Cellular metabolism of arsenic studied in

mammalian cells in vitro

Anna B. Fischer Institute of Hygiene, Justus-Liebig University, Giessen, Federal Republic of Germany

J.P. Buchet and R.R. Lauwerys Unit~ de Toxicologic Industrielle et M~dicale. University of Louvain, Clos Chapelle-aux-Champs, B- 1200 Brussels, Belgium

Abstract

The cytotoxicity of trivalent and pentavalent inorganic arsenic was studied in cultured mouse fibroblasts. Concentrations of As(Ill) in the I~M range and approximately 10-fold higher concentrations of As(V) led to a reduction of cellular proliferation and viability with a concomitant increase of LDH release and stimulation of lactate production. Cells pretreated with a low As(Ill) concentration were less sensitive to toxic doses of As(Ill) or As(V).

Uptake of As(Ill) by the cells was greater than that of As(V). Both forms of inorganic arsenic were converted intracellularly to monomethylarsonic (MMA) and dimethylarsinic (DMA) acids, which were subsequently released into the culture medium. In As- pretreated cells, which proved more resistant to As toxicity, biotransformation of inorganic to MMA and DMA was increased.

Introduction

After ingestion or parenteral application of trivalent or pentavalent inorganic arsenic (Asi), the organic metabolites monomethylarsonic (MMA) and dimethylarsinic (DMA) acids have been identified in human urine together with Asi. Numerous studies have conf'umed the capacity of several mammalian species to excrete methylated derivatives of arsenic in urine following treatment with Asi (Buchet et al., 1980, Odonaka et al., 1980, Vahter and Envall, 1983), Arsenic methylation is also documented in micro-organisms, but the contribution of the microflora of the gut to the biotransformation of As is insignificant (Rowland and Davies, 1982). It is by now established that the liver is the main site of As methylation and that methylation constitutes a detoxification process. In cytotoxicity studies with mammalian cells in vitro, the development of As adaption was observed in L- A mouse fibroblast cultures following chronic exposure to AsCI3 (Fischer et al., 1985). It thus appears that these cells possess the capacity for As detoxification. This study was designed to investigate the metabolic events in the fibroblasts.

Material and Methods

Static suspension cultures of L-A mouse fibroblasts, a subline of L 929 cells, were maintained in Medium 199 supplemented with 10% horse serum. Suspensions with - 10Ycells mL -1 (15 mL volume) were seeded into Breed-Demeter bottles and dosed with AsCI3 or

Na2HAsO4. At intervals aliquots of cell suspensions were withdrawn from the flasks for biological and chemical studies. Cell numbers were determined by electronic counting (Coulter Counter), viability assessed by the trypan blue test, and lactic dehydrogenase (LDI-1) and lactic acid levels were measured in the supernatants (test sets by Boehringer Mannheim, FRG). Asi and its methylated metabolites MMA and DMA were measured by atomic absorption spectrometry using a quartz tube as atomiser (MHS 20 device from Perkin-Elmer; Buchet and Lauwerys, 1981). The As compounds were reduced to their corresponding arsines with NaBH4, these were condensed in a liquid nitrogen trap and arsines were released by progressive heating and detected in the order of their boiling points. Samples for chemical analysis were prepared as outlined in Figure 1.

Arsenic-adapted cells were produced by culturing L-A cells in the usual growth medium containing AsC13 for up to 20 weeks. Adaptation was tested by exposing these cultures to acutely toxic concentrations of lrivalent or pentavalent As salts. The pretreated fibroblasts were centrifuged, resuspended in As-free medium and dosed according to the respective experimental protocols. There was no time interval between the adaptation period and the challenge tests. Studies on As cytotoxicity and adaptation were performed at least three times. Chemical analyses of As-exposed cells were done in four experiments.

Statistical evaluation of data was performed using the Student's t-test.

