TOXICOLOGICAL SCIENCES 79, 147–156 (2004)
DOI: 10.1093/toxsci/kfh108
Advance Access publication March 31, 2004
Ortho-Substituted but Not Coplanar PCBs Rapidly Kill Cerebellar
Granule Cells
Yuansheng Tan,* Renji Song,† David Lawrence,* ,† and David O. Carpenter‡ , 1
*University at Albany, School of Public Health, Department of Environmental Health and Toxicology, University at Albany, Rensselaer, New York 12144;
†Wadsworth Center for Laboratories and Research, New York State Department of Health, and School of Public Health, University at Albany, Albany, New
York 12201; and ‡University at Albany, Institute for Health and the Environment, and School of Public Health, Department of Environmental Health and
Toxicology, Rensselaer NY 12144
Received November 19, 2003; accepted February 20, 2004
Several PCB congeners were assessed for their cytotoxicity on
cerebellar granule cells in an attempt to compare their structureactivity relationship as potential neurotoxicants and to assess the
mechanisms associated with their toxicity. Flow cytometry was
used to monitor the changes of a number of biochemical endpoints:
membrane integrity, intracellular free calcium concentration
([Ca 2ⴙ] i), reactive oxygen species (ROS) production, mitochondrial membrane potential (⌬␺ m), and cell size. The non-coplanar,
ortho-substituted congeners, PCB 8 (2,4ⴕ-dichlorobiphenyl), PCB
28 (2,4,4ⴕ-trichlorobiphenyl), PCB 47 (2,4,2ⴕ,4ⴕ-tetrachlorobiphenyl), and PCB 52 (2,5,2ⴕ,5ⴕ-tetrachlorobiphenyl) (10 ␮M) killed
neurons to different degrees within 30 min. Loss of viability was
accompanied by increased [Ca 2ⴙ] i and decreased ⌬␺ m. No significant changes of ROS level were observed during exposure. The
coplanar congeners, PCB 77 (3,4,3ⴕ,4ⴕ-tetrachlorobiphenyl), PCB
80 (3,5,3ⴕ,5ⴕ-tetrachlorobiphenyl), and PCB 81 (3,4,5,4ⴕ-tetrachlorobiphenyl) (10 ␮M), had no effects on membrane integrity,
[Ca 2ⴙ] i or ⌬␺ m in this time period of exposure. In Ca 2ⴙ-free
Tyrode’s medium, there was no [Ca 2ⴙ] i increase after exposure to
the ortho-substituted congeners, but also no reduction in loss of
membrane integrity, suggesting Ca 2ⴙ influx was not the cause of
viability loss. The mitochondrial uncoupler, carbonyl cyanide
m-chlorophenyl hydrazone (CCCP) (1–2 ␮M), caused a large
decrease of ⌬␺ m, but only a slight loss of viability, which suggested
that ⌬␺ m is not the primary cause of PCB 52-induced cell death.
These studies show that ortho-substituted PCBs are toxic to cerebellar granule cells; however, their toxic action is not secondary to
elevation of intracellular calcium, a change in mitochondrial membrane potential, or free radical generation.
Key Words: PCBs; flow cytometry; cerebellar granule cells; intracellular calcium concentration; viability; membrane integrity.
Polychlorinated biphenyls (PCBs) are a class of persistent
industrial chemicals that have become distributed throughout
1
To whom correspondence should be addressed at School of Public Health,
University at Albany, One University Place, B242, Rensselaer, NY 12144.
Fax: (518) 525-2665. Email: [email protected].
Toxicological Sciences vol. 79 no. 1 © Society of Toxicology 2004; all rights
reserved.
the world. Because of their remarkable insulating capacity and
relatively inflammable nature, they gained widespread use for
much of this century as coolants and lubricants in transformers
and other electrical equipment where these properties are valued. Once released, PCBs are persistent in the ecosystem and
are bioconcentrated through the food chain due to their affinity
for lipids and resistance to metabolism. A major source of
human exposure to PCBs is from consumption of contaminated
fish, meat, and dairy products, but PCBs can also be absorbed
through the lungs or the skin (Agency for Toxic Substances
and Disease Registry, 2000).
PCBs are a family of 209 chemicals (congeners), each of
which consists of two biphenyl rings containing from one to
ten chlorine atoms. The properties of individual congeners
depend upon both the number of chlorines and their positions
around the biphenyl rings. For many years, concerns of PCB
toxicity have focused on dioxin-like actions (immunosuppression and carcinogenesis) mediated via activation of the aryl
hydrocarbon (Ah) receptor and induction of cytochrome P450
(see Safe, 1994, for review). These effects are primarily due to
congeners that have chlorines only in the meta and para
positions, which can assume a coplanar, dioxin-like configuration. Congeners with chlorines in the ortho position (closest
to the biphenyl bond) are energetically dissuaded from assuming a coplanar configuration.
There is accumulating evidence showing that PCBs are
neurotoxic. Studies of children exposed prenatally to PCBs
either in two massive poisoning incidents in Asia (Chen et al.,
1994; Guo et al., 1994; Lai et al., 1994; Yoshimura, 1978) or
U.S. children exposed by maternal consumption of contaminated fish (Jacobson and Jacobson, 1997; Lonky et al., 1996)
have shown that PCB exposure of the mother is associated with
a decrement of cognitive function in the child, which appears
to be irreversible. Adults who eat PCB-contaminated fish show
decrements in memory, but not in some other nervous system
functions (Schantz, et al. 2001), suggesting a selective toxicity
of memory circuits. Similar defects have been observed in
monkeys (Rice and Hayward, 1999; Schantz et al., 1997) and
other species (Tilson and Kodavanti, 1997).
147
148
TAN ET AL.
Laboratory studies have documented toxic actions of PCBs
on several cell types, including neurons, and have provided
evidence that at least some of these toxicities are mediated by
non-coplanar congeners (Brown and Ganey, 1995; Carpenter et
al., 1997; Fischer et al., 1996; Kodavanti and Tilson, 1997;
Nishihara and Utsumi, 1986; Seegal et al., 1990; Wong et al.,
1997). The cellular mechanisms of these actions remain elusive, but there are several hypotheses (see Fischer et al., 1998,
for review). Many of these hypotheses converge on one intracellular messenger, [Ca 2⫹] i. Nishihara and Utsumi (1982,
1985) reported that ortho-substituted PCBs alter calcium homeostasis by inducing changes in mitochondrial membrane
integrity, while Kodavanti and Tilson (1997) postulated inhibition of calcium sequestration by intracellular organelles.
Brown and Ganey (1995) have shown that non-coplanar PCBs
can induce oxygen radical (O 2– ) generation in neutrophils, and
this action is dependent on external calcium. Fischer et al.
