Increased formic acid excretion and the development of kidneytoxicity in rats following chronic dosing with trichloroethanol,

a major metabolite of trichloroethylene

Trevor Green *, Jacky Dow, John Foster

Syngenta Central Toxicology Laboratory, Alderley Park, Macclesfield, Cheshire SK10 4TJ, UK

Received 7 April 2003; accepted 16 May 2003

Abstract

The chronic toxicity of trichloroethanol, a major metabolite of trichloroethylene, has been assessed in male Fischer

rats (60 per group) given trichloroethanol in drinking water at concentrations of 0, 0.5 and 1.0 g/l for 52 weeks. The rats

excreted large amounts of formic acid in urine reaching a maximum after 12 weeks (�/65 mg/24 h at 1 g/l) and

thereafter declining to reach an apparent steady state at 40 weeks (15�/20 mg/24 h). Urine from treated rats was more

acidic throughout the study and urinary methylmalonic acid and plasma N -methyltetrahydrofolate concentrations were

increased, indicating an acidosis, vitamin B12 deficiency and impaired folate metabolism, respectively. The rats treated

with trichloroethanol developed kidney damage over the duration of the study which was characterised by increased

urinary NAG activity, protein excretion (from 4 weeks), increased basophilia, protein accumulation and tubular

damage (from 12 to 40 weeks), increased cell replication (at week 28) and evidence in some rats of focal proliferation of

abnormal tubules at 52 weeks. It was concluded that trichloroethanol, the major metabolite of trichloroethylene,

induced nephrotoxicity in rats as a result of formic acid excretion and acidosis.

# 2003 Elsevier Ireland Ltd. All rights reserved.

Keywords: Trichloroethylene; Formic acid; Kidney toxicity

1. Introduction

Trichloroethylene was first manufactured in

quantity in the 1920s and has had a wide variety

of uses since that time. Today, 90% of trichlor-

oethylene production is used for vapour degreas-

ing of metals (IARC, 1995). Industries where

trichloroethylene has been used have provided

large cohorts for epidemiology studies, which

have failed to establish a clear link between

trichloroethylene exposure and an increase in

cancer rates in the exposed populations. Contrary

to these findings, Henschler et al. (1995) and

Vamvakas et al. (1998) reported significant in-

creases in renal cancer in two small populations

exposed in factories in Germany. The apparent

discrepancy between the outcome of these studies

and the large cohort studies is attributed, by the

* Corresponding author. Tel.: �/44-1625-515-458; fax: �/44-

1625-586-396.

E-mail address: trevor.green@syngenta.com (T. Green).

Toxicology 191 (2003) 109�/119

www.elsevier.com/locate/toxicol

0300-483X/03/$ - see front matter # 2003 Elsevier Ireland Ltd. All rights reserved.

doi:10.1016/S0300-483X(03)00206-3

authors, to the uniquely high exposures thatoccurred in Germany at that time. Consistent

with the findings in the German studies, rats also

developed kidney cancer following exposure to

trichloroethylene (Maltoni et al., 1986, 1988; NTP,

1983, 1988, 1990). However, the incidences were

very low and two studies at similar dose levels

failed to find any increase in kidney cancer

(Fukuda et al., 1983; Henschler et al., 1980).Kidney toxicity is believed to be a pre-requisite

for the development of renal cancer following

exposure to trichloroethylene (Bruning and Bolt,

2000). Extensive kidney damage was seen in rats

exposed to trichloroethylene for a lifetime (Mal-

toni et al., 1986, 1988; NTP, 1983, 1988, 1990), two

of the studies being compromised by excessive

kidney toxicity (NTP, 1983, 1988). There is alsolimited evidence of renal toxicity in exposed

human populations (Bruning et al., 1996,

1997a,b, 1998, 1999). Kidney toxicity has, there-

fore, become a critical parameter in the evaluation

of the risks to humans from exposure to trichlor-

oethylene, not only as an indicator of acute

toxicity, but also of carcinogenic risk. Mechanistic

studies in rodents and to a lesser extent in humanshave identified metabolic pathways, which may

lead to renal toxicity, particularly at high dose

levels. One, a very minor pathway involving

conjugation of trichloroethylene with glutathione,

and subsequently leading to the formation of

isomers of S-(dichlorovinyl) cysteine (DCVC), a

known renal toxicant, has been identified in rats

and humans (Goeptar et al., 1995). An alternativemechanism, based on trichloroethylene induced

folate deficiency in the rat and the excretion of

large amounts of formic acid, has also been

proposed (Green et al., 1998; Dow and Green,

2000). The two major metabolites of trichloroethy-

lene, trichloroethanol and trichloroacetic acid,

were shown to be responsible for the increase in

formic acid excretion in trichloroethylene exposedrats (Dow and Green, 2000).

