(Received for publication, October 2, 1996)
From the Department of Microbiology and Immunology
and the § Department of Pathology, MCP Hahnemann School of
Medicine, Allegheny University of the Health Sciences,
Philadelphia, Pennsylvania 19102-1192
At present, approaches to studying mitochondrial
functions in malarial parasites are quite limited because of the
technical difficulties in isolating functional mitochondria in
sufficient quantity and purity. We have developed a flow cytometric
assay as an alternate means to study mitochondrial functions in intact erythrocytes infected with Plasmodium yoelii, a rodent
malaria parasite. By using a very low concentration (2 nM)
of a lipophilic cationic fluorescent probe,
3,3dihexyloxacarbocyanine iodide, we were able to measure
mitochondrial membrane potential(
m) in
live intact parasitized erythrocytes through flow cytometry. The
accumulation of the probe into parasite mitochondria was dependent on
the presence of a membrane potential since inclusion of carbonyl cyanide m-chlorophenylhydrazone, a protonophore, dissipated
the membrane potential and abolished the probe accumulation. We tested the effect of standard mitochondrial inhibitors such as myxothiazole, antimycin, cyanide and rotenone. All of them except rotenone collapsed the
m and inhibited respiration. The
assay was validated by comparing the EC50 of these
compounds for inhibiting
m and
respiration. This assay was used to investigate the effect of various
antimalarial drugs such as chloroquine, tetracycline and a broad
spectrum antiparasitic drug atovaquone. We observed that only
atovaquone collapsed
m and inhibited
parasite respiration within minutes after drug treatment. Furthermore, atovaquone had no effect on mammalian
m.
This suggests that atovaquone, shown to inhibit mitochondrial electron
transport, also depolarizes malarial mitochondria with consequent
cellular damage and death.
Plasmodium spp. are obligate intracellular parasites,
spending a major portion of their life cycle within erythrocytes and converting these relatively inactive cells into metabolically thriving
active hosts. At present, our knowledge of mechanisms by which the
parasite accomplishes these changes is limited, as is our understanding
of metabolic processes associated with parasitism. It is generally
believed that glycolysis is the main source of ATP in erythrocytic
stages of malarial parasites with little or no contribution by
mitochondria to the cellular ATP pool (1, 2). A lack of tricarboxylic
acid cycle enzymes (3-6) and an acristate mitochondrial morphology has
led to the suggestion that mitochondria in malaria parasite act mainly
to serve as an electron disposal sink for dihydroorotate dehydrogenase,
a critical enzyme in pyrimidine biosynthesis (7-9). It is well
established through studies in other systems that, in addition to
oxidative phosphorylation, mitochondria are also central to many other
physiological activities such as the metabolism of molecules such as
amino acids, lipids, and heme, as well as intracellular
Ca2+ homeostasis (10). These functions are achieved by the
action of gene products encoded by both mitochondrial and nuclear
genomes. Because most of the mitochondrial proteins are encoded by the nuclear genome and imported into mitochondria, an active import mechanism is necessary for mitochondrial functions. Both metabolites and protein transport require maintenance of membrane potential across
the inner mitochondrial membrane (11, 12). The mitochondrial electron
transport chain serves to generate this membrane potential (11). Hence,
maintenance of m1 is critical
not only for ATP synthesis but also for the maintenance of additional
metabolic activities of mitochondria. While it is not established which
of the non-ATP synthesis functions are present in malarial parasites,
it is safe to assume that these are likely to be critical for parasite
physiology.
Approaches for investigating mitochondrial functions in malarial
parasite are quite limited at present. Mitochondrial DNAs of various
Plasmodium spp. have been sequenced and found to encode at
least three components of the electron transport chain, viz. subunits 1 and 3 of cytochrome c oxidase, and apocytochrome
b (13-16). In addition, mitochondrial preparations have
been shown to contain ubiquinone cytochrome c oxidoreductase
(bc1 complex) (17), and cytochrome c
oxidase activities (18-20). However, detailed studies of mitochondrial
functions and their response to antimalarial drugs have been hampered
by the technical difficulties of obtaining workable quantities of
functional mitochondria. To circumvent these problems, we have explored
the possibility of studying mitochondrial functioning in intact
parasitized erythrocytes with a fluorescent activated cell scanner
(FACS). We have used the ability of the lipophilic cationic fluorescent
probe DiOC6(3) to partition into energized mitochondria as
a measure of m. By using this assay, we were able to
demonstrate that a new antiparasitic drug, atovaquone, rapidly
collapses
m in erythrocytes infected with a rodent
malaria parasite Plasmodium yoelii.