(1996) have shown that ortho-substituted PCBs (but not coplanars) increase [Ca 2⫹] i in beta cells of the pancreas, promoting insulin release, and that the calcium comes from extracellular sources. Wong et al. (1997) also suggested that a rise in
[Ca 2⫹] i was responsible for cytotoxicity, but proposed that the
calcium was released from intracellular ryanodine-sensitive
stores in the endoplasmic reticulum.
We performed experiments on cerebellar granule neurons
using flow cytometry in an attempt to understand the cellular
mechanism(s) of PCB-induced neurotoxicity. Cerebellar granule cells are useful because: (1) 99% of cerebellar cells are
granule cells (Zagon, 1977), so it is relatively easy to obtain a
large number of similar neurons; (2) they mature in the early
postnatal period, so they can be relatively easily dissociated as
single cells from the cerebelli of young animals; (3) the granule
cells are round and suitable for study by flow cytometry. Our
data show that ortho-substituted PCBs can rapidly kill neurons.
Cell death is not dependent upon calcium entry from extracellular sources, generation of reactive oxygen species (ROS), or
loss of mitochondrial membrane potential.
MATERIALS AND METHODS
Cerebellar granule cell preparation: Cerebellar granule neurons were
acutely dissociated from 10- to 15-day-old Sprague-Dawley (Taconic Inc.) rat
pups as described by Oyama et al. (1992, 1996). The cerebellum was cut into
about 1 mm-thick slices and enzymatically dissociated for about 30 min at
34°C in Tyrode’s solution (NaCl, 148 mM; KCl, 5 mM; CaCl 2, 2 mM; MgCl 2,
1 mM; D-glucose, 10 mM; HEPES, 10 mM; pH 7.3) containing 1.66 IU/ml
dispase. The slices were then washed three times with fresh Tyrode’s solution
and gently aspirated ten times using a fire-polished Pasteur pipette with a tip
diameter of 0.3– 0.5 mm. The suspension was then filtered by gravity through
a nylon mesh to remove large neurons and tissue fragments. The cell filtrate
was incubated at 37°C for 60 min and diluted to 1–2 ⫻ 10 6 cells/ml with
Tyrode’s solution. Aliquots were distributed to polystyrene tubes for incubation with reagents before flow cytometry. Neurons were preincubated with the
fluorescent probes for various periods of time, dependent upon the probe
(Carpenter, et al., 1997; Tan et al., 2003). Various PCB congeners at defined
concentrations were then added to the dye-loaded cells, and fluorescence
followed as a function of time.
Flow cytometry and data analysis: Using an EPICS ELITE ESP (Coulter)
flow cytometer, we preloaded cells with multiple fluorescent probes for simultaneous measurements. All probes used in the experiments are excited by an
argon laser (488 nm), and 10,000 cells per sample were analyzed. In order to
record from only granule cell neurons, subpopulations of the total cells were
selected (“gated”) on the basis of forward scatter (which reflects cell size) and
side scatter (which reflects cell granularity). With these techniques we could
identify a uniform population of cerebellar granule cells and exclude other
neurons, as well as distinguish living from dead cells (Carpenter et al., 1997;
Dyatlov et al., 1998). Data was acquired in a listmode for off-line analysis with
WinList software (Verity Software House, Inc.). Fluorescence of probes was
normalized as percentage of control. The fluorescence of all probes was
converted to a linear scale by the following formula provided by the WinList
program:
10 (decades⫻[parameter/resolution])
where decades are the number of decades for the log amplifiers in flow
cytometry (4 in this case); parameter is log listmode parameter to convert; and
resolution is ADC resolution of the parameter (256 in this case).
Three replicate tubes were used in most experiments, and each data point is
the result of at least three experimental replications. All values are reported as
mean ⫾ SEM (n ⫽ 3–9), unless specified. Multiple comparison tests were
performed using SAS software (SAS Institute Inc.). A p value of less than 0.05
was accepted as being significant.
Determination of membrane integrity and cell viability: The DNA-binding probes, 7-amino actinomycin-D (7-AAD) (5␮g/ml) or propidium iodide
(PI) (5␮g/ml), were used to determine loss of membrane integrity (Carpenter
et al., 1997; Tan et al., 2003). Either was preincubated for 3–5 min before
adding PCBs. Both PI and 7-AAD are excluded by healthy cells, but enter and
bind to DNA when membrane integrity is compromised, which we take to
indicate loss of viability. With extended incubations, viable cells will begin to
express more fluorescence, likely due to pinocytosis of these probes. The cells
were considered to be dead when the fluorescent intensity increased at least
ten-fold over the maximal fluorescence in the population of initial healthy
living cells (“PI-high”, as shown in Fig. 1). Dead cells were excluded from
measurements of intracellular calcium concentration, ROS, or mitochondrial
membrane potential. As shown in Figure 1, we grouped living cells into two
populations: PI-low neurons under the first bar (with maximal fluorescence less
than 27) and PI-intermediate neurons under the second bar (with fluorescence
between 27 and 294). While the cells under the second bar show greater PI (or
7-AAD) fluorescence than those under first bar, these cells have not become
readily permeable and are therefore considered to be viable but damaged. The
magnitude of the signal for fluorescence PI or 7-AAD is correlated with
membrane integrity (Maftah et al., 1993). There were no observed differences
between 7-AAD and PI other than the emission spectrum. We used PI for all
studies except in conjunction with JC-1, where we used 7-AAD to avoid
overlap of spectra.
Measurement of [Ca 2ⴙ] i and ROS level: The cellular ROS level was
determined using a new generation of 2⬘,7⬘-dichlorodihydrofluorescein, carboxylated DCF (DCF-DA), which has two negative charges at physiological
pH and two acetoxymethyl (AM) esters. The AM-ester group facilitates
permeability of the probe through the membrane. The dye was preincubated
with cells for 1 h prior to adding PCBs, as described by Oyama et al. (1992)
and Boldyrev, et al. (1999b). The additional negative charges enhance their
entrapment inside the cell and thus facilitate long-term studies. Cell-permeant
Fluo-3 was used to determine the changes in intracellular calcium concentration as previously described (Carpenter et al., 1997; Oyama et al., 1996).
Fluo-3 was preincubated with neurons for 1 h prior to adding the PCBs. Final
concentrations for Fluo-3 and DCF-DA were 300 nM and 100 ␮M,
respectively.
Detection of mitochondrial membrane potential (⌬␺ m) by JC-1: Cells
were loaded with JC-1 at a final concentration of 1 ␮M for about 30 min before
detection. ⌬␺ m was determined as the ratio of orange over green fluorescence,
149
PCBS KILL GRANULE CELLS
Statistical Analysis: Two-way ANOVA tests were used to determine
statistical significance of all results. Results are presented as mean ⫾ SEM. A
value of p ⬍ 0.05 was considered to be significant.