Exposure to formic acid has been associated

with kidney damage in a number of species

(Jacques, 1982; Liesivuori, 1986; Liesivuori et al.,

1987, 1992; Liesivuori and Savolainen, 1987,

1991). Although the exact mechanism of toxicity

is uncertain, a number have been proposed. These

include kidney damage resulting from cytochromec oxidase inhibition and histotoxic hypoxia (Ni-

cholls, 1976; Erecinska and Wilson, 1980; Zitting

et al., 1982), inhibition of ion transport between

the lumen and proximal tubular cells (Schild et al.,

1986; Karniski and Aronson, 1985), and toxicity as

a result of acidosis (Throssell et al., 1996).

It seems possible, therefore, that the excretion of

formic acid in rats exposed to trichloroethylenecould lead to the kidney toxicity which is believed

to be causally related to the low incidences of

kidney tumours seen in some of the 2 year

carcinogenicity studies. In order to test this

hypothesis, a 1 year chronic toxicity study was

conducted in which male Fischer F344 rats, one of

the strains used in the cancer bioassays, were given

trichloroethanol in their drinking water. Maleswere chosen for this study because, although

kidney toxicity was seen in both male and female

rats in the cancer bioassays, the low incidences of

renal tumours were largely confined to males.

2. Materials and methods

2.1. Chemicals

Trichloroethanol (99%) was obtained from

Sigma-Aldrich, UK. All other reagents were

obtained from commercial suppliers at the highest

purity available.

2.2. Animals

Male Fischer 344 rats (4�/6 weeks old) were

supplied by Harlan Olac, UK. The animals were

group housed in stainless steel cages in rooms

equipped with a 12-h light�/dark cycle. The

environment of the animal room was controlled

to provide a temperature within a target range of

19�/23 8C, a relative humidity within a target range

of 40�/70%, and between 25 and 30 air changes perhour. Food (pelleted RM1 diet, Special Diet

Services Ltd, Witham, Essex, UK) was provided

ad libitum.

Animals were randomly assigned to treatment

groups by a method based on individual body-

weights. Individual animals were identified by ear

T. Green et al. / Toxicology 191 (2003) 109�/119110

punching. All animals were observed to ensurethey were normal before the start of the study and

any showing adverse signs either before or during

the study were removed from the study.

2.3. Study design

Trichloroethanol was administered to rats in

their drinking water for up to 52 weeks. Theconcentrations of trichloroethanol at the start of

the study were 0.5 and 1.0 g/l. The lower concen-

tration was chosen to give urinary formate levels

similar to those reported during daily exposure of

rats by inhalation to 500 ppm trichloroethylene

(Green et al., 1998), a typical cancer bioassay

upper dose level. However, after 4 weeks of

dosing, both concentrations gave similar urinaryformate levels. A dose response was restored by

reducing the lower dose from 0.5 to 0.35 g/l and by

the addition of folic acid (25 mg/l) to the drinking

water (from 5 weeks). Water bottles, which were

protected from light, were topped up daily and

replaced twice weekly. Drinking water consump-

tion was monitored throughout the study. Clinical

observations were made daily and body weightsrecorded at 14 day intervals. Control rats were

given normal drinking water and all rats were

given control diet ad libitum.

Kidney toxicity was assessed at monthly inter-

vals during the study by analysis of urine samples

for markers of renal damage. Urine samples were

also analysed for formic acid and for methylma-

lonic acid, a marker of vitamin B12 deficiency. N -methyltetrahydrofolate, a marker of the function-

ing of the folate pathway was monitored in

plasma. Rats (five per time point per group) were

killed after 4, 12, 16, 28, 40 and 52 weeks for

histopathological examination of kidneys. Blood

samples were collected at termination for haema-

tology and biochemistry.

2.4. Urine biochemistry

Urine was collected at monthly intervals from

five rats from each dose group, and immediately

prior to sacrifice at the times given above. The rats

were placed in metabolism cages and urine col-

lected at �/70 8C for 24 h. During this period, rats

from the treated groups continued to receivedrinking water containing trichloroethanol. After-

wards, the rats were either returned to the study or

sacrificed. The urine samples were analysed for

formic acid as described previously (Green et al.,

1998). A number of markers of kidney damage;

creatinine, protein, alkaline phosphatase (ALP),

N -acetyl glucosaminidase (NAG) and gamma-

glutamyl transferase (GGT), were determined bystandard automated methods. Urinary pH was

also determined. Urinary methylmalonic acid was

measured at weeks 12, 28 and 52 using the method

described previously (Dow and Green, 2000).