88 Cellular metabolism of arsenic studied in mammalian cells in vitro

Cell suspension

I Centrifugation

Cells 2 washes (PBS)

I Centrifugation

Pellet + 5% TCA

Centrifugation

/ \ Pel le t Supernatant (discarded) ("culture medium")

(for analysis)

I00

Supernatant + 10% TCA (1:1)

60

I Gentrifuoation

/ \ t Pel le t Supernatant (discarded) ("cells")

(for analysis)

Figure 1 Diagram showing the processing of samples for chemical analysis.

Results

Cytotoxicity Following short term exposure AsCI3 and Na2HAsO4 inhibited cellular growth and decreased viability, but pentavalent As was less toxic than tfivalent As. In L-A cells the concentra t ions leading to a 50% reduct ion of proliferation (IC50) were 2.5-5.0 10- 6 M AsCI3 and 5.0-7.5

Cell number (~ of respective controls)

[ ] no As [ ] /,pM AsC[ 3 [ ] 8,uM AsCI 3

Co As Co As

+

day 4

i

§ I

day 6

Figure 2 Effect of As pretreatment on the S~.lrt-term cytotoxicity of arsenic. 110,000 L-A cells mL (three replicates/group) were exposed to 2.5, 5.0 and 10 p.M AsCl3 for 6 days and cell numbers were determined on days 2, 4 and 6. Co = previously unexposed cells; As = cells cultured in the presence of I p.M AsCl3 for I week. Statistically significant differences between pretreated and unpretreated cells are shown: + = p < 1%; ++ = p <0.1%.

10 -5 M Na2HAsO4 following a 7-day treatment. The corresponding LC50s were 1.0-2.0 10- 5 M for AsCI3 and 2.0-4.0 10 -4 M for Na2HAsO4. With reduced viability gross membrane damage could be demonstrated through the release of lactic dehydrogenase (LDH) into the medium. Lactic acid production was raised, indicating disturbed carbohydrate metabolism (Table 1, Figure 3: controls + Na2I-IAsO4).

Table 1 Effect of arsenic on L-A mouse fibroblasts. A suspension (15 mL) containing 110,000 cells mL "! was inoculated into Breed- Demeter flasks and dosed with varying concentrations of As salts. Cell number, viability, LDH release, and lactic acid production were determined 7 days later. There were four replicates per experimental group.

As added Cell number Viability LDH Lactic acid (106 mL -1 (% + SD) (mU 10 -6 cells (~ 10- 6 cells

(~bt) (rig As ml_, 1) + SD) + SD) + SD)

AsCI3 0 1.196 + 0.050 96.5 + 0.8 16.9 + 1.1 348 + 3.5 2.5 187.3 0.788 + 0.031 96.0 + 0.7 16.6 + 1.8 365 + 6.5 5 374.6 0.663 + 0.028 97.3 + 1.3 17.0 + 3.2 438 + 32.5 10 749.2 0.236 + 0.013 70.8 + 3.3 95.4 + 8.9 643 + 50.4 20 1498.4 0.089 + 0.002 27.3 + 6.3 200.0 + 11.2 n.d.

Na2HAsO4 0 1.254 + 0.026 96.8 + 0.8 18.1 + 1.2 348 + 3.5

75 5619.0 0.301 + 0.013 96.2 + 0.9 16.6 + 3.1 438 + 3.3 100 7492.0 0.175 + 0.009 85.3 + 3.3 95.4 + 8.9 643 + 8.0 200 14984.0 0.062 + 0.007 15.5 + 9.1 n.d. n.d.

n.d. = not determined

Anna B. Fischer, J. P. Buchet and R. R. Lauwreys 89

1.5

1.0.

0.5

[ ] without As [ ] 7.5 �9 10"5M NaHAsO, [] IO-~M N~H As~4 "

Cell number (10 6/ ml}

100

50

Viability {%)

100

Controls As-celIs Controls As-cells

LDH (mU/10 6cells)

Controls

1000

500

As-cells

Lactate { pg/10 6cells)

Controls As-cells

Figure 3 Effect of As pretreatment on the short-term cytotoxicity of arsenic. 115,000 L-A cells ml_, "1 (four replicates~group) were exposed to 75 and 100 p.M Na2HAs04, and cell numbers, viability, LDH release, and lactic acid production were determined on day 7. Controls = previously unexposed cells; As-cells = cells cultured for 14

weeks in the presence of I p.M AsCI3. Statistically significant differences between pretreated and unpretreated cells are shown: * = p < 1%; ** = p < 0.1%.