RESULTS
Ortho-Substituted PCBs Kill Neuronal Cells and Cause an
Elevation of [Ca 2⫹] i
FIG. 1. PI fluorescence from cerebellar granule cells. Two populations of
neurons are apparent in the control (0 min). The broad peak at the left reflects
cells that do not take up PI. A portion of these cells (16%) take up essentially
no PI and are reflected as the sharp line at channel number 1. Others show
small amounts of fluorescence, and we consider all cells with a maximal
fluorescence intensity of 27 (the right hand edge of the “PI-low” bar) or less to
be healthy and viable. The mean fluorescent intensity of these cells in this
experiment was 2. The sharp peak on the right reflects dead cells (“PI-high”)
with a minimal fluorescence of 294. In this particular experiment 10% of the
initial cells were dead. With time in the presence of 10 ␮M PCB 52 the number
of PI-low cells decreases, the number of dead cells increases, and a new
intermediate peak of cells (“PI-intermediate” cells) becomes apparent. The
bars over each of the peaks reflect our arbitrary distinction among the three
populations. We consider the intermediate population to be those neurons with
fluorescence intensity between 27 and 294 and assume that these are injured
but viable neurons. Please note the fluorescence in this figure is in log scale (4
decades from 1 to 10,000). MFI ⫽ mean fluorescence intensity.
detected separately, or as only orange fluorescence (Reers et al., 1991; 1995).
When membrane integrity was measured in the same cells as JC-1, it was
necessary to use 7-AAD so as to avoid overlap of emission spectra.
Reagents: PCB congeners (97⫹% pure) were obtained from Ultra Scientific (North Kingstown, RI). PCBs were dissolved in DMSO and diluted such
that the final DMSO concentration was never greater than 0.2%. Control
DMSO experiments were always run in parallel to PCB exposures. Ionomycin
and all fluorescent probes were purchased from Molecular Probes (Eugene,
OR). Dispase II (neutral protease) was ordered from Boehringer Mannheim
(GmbH, Mannheim, Germany). All other chemicals were purchased from
Sigma Aldrich, Inc. (St. Louis, MO) or Fisher Scientific (Pittsburgh, PA).
Figure 1 shows raw data from flow cytometric analysis of
acutely isolated cerebellar granule cells in the presence of PI
and the time dependence of changes in PI fluorescence in the
presence of PCB 52 (2,5,2⬘,5⬘-tetrachlorobiphenyl) at a concentration of 10 ␮M. The majority of cells do not demonstrate
significant PI fluorescence, indicating that membrane integrity
is intact (PI-low cells). We presume that these are healthy,
viable cells. A percentage of cells, usually less than about 15%,
show intense fluorescence (PI-high cells, seen as the sharp
peak on the right). These cells have lost impermeability to PI,
and therefore we presume they are dead cells. After exposure
to PCB 52, there is a time-dependent decrease in all components of the population of PI-low cells and a corresponding
increase in the peak of dead cells. In addition, an intermediate
peak becomes apparent (PI-intermediate cells). We presume
that these are injured but living cells that have partially lost
membrane integrity. The bars over the various peaks in the
upper right records indicate the distinctions between these
categories.
Figure 2 shows the effects of various PCB congeners, each
at a concentration of 10 ␮M, on membrane integrity (which we
assume to be related to cell viability) of cerebellar granule cells
(A) and [Ca 2⫹] i (Fluo-3 fluorescence) (B) as a function of
exposure time. The ortho-substituted, non-coplanar PCB congeners, PCB 8 (2,4⬘-dichlorobiphenyl), PCB 28 (2,4,4⬘-trichlorobiphenyl), PCB 47 (2,4,2⬘,4⬘-tetrachlorobiphenyl), and PCB
52, all caused loss of membrane integrity of cerebellar granule
cells with a potency of PCB 52 ⬎ PCB 47 ⬎ PCB 8 ⬇ PCB
28, while three coplanar congeners, PCB 77 (3,4,3⬘,4⬘-tetrachlorobiphenyl), PCB 80 (3,5,3⬘,5⬘-tetrachlorobiphenyl), and
PCB 81 (3,4,5,4⬘-tetrachlorobiphenyl), had no effect on cell
membrane integrity at the same concentration over this period
of time. As shown in Figure 2B, the ortho-substituted PCB
congeners also caused an increase in [Ca 2⫹] i, and the relative
potency of the different congeners paralleled that for loss of
membrane integrity. With the most cytotoxic congeners, there
was a decline of Fluo-3 fluorescence with time, presumably
reflecting loss of Fluo-3 from the cell because membrane
integrity is compromised.
Figure 3 shows the dose-response effects of PCB 52 on cell
membrane integrity and the concomitant change in [Ca 2⫹] i
after 20 min. In this experiment we gated on all PI-low plus all
PI-intermediate cells, but excluded all dead cells. As the
[Ca 2⫹] i in the living cells increased, there was a progressively
reduced percentage of cells with membranes that exclude PI.
150
TAN ET AL.
FIG. 2. Effects of seven PCB congeners (10 ␮M) on membrane integrity
(which we assume is correlated with viability) (A) and [Ca 2⫹] i (B) in cerebellar
granule cells. Membrane integrity was assessed by PI uptake (A). [Ca 2⫹] i was
measured using Fluo-3 fluorescence (B). The apparent [Ca 2⫹] i decrease after
30 min with PCB 52 and 60 min with PCB 47 presumably reflects leakage
of Fluo-3, as a result of loss of membrane integrity, and not a true decrease.
* indicates p ⬍ 0.05 compared to control (0 min). All results are mean values
⫾ SEM.
integrity in any of the three media (A), and the effects on
[Ca 2⫹] i (B) were not different from the DMSO controls. In
contrast, PCB 52 (10 ␮M) induced an increase in [Ca 2⫹] i
proportional to the external calcium concentration available
(A). There was no [Ca 2⫹] i increase in calcium-free medium
(B), but the lack of an increase of the [Ca 2⫹] i did not prevent
the loss of cell membrane integrity. The PCB 52-induced loss
of membrane integrity did not correlate with the extracellular
calcium concentration. Furthermore, there was no relationship
between the PCB-induced loss of membrane integrity (and,
by implication, cell viability) and the intracellular calcium
concentration.
Further evidence that the elevation of [Ca 2⫹] i is not responsible for the loss of membrane integrity is shown in Figure 5.