2.5. Haematology and blood biochemistry

Blood was collected by cardiac puncture at eachtermination. Part of the sample, collected in an

EDTA tube, was used for determination of red

blood cell count, mean cell haemoglobin, mean cell

haemoglobin concentration, haemoglobin, total

white cell count (WBC), haematocrit, neutrophils,

mean cell volume, platelet count, lymphocytes,

monocytes, eosinophils, basophils and leucocytes.

The remainder of the blood, collected in a lithiumheparin tube, was centrifuged to separate plasma

which was analysed for ALP, alanine transaminase

(ALT), aspartate transaminase (AST), gamma-

glutamyl transpeptidase (GGT), urea and creati-

nine by standard automated methods. Formic acid

was analysed as described above. Plasma N -

methyltetrahydrofolate, an indicator of the func-

tionality of the folate pathway, was measured atweeks 12, 28 and 52 using the method described

previously (Dow and Green, 2000).

2.6. Histopathology

Rats were sacrificed with an overdose of ha-

lothane and livers and kidneys removed and

weighed. The tissues were fixed in 10% (w/v)

neutral buffered formol saline, dehydratedthrough an ascending ethanol series and embedded

in paraffin wax. Sections (5�/7 mm) were cut and

stained with haematoxylin and eosin.

Kidney sections were also immunostained for a-

2m-globulin. 5�/7 mm sections were mounted onto

glass slides, dried overnight at 57 8C and then

T. Green et al. / Toxicology 191 (2003) 109�/119 111

dewaxed in xylene and rehydrated through des-cending grades of ethanol and stored in water for

10�/15 min. The sections were then placed into

phosphate-buffered saline (PBS) and covered with,

and exposed overnight to, the rabbit antibody to

rat a-2m-globulin, used at a dilution of 1:2000. The

antibody was washed from the sections with PBS

and the sections were then exposed to a commer-

cial visualisation kit, the Dako Strept AB complex/HRP Duet Kit. This kit is composed of a

biotinylated anti-rabbit IgG antibody and a strep-

tavidin conjugated horse-radish peroxidase en-

zyme system. After the prescribed incubation

times, the complex was washed off the slides which

were then incubated with 3,3-diaminobenzidine

tetrahydrochloride for 5 min to develop the colour

reaction indicating the localisation of the a-2m-globulin protein. The sections were then counter-

stained with haematoxylin, dehydrated in an

ascending concentration of ethanol, cleared in

xylene and mounted in DPX.

2.7. Cell replication

Three days prior to sacrifice at weeks 28 and 40,rats (five per group) were surgically implanted

with Alzet 7 day mini-pumps containing 2 ml 5-

bromo-2-deoxyuridine (BrdU) at a concentration

of 15 mg/ml in 0.9% saline. Sections from kidneys

prepared as above were stained to detect the

presence of bromodeoxyuridine (BrdU) as de-

scribed by Soames et al. (1994). Two areas were

selected in each kidney: the proximal tubules of thepars recta and the tubules of the pars convoluta. In

each region, 800 nuclei were examined and the

number containing BrdU recorded. The labelling

index was determined as the number of BrdU-

labelled nuclei per 100 nuclei. The labelling indices

from treated and control groups were compared

by a 2-sided Student’s t-test.

3. Results

3.1. Study conduct

Water consumption in the dosed groups was

comparable with that in controls throughout the

study. Within a few weeks of the start of the study,the dose response for formic acid excretion was no

longer apparent. Consequently, the 0.5 g/l dose

was reduced to 0.35 g/l after 4 weeks. It was also

found that formic acid excretion at this dose level

could be more accurately controlled by the addi-

tion of folic acid (25 mg/l) to the drinking water

from week 5 onwards. Folic acid had been shown

previously to modulate urinary formic acid levels(Dow and Green, 2000). The mean dose of

trichloroethanol received over the duration of the

study, was 18.3 and 54.3 mg/kg per day for the 0.5

(0.35) and 1.0 g/l dose groups, respectively (Fig. 1).

Body weight gain was slightly reduced at 52 weeks

by up to 5% at the top dose level (data not shown).

Kidney to body weight ratios were elevated

throughout the study (Fig. 2) whereas liver bodyweight ratios were unchanged except for a small

reduction (�/7%) after 52 weeks in rats receiving

the top dose level (data not shown).

3.2. Urine and plasma formate levels

The levels of formic acid in urine are shown for

the duration of the study in Fig. 3. The levelsincreased over the first 12 weeks of the study to

reach a maximum of approximately 30 mg/24 h at

the 0.35 g/l dose level and 60 mg/24 h in rats

receiving 1 g/l trichloroethanol. The urinary ex-

cretion of formate remained dose-dependent up to

40 weeks from which point the levels of formate

were similar (15 mg/24 h) at both dose levels. The

pH of urine was consistently more acidic through-out the study with a clear dose dependant change

up to 40 weeks (Fig. 3). Plasma formate levels

followed the same pattern as those in urine reach-

ing peaks of 0.16 and 0.07 mg/ml after 12 weeks at

the higher and lower dose levels respectively.

