the effects of dimethylnitrosamine and allyl alcohol on primary maintenance cultures of adult rabbit...
TRANSCRIPT
Toxicology, 24 (1982) 159--167 Elsevier Scientific Publishers Ireland Ltd.
THE EFFECTS OF DIMETHYLNITROSAMINE AND ALLYL ALCOHOL ON PRIMARY MAINTENANCE CULTURES OF ADULT RABBIT HEPATOCYTES
ALAN POOLE a'* , L.F. CHASSEAUD a and JAMES W. BRIDGES b
aHuntingdon Research Centre, Huntingdon, Cambs, PE18 6ES and bRobens Institute of Industrial and Environmental Health and Safety, University of Surrey, Guildford, Surrey, GU2 5XH (Great Britain)
(Received October 9th, 1981) (Accepted April 18th, 1982)
SUMMARY
Cultures of adult rabbit hepatocytes have been used to study the early toxic effects of 2 model hepatotoxins, dimethylnitrosamine and allyl alcohol. Leakage of glutamate oxaloacetate transaminase and glutamate pyruvate transaminase into the cell culture medium was a sensitive indicator of plasma membrane damage by these compounds and a dose-response relationship was observed. By contrast, 7-glutamyltranspeptidase and alkaline phosphatase were insensitive markers. The effects of dimethylnitrosamine were slower to develop. Dimethylnitrosamine also produced a dose-related inhibition of protein synthesis after 4 h, a decrease in NADPH diaphorase and an increase in non-specific esterase after 20 h. Dimethylnitrosamine, unlike allyl alcohol, caused extensive disruption of ribosome association with the endoplasmic reticulum.
INTRODUCTION
The liver is highly susceptible to chemical damage, partly because of the ability of hepatocytes to concentrate chemicals intracellularly and to con- vert them to active metabolites. Isolated hepatocytes are therefore an obvious model for investigating the toxic properties of chemicals [1].
*Present address: MRC Pneumoconiosis Unit, Llandough Hospital, Penarth, South Glamorgan, Great Britain. Abbreviations: DMNA, dimethylnitrosamine; GOT, glutamate oxaloacetate transaminase; GPT, glutamate pyruvate transaminase; -~-GT, 7-glutamyl-transpeptidase; LAP, leucine aminopeptidase.
0300-483X/82/0000--0000/$02.75 © 1982 Elsevier Scientific Publishers Ireland Ltd. 159
Hepatic cell lines tend to lack cytochrome P-450 and are therefore poor models while freshly isolated cells may respond atypically to chemicals due to trauma suffered in the isolation procedure [2]. An alternative approach is the use of cells maintained in short-term monolayer culture which retain many of the functional characteristics of hepatocytes. Rat hepatocytes in culture tend to lose much of their cytochrome P-450 content [1,3]. Cytochrome P-450 levels in cultures of rabbit hepatocytes are much more stable than those of rats and hence rabbit hepatocyte cultures may provide a particularly appropriate in vitro model for studying the early toxic effects of chemicals on the liver [4].
In the present study, the effects of 2 model hepatotoxins, allyl alcohol, which produces mainly periportal damage, and dimethylnitrosamine (DMNA) which acts particularly in the centrilobular region via the aegis of an "un- known" active metabolites(s) [6], on adult rabbit hepatocytes are investi- gated.
METHODS
Culture Adult rabbit hepatocytes were prepared and cultured in either Nunc
culture flasks (area 21 cm:) containing sterile glass coverslips (22 mm in diameter) or in larger culture dishes (area 57 cm:) as described previously [4].
Enzyme leakage These experiments were conducted after the cells had been maintained
for 3 days in rotary culture [4]. The old culture medium was removed and replaced with fresh phenol red free medium containingvarious concentrations of dimethylnitrosamine (0--500 ug/ml) or ailyl alcohol (0--500 ug/ml). At various time intervals over a 60-h period, samples of medium were removed and assayed for enzyme activity. The activity of glutamate oxaloacetate transaminases (GOT) and glutamate pyruvate transaminases (GPT) were determined with an LKB 8600 Reaction Rate Analyser using a Diamed Test kit (J.T. Baker Chemicals Ltd., Holland). Alkaline phosphatase (AKP) activity was determined using Technicon Autoanalyser method and activities of leucine aminopeptidase (LAP) and 7-glutamyltranspeptidase (7-GT) using commercial test combination kits (Boehringer Mannheim).