In this figure, viability was determined by PI uptake and
changes in [Ca 2⫹] i, by determination of Fluo-3 fluorescence in
normal Tyrode’s. Ionomycin (10 ␮M) was added after 30 min
(at the arrow). Ionomycin forms plasma membrane channels
through which calcium (and other divalent cations) can enter,
and it induced an immediate and much more robust increase of
[Ca 2⫹] i than that caused by the non-coplanar congeners. However, the slow, time dependent decline of cell membrane integrity did not change after addition of ionomycin. These
observations are consistent with a previous report (Pearson et
al., 1992), which indicated that cerebellar granule cells from
young animals were relatively resistant to elevations of [Ca 2⫹] i
and provide evidence that the loss of viability is not secondary
to elevation of [Ca 2⫹] i.
PCBs Do Not Induce ROS Generation in Cerebellar Granule
Cells
An alternate possible cause of cell death induced by orthosubstituted PCBs is through production of ROS. Therefore, we
The Source of PCB-Induced Elevation of [Ca 2⫹] i and Its
Relation to Cell Death
Elevation of [Ca 2⫹] i is one possible cause of rapid cell death
(Berridge et al., 1998; Siesjo, 1990). Increased [Ca 2⫹] i may
result either from entry of calcium from the external medium
through the membrane or from release from intracellular storage sites. Therefore, we performed experiments to determine
the source of the calcium increase and the effect of removal of
extracellular calcium on the membrane integrity changes. We
incubated cells in Tyrode’s solution containing three different
calcium concentrations: the normal concentration of 2 mM,
calcium-free (no added calcium plus 2 mM EGTA), and eight
times normal calcium (16 mM). As shown in Figure 4, the
DMSO control cells had no change in membrane integrity for
periods up to 30 min (A), but their Fluo-3 fluorescence was
reduced at 10 min with the zero-calcium solution (B). There
was only slight further reduction at 30 min. The coplanar
congener, PCB 80 (10 ␮M), also had no effect on membrane
FIG. 3. Dose-dependence of effects of PCB 52 on cell membrane integrity
(which we assume correlates with viability) (solid circles) and intracellular
calcium concentration (open circles); measurements made at 20 min after
addition of PCB 52. * indicates p ⬍ 0.05 compared to control (0 min). All
points show mean ⫾ SEM.
PCBS KILL GRANULE CELLS
151
FIG. 5. Effects of ionomycin on membrane integrity (assumed to reflect
viability) and Fluo-3 fluorescence. With addition of 10 ␮M ionomycin (at
arrow) there was a change in [Ca 2⫹] i (open circles) with no change in the slight,
time dependent loss of membrane integrity (closed circles) over 20 min.
solution. Both compounds caused a marked increase in
DCF-DA fluorescence, as expected (data not shown), and
about 15% loss of viability after 80 min. We conclude that
utilization of DCF-DA for ROS detection was appropriate, and
that ortho-substituted PCBs do not induce DCF-DA-sensitive
FIG. 4. Effects of PCB 52 (10 ␮M) and PCB 47 (10 ␮M) on membrane
integrity (assumed to reflect viability) (A) and [Ca 2⫹] i (B) in normal Tyrode’s
solution containing 2 mM CaCl 2, in Tyrode⬘s without added calcium plus 2
mM EGTA, or in Tyrode⬘s containing 16 mM CaCl 2. The numbers in parentheses indicate the calcium concentration (mM). PCB 47 produced no effect on
membrane integrity and [Ca 2⫹] i in any medium. In contrast, PCB 52 caused a
loss of membrane integrity independent of external calcium concentration.
* indicates p ⬍ 0.05 compared to control (0 min). All results are mean ⫾ SEM.
used the ROS-sensitive probe, DCF-DA, to determine whether
representative non-coplanar and coplanar congeners (10 ␮M)
induced ROS generation, assessed in Tyrode’s solution containing 0, 2, or 16 mM external Ca 2⫹ concentration. There was
no indication of ROS production, regardless of the extracellular Ca 2⫹ concentration (Fig. 6; data for 16 mM not shown). The
upper records (A) were obtained in normal Ca 2⫹, while those in
B were obtained with a Ca-free solution. The loss of membrane
integrity, measured simultaneously, showed a pattern similar to
that shown in Figure 2. In order to test the reliability of the
ROS assessment, we performed control experiments in which
we applied 3-morpholinosydnonimine hydrochloride (SIN-1)
or 1-hydroxy-2-oxo-3-(N-3-methyl-aminopropyl)-3-methyl-1triazene (NOC-7), both at a concentration of 100 ␮M. These
substances are known to produce free radicals in physiological
FIG. 6. Effects of PCB 52 (10 mM) and PCB 47 (10 mM) on DCF-DA
fluorescence, a measure of ROS generation and peroxidase activity, in Tyrode’s solution containing the normal calcium concentration (2 mM) (A) or no
free calcium (B). Results are shown as mean ⫾ SEM.
152
TAN ET AL.
ROS over the period of time we have studied. In other studies
we have demonstrated significant ROS production in these
neurons upon application of excitatory amino acid agonists,
and this was associated with increased cell death (Boldyrev et
al., 1999a,b).
PCB Effects on Mitochondrial Membrane Potential
Since previous studies by Nishihara (1984), Nishihara and
Utsumi (1985), and Kodavanti et al. (1996) have implicated
mitochondria in PCB-induced cytotoxicity, we investigated the
actions of the mitochondrial uncoupler, carbonyl cyanide mchlorophenyl hydrazone (CCCP), and two representative congeners on mitochondrial membrane potential (⌬␺ m) and viability, using JC-1, a fluorescent dye which is sensitive to ⌬␺ m,
in conjunction with 7-AAD to monitor cell viability. Upon
excitation, JC-1 emits a different spectral pattern dependent
upon local membrane potential. Changes in ⌬␺ m can be measured either by assessment of the orange mean fluorescent
intensity (MFI) at 570 nm or determining the ratio of orange
over green fluorescence (MFI at 570 nm/MFI at 520 nm).
Similar results were obtained using either measurement. Results with CCCP (2 ␮M), PCB 80 (10 ␮M) and PCB 52 (10
␮M) are shown in Figure 7. CCCP caused a significant decline
in the ratio of orange to green fluorescence, indicating significant mitochondrial depolarization, but this was accompanied
by only a small loss of cell membrane integrity at 30 min.
There was no effect of PCB 80, but PCB 52 caused substantial
loss of membrane integrity (60%), with a lesser but significant
decrease of ⌬␺ m. These observations indicate that, while the
non-coplanar PCBs may have actions on mitochondria, their
ability to induce cytotoxicity is not solely correlated with
mitochondrial membrane potential.
