(CANCER RESEARCH 50. 7564-7570. December I. 1990]
Carcinogenic Sulfide Salts of Nickel and Cadmium Induce H2O2 Formation by
Human Polymorphonuclear Leukocytes1
Zhaojing Zhong, Walter Troll, Karen L. Koenig, and Krystyna Frenkel2
Departments of Environmental Medicine ¡Z.Z., W. T.. K. L. K., K. F.] and Pathology ¡K.F.], New York University Medical Center, New York, New York 10016
ABSTRACT
Some derivatives of nickel, cadmium, and cobalt are carcinogenic in
humans and/or animals but their mechanisms of action are not known.
We show that they are capable of stimulating human polymorphonuclear
leukocytes (PMNs), as measured by H2O2 formation, a known tumor
promoter. Most effective were the carcinogens nickel subsulfide, which
caused a 550% net increase in IM)., over that formed by resting PMNs,
followed by cadmium sulfide, 400%, and nickel disulfide, 200%. Nickel
sulfide and cobalt sulfide caused statistically nonsignificant increases of
45 and 20%, respectively. Noncarcinogenic barium and manganese sulfides, and sulfates of nickel, cadmium, and cobalt were inactive. The
enhancement of 11..(>..formation by CdS and Ni,S.; (1 «<mol/2.5
x 10;
PMNs) was comparable to that mediated by the potent tumor promoter
12-0-tetradecanoylphorbol-13-acetate,
used at 0.5 and l UM, respec
tively. Concurrent treatment of 12-O-tetradecanoylphorbol-13-acetatestimulated PMNs with Ni ,S. or NiS caused a decrease in 11.(). accu
mulation from that expected if the effects were additive. Including catalase in the reaction mixture proved that the oxidant formed by stimulated
PMNs was 11..()...whereas adding Superoxide dismutase showed that
Superoxide was also present ¡nPMN samples treated with NiS but not
with Ni ,S...Since nickel- and cadmium-containing particulates are depos
ited in the lungs and cause infiltration of PMNs, the ability to activate
those cells and induce 11.<)..formation may contribute to their carcinogenicity.
INTRODUCTION
A number of metal ions and some of their water-insoluble
derivatives are carcinogenic in humans and animals (reviewed
in Refs. 1-5). These include nickel, cadmium, and cobalt, all of
which can cause lung cancer in animals and some in humans
(1,2, 6-8). Workers in nickel refineries and processing indus
tries who are exposed to nickel by inhalation, are at the highest
risk for lung and sinonasal cancers. The most prevalent are
bronchogenic carcinomas, 90% of which are squamous cell
carcinomas, and 5% each of oat cell carcinomas and adenocarcinomas. There also is limited evidence that exposure of people
to cadmium compounds may lead to cancer. The difficulty in
assessing cadmium carcinogenicity in humans lies in the fact
that cadmium exposure is usually combined with other metal
compounds, i.e., in the case of cadmium-nickel battery and
cadmium-copper alloy workers. However, it appears that there
is cadmium-mediated increased mortality from prostatic can
cers. The carcinogenicity of nickel, cadmium, and cobalt deriv
atives in experimental animals is better established. Ni.,S2 in
duces lung carcinomas and local sarcomas, when administered
to mice or various strains of rats i.m., i.v., or by inhalation, and
to cats by implantation into nasal sinuses. CdS and CdCl2 cause
Received 5/15/90; accepted 8/27/90.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1This work was supported in part by National Institute of Environmental
Health Sciences Grants ES 04895 and ES 00260; and by Public Health Service
Grant CA 37858, awarded by the National Cancer Institute. Department of
Health and Human Services.
2To whom all requests for reprints should be addressed, at Department of
Environmental Medicine, New York University Medical Center. 550 First Ave
nue, New York, NY 10016.
sarcomas at the sites of i.m. or s.c. injection, and the latter also
induces testicular tumors. Exposure to CoS is carcinogenic in
rabbits and rats, in which it causes sarcomas.
Both nickel and cadmium are chemotactic for phagocytic
cells and cause inflammatory responses (8, 9). Alveolar rat
macrophages are particularly susceptible to the toxic effects of
NiCl2 (10). These changes are characterized by a rapid increase
in cyclic AMP, plasma membrane ruffling, and a decrease in
S'-nucleotidase activity, which are markers of macrophage ac
tivation. They are followed by a decline in phagocytosis and an
increase in lipid peroxidation after 24 and 48 h, respectively.
Human PMNs' respond with Superoxide anión radical (-O2~)
production to treatment with a Nr*-glycylglycyl-i.-histidine
complex (2). PMNs activated by digitonin can be stimulated
additionally by low concentrations of CdCl2, as measured by
formation of -O2~, but they are inhibited by higher amounts of
this agent (11).
Accumulating evidence shows that inflammation involves
formation of active oxygen species and that phagocytic cells
play a major role (12). The active OS generated by PMNs are
mutagenic and carcinogenic (13-15). Those OS can cause DNA
base damage (16-18), some of which may be heritable (19-21).
Inflammation, production of oxidants, and oxidative DNA base
damage have been implicated as being necessary in tumor
promotion (12, 16, 18, 22-25). Inflammation causes vasodilation, which facilitates infiltration of PMNs into the affected
sites where PMN-generated active OS cause damage to the
neighboring cells.
We have shown that H2O2, produced by PMNs activated with
the potent tumor promoter TPA, migrates into coincubated
HeLa cells where it causes formation of HMdUrd in cellular
DNA (26). HMdUrd, a product of thymidine oxidation, induces
a mutagenic response in Chinese hamster cells (19, 27). We
found that treatment of HeLa cells with TPA induces formation
of HMdUrd in cellular DNA even in the absence of PMNs,
although at lower levels than in their presence (26). Recently,
we showed that H2O2 is actually produced by TPA-treated HeLa
cells (28). Hence, tumor promoters may be capable of oxidative
activation of cells, characterized by generation of active OS that
can cause genetic damage.
