mutagenic metabolite in Salmonella typhirnurium. Mutat Res 68:15-22.

Owais WM, Kleinhofs A, Nilan RA (1981a): Effects of L-cysteine and 0-acetyl-L-serine in the synthesis and mutagenicity of azide metabolite. Mutat Res 80:99-104.

Owais WM, Kleinhofs A, Ronald RC, Nilan RA (1981b): Isolation of an azide mutagenic metabolite in Salmonella typhimurium. Mutat Res 91:155-161.

Owais WM, Ronald RC, Kleinhofs A, Nilan RA (1986): Synthesis and mutagenicity of the two stereoisomers of azide metabolite (azidoalanine). Mutat Res 175:121-126.

Owais WM, Rosichan JL, Ronald RC, Kleinhofs A, Nilan RA (1983): A mutagenic metabolite synthesized in Salmonella typhimurium grown in the presence of azide is azidoalanine. Mutat Res 1 I8:229- 239.

Owais WM, Zarowitz MA, Gunovich RA, Hodgdon AL, Kleinhofs A, Nilan RA (1978): A mutagenic in vivo metabolite of sodium azide. Mutat Res 53:355-358.

Perry JM, Wolfe S (1974): New giemsa method for the differential staining of sister chromatids. Nature 251:156-158.

Rosichan JL, Blake N, Stallard R, Owais WM, Kleinhofs A, Nilan RA (1983): 0-Acetylserine(thio1)-lyase from barley converts sodium azide to a mutagenic metabolite. Biochim Biophys Acta 748:367- 373.

Sanders C, Nilan RA, Kleinhofs A, Vig BK (1978): Mutagenic and Chro- mosome breaking effect of azide in barley and human leucocytes. Mutat Res 50:67-75.

27-33.


Slamenova D, Gabelova A (1980): The effects of sodium azide on mammalian cells cultivated in vitro. Mutat Res 71:253-261.

Szwacka M, Ciesla Z, Klopotowski T (1979): Azide-induced mutagenesis in gram-negative bacteria is recA- and IexA-independent. Mutat Res 62~221-225.

Velerninsky J , Satava J , Kleinhofs A, Nilan RA, Gichner T (1985): In-

duction of proteinase K sensitive sites in M. luteus endonuclease sensitive sites in DNA of barley embryos by sodium azide. Mutat Res 149:431-436.

Veleminsky J, Silhankova L, Smiovska V, Gichner T (1979): Mutagenesis of Socchoromyces cerevisiae by sodium azide activated in barley. Mutat Res 61:197-205.

genotoxicity of azidoalanine in mammalian cells

Documents