The effects of pretreatment were tested by culturing L-A cells for up to 20 weeks in medium containing 1 lxM AsCI3. Thereafter, the cells were exposed for 6 days to toxic levels of AsCI3 and Na2HAsO4. In both cases As cytotoxicity was decreased in the conditioned cells, as evidenced by improved growth and viability and by normalised LDH, and lactate (Figures 2, 3). An adaptatlon period of 1 week was sufficient to convey significant protection (Figure 2).

Arsenic uptake and metabolism The As contents of L-A cells were analysed after incubation with AsC13. Figure 4 shows the cellular contents on days 2, 4 and 6 of the experiment. With increasing As dosage the intracellular concentrations of Asi rose, but did not differ consistently between the different days. The cells metabolised the Asi to MMA and DMA. Cells not previously exposed did so only on day 6, whereas the metabolites could already be detected in the As- adapted cells on day 4. Cells treated with 10 lxM AsCI3 were unable to produce MMA and DMA and the As metabolism of the pretreated cells was also strongly inhibited by the high As concentration.

The methylated compounds were released into the medium, and only a fraction could be detected intracellularly (Figure 5). Since the MMA and DMA in the culture medium were produced by the fibroblasts, the total As taken up and metabolised included the intracellular As as well as the methylated species in the culture supernatant. In the course of 6 days the conlrols handled 32, 17 and 1% of the added 2.5, 5 and 10 lxM AsC13, respectively, i.e. there was a dose-dependent decrease in As uptake and

metabolism. The pretreated cells performed slightly better, but showed the same trend with 39, 24 and 5%. The As-conditioned fibroblasts methylated As more efficiently. The controls exposed to 2.5, 5.0 and 10 laM AsCI3 metabolised 98.7, 97.3 and 0% of the As taken up, whereas the adapted cells methylated 99.1, 98.9 and 87.3% of the accumulated As. The ceils produced more DMA than MMA, but DMA was proportionally higher at low exposure levels and decreased at higher concentrations.

The biotransformation of the pentavalent compound Na2HAsO4 was similar to that of the trivalent AsCI3. However, it was found that the pentavalent salt was accumulated and metabolised by the fibroblasts only to a small extent, which explains its lower cytotoxicity (Figure 5).

Discussion

The trivalent and pentavalent arsenic compounds produce cytotoxic effects over a wide concentration range. The minimal concentration of AsCI3 inhibiting proliferation of L-A mouse fibroblasts exposed for 7 days is > 1 lxM, 50% growth inhibition occurs at 2.5-5.0 ptd and significant cell death with the concomitant leakage of cytoplasmic LDH through damaged membranes is noted only at___ 1 lxM. Thus AsCI3 is nearly as cytotoxic as CdCI2, but differs from CdC12 in that the growth- inhibi t ing and lethal concentrations are much farther apart from each other (Fischer and Skreb 1980; Fischer 1981). Na2HAsO4 is over one log factor less effective, which can be explained by the low uptake of the arsenate anion and is in agreement with in vivo observations of Vahter and Norin (1980).

90 Cellular metabolism of arsenic studied in mammalian cells in vitro

10

8

4

2

ng

[ ] DMA

[ ] MMA As/10 v ceils I]]] As i

Controts

day 2 day 4 02.5510 02.5510

As -ce l l s

doy 6 02.5 510

)uM AsCI 3

day 2 0 2.5 5 l0

+

+

day 4 02.5510

O C

�9

o o o 0

day 02.5 5

~ C o

r D

r o

D

6 10

Figure 4 Arsenic content of L-A mouse fibroblasts (ng 10 -6 cells) cultured for 6 days in the presence of Asi. 110,000 L-A cells mL "1 (4 replicates per group) were exposed to O, 2.5, 5.0 and 10 p.M AsCI3. Co = control cultures; As =

cultures pretreated with i laM AsCl3for 20 weeks; + = traces.