Thereafter the concentrations declined, at both

dose levels, to approximately 0.02 mg/ml at 40 and

52 weeks.

3.3. Urine biochemistry

Urinary N -acetyl glucosaminidase levels were

increased in both dose groups throughout the

study, and urinary protein levels were increased

at the earlier time points, suggesting renal prox-

T. Green et al. / Toxicology 191 (2003) 109�/119112

imal tubular damage (Fig. 4). The other markers

of kidney damage were unchanged from control

values. Methylmalonic acid, an indicator of vita-

min B12 deficiency, was markedly increased at 12

weeks in both dose groups (control, B/1.0; 0.35 g/l,

5.149/3.0; 1 g/l, 8.859/1.69 mg/24 h). At 28 weeks

an increase was seen only in the top dose group

(2.549/0.95 mg/24 h) and by 52 weeks the levels

had returned to control values.

3.4. Haematology and blood biochemistry

Small, but significant decreases in mean cell

volume and mean cell haemoglobin concentrations

were seen during the study. These changes weretypical of a very mild microcytic anaemia. At 52

weeks, there was also a small increase in neutro-

phils. All other haematological parameters were

normal. Clinical markers of liver damage (ALP,

ALT, AST, GGT) were unchanged throughout the

study (data not shown). Plasma levels of N -

methyltetrahydrofolate were increased up to 3-

fold in the dosed groups throughout the studydemonstrating that the methionine salvage path-

way was impaired in these animals. The largest

increases were seen in the 0.35 g/l dose group (Fig.

5).

3.5. Cell replication

Cell replication was assessed in the kidneys

taken from rats at weeks 29 and 40. At 29 weeks,

there were focal increases in cells in S-phase in the

outer cortex, but not in the inner cortex, of kidneys

from rats treated with 1 g/l trichloroethanol

(control 2.39/0.7; 1 g/l 16.59/2.9** P B/0.01).

Fig. 1. Trichloroethanol drinking water study: dose in mg/kg per day calculated from water consumption and bodyweight (NB. The

lower dose level was initially 0.5 g/l and was reduced to 0.35 g/l containing folic acid (25 mg/ml) after 4 weeks).

Fig. 2. Kidney:bodyweight ratios in rats receiving trichlor-

oethanol in drinking water for 52 weeks. *P B/0.05, **P B/0.01.

T. Green et al. / Toxicology 191 (2003) 109�/119 113

Fig. 3. Formic acid concentrations and pH values of urine from rats receiving trichloroethanol in drinking water for 52 weeks. *P B/

0.05, **P B/0.005.

Fig. 4. N -acetylglucosaminidase (NAG) activity in the urine of rats receiving trichloroethanol in drinking water for 52 weeks. *P B/

0.05, **P B/0.01.

T. Green et al. / Toxicology 191 (2003) 109�/119114

There were no significant changes in cell replica-tion in the kidneys of rats that had been treated

with 0.5 g/l trichloroethanol for 29 weeks or at

either dose level at 40 weeks.

3.6. Renal pathology

There were no treatment related macroscopic

changes at any time point in the study nor were

there any microscopic changes in the kidneys ofrats killed after 4 weeks of treatment. Livers were

normal at all time points. A dose and time

dependent increase in renal tubular basophilia

was seen between weeks 12 and 28. At week 12,

basophilia was seen in 4/5 rats receiving 1 g/l

trichloroethanol, and at 16 and 28 weeks basophi-

lia was present in high incidences and with similar

severity in both dose groups. Over this periodthere was also a dose-related increase in both the

incidences and severities of hyaline droplet accu-

mulation. The kidneys were also immunostained

for a-2m-globulin and, although treatment-related

increases in a-2m-globulin were seen in male rats at

both dose levels of trichloroethanol, they were

considered insufficient, on their own, to account

for the magnitude of the increase in hyalinedroplets observed in the H&E stained sections.

By 40 weeks, tubular degeneration, consisting of

increased cellular eosinophilia, tubular vacuola-

tion and intra-tubular cast formation, was noted in

the kidneys of all rats given 1 g/l trichloroethanol

(Fig. 6). In 4/5 rats at the 0.35 g/l dose level and in

5/5 rats at the 1 g/l dose levels there was anincrease in hyaline droplet accumulation and an

increased amount of pigmentation in the S2

portions of the proximal tubules. The levels of a-

2m-globulin in the kidney showed a dose-depen-

dent increase at this time point although, as at the

earlier times, the increase was not considered

sufficient to account for the large increase in

hyaline droplets observed in H/E stained sections(Fig. 6).