An animal colony of BALB/cByJ mice was maintained in our American Association of Laboratory Animal Care accredited animal facility.
ParasitesP. yoelii 17XL was maintained in vivo in either male or female BALB/cByJ mice. Blood was collected in Hanks' balanced salt solution (HBSS) containing 10 units of heparin/ml of HBSS at approximately 60% parasitemia. Red blood cells were washed three times with HBSS at low speed centrifugation (800 × g for 10 min each time). Washed erythrocytes were diluted 1:1 in phosphate-buffered saline, pH 7.4, and passed over a microcrystalline cellulose column to remove leukocytes and platelets (21). Red blood cells depleted of leukocytes and platelets were washed and resuspended in phosphate-buffered saline, and infected red blood cells were enriched by centrifugation over a discontinuous Percoll density gradient as described elsewhere (22). Infected erythrocytes within 65% Percoll and 65-75% Percoll interphase were collected and contained predominantly trophozoites and schizonts as determined by Giemsa-stained thin blood smears. These preparations were free of leukocytes and platelets as judged by Giemsa staining. These fractions were pooled, washed twice with RPMI 1640 medium containing 1% fetal calf serum and used in all of the experiments described here.
InhibitorsAll of the mitochondrial respiratory chain inhibitors, e.g. antimycin, myxothiazole, rotenone, and cyanide; uncoupler carbonyl cyanide m-chlorophenyl hydrazone (CCCP); and antimalarials tetracycline and chloroquine, were purchased from Sigma. The antimalarial atovaquone was a gift from Glaxo Wellcome, Research Triangle Park, NC.
Flow Cytometric Assay forFlow cytometric
assays were carried out using a FACScan (Becton Dickinson Cellular
Imaging). The flow cytometer was adjusted for forward scattering
profile, side scattering profile and fluorescence detection of infected
red blood cells at channel 1. Parasitized cells at a concentration of
5 × 106/ml in RPMI 1640 medium containing 1% fetal
calf serum were incubated with a 2 nM final concentration
of DiOC6(3) for 20 min at 37 °C. At the end of the
incubation period the suspension was aliquoted into eight tubes of 250 µl each. Different concentrations of the compound to be tested were
added, and the mixture was incubated for an additional 20 min. At the
end of the incubation period, each sample was subjected to flow
cytometric analysis. For each sample, 10,000 events were counted at the
same flow cytometer setting. The results were calculated from the mean
fluorescence of 104 cells in a histogram. In each
experiment, measurements of fluorescence in infected erythrocytes in
the presence of dye (F+D) and in absence of dye
(FD) were carried out to establish baselines.
Mitochondrial uptake of O2 was measured in a closed system using a Clark's oxygen electrode and K-IC oxygraph (Gilson Medical Electronics Inc., Middleton, WI) in a reaction volume of 1.5 ml at 37 °C following the method of Chance and Williams (23). The instrument was calibrated following the manufacturer's instructions. Briefly, 1.5 × 108 parasitized cells in 1.5 ml of RPMI 1640 medium containing 1% fetal calf serum were added into the chamber, and the rate of O2 consumption by infected erythrocytes was followed for 5 min. A desired concentration of the compound to be tested was added into the chamber using a Hamilton syringe. The rate of O2 consumption by infected erythrocytes was followed for the next 5 min. The rate of O2 consumption was expressed as nAO/108 infected erythrocytes/min. The difference in the rate of O2 consumption by infected erythrocytes in the presence versus the absence of the compound was calculated as the measure of respiration inhibition. All the compound concentrations were tested individually in a separate set of experiments.
DiOC6(3) is a cationic, lipophilic
fluorescent compound. When incubated with infected erythrocytes, it
diffuses into cells and is concentrated several orders of magnitude
into negative-inside mitochondria. A collapsed m will
result in diffusion of the probe out of the mitochondria resulting in
dissipation of the signal. As shown in Fig.
1A, incubation of parasitized erythrocytes
with 2 nM DiOC6(3) led to the accumulation of
the probe into parasites, which was reflected by the increase in
fluorescence intensity. The fact that the accumulation of the probe
within the mitochondria was dependent upon a membrane potential was
shown by the dissipation of the signal intensity after incubation of infected erythrocytes with the protonophore CCCP (Fig. 1B).