Nickel compounds disrupt cell-to-cell communication in a
manner similar to TPA, hence, the suggestion was made that
nickel can also act as a tumor promoter (29-31). That this may
be the case is shown by the fact that some metal salts, particu
larly nickel, can act in a tumor-promoting capacity in a 2-stage
carcinogenesis model (32). Although these biological effects are
caused by treatment with soluble nickel compounds, it is not
known whether they act in this form or perhaps as complexes
with proteins or peptides. In this report we show that, similar
to TPA and other tumor promoters (18), certain nickel and
'The abbreviations used are: PMNs, polymorphonuclear leukocytes; OS,
oxygen species; HMdUrd, 5-hydroxymethyl-2'-deoxyuridine;
8-OHdGuo, 8-hydroxy-2'-deoxyguanosine;TPA,
12-O-tetradecanoylphorbol-l3-acetaie;SOD.
superoxide dismutase; BSS, balanced salt solution (137 mM NaCl, 5 mM KCI, 8.5
mM Na,HPO4 NaH¡PO4, 0.8 mM MgSO«,5 mM glucose. pH 7.4); HRPO,
horseradish peroxidase.
7564
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.
CARCINOGENIC
METALS AND NEUTROPH1LS
cadmium compounds are able to stimulate human PMNs, as
measured by H2O2 formation.4 The levels of H2O; produced
depend on the type of metal derivative used as well as treatment
time. This finding may be important because of the propensity
of some of these agents that accumulate in the target tissues to
be chemotactic for phagocytic cells. These cells would then be
stimulated by the same agents and generate active OS that
could promulgate the carcinogenic processes.
MATERIALS
AND METHODS
metal salts were applied either as a solution or as a suspension in BSS
after the sonic burst to disperse the particles uniformly.
Development of Assay Conditions. First, the number of PMNs appro
priate for these assays was determined by using (0.1-10) x IO5PMNs/
ml and activated with 25 n\i TPA, the concentration of TPA shown
before to cause maximal stimulation of these cells (18). As Fig. \A
shows, H2O2 formation was linear between 0.5 x 10s and 5 x IO5
PMNs/ml (semilog scale) during a 30-min incubation at 37°C;2.5 x
IO5 PMNs/ml were used in all subsequent experiments. Incubation of
2.5 x IO5 PMNs/ml with 25 niviTPA for 10 to 60 min resulted in the
time-dependent increase in H2O2. Formation of H2O2 was linear up to
30 min followed by a slower rate of increase (Fig. Iß).For this reason,
Chemicals. NiS2, NiS, CdS, CoS, MnS, and BaS were obtained in
the highest purity available (Alfa Co., Danver, MA) and used as such,
except that crystalline 0NÃŒS
was further ground by using a Spex grinder
(33). The sample was placed in a grinder puck and ground for 30 min.
The powder was removed from the puck by washing with deionized
water and the suspension was filtered through a 5-/jm polycarbonate
filter (Nucleopore). The particles passing through the filter were col
lected by centrifugation at 400 x g for 5 min and resuspended in
acetone. The suspension was recentrifuged and the pelleted particles
were resuspended in acetone. Small samples were aliquoted, acetone
was evaporated under N2 and stored under N2. The first sample of NiS
was a kind gift from Dr. M. Costa (Environmental Medicine, New
York University Medical Center). The first sample of crystalline «Ni3S2
[purchased from INCO, Ltd. (Toronto, Ontario, Canada) as <10->¿m
particles and further ground as described in Ref. 33] was also a gift
from Dr. M. Costa. Subsequent batches of Ni3S2were ground by mixing
with glass beads and allowing the sample to rotate on a ball mill for 7
days to obtain particles <5 ^m. The resulting fine powder was sus
pended in deionized water and filtered as described above for crystalline
NiS. Ni.,S2 and NiS are referred to as insoluble sulfides because of their
very limited solubility in water; one-half of the former would dissolve
within 10.3 years and of the latter within 3.3 years (reviewed in Ref.
4). CdS, MnS, and NiS can be dissolved at 0.13, 0.60. and 0.36 mg/
100 ml, respectively (34). NiSO4, CdSO4, CoSO4, TPA, phenol red,
H2O2, horseradish peroxidase, Superoxide dismutase, and catalase were
obtained from Sigma Chemical Co. (St. Louis, MO).
Preparation of PMNs. Blood was obtained by venipuncture from
informed, healthy volunteers and collected into BD vacutainer tubes
containing EDTA as an anticoagulant. RBC were removed by dextran
sedimentation followed by treatment with buffered hypotonie ammo
nium chloride, as previously described (16). WBC were washed twice
and suspended in BSS containing glucose at a ratio of 10 ml of blood/
ml BSS. This type of purification results in at least 98% viable cells,
which remain viable for 5 h, as determined by trypan blue exclusion.
However, after 5 h, the ability of the cells to produce H2O2 steadily
declines and, therefore, all of the experiments were carried out within
this time limit. The whole WBC population was used for H:O2 deter
mination because presence of other WBC does not interfere with the
oxidative burst of human PMNs (35), and their removal would increase
purification time. The number of PMNs was determined by examina
tion of whole blood smears by using Wright's stain (Fisher Scientific
0.6
0.2
IxlO1*
0.8
4 Presented in part al the 81st Annual Meeting of the American Association
for Cancer Research, Washington, DC, May 23-26, 1990 (55).
10
0.4
(C)
1 SOD
1.5
Co., Fair Lawn, NJ).