The treatment-related increase in hyaline dro-

plets and the tubular degeneration seen at 40

weeks was no longer present in the kidneys of

rats killed at 52 weeks. The treatment-related

increase in tubular pigmentation seen at 40 weeks

was still present at both dose levels and is thought

to reflect a previous or continued degree of tubulardamage and repair. Foci of ‘atypical’ tubule

hyperplasia (Fig. 7) were seen in two rats from

each trichloroethanol treated group, which were

not seen in control rats.

4. Discussion

The mode of action of trichloroethylene as arenal toxin in rodents following prolonged expo-

sure to trichloroethylene is uncertain. The toxicity,

which is only seen after prolonged exposure, has

been attributed to S -(1,2-dichlorovinyl)-L-cysteine

(DCVC), a nephrotoxic and potentially mutagenic

metabolite of trichloroethylene. However, this

Fig. 5. N -Methyltetrahydrofolate concentrations in plasma of rats receiving trichloroethanol in drinking water for 52 weeks. *P B/

0.05, **P B/0.01.

T. Green et al. / Toxicology 191 (2003) 109�/119 115

metabolite is formed in very small amounts (B/

0.005% of the dose), its formation fails to explain

the male rat specific carcinogenicity of trichlor-

oethylene and consequently its role in the devel-

opment of trichloroethylene induced

nephrotoxicity has been questioned (Birner et al.,

1993; Eyre et al., 1995a,b; Goeptar et al., 1995;

Green et al., 1997). This uncertainty led to

additional research and the identification of an

alternative mechanism.Trichloroethylene, via its two major metabolites

trichloroethanol and trichloroacetic acid, has been

shown to stimulate the excretion of large amounts

of formic acid in rat urine (Green et al., 1998). The

mechanism involves an interaction between these

metabolites and vitamin B12 which results in aninhibition of the methionine salvage pathway, a

key pathway in the recovery of tetrahydrofolate in

the rat (Dow and Green, 2000). As a consequence,

rats treated with trichloroethylene, or its two

major metabolites, become folate deficient and

can no longer completely utilise the large amounts

of formic acid produced from tryptophan and

normally used in the folic acid pathway. Theexcess formate is excreted in urine. The reported

connection between formic acid excretion, acido-

sis, and the development of kidney damage

(Jacques, 1982; Liesivuori, 1986; Liesivuori et al.,

1987, 1992; Liesivuori and Savolainen, 1987, 1991;

Throssell et al., 1996) suggested that the renal

toxicity and low incidence of renal cancer seen in

rats exposed to trichloroethylene for 2 years maybe related to this phenomenon.

Trichloroethanol was chosen to test this hypoth-

esis because it induces formic acid excretion, it is

the major metabolite of trichloroethylene in the rat

(Green and Prout, 1985), and it is not metabolised

to DCVC, a potential confounder if the study had

been conducted with trichloroethylene itself. The

dosimeter for the study was urinary formic acidconcentration, rather than the internal dose of

trichloroethanol itself. This was done in order to

determine the toxicity of formic acid to the kidney

at concentrations comparable to those produced

from trichloroethylene under the conditions of the

carcinogenicity studies. To that end, the lower

dose was selected on the basis that it gave the same

concentration of formic acid in urine as that seenin rats exposed to 500 ppm trichloroethylene by

inhalation, a typical top dose level in the trichlor-

oethylene cancer bioassays (Maltoni et al., 1988).

The higher dose level was based on previous

studies which had shown that 1 g/l trichloroetha-

nol gave maximal excretion of formic acid (Dow

and Green, 2000). Over the duration of the study,

a reduction in the amount of water consumed withage resulted in a reduction in the dose of trichlor-

oethanol which was reflected in the urine and

plasma formic acid levels, and in the magnitude of

the urinary pH change. Reductions of the same

magnitude would not be expected following either

gavage dosing or inhalation exposure, the two

routes used in the cancer bioassays (Maltoni et al.,

Fig. 7. Kidney cortex, showing abnormal focus of hyaline

droplet containing cells, taken from a rat receiving 1 g/l

trichloroethanol in drinking water for 52 weeks. H&E.

Fig. 6. Kidney cortex taken from a rat receiving 1 g/l

trichloroethanol in drinking water for 40 weeks showing an

accumulation of hyaline droplets and tubular degeneration.

T. Green et al. / Toxicology 191 (2003) 109�/119116

1986, 1988; NTP, 1983, 1988, 1990). Nevertheless,the present study has shown that formic acid

excretion is sustained over a 1 year period and

that the urine of treated animals was consistently

more acidic than that of controls.