The arsenic salts cause elevated lactate production, which is interpreted as a compensating mechanism counteracting effects on cellular respiration. It has been shown that As alters hepatic mitochondrial respiration and morphology (Fowler and Woods, 1979; Ghafgazi et al., 1980).

In the course of a 6 day incubation with AsCI3 the cellular contents of Asi hardly change, but MMA and DMA can be detected eventually. This means that the inorganic species is converted into the methylated forms, which are then rapidly released into the medium. The cells need a certain induction time, which explains why the metabolites are detected earlier in the pretreated cells. Methylation is most effective at low levels of As exposure, while the bioconversion of Asi, especially into DMA, decreases at higher concentrations. A similar observation was made with adequately fortified rat liver homogenate: the production of DMA but not of MMA is inhibited by a high concentration of Asi (Buchet and Lauwerys, 1985). At comparable levels of As uptake [see, e.g. cells treated with 2.5 I.taM As(III) or 50 lxM As(V), Figure 5], methylation of As(III) was more efficient than that of As(V). This strongly suggests that As(V) must be reduced to As(III) before being methylated and agrees well with in vivo findings by Vahter and Envall (1983).

Conditioned L-A cells grow better in the presence of Asi than previously unexposed cells; they also exhibit more efficient methylation. Pretreatment with As(III) causes adaptation to toxic levels of trivalent and pentavalent As salts. This means that in both cases Asi is converted

intracellularly to the same chemical form. Furthermore, the experiments show that cells adapt to As by increasing the biotransformation of Asi to the methylated forms MMA and DMA. Thus cultured cells provide a model system to elucidate the cellular events involved in arsenic tolerance.

In summary, the toxicity of inorganic arsenic compounds depends on their valency and on a specific cell's, tissue's or organism's capacity for detoxification by methylation. The uptake of As(III) considerably exceeds that of As(V). The process of methylation, which is localised in the cytoplasm, requires As(III) as a substrate so that As(V) must first be reduced (Vahter and Envall, 1983). Methylation appears to involve two separate enzymatic pathways as shown by the observations that excess Asi or addition of mercuric ions inhibits the formation of DMA but not MMA and that it was not possible to produce DMA in vitro using MMA as a substrate (Buchet and Lauwerys, 1985). Furthermore liver insufficiency leads to reduced excretion of MMA and increased excretion of DMA (Buchet et al., 1984). Possibly MMA and DMA have a common precursor. Methylation depends on the production of high energy cofactors and the functioning of mitochondria. S- adenosylmethionine is the essential methyl group donor and hydroxocobalamine stimulates the reaction in vitro (Buchet and Lauwerys, 1985). Inhibition of methyltransferase inhibits methylation and increases tissue retention of Asi (Marafante and Vahter, 1984). Reduced glutathion (GSH) plays an important role, but its exact function is not known. One possibility is that

Anna B. Fischer, J. P. Buchet and R. R. Lauwreys 91

D ex t race l lu la r DMA

I~1 " MMA

l-El i n t race t l u la r D M A

I I ', MMA

I]71 " As i

100

50

30 20

10.

5

3 2

1

0.5 0.3 0.2

0.1

ng As/mr cell suspension

Controls As -ce i l s

0 2.5 5 10 0 2.5 5 10 50 pM ' As III ' As V

Figure 5 Arsenic content of L-A fibroblasts (ng 10 -6 cells) and of culture medium (ng mL "1) after 6 days incubation in the presence of Asi. 110.000 L-A cells mL "1 (4 replicates per group) were exposed to O, 25, 5.0 and 10 BM AsCl3 or

50 BM Na2HAs04. Co = control cultures; As = cultures pretreated with i BM AsCl3 for 20 weeks.

GSH prevents the oxidation of As(III) to As(V) by maintaining reducing conditions in the cell (Buchet and Lauwerys, 1987). The final step of As detoxication is the

rapid excretion of the methylated metabolites, which are less toxic and have a much lower affinity for cellular constituents (Vahter and Marafante, 1983).