FIG. 7. Effects of two PCB congeners (10 ␮M) and CCCP (30 min
incubation) on ⌬␺ m, where depolarization was measured as a change in the
JC-1 ratio, and membrane integrity (assumed to reflect viability) was measured
by uptake of 7-AAD. * indicates p ⬍ 0.05 compared to control (0 min). All
results shown are mean ⫾ SEM.
Effects of PCBs on Cell Size
Rothman (1985) has described a form of neuronal cell death
that results from anionic and cationic influxes induced by high
concentrations of the neurotransmitter glutamate, which causes
significant cell swelling leading to cell lysis. The cell swelling
and death is rapid and independent of changes in [Ca 2⫹] i. To
test whether ortho-substituted PCBs induce cell death via this
mechanism, we monitored cell size upon exposure to several
congeners (Fig. 8). Coplanar congeners did not alter cell size,
but the toxic ortho-substituted congeners caused a time-dependent decrease in cell size. In this experiment only the PI-low
neurons were gated for analysis (indicated by the left bar in
Fig. 1). These observations indicate that cell swelling, as a
result of NaCl accumulation, is not the mechanism of PCBinduced cell death.
DISCUSSION
These studies demonstrate that ortho-substituted but not
coplanar PCB congeners cause rapid loss of membrane integrity, which we assume correlates with cell death, of rat cerebellar granule cell neurons. The relative toxicity of the orthosubstituted congeners varied with the number of ortho
chlorines.
Rapid cell death of neurons is usually ascribed to one of four
mechanisms: increased toxic level of [Ca 2⫹] i (Berridge et al.,
1998; Siesjo, 1990), generation of a toxic amount of ROS
(Olanow, 1993), mitochondrial dysfunction (Cassarino and
Bennett, 1999; Wallace, et al. 1997), or acute swelling due to
entry of sodium and chloride (Rothman, 1992). The observations reported here indicate, however, that none of these four
mechanisms is responsible for the observed PCB-induced cell
death.
Two other groups have reported similar cytotoxic actions of
ortho-substituted PCBs. Kodavanti and colleagues (1993,
1994, 1996, 1997) have applied somewhat different techniques
to cultured cerebellar granule cell neurons and have shown that
non-dioxin-like congeners are cytotoxic, potentiate phosphoinositide hydrolysis, and translocation of protein kinase C.
These actions are apparent hours after exposure, a considerably
longer time than used in our studies, perhaps reflecting differences between acutely dissociated and cultured neurons, and
the effects were seen with PCB concentrations in the range of
10 to 100 ␮M (Kodavanti et al., 1993). They attributed these
actions to altered Ca 2⫹ homeostasis (Shafer et al., 1996) and
suggested that mitochondrial and microsomal calcium transport systems were made dysfunctional. Pessah and colleagues
(Wong and Pessah, 1997; Wong et al., 1997) have also demonstrated interference with calcium homeostasis by non-dioxin-like congeners, but they suggested the site of action to be the
ryanodine receptor in the endoplasmic reticulum in neurons or
the sarcoplasmic reticulum of muscle. Our previous investigations using flow cytometry of granule cells have also reported
PCBS KILL GRANULE CELLS
FIG. 8. Effects of different PCB congeners on cerebellar granule cell size,
represented by the mean channel number of forward scatter (FSC), as a
function of time of incubation. Forward scatter reflects cell size, and therefore
a decrease in forward scatter indicates a shrinkage of average cell size. All data
were obtained by gating on PI-low cells, and the values are means ⫾ SEM.
* indicates p ⬍ 0.05 compared to control (0 min). The reduction in forward
scatter indicates shrinkage of the cells.
an elevation of [Ca 2⫹] i induced by ortho-substituted congeners
(Carpenter et al., 1997). Our previous and present results are
therefore not incompatible with either of these explanations for
the elevation of intracellular calcium. However, our results also
indicate that the non-coplanar PCB cytotoxicity is not dependent upon the elevation of [Ca 2⫹] i.
Calcium elevation has been widely viewed as a common
final mechanism of cell death (Siesjo, 1990). The observation
that there was no PCB-induced elevation of intracellular calcium in absence of external calcium provides evidence that the
extracellular calcium is the source of the elevation, but this
evidence does not rule out the possibility that ortho-substituted
PCBs also act by disruption of calcium regulation at intracellular storage sites. However, even if this is the case, PCBs do
not appear to kill cells by inducing a massive release of
calcium from intracellular stores, since such release should be
reflected in a rise of intracellular calcium in calcium-free
external medium. Voie et al. (1998) reported that elevation of
[Ca 2⫹] i in human granulocytes by PCB 4 (2,2⬘-dichlorobiphenyl) was dependent on the extracellular calcium concentration.
Cerebellar granule cells from young animals are known to be
more tolerant of significant elevations of intracellular calcium
than are tolerated by most mature cells, and indeed it has been
shown that application of NMDA (Balazs et al., 1988) or
ionomycin (Pearson et al., 1992), both of which increase
intracellular calcium concentration, actually promotes proliferation of granule cells. In addition, a promotion of cell survival
by agents that raise calcium has been reported in other types of
immature neurons (Collins, 1991).
While in our analyses the elevation of the [Ca 2⫹] i is not
153
directly responsible for the cell death, this does not mean that
the calcium elevation is without physiological importance in
either these or other kinds of cells. Fischer et al. (1999) have
shown that ortho-substituted PCBs, but not coplanars, cause
release of insulin from RINm5F cells, a pancreatic cell line,
and that this release is dependent upon entry of extracellular
calcium. It appears likely that the mechanism responsible for
calcium entry is similar to that described here, since our results
clearly demonstrate that ortho-substituted PCBs cause an elevation of [Ca 2⫹] i coming from extracellular sources, and elevation of [Ca 2⫹] i is known to be the trigger for release of
insulin.
Like elevations of intracellular calcium, oxygen radicals are
widely believed to be common mediators of neurotoxicity
(LeBel and Bondy, 1991). Non-coplanar PCBs induce superoxide anion (O 2– ) generation in neutrophils (Ganey et al.,
1993), and this activation of neutrophils is dependent on calcium (Brown and Ganey, 1995). In addition, Oakley et al
(1996) have shown ROS generation from the dihydroxy metabolites of PCBs. Hennig et al. (1999) have shown that some
coplanar congeners damage endothelial cells, and that these
actions are potentiated by certain unsaturated fatty acids, presumably via oxidative stress. However, using DCF-DA, which
is a relatively broad spectrum ROS detector that reacts with
most but not all ROS, we did not detect any free radical
generation by cerebellar granule cells treated with PCB 52 or
PCB 80 (10 ␮M) in three different media. The generation of
ROS in neutrophils and endothelial cells by PCBs may reflect
specific characteristics of these cell types. In the case of the
neutrophils, this makes particular sense because they are designed to generate free radicals to defend the host from destruction by pathogens.