Determination of H2O2. Formation of H2O2 was measured by horse
radish peroxidase-mediated oxidation of phenol red, as previously
described (18). In short, the reaction mixtures containing human
PMNs, 100 Mgphenol red, and 50 MgHRPO/1 ml BSS were incubated
with TPA and/or metal derivatives at 37°Cfor 30 min. Reactions were
stopped by adding 20 n%of catalase (10 /jl). followed by l N NaOH (10
^1) 5 min later. After centrifugation, absorbance of supernatants was
determined spectrophotometrically at 598 nm (Response II, Gilford
Instruments) 40 min after the end of incubation, and the amount of
H2O2 generated was calculated from a standard curve (18). Blanks
contained everything except PMNs, whereas controls contained PMNs
but no TPA or metal derivatives. TPA was dissolved in dimethyl
sulfoxide (0.001 % final concentration in the reaction mixture), whereas
IxlO6
NUMBER
OF PMNs/ML
1.0
0.5
20
30
15
INCUBATION
TINE (HIN)
Fig. 1. Effects of a number of PMNs (A), incubation times (B), and SOD (C)
on H2O2 formation by TPA-aclivated human PMNs. as measured by horseradish
peroxidase-mediated oxidation of phenol red at 598 nm. Left ordinale, absorbance
at 598 nm; right ordinale, ¿IM
H2O2 formed; abscissa, number of PMNs/ml (log
scale). A, PMNs were treated with 25 nM TPA at 37°Cfor 30 min. B, 2.5 x 10*
PMNs/ml were incubated with 25 nM TPA at 37°Cfor 10 to 60 min. C, 2.5 x
IO5PMNs/ml were incubated with 25 nM TPA at 37°Cfor 30 min in the presence
(d) and absence (•)of 100 /jg SOD. All values are given as means ±SE of two
to six experiments, each carried out in duplicate on different PMN preparations,
except in C, which shows the results of one paired experiment.
7565
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.
CARCINOGENIC
METALS AND NEUTROPH11.S
Table 2 H'¡O2
formation by human PMNs treated with metal sulfides and
sulfates
Human PMNs were incubated with metal derivatives (1 (jmol/2.5 x 10s PMNs)
at 37°Cfor 30 min. H¡O¡
formation was measured by horseradish peroxidasemediated oxidation of phenol red at 598 nm, as described in "Materials and
Methods." All experiments were carried out in duplicate on different PMN
a 30-min incubation time was used in other experiments. To determine
the total capacity for H2O2 production in this system, PMNs were
incubated in the absence and presence of SOD, an enzyme which rapidly
dismutates O2~ to H2O2. SOD increased H2O2 by about 160, 115, and
100% during the 10-, 20-, and 30-min incubations, respectively (Fig.
1C), showing that the chosen assay conditions allow us to see the
increases and decreases in H2O2 formation due to the various treat
ments.
preparations.
Mean absorbance ±SE (no. of determinations)
RESULTS
Effects of Nickel Derivatives on Human PMNs in Presence
and Absence of TPA. To determine whether nickel compounds
have an effect on TPA-induced H2O2 formation by PMNs, the
cells were preincubated with Ni3S2, NiS, and NiSO4 (1 ¿¿mol/
2.5 x IO5 PMNs) at 37°Cfor 15 min, then incubated with or
without 2.5 HMTPA for 30 min (Table 1). Of the three nickel
derivatives, the most effective inducer of H2O2 formation was
NijS2. This is the first demonstration that Ni.,S2 is able to
activate PMNs which results in production of H2O2. This
finding contrasts with those using latex beads, which do not
activate PMNs in the absence of serum and do not interfere
with NiiS2-induced H2O2 formation (data not shown). Interest
ingly, even in the absence of PMNs (blanks), the two insoluble
Sulfides, Ni.,S2 and NiS, caused formation of some H2O2, al
though to a lower extent than in the presence of PMNs (Table
2). Inhibition by catatase proved that it was indeed H2O2 that
was produced by the nickel sulfides in the absence of PMNs,
whereas SOD did not increase H2O2 levels, showing that -O2~
was not formed (data not shown). Table 1 also shows that when
PMNs were treated with TPA in the presence of any of the
three nickel derivatives, H2O2 accumulation generally de
creased, in comparison to that occurring due to the action of
TPA alone, and in all cases the effects were less than additive.
Ni,S2 induced a significant increase (~500%) in H2O2 pro
duction by PMNs over that of the control, NiS caused a 65%
increase, whereas NiSO4 was virtually without effect. When
combined with 2.5 HM TPA, NiSO4 mediated the greatest
decrease in the accumulation of H2O2 followed by NiS, while
NijS2 showed a statistically significant decrease (P < 0.05) from
the sum of the separate Ni.,S2 and TPA effects. Concurrent
incubation of PMNs (obtained from two additional prepara
tions) with 25 nM TPA and Ni,S2 caused a 59.2% ±0.2 (SE)
decrease in H2O2 formation, compared to that caused by the
sum of the separate Ni.,S2 and TPA effects (data not shown).
These results show that insoluble Ni.,S2 is capable of stimulating
human PMNs, whereas soluble NiSO4 rather inhibits H2O2
Table 1 Effects of nickel derivatives on f!2O¡formation by human PMNs in
absence and presence of TPA
Human PMNs were treated with nickel derivalives (1 nmol/2.5 x 10s PMNs).
with TPA (2.5 and 25 nM). or with mixtures of nickel and TPA at 37'C for 30
min (samples). H¡O2formation was determined by horseradish peroxidase-mediated oxidation of phenol red at 598 nm, as described in "Materials and
Methods." All experiments involving nickel salts were carried out in duplicate on
matched PMN preparations.
Net absorbance" ±SE (no. of experiments)
TPA0.024
TPA0.238
nM
+ 0.006(14)'
No nickel
0.306'' ±
±0.090
0.082 (3)
(3)
Ni,S2
0.1 63 ±0.024 (3)
NiS
0.045 ±0.045 (3)
0.167 ±0.072 (3)
0.079 ±0.028 (2)
N ¡SO,No
0.018 ±0.010(2)2.5
" Net = samples - nickel blank; blanks contained nickel derivatives but no
PMNs and are listed in Table 2.
h C'ontrols contained PMNs but no nickel derivatives and no TPA.
r Statistically lower (P< 0.05) (by the paired t test) than the sum of the separate
NijS; and TPA effects.
Blanks-
Net*
Compounds
Ni,S2
NiS;
NiS
NiSO4
Samples
0.258 ±0.012 (5)
0.064 + 0.007(3)
0.105 + 0.015(9)
0.007 + 0.004(4)
0.099
-0.012
0.069
-0.015
CdS
CdSO4
0.203 + 0.022(4)
0.000 + 0.001 (2)
0.080 + 0.012(7)
-0.022 + 0.010 (2)
0.123r± 0.025
0.022 ±0.011
CoS
CoSO.,
0.056 ±0.010(6)
-0.020 ±0.028 (2)
0.027 ±0.008 (6)
0.004 ±0.012(2)
0.029 + 0.013
-0.024 ±0.030
-0.011 ±0.012 (4) -0.005 + 0.009 (6)
BaS
0.136 ±0.040(2)
0.176 ±0.059(2)
MnS
* Blanks contained metal derivatives but no PMNs.