Changes in enzyme histochemistry After 2 days in culture, the cells growing on the sterile glass coverslips
received fresh culture medium containing various concentrations of either allyl alcohol or DMNA. During a further 48-h incubation period, samples were removed for histochemical detection of non-specific esterase and diaphorase, using methods described by Pearse [7]. A Vickers M85 Scanning Microdensitometer was used for quantification in this study. The value obtained for the integrated density reading of the area of cells being ex-
160
amined was used as a measure of enzyme activity. Ten separate areas of each specimen were examined.
Electron microscopy The cells were fixed in situ for 2 h in pre-warmed (37°C) 5% glutaraldehyde
in 0.2 M cacodylate buffer. (They were washed twice with buffer, post- fixed for 1 h at 4°C in cacodylate buffered osmium tetroxide, left either in situ or gently removed from the surface of the culture vessel with a camel hair brush, dehydrated in graded alcohol and then embedded in Epon resin.) Ultra-thin sections cut on a Reichart OMU3 microtome were mounted on 200-mesh copper grids and stained for 3 min at room temperature with lead citrate. The washed sections were allowed to dry and examined with a Philips EM300 electron microscope operating at 60 kV.
Changes in protein synthesis Two days after the initiation of the culture, cells growing on 57-cm 2
culture dishes received fresh culture medium containing various concentra- tions of DMNA and L-[4,5-3H]leucine (130 Ci/mmol) at a concentrat ion of 5 ~Ci/ml. At various time intervals, the cells were harvested with 1% trypsin and pelleted by centrifugation at 1000 rev./min for 10 min. The cellular protein was then precipitated with cold 20% trichloroacetic acid and the resulting precipitate washed 3 times in 10% trichloroacetic acid by centri- fugation at 1000 rev./min for 10 min. The precipitates were dissolved in toluene-tri ton X-100 based scintillator and the radioactivity was measured in a Philips Liquid Scintillation Analyser with automatic external standard channels ratio quench correction.
RESULTS
Enzyme leakage Dimethylnitrosamine at concentrations of 50, 100, 200, and 500 pg/ml
caused a dose-related leakage of the enzymes GOT and GPT from hepato- cytes into the growth medium. However, an exposure time of up to 20 h was required before appreciable levels of these enzymes appeared in the culture medium (Table I).
Exposure of the hepatocytes to DMNA also resulted in AKP and LAP leaking into the culture medium and a distinct dose-related response was evident for DMNA treated cultures (Table I).
Addition of allyl alcohol to the cells at concentrations of 10, 100 and 500 pg/ml led to a dose-related leakage of GOT and GPT into the growth medium; this occurred within 12 h of exposure. 7-GT levels were also raised in the growth medium of cultures treated with aUyl alcohol, although not as markedly as those of the GOT and GPT (Table II).
Histochemical changes Af te r 24 h t reatment with 1 or 100 pg/ml of DMNA, the hepatocytes
161
O~
tO
TA
BL
E I
EF
FE
CT
OF
DM
NA
ON
3-D
AY
CU
LT
UR
ES
OF
AD
UL
T R
AB
BIT
HE
PA
TO
CY
TE
S
Th
ere
was
ev
iden
ce o
f si
gn
ific
ant
tren
ds
wit
h c
on
ce
ntr
ati
on
s fo
r G
PT
at
36
an
d 6
0 h
an
d G
OT
at
60
h (
P
1.6
%).
Th
e se
qu
ence
of
resu
lts
wit
h L
AP
an
d A
KP
did
no
t at
tain
th
e 5%
iev
el o
f si
gn
ific
ance
. C
ellu
lar
en
zy
me
ch
ang
es d
id n
ot
atta
in t
he
5% l
evel
of
sig
nif
ican
ce.