92 Cel lu lar me tabo l i sm o f arsen ic s tud ied in m a m m a l i a n cel ls in vi tro

R e f e r e n c e s

Buchet, J.P. and Lauwerys, R. 1981. Evaluation of exposure to inorganic arsenic in man. In: Forechetfi, S. (ed.), Analytical Techniques for Heavy Metals in Biological Fluids. Elsevier Science Publishers BV, Amsterdam.

Buchet, J.P. and Lauwerys, R. 1985. Study of inorganic arsenic methylation by rat liver in vitro: relevance for the interpretation of observations in man. Arch. Toxicol., 57, 125- 129.

Buchet, J.P., Lauwerys, R. and Roels, H. 1980. Comparison of several methods for the detemfination of arsenic compounds in water and urine. Int. Arch. Occup. Environ. Health, 46, 11-29.

Buchet, J.P., Lauwerys, R. and Roels, H. 1981. Comparison of the urinary excretion of arsenic metabolites after a single oral dose of sodium arsenite, monomethylarsonate, or demethylarsinate in man. Int. Arch. Occup. Environ. Health, 48, 71-79.

Buchet, LP., Geubel, A., Pauwels, S., Mahien, P. and Lauwerys, R. 1984. The influence of liver disease on the methylafion of arsenite in humans. Arch. Toxicol., 55, 151-154.

Buchet, J.P. and Lauwerys, R. 1987. Study of factors influencing the in vivo methylation of inorganic arsenic in rats. Toxicol. Appl. Pharmacol., 91, 65-74.

Fischer, A.B. 1981. Comparative studies of heavy metal toxicity at the cenular level In: International Conference on Heavy Metals in the Environment, pp.486-489. CEP Consultants, Edinburgh.

Fischer, A.B., Buchet, J.P. and Lauwerys, R.R. 1985. Arsenic uptake, cytotoxicity and detoxification studied in mammalian ceils in culture. Arch. Toxicol., 57, 168-172.

Fischer, A.B. and Skreb, Y. 1980. Cytotoxicity of manganese for mammalian cells in vitro - comparison with lead, mercury and

cadmium. Zbl. Bakt. Hyg. I. Abt. Orig. B, 171, 525-537. Fowler, B.A. and Woods, J.S. 1979. The effects of prolonged oral

arsenate exposure on liver mitochondria of mice: morphometric and biochemical studies. Toxicol. Appl. Pharmacol., 50, 177-187.

Ghafgazi, T., Ridlington, J.W. and Fowler B.A. 1980. The effects of acute and subacute sodium arsenite administration on carbohydrate metabolism. Toxicol. Appl. Pharmacol., 55, 126-130.

Marafante, E. and Vahter, M. 1984. The effect of methyl transferase inhibition on the metabolism of [74As] arsenite in mice and rabbits. Chem.-Biol. Interactions, 50, 49-57.

Odonaka, Y., Matano, O. and Goto, S. 1980. Biomethylation of inorganic arsenic by the rat and some laboratory animals. Bull. Environ. Contain. Tox&ol., 24, 452-459.

Rowland, LR and Davies, M.J. 1981. In vitro metabolism of inorganic arsenic by the gastrointestinal microflora of the rat. J. Appl. Toxicol., 1,278-283.

Vahter, M. and Envall, J. 1983. In vivo reduction of arsenate in mice and rabbits. Environ. Res., 32, 14-24.

Vahter, M. and Marafante, E. 1983. Intracellular interaction and metabolic fate of arsenite and arsenate in mice and rabbits. Chem. Biol. Interact., 47, 29-44.

Vahter, M. and Norin, H. 1980. Metabolism of 74As-labelled trivalent and pentavalent inorganic arsenic in mice. Environ. Res., 2 I, 446-457.

(Manuscript No. 196: received July 24, and accepted for publication

September 25, 1989.)