* Net = (sample - blank) values ±SE calculated as:
-0.006 + 0.014
-0.040 + 0.072
±0.022 (7)
±0.006 (4)
+ 0.010(9)
±0.007 (4)
0.159r
0.076r
0.035
0.022
±0.025
±0.009
±0.018
±0.009
((sample SE)J + (blank SE)1)0'.
'Statistically
different (P < 0.05) from control mean of 0.024 ±0.006 (14
determinations). Statistical significance was assessed by means of the unpooled t
and Mann-Whitney tests.
formation. Interestingly, although TPA or one of the nickel
sulfides alone induce PMNs to generate H2O2, activation of
PMNs in the presence of both TPA and one nickel sulfide
results in overall suppression of H2O2 accumulation, since H2O2
formation was less than the sum of those induced by TPA and
either of the nickel sul fulos.
Effects of Catalase and SOD on H2O2 Production by PMNs.
In order to prove that H2O2 is the oxidant produced in response
to treating PMNs with NiS and Ni,S2, those incubations were
carried out in the absence and presence of catalase. As Fig. 2
shows, catalase caused a dose-dependent decrease in the amount
of oxidant formed by the action of both nickel sulfides and of
TPA. The degree of inhibition was inversely proportional to
the actual amounts of H2O2 generated, since of the three agents
used, NiS induced formation of the lowest H2O2 levels (Table
1) and catalase (500 ^g/ml)-mediated inhibition was the great
est, 90%. Ni,S2 caused intermediate and TPA the highest H2O2
generation, and catalase decreased them by 65 and 45%, re
spectively. The presence of SOD in the reaction mixture nearly
doubled the amount of H2O2 generated by PMNs treated with
NiS and doubled that formed by TPA-activated PMNs (Fig. 2).
These results show that -O2 is generated by PMNs stimulated
with either NiS or TPA. However, SOD was virtually without
effect on the Ni,S2-induced production of H2O2. At present, it
is not known whether Ni,S2 enhances dismutation of the -O2~
initially produced or acts by a different mechanism.
Effect of Incubation Time on H2O2 Formation by NiS- or
NiS3S2-treated PMNs. Fig. 3 shows that formation of H2O2 by
NiS- or Ni3S2-stimulated PMNs increased with increased in
cubation times, whereas that produced by NiS-activated PMNs
reached the maximum within 20 min, followed by a decline.
This pattern became obvious only after subtracting the amounts
of H2O2 generated by NiS in the absence of PMNs, which
gradually increased with incubation time, as did that produced
by Ni.,S2. These results suggest that the mechanisms of inter
action of NiS and NijS2 with PMNs differ from each other,
particularly since neither of these agents was capable of decom
posing H2O2 when a known amount of H2O2 was added to
PMNs (data not shown).
7566
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.
CARCINOGENIC
METALS AND NEUTROPHII.S
H2O2 Formation by Human PMNs Treated with Various
Metal Derivatives. In order to establish whether other metal
derivatives are capable of activating human PMNs and of
generating H2O2, cadmium, cobalt, nickel, barium, and man
ganese sulfides and cadmium, nickel, and cobalt sulfates were
assayed. PMNs were incubated with sulfides and sulfates (1
Ã-Ã-mol/2.5x IO5 PMNs) at 37°C for 30 min and H2O2 was
determined
as described in "Materials
and Methods."
The
5.0
0.2
results are shown in Table 2. In addition to NijSi, CdS, and to
a lesser extent, NiS2 also induced formation of H2O2. Similar
to NiS and Ni.,S2 (Table 1), CdS produced H2O2 in the absence
of PMNs as well (Table 2), the formation of which was abolished by catalase (data not shown). Interestingly,
MnS gener-
J
££
¡-
ated higher amounts of H2O2 in the absence of PMNs than in
their presence (Table 2). Of all of the metal sulfides and sulfates
tested, the most effective in forming H2O2 was Ni,S2, a subsulfide which caused a 550% net (sample—blank—control) increase
m
0.1
=¡
|
jg
i
2.5
o
SE
ro
,5a
over that generated by control PMNs in the absence of metals
[0.024 ±0.006 (n = 14)] (Fig. 4). CdS, which was the most
effective among the sulfides, enhanced
H2O2 formation
by
400%, followed by NiS2 with a 200% increase, whereas NiS
and CoS caused only statistically insignificant increases, 45 and
20%, respectively. The enhancement
of H2O2 formation in-
I
NiS
/
200100n_G
ALONE-'/////////////////////s1'///////YY///////////////////////AE3
- AGENT
20
- AGENT+ SOD
40
60
INCUBATION
TIME (MIN)
Fig. 3. Effect of incubation time on the formation of H2O2 by human PMNs
treated with NiS or Ni,S2 (1 »imol/2.5x 10! PMNs) at 37'C. Values are the
O - AGENT+ 50 yc CAT
means of duplicate determinations carried out on two (±SE) or
CAT1-Iv///py///////////y////Yy/////////s\1w////////////////////.1--H - AGENT+ 500yG
PMN preparations from which the values of appropriate blanks (no
one (no SE)
PMNs) were
subtracted. Both NiS and Ni,S2 were analyzed at the same time on the same
PMN preparations (paired experiments). Left ordinale, absorbance at 598 nm;
right ordinate, concentration of H2O2 formed (MM).
duced by CdS and Ni,S2 was comparable to those mediated by
0.5 and 1 nM TPA, respectively, which are included for com
parison (Fig. 4). Two noncarcinogenic sulfides of barium and
manganese as well as cadmium, nickel, and cobalt sulfates had
no effect or decreased H2O2 production below those of the
controls (PMNs in the resting state).
c?"