Inc
ub
ati
on
tim
e
12
(b
) an
d c
on
ch
(gg
/ml)
0
1
20
10
5
0
10
0
20
0
50
0
0 1
10
5
0
10
0
20
0
50
0
36
6
0
0 1
10
5
0
10
0
20
0
50
0
0 5
0
10
0
20
0
50
0
En
zym
e le
akag
e a
Leu
cin
e am
ino
- p
epti
das
e a
13
A
lkal
ine
ph
osp
hat
ase
a 1
0
Glu
tam
ate
p
yru
vat
e tr
ausa
min
ase
a 6
0
Glu
tam
ate
ox
alo
- ac
etat
e tr
ans-
am
inas
e a
10
0
Cel
lula
r en
zym
e ch
ange
s N
AD
PH
d
iap
ho
rase
b
28
0
29
8
27
0
27
0
No
n-s
pec
ific
es
tera
se b
1
95
2
00
2
10
--
1
90
16
1
3
14
1
5
14
10
1
2
12
1
2
10
58
6
5
62
1
05
6
0
10
0
10
0
12
5
11
5
95
17
1
7
18
1
9
15
1
6
17
2
2
21
14
1
9
21
25
2
7
12
1
2
13
1
5
12
1
4
15
1
5
17
1
5
16
17
1
8
19
58
7
7
10
2
15
4
60
7
0
85
1
30
2
30
4
5
23
5
26
5
34
0
35
0
95
1
21
1
45
1
50
1
15
--
1
45
1
35
1
60
2
75
1
35
1
75
2
50
2
95
3
40
26
5
29
0 2
40
1
95
20
5
20
01
95
--
1
80
--
--
26
0
28
5
24
0
--
14
0
19
0
19
0
20
5
17
5
aIn
in
tern
atio
nal
un
its/
litr
e.
bE
ach
fig
ure
rep
rese
nts
th
e av
erag
e en
zym
e ac
tiv
ity
in
10
cel
ls;
read
ing
s in
arb
itra
ry u
nit
s.
TA
BL
E I
I
EF
FE
CT
O
F A
LL
YL
A
LC
OH
OL
O
N 3
-DA
Y C
UL
TU
RE
S
OF
AD
UL
T
RA
BB
IT H
EP
AT
OC
YT
ES
Th
ese
re
sult
s ar
e re
pre
sen
tati
ve
of
a se
ries
of
ex
pe
rim
en
ts u
nd
ert
ak
en
d
uri
ng
th
e c
ou
rse
of
the
se s
tud
ies.
Pa
ram
ete
r In
cu
ba
tio
n t
ime
(h
) a
nd
co
nc
n
(ug
/ml)
6 1
2
24
4
8
0 1
0
10
0
50
0
0 1
10
1
00
5
00
0
1 1
0
10
0
50
0
0 1
10
1
00
5
00
En
zym
e le
akag
e 7
-Glu
tam
yl
tra
nsp
ep
tid
ase
a G
luta
ma
te p
yru
va
te
tra
nsa
min
ase
a
Glu
tam
ate
ox
alo
ac
eta
te
tra
nsa
min
ase
a
6
6
6.4
6
.9
6.1
--
50
60
1
15
1
55
6
0
--
85
9
5
19
0
18
5
10
0
--
6.1
7.1
7
.3
6.2
--
14
0
19
5
27
0
80
--
13
0
24
5
34
0
12
0
--
6.8
7
.2
8.1
6
.1
--
17
0
33
0
37
0
10
0
--
15
5
34
0
38
5
12
0
--
7.0
7
.4
8.3
19
5
39
5
43
0
20
0
48
0
50
5
Cel
lula
r en
xym
e ch
ange
s N
AD
PH
Dia
ph
ora
se b
--
--
N
on
-sp
ec
ific
est
era
se b
--
--
m
35
0
37
0
23
5
21
6
--
20
0
22
2
21
5
22
8
--
34
0
31
0
22
5
21
2
--
23
5
21
5
26
0
29
0
--
28
0
28
5
21
3
20
5
--
23
0
20
5
22
0
27
5
--
aIn
in
tern
ati
on
al
un
its/
litr
e.
b E
ach
fig
ure
re
pre
sen
ts t
he
av
erag
e e
nz
ym
e a
cti
vit
y i
n 1
0 c
ells
re
ad
ing
s in
arb
itra
ry u
nit
s.
b-L
showed little change in morphology or esterase activity. At 48 h treatment, non-specific esterase activity could still be detected in the cytoplasm (Table I). In the cultures receiving 10 ~g/ml there was a slight increase in activity compared with control cultures although the 5% level of significance was not attained.