The biochemical changes indicative of the mode

of action of trichloroethanol, which have been

reported previously in the short term (Dow and

Green, 2000), were also seen in this study. Urinaryexcretion of methylmalonic acid, a marker of

vitamin B12 deficiency was increased at 12 weeks

but thereafter declined, returning to control levels

by week 52. In contrast, plasma N -methyltetrahy-

drofolate levels remained elevated throughout the

study, indicating that the methionine salvage

pathway is still impaired, which is consistent with

the continuing excretion of formic acid. Themethionine salvage pathway was shown previously

to be far more sensitive to the effects of trichlor-

oethanol than the conversion of methylmalonyl

CoA to succinyl CoA. The recovery of the latter

pathway, but not the former, probably reflects this

difference in sensitivity and occurs because of the

reduction in the dose of trichloroethanol (in mg/kg

per day) as the study progressed.The haematological changes seen over the

duration of this study were typical of a very mild

microcytic anaemia. This effect is in contrast to the

megaloblastic anaemia normally associated with

folate deficiency in humans. However, other

studies using folate deficient rats have also failed

to induce megaloblastic anaemia, suggesting that

the rat is a poor model for this end-point,presumably due to the higher folate status in

control rats compared with humans (Johlin et al.,

1987).

The biochemical and morphological changes

seen in the kidneys of male rats in these studies

are consistent with the nephrotoxicity seen at the

end of the 2 year cancer bioassays with trichlor-

oethylene. Furthermore, the chronology of thesechanges is also consistent with its known acute and

sub chronic toxicity. Previous studies have failed

to find anything other than minor changes in the

kidneys of rats treated with high doses of trichlor-

oethylene for up to 90 days (Stott et al., 1982;

Goldsworthy et al., 1988; Green et al., 1997; NTP,

1988). The earliest changes in this study were small

increases in urinary NAG and protein from 4weeks onwards. Morphological changes were not

seen until 12 weeks when tubular hyaline droplet

formation and an increased incidence of basophilic

tubules in the cortex of the kidneys were observed.

The observation of increased basophilia is gener-

ally considered to represent newly divided cells,

which do not yet show the fully differentiated

character of normal proximal tubules, which areeosinophilic in staining. The finding occurs com-

monly in situations where proximal tubule damage

is occurring and it represents tubular regeneration

in response to the damage. Nephrotoxic chemicals

commonly induce this change in the rat kidney and

it is also seen with the spontaneous damage that

occurs in ageing rats.

Exposure to trichloroethanol also caused a non-specific accumulation of protein within the prox-

imal tubules of the male rat kidney. Similar

findings have been reported in the rat kidney as

a consequence of acidosis impaired proteolysis

(Throssell et al., 1996) supporting the view that

protein accumulation in this study is a result of

formic acid induced acidosis. If this is the case, the

reason for protein accumulation in the kidneydiffers from that known for a-2m-globulin where a

chemical or its metabolites are known to bind to

the protein. Nevertheless, there is evidence that the

consequences, in terms of the long-term effects on

the kidney, may be similar. At 40 weeks, protein

accumulation was accompanied by a significant

degree of tubular damage and at 52 weeks there

was evidence in two out of five rats examined ineach dose group of focal proliferations of abnor-

mal tubules which may represent pre-neoplastic

development.

In conclusion, administration of trichloroetha-

nol to rats for 52 weeks caused a marked increase

in basophilic tubules in the cortex of the kidneys,

an increase in cell replication rates, and provided

evidence of focal proliferation, changes consistentwith the early development of the nephropathy

normally seen in 2-year-old rats. The accompany-

ing high excretion of formic acid, urinary pH

change and protein accumulation suggests that the

morphological changes are a result of formic acid

induced acidosis. Overall, the changes in the rat

kidney induced by trichloroethanol and formic

T. Green et al. / Toxicology 191 (2003) 109�/119 117

acid excretion are consistent in nature with theknown acute and chronic renal toxicology of

trichloroethylene in the rat.

Acknowledgements

These studies were sponsored by the European

Chlorinated Solvents Association, Brussels, Bel-

gium.

References

Birner, G., Vamvakas, S., Dekant, W., Henschler, D., 1993.

The nephrotoxic and genotoxic N -acetyl-S -dichlorovinyl-L-

cysteine is a urinary metabolite after occupational 1,1,2-

trichloroethene exposure in humans: implications for the

risk of trichloroethene exposure. Environ. Health Perspect.

99, 281�/284.

Bruning, T., Bolt, H.M., 2000. Renal toxicity and carcinogeni-

city of trichloroethylene: key results, mechanisms, and

controversies. Crit. Rev. Toxicol. 30, 253�/285.