DISCUSSION
NiS
N,3S2
TPA
Fig. 2. Effects of SOD and catalase (dr) on H2O2 production by PMNs
treated with NiS, Ni3S2, or TPA. PMNs (2.5 x 105/ml BSS) were incubated with
sulfides (1 Mmol/2.5 x 10' PMNs) or TPA (25 nM) at 37'C for 30 min in the
absence or presence of SOD (100 ^g) or catalase (50 or 500 /ig), and the H2O2
generated was measured by horseradish peroxidase-mediated oxidation of phenol
red. Results are expressed as the means of two to three experiments, each in
duplicate, carried out on different PMN preparations. Formation of H2O2 by
PMNs activated with sulfides or TPA in the absence of SOD and catalase is
expressed as 100%.
It is known that derivatives of nickel and cadmium possess
carcinogenic properties (1, 3, 4, 7); however, it is not clear
which type(s) of action(s) is(are) responsible for those proper
ties. Our results point to the possibility that one of the modes
of action could be through the ability of some of these deriva
tives to activate phagocytic cells. Particularly, since nickel and
cadmium compounds are chemotactic when present in the
lungs, tissue that is susceptible to nickel- and cadmium-induced
tumors (1,8, 36, 37). We do not know whether activation of
PMNs occurs through the internalization of the particles or by
extracellular binding. However, our data show that activation
of PMNs by insoluble salts could not be due to indiscriminate
phagocytosis or binding because insoluble BaS and MnS do not
activate them (Fig. 4). It appears that (of those tested) only
certain carcinogenic insoluble sulfides stimulate PMNs to form
active OS. The fate of those OS may differ depending on the
type of sulfide used, since NiS produced both -O2~ and H2O2,
whereas Ni,S2 seemed to generate H2O^ with virtually no -O2
(Fig. 2).
7567
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.
CARCINOGENIC
METALS AND NEUTROPH1LS
However, the accumulated H2O2 is likely the cause of the
subsequent oxidative damage in neighboring cells because, of
the oxidants formed, it is the only one that can migrate through
cellular and nuclear membranes (12, 25, 38), and reach the
target cellular macromolecules, including the genetic material.
Hence, two metal derivatives that mediate formation of the
highest levels of H2O2 by PMNs have a potential of causing the
greatest damage, ¡nthis case Ni,S2 and CdS, both potent metal
carcinogens.
Interestingly, we found that certain metal sulfides are capable
of generating some H2O2 even in the absence of PMNs (Table
2). Indirect evidence for H2O2 production by Ni,S2 was pre
sented (39) by showing that this compound induces slow hydroxylation of 2'-deoxyguanosine (presumably caused by H2O2)
to 8-OHdGuo, a process that was greatly enhanced by addition
of H2O2. Hence, formation of H2O2 by metal sulfides alone and
due to activation of PMNs may be partly responsible for their
tumor-promoting (32) and cocarcinogenic (1, 3) properties, as
has been proposed for other types of tumor promoters (18, 2226).
In contrast to carcinogenic metal sulfides, we found that
MnS, a noncarcinogen, does not activate PMNs (Table 2).
Actually, the presence of PMNs even suppresses H2O2 genera
tion by MnS alone. We do not know the mechanism of this
inhibition. However, manganese dust counteracts nickel-in
duced toxicity and carcinogenicity (reviewed in Refs. 1 and 4),
whereas manganese in various soluble forms inhibits hydroxyl
radical-mediated damage, providing that both -O2~ and H2O2
NifSI
CO£-Y///////////////////////////^^^^C*C/3Il/Ì1LJ
(it1
S
1
CIèCrB*t
1¿fOCoCO
otoCDt
^
^H-
-2
C1il1TPA1'Õ_-VfIT>0
Fig. 4. Change in H2O2 formation by human PMNs treated with Sulfides or
sulfates of nickel, cadmium, cobalt, manganese, and barium (1 ^mol/2.5 x 10*
PMNs) or with 0.5 and I nM TPA over that generated by resting PMNs (controls).
PMNs were incubated with metal derivatives or TPA at 37"C for 30 min, and the
H2O2 was determined by horseradish peroxidase-mediated oxidation of phenol
red. Values of the controls [Asm = 0.024 ±0.006 (n = 14)] and blanks were
subtracted from those of samples (Table 2). and divided by controls.
This lower generation of H2O2 by NiS-stimulated PMNs
than by Ni.,S2 may be due to a number of factors. First, as
mentioned above, NiS induces formation of both -O2~ and
H2O2, whereas Ni.,S2 produces only H2O2. This conclusion was
reached by determining H2O2 formation in the absence and
presence of SOD. The estimate of -O2 generated with this
method should be generally the same as that obtained by the
reduction of cytochrome c, which also is carried out in the
absence and presence of SOD. The major difference is that the
former method measures the appearance of H2O2, whereas the
latter measures disappearance of -O2~ and, thus, could have led
to the erroneous conclusion that NbS2 does not activate PMNs.
Moreover, it previously was shown that it is the production of
H2O2 that is related to tumor promotion, not that of -O2" (18,
25). There are also time-dependent differences in the effects of
Ni.iSj and NiS on PMNs. During a 30-min incubation, NiS
apparently induces only about one-half the amount of H2O2
than produced at the maximum (20-min incubation). At the
same time, Ni,S2 causes formation of much more H2O2, which
is only a small fraction of the H2O2 produced in 60 min (Fig.
3). Since "aging" of NiS decreases its phagocytosis by cultured
cells (33), it might also decrease the ability of NiS to activate
PMNs. All of these parameters may contribute to the appar
ently lower H2O2 formation by NiS, which would result in a
potential underestimation of the NiS effects on human PMNs.
At present, it is not known what the significance is of the
differences in the types and amounts of active OS produced.
are also present (40).
It was suggested that water-soluble metal derivatives are less
effective as carcinogens because they are readily excreted (1).
However, these salts are carcinogenic in the tissues that accu
mulate them, as shown for Nr+ (3, 41, 42) and Cd2+ (5, 7, 8),
but it is not known in what form they persist in those tissues.
If they form complexes with proteins in the target tissues, as
they readily do in vitro (2), their clearance from those tissues
may be substantially reduced. Formation of such complexes
could also change the reactivity of those metals with active OS.