Diaphorase activity was not appreciably affected in cultures incubated with 1 and 10 pg/ml of DMNA. In cultures treated with 100 pg/ml, there
Fig. 1. E lec t ron micrograph o f a rabb i t h e p a t o c y t e incuba ted in cul ture m e d i u m con- ta ining DMNA at a concen t r a t i on o f 10 ug/ml (x 5600). n = nucleus; rer = rough endo- plasmic re t icu lum; set = h y p e r t r o p h y of s m o o t h endoplasmic re t iculum.
164
was no appreciable alteration in activity for up to 12 h, but after this time the cells began to lose their activity, the depression remaining moderate in extent up to 48 h after exposure (Table I).
Hepatocytes treated with 10 or 100 ~g/ml or allyl alcohol for 12 h showed a decrease in diaphorase activity compared to the controls (Table II). This reduction in enzyme activity was maintained for at least a further 36 h. A slight but insignificant depression of enzyme activity was also noted in the control cultures and those treated with 1 ~g/ml.
Esterase activity was slightly increased in cultures incubated with aUyl alcohol (100 ug/ml) for 24 h, and these higher levels were maintained for at least a further 24 h. (Table II). A small reduction in esterase activity was noted in cultures incubated with 1 and 10 ~g/ml allyl alcohol. Although DMNA and allyl alcohol caused alterations in levels of the examined intra- cellular enzymes, none attained the 5% level o f significance.
Protein synthesis A 4-h incubation of cultured hepatocytes in growth medium containing
DMNA concentrations up to 500 ~g/ml had no significant effect on the ability of the cells to incorporate L-[4,5-3H] leucine (data not presented). The incorporation of the labelled amino acid by hepatocytes was reduced on more prolonged incubation (> 4 h) in the presence of 100 and 500 ~g/ml of DMNA. The decrease in the amount of incorporation of [SH] leucine into hepatocytes treated with 500 ~g/ml of DMNA for 24 h was presumably due to the removal of large numbers of dead or dying cells when the culture was washed prior to its processing for measurement of radioactivity.
Ultrastructure Incubation of 3-day-old cultures with DMNA (1 and 10 ~g/ml) resulted
in no observable changes in nuclei, mitochondria, lipid vacuoles and gly- cogen. The treated cultures, however, did show a dose-related hyperplasia of the smooth endoplasmic reticulum and disorganisation, dilation and fragmentation of the rough endoplasmic reticulum with a loss of ribosomes from the latter (see Fig. 1).
DISCUSSION
Certain chemicals induce damage to the plasma membranes of hepato- cytes thereby causing leakage of intracellular enzymes. Elevations in plasma levels of GOT, GPT, AKP, LAP and 7-GT are commonly used as indicators of acute or chronic damage to the liver. In the present study, the trans- aminases proved the most sensitive indicators of plasma membrane damage while the other 3 enzymes were less effective markers in this respect.
Previous workers [8--10] have detected differences in the hepatotoxici ty of closely related chemicals by measuring enzyme leakage from either freshly isolated adult rat hepatocytes or Chang liver cells. There are, however, problems associated with using liver cell lines or freshly isolated hepato-
165
cytes. Long-term cultures of liver cells seldom retain the properties of differentiated hepatic parenchymal cells and thus may not be capable of metabolising the test compound to its ultimate toxic form while freshly isolated cells, although possessing the majority of hepatic functions, may suffer from membrane trauma as a result of the isolation procedure [11,12] and as a consequence may respond atypically to toxins. Short-term culturing allows the cells to recover from the trauma. Rabbi t hepatocytes in culture preserve their in vivo drug metabolising capability for several days in contrast to rat hepatocytes [1,3]. For these reasons, rabbit hepatocytes in short- term culture may provide a better model for studying chemically induced plasma membrane damage than those of the rat.
In contrast to the in vivo situation in which zonal differences in enzymic staining reactions are evident, the distribution and activity of esterase and diaphorase were uniform in all cultures examined. The reason for the homo- geneity in staining reactions in vitro could be due to either culture condit ions favouring the selection of cells derived from similar zones in the liver or to the environment of the cells in vitro being more uniform than that in vivo. Non-specific esterase appeared to react less to allyl alcohol and DMNA treatment than did NADPH diaphorase.