Bruning, T., Golka, K., Makropoulos, V., Bolt, H.M., 1996.

Preexistence of chronic tubular damage in cases of renal cell

cancer after long and high exposure to trichloroethylene.

Arch. Toxicol. 70, 259�/260.

Bruning, T., Lammert, M., Kempes, M., Their, R., Golka, K.,

Bolt, H.M., 1997a. Influence of polymorphisms of GSTM1

and GSTT1 for risk of renal cell cancer in workers with

long-term and high occupational exposure to trichloroethy-

lene. Arch. Toxicol. 71, 596�/599.

Bruning, T., Mann, H., Metzler, H., Sundberg, A.G.M., Bolt,

H.M., 1997b. Subclinical changes in urinary protein excre-

tion of renal cell cancer patients: effect of high occupational

exposure to trichloroethylene. Centr. Eur. J. Occup. En-

viron. Med. 3, 167�/174.

Bruning, T., Vamvakas, S., Makropoulos, V., Birner, G., 1998.

Acute intoxication with trichloroethylene: clinical symp-

toms, toxicokinetics, metabolism and development of bio-

chemical parameters for renal damage. Toxicol. Sci. 41,

157�/165.

Bruning, T., Sundberg, A.G.M., Birner, G., Lammert, M., Bolt,

H.M., Appelkvist, E.E.., Nilsson, R.., Dallner, G., 1999.

Glutathione S-transferase a as a marker for tubular damage

after trichloroethylene exposure. Arch. Toxicol. 73, 246�/

254.

Dow, J., Green, T., 2000. Trichloroethylene induced vitamin

B12 and folate deficiency leads to increased formic acid

excretion in the rat. Toxicology 146, 123�/136.

Erecinska, M., Wilson, D.F., 1980. Inhibitors of cytochrome c

oxidase. Pharmacol. Ther. 8, 1�/20.

Eyre, R.J., Stevens, D.K., Parker, J.C., Bull, R.J., 1995a. Acid-

labile adducts to protein can be used as indicators of the

cysteine S-conjugate pathway of trichloroethene metabo-

lism. J. Toxicol. Environ. Health 46, 443�/464.

Eyre, R.J., Stevens, D.K., Parker, J.C., Bull, R.J., 1995b. Renal

activation of trichloroethene and S -(1,2-dichlorovinyl)-L-

cysteine and cell proliferative responses in the kidneys of

F344 rats and B6C3F1 mice. J. Toxicol. Environ. Health 46,

465�/481.

Fukuda, K., Takemoto, K., Tsurata, H., 1983. Inhalation

carcinogenicity of trichloroethylene in mice and rats. Ind.

Health 21, 243�/254.

Goeptar, A.R., Commandeur, J.N.M., van Ommen, B., van

Bladeren, P.J., Vermeulen, N.P.E., 1995. Metabolism and

kinetics of trichloroethylene in relation to toxicity and

carcinogenicity. Relevance of the mercapturic acid pathway.

Chem. Res. Toxicol. 8, 3�/21.

Goldsworthy, T.L., Lyght, O., Burnett, V.L., Popp, J.A., 1988.

Potential role of alpha-2m-globulin, protein droplet accu-

mulation and cell replication in the renal carcinogenicity of

rats exposed to trichloroethylene, perchloroethylene and

pentachloroethane. Toxicol. Appl. Pharmacol. 96, 367�/379.

Green, T., Prout, M.S., 1985. Species differences in response to

trichloroethylene. II. Biotransformation in rats and mice.

Toxicol. Appl. Pharmacol. 79, 401�/411.

Green, T., Dow, J.L., Ellis, M.K., Foster, J.R., Odum, J., 1997.

The role of glutathione conjugation in the development of

kidney tumours in rats exposed to trichloroethylene. Chem-

Biol. Interact. 105, 99�/117.

Green, T., Dow, J., Foster, J.R., Hext, P.M., 1998. Formic acid

excretion in rats exposed to trichloroethylene: a possible

explanation for renal toxicity in long-term studies. Toxicol-

ogy 127, 39�/47.

Henschler, D., Romen, W., Elsasser, H.M., Reichert, D., Eder,

E., Radwan, Z., 1980. Carcinogenicity study of trichlor-

oethylene by long term inhalation in three animal species.

Arch. Toxicol. 43, 237�/248.

Henschler, D., Vamvakas, S., Lammert, M., Dekant, W.,

Kraus, B., Thomas, B., Ulm, K., 1995. Increased incidence

of renal cell tumours in a cohort of cardboard workers

exposed to trichloroethylene. Arch Toxicol. 70, 130�/133.