For example, Ni2+ complexed to histidine-containing oligopeptide reacts with -O2~ and H2O2, while separately neither Ni2+
nor oligopeptide does (43, 44). In contrast, Ni2+ and Co2+
gradually decompose a nucleoside hydroperoxide (45). Many
water-soluble metal salts also cause infidelity of DNA synthesis
in vitro (46), and modulate DNA synthesis and repair (41, 47;
reviewed in Refs. 1 and 3), as well as the activity of antioxidant
enzymes (8). Impairment in any of these processes may con
tribute to the carcinogenicity of those compounds.
We found that PMNs incubated with the soluble salts of
nickel, cadmium, and cobalt do not generate H2O2. If those
salts act on PMNs it is in a rather inhibitory manner, resulting
in production of H2O2 at levels lower than those generated by
resting (not treated) PMNs (Fig. 4). Cd2+ and Zn2+ (1 mM) are
known to inhibit the TPA-mediated respiratory burst of PMNs
by blocking the H+ channel (48, 49). Normally, such a channel
provides compensatory charge migration from the inside to
outside of the cell where -O2~ is produced. Blocking that
channel causes a decrease in internal pH and depolarization of
the membrane potential, which is known to suppress the activity
of NADPH oxidase, an enzyme mediating the oxidative burst
of PMNs. Others found that Co2+ also inhibits the TPAmediated respiratory burst but concluded that it is due to the
interaction with the Ca2+-dependent intracellular target (50).
Thus, our results show that in addition to cadmium and cobalt,
7568
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.
CARCINOGENIC
METALS AND NEUTROPHILS
nickel ions can also inhibit at least one of the steps necessary
for the action of NADPH oxidase.
The effects of the insoluble metal sulfides described in this
communication may be relevant to understanding the human
carcinogenicity of these agents since we show that they can
activate human phagocytic cells known to accumulate in the
lungs, the target tissue. The fact that very potent carcinogenic
agents also cause formation of high levels of H2O2 by PMNs
suggests that these compounds may also act in the capacity of
tumor promoters. It appears that differences exist among these
agents in the way they mediate PMN activation and the result
ant oxidant formation (Figs. 2 and 3). However, they are
probably phagocytized by PMNs, since sulfides such as Ni,S2
are phagocytized even by cultured cells (4, 52).
When PMNs are incubated with TPA or nickel sulfides,
H2O2 is generated (Table 1). However, cotreatment of PMNs
with TPA and one of the sulfides causes apparent inhibition of
H2O2 production because it is lower than the additive effects
expected. The reasons for such an outcome are unknown. It
could be that TPA, which activates PMNs within 1 min of
application, induces PMNs to generate -O2~ and H2O2 well
ahead of the action by a sulfide and in greater amounts. Since
activation of the NADPH oxidase of PMNs by TPA causes a
rapid (within about 1 min) decrease in the internal pH of the
cells (48), dissolution of phagocytized sulfides may be acceler
ated (52). This would generate enough soluble metal ions to
suppress passage of H+ through the proton channels, leading
to the further fall in pH that is followed by inhibition of
NADPH oxidase activity. Regardless of the mechanism respon
sible for such an outcome, our results show that simultaneous
treatment of PMNs by two different agents is likely to cause
different biological responses than when used separately.
Active OS generated by TPA-stimulated PMNs cause malig
nant transformation of cells and growth of tumors in nude mice
given injections of those cells (15, 53). They also cause forma
tion of HMdUrd (an oxidized thymidine) in the DNA of
coincubated cells (25, 26). Recently, an increase in the 8OHdGuo in DNA isolated from kidneys of rats treated with
nickel acetate was noted and this correlates with the initiation
of renal carcinogenesis (42). HMdUrd is mutagenic, whereas
the 8-OHdGuo present in oligomers causes misincorporation
of bases during replication (19-21, 54). Although nickel acetate
is water soluble, the kidney is known to accumulate such salts
(1) and 16-48 h after exposing the rats there is still a substantial
amount of nickel present. Hence, direct (in cells) or indirect
(through phagocytes) oxidant formation induced by nickel com
pounds in target tissues such as kidneys and lungs, and a
subsequent oxidative modification of bases in their DNA are
likely to initiate and/or promote the carcinogenic process.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
ACKNOWLEDGMENTS
25.
The authors wish to thank Diane Doucette and Dr. Katherine Squibb
for NiS and Ni.iS^ sample preparations.
26.
REFERENCES
27.
1. Sum lu man, F. W., Jr. Recent advances in metal carcinogenesis. Ann. Clin.
Lab. Sci., 14: 93-122, 1984.
2. Nieboer, E., Maxwell, R. !.. Rossetto. F. E., Stafford. A. R., and Stetsko, P.
1. Concepts in nickel carcinogenesis. In: A. V. Xavier (éd.).Frontiers in
Bioinorganic Chemistry, pp. 142-151. Weinheim, Federal Republic of Ger
many: VCH Verlagsgellschaft mbH, 1986.
3. Sunderman, F. W., Jr. Mechanisms of nickel carcinogenesis. Scand. J. Work
environ. Health. 15: 1-12. 1989.
28.
29.
30.
31.
Coogan, T. P., Latta. D. M., Snow, E. T., and Costa, M. Toxicity and
carcinogenicity of nickel compounds. CRC Crii. Rev. Toxicol., 19: 341-384,
1989.
Waalkes. M. P., Rehm, S., Riggs, C. W., Bare, R. M., Devor, D. E., Poirier,
L. A., Wenk, M. L., and Henneman, J. R. Cadmium carcinogenesis in male
Wistar |Crl:(WI)BR¡rats: dose-response analysis of effects of zinc on tumor
induction in the prostate, in the testes, and at the injection site. Cancer Res.,
49:4282-4288, 1989.
Gilman, J. P. W., and Ruckerbauer, G. M. Metal carcinogenesis. I. Obser
vations on the carcinogenicity of a refinery dust, cobalt oxide, and colloidal
therium dioxide. Cancer Res., 22: 152-157, 1962.