The slight increases in esterase activity noted in some hepatocytes treated with allyl alcohol, may be due to damaged cells producing an augmented quanti ty of enzymes which are then liberated into the surrounding medium [14]. NADPH diaphorase showed a dose-related decrease in activity; similar changes have been observed in cultures of rat hepatocytes [ 15].
The decreased capacity of cultured hepatocytes treated with DMNA to incorporate [3H]leucine into protein is in agreement with earlier in vitro investigations using liver slices [16]. The fact that this inhibition occurred slowly (i.e. not detectable before 4 h) while enzyme leakage did not occur before 20 h and significant intracellular enzyme level changes took more than 24 h to develop, may indicate that a metaboli te of DMNA rather than the amine itself may be the toxic species responsible for the observed toxic effects.
Evidence that the endoplasmic reticulum is a favoured site of DMNA action came from the ultra-structural studies. The degeneration of the rough endoplasmic reticulum caused by DMNA is of particular interest since several authors have suggested [17,18] that interference in the association of ribo- somes with membranes is a feature of carcinogens. Such an effect was not observed with allyl alcohol.
The fact that the effects of allyl alcohol were more rapid than those of DMNA, suggests either that the former is toxic, or that it is converted more rapidly to its toxic metabolite(s) (e.g. acrolein) than is DMNA.
REFERENCES
1 J.W. Bridges and J.R. Fry, in J. Bend, E. Hodgson and R. Philpot (Eds.) Reviews in Biochemical Toxicology, Vol. 1, Elsevier/North-Holland, 1979, p. 201.
166
2 F.B. Bang and A.G. Warwich, in: E.N. Willmer (Ed.), Cells and Tissues in Culture, Vol. 2, Academic Press, New York, 1965, pp. 607--627.
3 D.M. BisseU and P.S. Guzelian, in: L.E. Gerschenson and E.B. Thompson (Eds.), Gene Expression and Carcinogenesis in Cultured Liver, Academic Press, London, 1975, p. 119.
4 A. Poole, L.F. Chasseaud and J.W. Bridges, Toxicology, 24 (1982) 000. 5 M.K. Dunlop, J.K. Kodama, J.W. Wellington, H.H. Anderson and C.H. Hine, Arch
Ind. Health, 18 (1958) 303. 6 P.N. Magee, in H.H. Hiatt, H.D. Watson and J.A. Winsten (Eds.), Origins of Human
Cancer, Cold Spring Harbor Laboratory, 1977, p. 629. 7 A.G.E. Pearse, Histochemistry, Theoretical and Applied, Churchill, London, 1972,
pp. 218--221. 8 C.A. Dujovne, R. Levy and H.J. Zimmerman,Proc. Soc. Exp. Biol. Med., 131 (1969)
583. 9 H.J. Zimmerman, J. Kendler, S. Libber and L. Lukus, Biochem. Pharmacol., 23
(1974) 2187. 10 C.O. Abernarthy and H.J. Zimmerman, Proc. Soc. Exp. Biol. Med., 150 (1975)
385. 11 R.J. Bonney, J.E. Becker, P.R. Walker and V.R. Potter, In Vitro, 9 (1974) 399. 12 R.J. Bonney and F. Maley, in L.E. Gerschenson and E.B. Thompson, Gene, Expres-
sion and Carcinogenesis in Cultured Liver, Academic Press, New York, 1975. 13 M. Wachstein, Gastroenterology, 37 (1959) 525. 14 K.R. Rees and K.P. Sinha, J. Pathol. Bacteriol., 87 (1961) 107. 15 J.W. Bridges, J.R. Fry and C.A. Jones, in J.R. Pattison, L. Bitensky and J. Chayen
(Eds.), Quantitative cytochemistry and its applications, Academic Press, 1979. 16 T. Hulton, E. Arrhennius, H. Low and P.N. Magee, Biochem. J., 16 (1960) 109. 17 W.H. Butler, Am. J. Pathol., 49 (1966) 113. 18 D.J. Williams and B.R. Rabin, Nature, 230 (1968) 113.
167