IARC, 1995. IARC Monographs on the Evaluation of Carci-

nogenic Risks to Humans. Vol. 63, Dry Cleaning, Some

Chlorinated Solvents and Other Industrial Chemicals.

IARC, Lyon.

Jacques, S., 1982. Acute renal failure following ingestion of

formic acid. Nursing Times, August 4, 1312�/1315.

Johlin, J.C, Fortman, C.S., Nghiem, D.D., Tephly, T.R., 1987.

Studies on the role of folic acid and folate dependant

enzymes in human methanol poisoning. Biochem. Pharma-

col. 31, 557�/561.

Karniski, P.L., Aronson, P.S., 1985. Chloride/formate exchange

with formic acid recycling: a mechanism of active chloride

transport across epithelial membranes. Proc. Natl. Acad.

Sci. 82, 6362�/6365.

Liesivuori, J., 1986. Slow urinary elimination of formic acid in

occupationally exposed farmers. Ann. Occup. Hyg. 27,

327�/329.

T. Green et al. / Toxicology 191 (2003) 109�/119118

Liesivuori, J., Savolainen, H., 1987. Effect of renal formic acid

excretion on urinary calcium and ammonia concentrations.

Klin. Wochenschr. 65, 860�/863.

Liesivuori, J., Savolainen, H., 1991. Methanol and formic acid

toxicity: biochemical mechanisms. Pharmacol. Toxicol. 69,

157�/163.

Liesivuori, J., Kosma, V.M., Naukkarinen, A., Savolainen, H.,

1987. Kinetics and toxic effects of repeated intravenous

dosage of formic acid in rabbits. Br. J. Exp. Pathol. 68,

853�/861.

Liesivuori, J., Laitinen, J., Savolainen, H., 1992. Kinetics and

renal effects of formic acid in occupationally exposed

farmers. Arch. Toxicol. 66, 522�/524.

Maltoni, C., Lefemine, G., Cotti, G., 1986. Experimental

research on trichloroethylene carcinogenesis. In: Maltoni,

C., Mehlman, M.A. (Eds.), Archives of Research on

Industrial Carcinogenesis, vol. 5. Princeton Sci. Publ. Co.

Inc, Princeton, NJ.

Maltoni, C., Lefemine, G., Cotti, G., Perino, G., 1988. Long-

term carcinogenicity bioassays on trichloroethylene admi-

nistered by inhalation to Sprague�/Dawley rats and Swiss

and B6C3F1 mice. Ann. New York Acad. Sci. 534, 316�/342.

Nicholls, P., 1976. The effect of formate transport on cyto-

chrome aa3 and on electron transport in the intact

respiratory chain. Biochim. Biophys. Acta 430, 13�/29.

NTP, 1983. National Toxicology Program Carcinogenesis

bioassays of trichloroethylene in F344 rats and B6C3F1

mice. US Dept. of Health and Human Services, NIH

Publication No. 82-1799.

NTP, 1988. National Toxicology Program technical report on

the toxicology and carcinogenesis studies of trichloroethy-

lene in four strains of rats. NTP TR 273, Research Triangle

Park, NC.

NTP, 1990. National Toxicology Program Carcinogenesis

studies of trichloroethylene (without epichlorohydrin) in

F344 rats and B6C3F1 mice. NTP TR 243, Research

Triangle Park, NC.

Schild, L., Giebisch, G., Karniski, L., Aronson, P.S., 1986.

Chloride transport in the mammalian proximal tubule.

Pflugers Arch. 407, S156�/S159.

Soames, A.R., Lavender, D., Foster, J.R., Williams, S.M.,

Wheeldon, EB, 1994. Image analysis of S-phase in rat and

mouse liver. J. Histochem. Cytochem. 42, 939�/944.

Stott, W.T., Quast, J.F., Watanabe, P.G., 1982. The pharma-

cokinetics and macromolecular interactions of trichlor-

oethylene in mice and rats. Toxicol. Appl. Pharmacol. 62,

137�/151.

Throssell, D., Harris, K.P.G., Bevington, A., Furness, P.N.,

Howie, A.J., Walls, J., 1996. Renal effects of metabolic

acidosis in the normal rat. Nephron 73, 450�/455.

Vamvakas, S., Bruning, T., Thomasson, B., Lammert, M., Bolt,

H.M., Dekant, W., Birner, G., Henschler, D., Ulm, K.,

1998. Renal cell cancer correlated with occupational ex-

posure to trichloroethylene. J. Cancer Res. Clin. Oncol. 124,

374�/382.

Zitting, A., Savolainen, H., Nickels, J., 1982. Biochemical and

toxicological effects of single and repeated exposure to

polyacetal thermodegradation products. Environ. Res. 29,

287�/296.

T. Green et al. / Toxicology 191 (2003) 109�/119 119