Ortho-substituted PCBs alter liver mitochondria by inhibition of the electron transport chain and act as an uncoupler
(Nishihara and Utsumi, 1987; Nishihara et al., 1986, 1987).
Thus, one predicted effect of ortho-substituted PCBs would be
a reduction of mitochondrial potential. We tested this hypothesis using JC-1 (Reers et al., 1991, 1995). Treatment with
non-coplanar PCBs, but not dioxin-like PCBs, caused a significant drop in ⌬␺ m, as did the uncoupler, CCCP (2 ␮M).
However, in spite of a dramatic decrement of ⌬␺ m, CCCP
caused only a slight decline in cell membrane integrity. Thus,
the ⌬␺ m alone cannot be the major factor for the rapid
cytotoxicity.
Traditional uncouplers, including CCCP, dissipate ⌬␺ m by
shuttling protons across mitochondrial membranes with an
acid-dissociable group within the molecule (Budd and
Nicholls, 1996). It is not clear how PCBs decrease ⌬␺ m. The
mechanism of PCB uncoupling must be different, since PCBs
are neutral molecules. LaRocca and Carlson (1979) have evaluated the relationship between inhibition of Mg-ATPase and
the lipophilic property of PCBs and observed a strong negative
correlation between PCB-induced inhibition of ATPases and
solubility of the PCB congeners. Kodavanti et al. (1996) have
154
TAN ET AL.
reported inhibition of microsomal and mitochondrial Ca 2⫹ sequestration in granule cells by PCB congeners. The IC 50 values
for inhibition of microsomal and mitochondrial Ca 2⫹ sequestration were quite similar for both PCB mixtures and single
congeners. Since mitochondria and microsomes employ totally
different mechanisms for calcium sequestration (uniporter and
Ca-ATPase, respectively), one would expect distinct IC 50 values if PCBs acted directly on these transport systems. Nishihara and Utsumi (1986) reported that, when the uniporter was
blocked by ruthenium red, administration of PCB 52 still
induced calcium release from mitochondria accompanied by
mitochondrial swelling. Compromised mitochondrial membrane integrity was further evidenced by the release of endogenous K ⫹ (Nishihara, 1984) and altered permeability to large
molecules such as NADH (Nishihara, 1984), induced by Kanechlor-400, in which tetrachlorobiphenyls are the major congeners.
Rothman (1985) has described a form of neuronal cell death
that results from anionic and cationic influxes induced by high
concentrations of the neurotransmitter, glutamate, which
causes significant cell swelling leading to cell lysis. The cell
swelling and death is rapid and independent of changes in
[Ca 2⫹] i. To test whether non-planar PCBs induce cell death via
this mechanism, we monitored cell size upon exposure to
several congeners (Fig. 8). Coplanar congeners did not alter
cell size, but the ortho-substituted congeners caused a timedependent decrease in cell size. These observations indicate
that cell swelling, as a result of NaCl accumulation, is not the
mechanism of PCB-induced cell death.
Cell shrinkage is characteristic of apoptotic cell death, not
necrotic (Bonfoco et al. 1995), but apoptotic cell death is
usually a more delayed response than we see here. Our present
data do not allow one to characterize cerebellar granule cell
death upon exposure to PCBs as being either necrotic or
apoptotic, but the cell shrinkage is more consistent with apoptotic cell death.
Ortho-substituted but not coplanar PCB congeners alter
plasma membrane permeability to the relatively large probes,
7-AAD (MW ⫽ 1270) and PI (MW ⫽ 668), as well as to
calcium, of living or injured granule cells. Under normal circumstances these probes are not able to penetrate the impermeable barrier of the plasma membrane. However, after exposure to PCBs there is a degree of accumulation of these
compounds, as indicated by the broadening of the peak under
“PI-low” cells, and the appearance of injured (“PI-intermediate”) cells (Fig. 1), and an increase in [Ca 2⫹] i (Fig. 3).
These results from cerebellar granule cells are very similar
to results previously obtained using thymocytes (Tan et al.,
2003). Thymocytes are rapidly killed by ortho-substituted
PCBs in a dose-dependent fashion. The relative potency of the
various congeners is similar to that we describe in this study.
The cell death is accompanied by elevations of [Ca 2⫹] i and
reduction of ⌬␺ m. There were some differences, in that thymocytes were somewhat more sensitive than granule cells,
showing significant cell death at 1 ␮M concentrations of PCB
52, and there was a significant ROS generation at higher
concentrations.
The general similarity of effects of ortho-substituted PCBs
on cerebellar granule cell neurons and thymocytes, plus the
evidence for disruption of both plasma membrane integrity and
intracellular organelles suggests that these compounds may be
altering membrane structure. We have examined this possibility using fluorescence polarization dyes. These results are
described in the accompanying paper, where evidence is presented that the non-planar PCBs alter the structure and function
of several (and likely all) cellular membranes (Tan et al.,
2004).
ACKNOWLEDGMENTS
This study was supported by a grant received from the National Institute of
Environmental Health Sciences, #ES04913 to D.O. Carpenter.
REFERENCES
Agency for Toxic Substances and Disease Registry. (2000). Toxicological
Profile for Polychlorinated Biphenyls. US Dept. of Health and Human
Services, prepared by Syracuse Research Corporation, contract number
205-1999-00024.
Balazs, R., Hack, N., and Jorgensen, O. S. (1988). Stimulation of the Nmethyl-D-aspartate receptor has a trophic effect on differentiating cerebellar
granule cells. Neurosci. Lett. 87, 80 – 86.
Berridge, M. J., Bootman, M. D., and Lipp, P. (1998). Calcium—a life and
death signal [news]. Nature 395, 645– 648.
Boldyrev, A., Song, R., Lawrence, D., and Carpenter, D. O. (1999b). Carnosine protects against excitotoxic cell death independently of effects on
reactive oxygen species. Neuroscience 94, 571–577.
Boldyrev, A. A., Johnson, P., Wei, Y., Tan, Y., and Carpenter, D. O. (1999a).
Carnosine and taurine protect rat cerebellar granular cells from free radical
damage. Neurosci. Lett. 263, 169 –172.
Bonfoco, E., Krainc, D., Ankarcrona, M., Nicotera, P., and Lipton, S. A.
(1995). Apoptosis and necrosis: Two distinct events induced, respectively,
by mild and intense insults with N-methyl-d-aspartate or nitric oxide/
superoxide in cortical cell cultures. Proc. Natl. Acad. Sci. U.S.A. 92, 7162–
7166.