Takenaka, S., Oldiges, H., König,H. Hochrainer, D., and Oberdörster, G.
Carcinogenicity of cadmium chloride aerosols in W rats. J. Nati. Cancer
Inst., 70: 1367-371, 1983.
Cross, C., Omaye, S., Rifas, D., Hasegawa, G. K., and Reddy, K. A. Bio
chemical effects of intratracheal instillation of cadmium chloride on rat lung.
Biochem. Pharmacol., 28: 381-388, 1979.
Waalkes, M. P., Rehm, S., Kasprzak, K., and Issaq, H. J. Inflammatory,
proliferative, and neoplastic lesions at the site of metallic identification ear
tags in Wistar [Crl:(WI)BR] rats. Cancer Res., 47: 2445-2450, 1987.
Sunderman, F. W., Jr., Hopfer, S. M., Lin, S-M., Plowman, M. C, Stojanovic. T., Wong, S. H-Y., Zacharia, O., and Ziebka, L. Toxicity to alveolar
macrophages in rats following parenteral injection of nickel chloride. Toxicol.
Appi. Pharmacol., 100: 107-118, 1989.
Amoruso, M. A., Witz, G., and Goldstein, B. D. Enhancement of rat and
human phagocyte Superoxide aniónradical production by cadmium in vitro.
Toxicol. Lett. (Amst.), 10: 133-138, 1982.
Fantone, J. C., and Ward, P. A. Role of oxygen-derived free radicals and
metabolites in leukocyte-dependent inflammatory reactions. Am. J. Pathol.,
107: 397-418, 1982.
Weitzman, S. A., and Stessei, T. P. Effects of oxygen radical scavengers and
antioxidants on phagocyte-induced mutagenesis. J. Immunol.. 128: 27702772, 1982.
Barak, M., Ulitzur, S., and Merzbach, D. Phagocytosis-induced mutagenesis
in bacteria. Mutât.Res., 121: 7-16, 1983.
Weitzman, S. A., Weitberg, A. B., Clark, E. P., and Stossel, T. P. Phagocytes
as carcinogens: malignant transformation produced by human neutrophils.
Science (Washington DC), 227: 1231-1233, 1985.
Frenkel, K., Chrzan, K., Troll, W., Teebor, G. W., and Steinberg, J. J.
Radiation-like modification of bases in DNA exposed to tumor promoteractivated polymorphonuclear leukocytes. Cancer Res., 46: 5533-5540, 1986.
Floyd, R. A., Watson, J. J., Harris, J., West, M., and Wong, P. K. Formation
of 8-hydroxydeoxyguanosine, hydroxyl free radical adduci of DNA in granulocytes exposed to the tumor promoter, tetradecanoylphorbolacetate.
Biochem. Biophys. Res. Commun., 137: 841-846, 1986.
Frenkel, K., and Chrzan, K. H2O¡formation and DNA base modification by
tumor promoter-activated polymorphonuclear leukocytes. Carcinogenesis
(Lond.), 8: 455-460, 1987.
Shirnamé-Moré,
L., Rossman, T. G., Troll, W., Teebor, G. W., and Frenkel,
K. Genetic effects of 5-hydroxymethyl-2'-deoxyuridine, a product of ionizing
radiation. Mutât.Res., 178: 177-186, 1987,
Itilimona. M. H., and Gupta, S. V. Comparison of the mutagenic activity of
5-hydroxymethyldeoxyuridine with 5-substituted 2'-deoxyuridine analogs in
the Ames Salmonella/microsome test. Mutât.Res., 169: 123-127, 1986.
Kuchino, Y., Mori, F., Kasai, H., Inouc, H.. Iwai, S.. Miura, K., Ohtsuka,
E., and Nishimura, S. Misreading of DNA templates containing 8-hydroxy
deoxyguanosine at the modified base and at adjacent residues. Nature
(Lond.), 327: 77-79, 1987.
Zimmerman, R., and Cerutti, P. Active oxygen acts as a promoter of
transformation in mouse embryo C3H/IOT'/2/C18 fibroblasts. Proc. Nati.
Acad. Sci. USA, 81: 2085-2087, 1984.
Leadon, S. A. Production of thymine glycols in DNA by radiation and
chemical carcinogens as detected by a monoclonal antibody. Br. J. Cancer,
55(Suppl. 8): 113-117, 1987.
Frenkel, K., Donahue, J. M., and Banerjee, S. Benzo(a)pyrene-induced
oxidative DNA damage: a possible mechanism for promotion by complete
carcinogens. In: P. Cerutti, I. Fridovich, and J. McCord, (eds.), Oxy-radicals
in Molecular Biology and Pathology. UCLA Symposia on Molecular and
Cellular Biology, New Series, Vol. 82, pp. 509-524. New York: Alan R. Liss,
Inc., 1988.
Frenkel, K. Oxidation of DNA bases by tumor promoter-activated processes.
Environ. Health Perspect., 81: 45-54, 1989.
Frenkel, K., and Chrzan, K. Radiation-like modification of DNA and HA
formation by activated human polymorphonuclear leukocytes (PMNs). In:
P. Cerutti, O. Nygaard, and M. Simic (eds.), Anticarcinogenesis and Radia
tion Protection, pp. 97-102. New York: Plenum Publishing Corp., 1987.
Boorstein, R. J., and Teebor, G. W. Mutagenicity of 5-hydroxymethyl-2'deoxyuridine to Chinese hamster cells. Cancer Res., 48: 5466-5470, 1988.
Frenkel, K., and Gleichauf, C. Hydrogen peroxide formation by cells treated
with a tumor promoter. Free Rad. Res. Comms., in press. 1990.
Loch-Caruso. R.. Coreos, I. A., and Trosko, J. E. Inhibition of metabolic
cooperation by metals. Toxicologist, 6: 267, 1987.
Miki, H.. Kasprzak, K. S., Kenney, S., and Heine, U. I. Inhibition of
intercellular communication by nickel2*: antagonistic effect of magnesium.
Carcinogenesis (Lond.), S: 1757-1760, 1987.
Hamel, E., Katoch, F., Mueller. G.. Birchmeier, W., and Yamasaki, H.