Brown, A. P., and Ganey, P. E. (1995). Neutrophil degranulation and superoxide production induced by polychlorinated biphenyls are calcium dependent. Toxicol. Appl. Pharmacol. 131, 198 –205.
Budd, S. L., and Nicholls, D. G. (1996). A reevaluation of the role of
mitochondria in neuronal Ca 2⫹ homeostasis. J. Neurochem. 66, 403– 411.
Carpenter, D. O., Stoner, C. R., and Lawrence, D. A. (1997). Flow cytometric
measurements of neuronal death triggered by PCBs. Neurotoxicology 18,
507–513.
Cassarino, D. S., and Bennett, J. P. (1999). An evaluation of the role of
mitochondria in neurodegenerative diseases: Mitochondrial mutations and
oxidative pathology, protective nuclear responses, and cell death in neurodegeneration. Brain Res. Rev. 29, 1–25.
Chen, Y. C., Yu, M. L., Rogan, W. J., Gladen, B. C., and Hsu, C. C. (1994).
A 6-year follow-up of behavior and activity disorders in the Taiwan Yucheng children. Am. J. Public Health 84, 415– 421.
Collins, F. (1991). Sustained increase in intracellular calcium promotes neuronal survival. J. Neurosci. 11, 2582–2587.
Dyatlov, V. A., Dyatlova, O. M., Parsons, P. J., Lawrence, D. A., and
PCBS KILL GRANULE CELLS
Carpenter, D. O. (1998). Lipopolysaccharide and interleukin-6 enhance lead
entry into cerebellar neurons: Application of a new and sensitive flow
cytometric technique to measure intracellular lead and calcium concentrations. Neurotoxicology 19, 293–302.
Fischer, L. J., Seegal, R. F., Ganey, P. E., Pessah, I. N., and Kodavanti, P. R.
(1998). Symposium overview: Toxicity of non-coplanar PCBs. Toxicol. Sci.
41, 49 – 61.
Fischer, L. J., Wagner, M. A., and Madhukar, B. V. (1999). Potential involvement of calcium, CaM kinase II, and MAP kinases in PCB-stimulated
insulin release from RINm5F cells. Toxicol. Appl. Pharmacol. 159, 194 –
203.
Fischer, L. J., Zhou, H. R., and Wagner, M. A. (1996). Polychlorinated
biphenyls release insulin from RINm5F cells. Life Sci. 59, 2041–2049.
Ganey, P. E., Sirois, J. E., Denison, M., Robinson, J. P., and Roth, R. A.
(1993). Neutrophil function after exposure to polychlorinated biphenyls in
vitro. Environ. Health Perspect. 101, 430 – 434.
Guo, Y. L., Chen, Y. C., Yu, M. L., and Hsu, C. C. (1994). Early development
of Yu-Cheng children born seven to twelve years after the Taiwan PCB
outbreak. Chemosphere 29, 2395–2404.
Hennig, B., Slim, R., Toborek, M., and Robertson, L. W. (1999). Linoleic acid
amplifies polychlorinated biphenyl-mediated dysfunction of endothelial
cells. J. Biochem. Mol. Toxicol. 13, 83–91.
Jacobson, J. L., and Jacobson, S. W. (1997). Evidence for PCBs as neurodevelopmental toxicants in humans. Neurotoxicology 18, 415– 424.
Kodavanti, P. R., Shafer, T. J., Ward, T. R., Mundy, W. R., Freudenrich, T.,
Harry, G. J., and Tilson, H. A. (1994). Differential effects of polychlorinated biphenyl congeners on phosphoinositide hydrolysis and protein
kinase C translocation in rat cerebellar granule cells. Brain Res. 662,
75– 82.
Kodavanti, P. R., Shin, D. S., Tilson, H. A., and Harry, G. J. (1993). Comparative effects of two polychlorinated biphenyl congeners on calcium
homeostasis in rat cerebellar granule cells. Toxicol. Appl. Pharmacol. 123,
97–106.
Kodavanti, P. R., and Tilson, H. A. (1997). Structure-activity relationships of
potentially neurotoxic PCB congeners in the rat. Neurotoxicology 18, 425–
441.
Kodavanti, P. R., Ward, T. R., McKinney, J. D., and Tilson, H. A. (1996).
Inhibition of microsomal and mitochondrial Ca 2⫹-sequestration in rat cerebellum by polychlorinated biphenyl mixtures and congeners. Structureactivity relationships. Arch. Toxicol. 70, 150 –157.
LaRocca, P. T., and Carlson, G. P. (1979). The effect of polychlorinated
biphenyls on adenosine triphosphatase activity. Toxicol. Appl. Pharmacol.
48, 185–192.
Lai, T. J., Guo, Y. L., Yu, M. L., Ko, H. C., and Hsu, C. C. (1994). Cognitive
development in Yucheng children. Chemosphere 29, 2405–2411.
LeBel, C. P., and Bondy, S. C. (1991). Oxygen radicals: Common mediators
of neurotoxicity. Neurotoxicol. Teratol. 13, 341–346.
Lonky, E., Reihman, J., Darvill, T., Mather, J., and Daly, H. (1996). Neonatal
behavioral assessment scale performance in humans influenced by maternal
consumption of environmentally contaminated lake Ontario fish. J. Great
Lakes Res. 22, 198 –212.
Maftah, A., Huet, O., Gallet, P. F., and Ratinaud, M. H. (1993). Flow cytometry’s contribution to the measurement of cell functions. Biol. Cell 78,
85–93.
Nishihara, Y. (1984). Uncoupling action of polychlorinated biphenyls (Kanechlor- 400) on oxidative phosphorylation in rat liver mitochondria. Arch.
Environ. Contam. Toxicol. 13, 225–229.
Nishihara, Y., Robertson, L. W., Oesch, F., and Utsumi, K. (1985). Interaction
of tetrachlorobiphenyls with isolated rat liver mitochondria. J. Pharmacobiodyn. 8, 726 –732.
155
Nishihara, Y., Robertson, L. W., Oesch, F., and Utsumi, K. (1986). The effects
of tetrachlorobiphenyls on the electron transfer reaction of isolated rat liver
mitochondria. Life Sci. 38, 627– 635.
Nishihara, Y., Robertson, L. W., and Utsumi, K. (1987). Studies of the
influence of chloro-substituent sites in tetrachlorobiphenyls on the uncoupling of oxidative phosphorylation in isolated rat liver mitochondria. Chem.
Pharm. Bull. (Tokyo) 35, 587–1595.
Nishihara, Y., and Utsumi, K. (1982). Effects of polychlorinated biphenyls
(Kanechlor-400) on the potassium compartmentation and glucose permeability of human erythrocyte membranes. Bull. Environ. Contam. Toxicol.
29, 208 –213.
Nishihara, Y., and Utsumi, K. (1985). Studies on the response of isolated rat
liver mitochondria to polychlorinated biphenyls (Kanechlors). Arch. Environ. Contam. Toxicol. 14, 65–71.
Nishihara, Y., and Utsumi, K. (1986). 2,5,2⬘,5⬘-Tetrachlorobiphenyl impairs
the bioenergetic functions of isolated rat liver mitochondria. Biochem.
Pharmacol. 35, 3335–3339.
Nishihara, Y., and Utsumi, K. (1987). 4-Chloro-4⬘-biphenylol as an uncoupler
and an inhibitor of mitochondrial oxidative phosphorylation. Biochem.
Pharmacol. 36, 3453–3457.
Oakley, G. C., Devanaboyina, U.-S., Robertson, L. W., and Gupta, R. C.
(1996). Oxidative DNA damage induced by activation of polychlorinated
biphenyls (PCBs): Implications for PCB-induced oxidative stress in breast
cancer. Chem. Res. Toxicol. 9, 1285–1292.
Olanow, C. W. (1993). A radical hypothesis for neurodegeneration. Trends
Neurosci. 16, 439 – 444.
Oyama, Y., Carpenter, D. O., Chikahisa, L., and Okazaki, E. (1996). Flowcytometric estimation on glutamate- and kainate-induced increases in intracellular Ca 2⫹ of brain neurons: A technical aspect. Brain Res. 728, 121–124.
Oyama, Y., Ueha, T., Hayashi, A., Chikahisa, L., and Noda, K. (1992). Flow
cytometric estimation of the effect of Ginkgo Biloba extract on the content
of hydrogen peroxide in dissociated mammalian brain neurons. Japan
J. Pharmacol. 60, 385–388.
Pearson, H., Graham, G. E., and Burgoyne, R. D. (1992). Relationship between
intracellular free calcium concentration and NMDA-induced cerebellar
granule cell survival in vitro. Eur. J. Neurosci. 4, 1369 –1375.
Reers, M., Smiley, S. T., Mottola, H. C., Chen, A., Lin, M., and Chen, L. B.
(1995). Mitochondrial membrane potential monitored by JC-1 dye. Methods
Enzymol. 260, 406 – 417.
Reers, M., Smith, T. W., and Chen, L. B. (1991). J-aggregate formation of a
carbocyanine as a quantitative fluorescent indicator of membrane potential.
Biochemistry 30, 4480 – 4486.
Rice, D. C., and Hayward. S. (1999). Effects of postnatal exposure of monkeys
to a PCB mixture on concurrent random interval-random interval and
progressive ratio performance. Neurotoxicol. Teratol. 21, 47–58.
Rothman, S. M. (1985). The neurotoxicity of excitatory amino acids is produced by passive chloride influx. J. Neurosci. 5, 1483–1489.
Rothman, S. M. (1992). Excitotoxins: Possible mechanisms of action. Ann.
N.Y. Acad. Sci. 648, 132–139.
Safe, S. H. (1994). Polychlorinated biphenyls (PCBs): Environmental impact,
biochemical and toxic responses, and implications for risk assessment. Crit.
Rev. Toxicol. 24, 87–149.
Schantz, S. L., Seo, B. W., Wong, P. W., and Pessah, I. N. (1997). Long-term
effects of developmental exposure to 2,2⬘,3,5⬘,6-pentachlorobiphenyl (PCB
95) on locomotor activity, spatial learning and memory and brain ryanodine
binding. Neurotoxicology 18, 457– 467.
Schantz, S. L., Gasior, D. M., Polverejan, E., McCaffrey, R. J., Sweeney,
A. M., Humphrey, H. E., and Gardiner J. C. (2001). Impairments of memory
and learning in older adults exposed to polychlorinated biphenyls via consumption of Great Lakes fish. Environ. Health Perspect. 110, A70 –A72.
156
TAN ET AL.
Seegal, R. F., Bush, B., and Shain, W. (1990). Lightly chlorinated orthosubstituted PCB congeners decrease dopamine in nonhuman primate brain
and in tissue culture. Toxicol. Appl. Pharmacol. 106, 136 –144.
Shafer, T. J., Mundy, W. R., Tilson, H. A., and Kodavanti, P. R. (1996).
Disruption of inositol phosphate accumulation in cerebellar granule cells by
polychlorinated biphenyls: A consequence of altered Ca 2⫹ homeostasis.
Toxicol. Appl. Pharmacol. 141, 448 – 455.
Siesjo, B. K. (1990). Calcium in the brain under physiological and pathological
conditions. Eur. Neurol. 30, Suppl 2:3–9; discussion 39 – 41, 3–9.
Tan, Y., Chen, C-H., Lawrence, D., and Carpenter, D. O. (2004). Ortho-substituted
PCBs kill cells by altering membrane structure. Toxicol. Sci. (in press).
Tan, Y., Li, D., Song, R., Lawrence, D., and Carpenter, D. O. (2003).
Ortho-substituted PCBs kill thymocytes. Toxicol. Sci. 76, 328 –337.
Tilson, H. A., and Kodavanti. P. R. (1997). Neurochemical effects of polychlorinated biphenyls: An overview and identification of research needs.
Neurotoxicology 18, 727–743.
Voie, O. A., Wiik, P., and Fonnum, F. (1998). Ortho-substituted polychlori-
nated biphenyls activate respiratory burst measured as luminol-amplified
chemiluminescence in human granulocytes. Toxicol. Appl. Pharmacol. 150,
369 –375.
Wallace, K. B., Eells, J. T., Madeira, V. M.C., Cortopassi, G., and Jones, D. P.
(1997). Mitochondria-mediated cell injury. Fund. Appl. Toxicol. 38, 23–37.
Wong, P. W., Brackney, W. R., and Pessah, I. N. (1997). Ortho-substituted
polychlorinated biphenyls alter microsomal calcium transport by direct
interaction with ryanodine receptors of mammalian brain. J. Biol. Chem.
272, 15145–15153.
Wong, P. W., and Pessah, I. N. (1997). Noncoplanar PCB 95 alters microsomal
calcium transport by an immunophilin FKBP12-dependent mechanism. Mol.
Pharmacol. 51, 693–702.
Yoshimura, T. (1978). Growth of school children with polychlorinated biphenyl poisoning or yusho. Environ. Res. 17, 416 – 425.
Zagon, I. S., McLaughlin, P. J., and Smith. S. (1977). Neural populations in the
human cerebellum: Estimations from isolated cell nuclei. Brain Res. 127,
279 –282.