7569
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.
CARCINOGENIC METALS AND NEUJ KOI'llll-S
Transforming growth factor f> as a potent promoter in two-stage BALB/c
3T3 cell transformation. Cancer Res., 48: 2832-2836. 1988.
32. Kurokawa, V., Matsushima. M.. Imazawa. T., Takamura, N., Takahashi.
M.. and Hayashi. V. Promoting effect of metal compounds on rat renal
tumorigenesis. J. Am. Coll. Toxicol.. 4: 321-330, 1985.
33. Heck, J. D.. and C'osta, M. Influence of surface charge and dissolution on
34.
35.
36.
37.
38.
39.
40.
the selective phagocytosis of potentially carcinogenic paniculate metal com
pounds. Cancer Res., 43: 5652-5656. 1983.
Weast. R. C. (ed.). Handbook of Chemistry and Physics. Ed. 62. Boca Raton.
KL: CRC Press. 1981-1982.
Ambruso, D. R.. and Johnston, R. B. Laetoferrin enhances hydroxyl radical
production by human neutrophils. neutrophil paniculate fractions, and an
enzymatic generating system. J. Clin. Invest. 67: 352-360, 1981.
Katsnelson. B. A., and Privalova, L. I. Recruitment of phagocyti/ing cells
into the respiratory tract as a response to the cytotoxic action of deposited
particles. Environ. Health Perspect., 55: 313-325, 1984.
Lynn, W. S. Control of the cellular influx in lung and its role in pulmonary
toxicology. Environ. Health Perspect., 55: 307-311, 1984.
Meneghini, R. Genotoxicity of active oxygen species in mammalian cells.
Mutai. Res., 195: 215-230, 1988.
Kasprzak. K. S., and Hernandez. L. Enhancement of hydroxylation
and
deglycosylation of 2'-deoxyguanosine
by carcinogenic nickel compounds.
Cancer Res., 49: 5964-5968. 1989.
C'heton. P. L-B.. and Archibald. E. S. Manganese complexes and the gener
ation and scavenging of hydroxyl tree radicals. Eree Radical Biol. Med., 5:
325-333, 1988.
41. Kasprzak, K. S.. and Poirier, L. A. Effects of calcium(ll) and magnesium(ll)
on nickel(ll) uptake and thymidinc incorporation into DNA in the lungs of
strain A mice. Carcinogenesis (Lond.), 6: 1819-1821. 1985
42. Kasprzak. K. S., Diwan. B. A.. Konishi, N., Misra. M.. and Rice, J. M.
Initiation by nickel acetate and promotion bv sodium barbila! ot renal cortical
epithelial tumors in male E344 rats. Carcinogenesis (Lond.). //: 647-652,
1990.
43. Nieboer, E.. Tom, R. T., and Rossetto. F. E. Superoxide dismutase activity
and novel reactions with hydrogen peroxide of hislidine-containing
nickel(ll)oligopeptide complexes and nickel( 11) induced structural changes in synthetic
DNA. Biol. Trace Element Res.. 21: 23-33, 1989.
44. [noue. S., and Kawanishi. S. ESR evidence for Superoxide, hydroxyl radicals
and singlet oxygen produced from hydrogen peroxide and nickel(ll) complex
of
glycylglycyl-i
-histidine. Biochem. Biophys. Res. Commun.. /5V:445-451,
1989.
"
45. Tofigh. S., and Erenkel, K. Effect of metals on nucleoside hydroperoxide, a
produci of ionizing radiation in DNA. Eree Radical Biol. Med., 7: 131-143,
1989.
46. Sirover. M. A., and Loeb. L. A. Infidelity of DNA synthesis in vilro: screening
for potential metal mutagens or carcinogens. Science (Washington DC'). 194:
1434-1436. 1976.
47. Nocentini. S. Inhibition of DNA replication and repair by cadmium in
mammalian cells. Protective interaction of zinc. Nucleic Acids Res., 15:
4211 4225, 1987.
48 Henderson, L. M.. C'happell. J. B., and Jones, O. T. Ci. Superoxide generation
by the electrogenic NADPH oxidase of human neulrophils is limited by the
movement of a compensating charge. Biochem. J.. 255: 285-290. 1988.
49. C'ross. A R. Inhibitors of the leukocyte Superoxide generating oxidase:
mechanisms of action and methods for their elucidation. Eree Rad. Biol.
Med.. S: 71 -93. 1990.
50. Elferink, J. (J. R.. and Deierkauf. M. Suppressive action of cobalt on
exocylosis and respiratory burst in neutrophils. Am. J. Physiol., 257: C859C 864, 1989.
51. C'osta. M., and Mollenhauer. H. H. Carcinogenic activity of paniculate nickel
compounds is proportional to their cellular uptake. Science (Washington
DC'), 209: 515-517. 1980.
52. Evans, R. M.. Davies. P. J. A., and C'osta. M. Video time-lapse microscopy
of phagocytosis and inlracellular fate of crystalline nickel sulfidc particles in
cultured mammalian cells. Cancer Res., 42: 2729-2735, 1982.
53. Weitberg, A. B., and Córvese. D. Oxygen radical induced malignant trans
formation of cultured human lung cells: An in vitro model. Proc. Am. Assoc.
Cancer Res.. _</:l47. 1990.
54. Shibutani. S.. Bodepudi. V., Johnson. I-.. Moriva, M., and Grollnian. A. P.
Nucleotide misincorporation
directed by 8-oxodeoxyguanosine. Proc. Am.
Assoc. Cancer Res., 31: 93. 1990.
55. Zhong. /., Troll, W., and Frenkel, K. Carcinogenic nickel sullides activate
human polymorphonuclear leukocytes (PMNs) and induce M ( ) formation.
Proc. Am. Assoc. Cancer Res...?/: 84. 1990.
7570
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.
Carcinogenic Sulfide Salts of Nickel and Cadmium Induce H2O2
Formation by Human Polymorphonuclear Leukocytes
Zhaojing Zhong, Walter Troll, Karen L. Koenig, et al.
Cancer Res 1990;50:7564-7570.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/50/23/7564
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.