As shown in Fig. 2A), cell-associated
DiOC6(3) fluorescence intensity increased as a function of
the probe concentration up to 50 nM. Incubation with CCCP
dissipated cell-associated DiOC6(3) fluorescence. Maximal
dissipation of about 70% was obtained at the probe concentration of 2 nM. At higher probe concentrations the cell-associated
fluorescence became relatively resistant to CCCP-mediated dissipation.
The inhibition of the fluorescence above 50 nM appears to
be the result of the inhibition of electron transport by the dye (see
below). The decrease in the extent of quenching by CCCP above 2 nM appears to result from the self-quenching of the dye on
the interface of the mitochondrial membrane, which thus saturates the
signal. Hence, we decided to use 2 nM DiOC6(3)
as the final probe concentration for the
m assay for
malarial parasite. We also measured the kinetics of probe accumulation
at 2 nM concentration in infected erythrocytes as shown in
Fig. 2B. A rapid accumulation of the probe in infected
erythrocytes was observed, reaching maximum levels within 30 min. The
slight reduction of fluorescent intensity after 30 min suggested that
some self quenching occurred even at 2 nM dye
concentration. Continued incubation of infected erythrocytes with the
probe for longer than 60 min did not result in any significant change
of fluorescence profile of the probe, indicating that partitioning of
the probe into mitochondria had reached its equilibrium.
Because incubation with lipophilic compounds can also affect parasite
physiology, we determined the effect of DiOC6(3) on mitochondrial functioning by measuring the rate of respiration in the
presence and absence of DiOC6(3). As shown in Fig.
3, there was no significant effect of the probe at 2 nM upon the rate of O2 consumption. However, at
the higher concentrations inhibition of respiration became apparent,
reaching up to 50% at 150 nM DiOC6(3) concentration. Hence, for the rest of the study, DiOC6(3)
was used at the final concentration of 2 nM.
Effect of Mitochondrial Inhibitors on
Having established a way to assay m in
intact parasitized erythrocytes, we tested the effect of various known
mitochondrial inhibitors on
m. Inhibitors used were a
protonophore (CCCP), a site I inhibitor (rotenone), site II inhibitors
(antimycin and myxothiazole), and a site III inhibitor (cyanide). Fig.
4 shows representative histograms showing effects of
these inhibitors (at the highest concentration tested) on mitochondrial
accumulation of DiOC6(3) in infected erythrocytes. Fig.
5A shows the concentration-dependent dissipation of the fluorescence for each of these inhibitors. Rotenone
did not have any significant effect on
m, which is
consistent with the earlier findings regarding the lack of
NADH-ubiquinone oxidoreductase (site I) (17) in malarial parasites.
Myxothiazole and antimycin collapsed
m with an
EC50 of 2 × 10
7 M and
6 × 10
7 M, respectively, consistent
with the presence of the bc1 complex (site II)
in malarial parasites. A much higher concentration of cyanide (6 mM) was required to completely collapse the
m and dissipate the DiOC6(3) accumulation with an
EC50 of 3 × 10
4 M.
We also tested the effect of these inhibitors on rate of respiration by
infected erythrocytes as shown in Fig. 5B. Both antimycin and
myxothiazole inhibited respiration with EC50 of 2.5 × 107 M and 9 × 10
8
M, respectively. At higher concentrations, i.e.
1 × 10
6 M and above, both these drugs
almost completely inhibited respiration. Rotenone had no apparent
effect upon the rate of O2 consumption in the range of
concentrations tested in this study. Surprisingly, CCCP did not have
any appreciable effect on the respiration rate of infected red blood
cells. This observation is counter to the classical effect of CCCP
where a collapsed membrane potential leads to a release from
respiration control over electron transport, resulting in an increased
rate of O2 consumption.
Having demonstrated a good correlation
between the DiOC6(3) assay of m and
respiration by intact infected erythrocytes, we decided to assess the
effect of various antimalarial drugs on mitochondrial physiology. Three
different drugs were tested: atovaquone, a new broad spectrum
antiparasitic drug shown to inhibit the bc1
complex of the malarial parasite (24); tetracycline, an antibiotic
thought to inhibit organellar protein synthesis (9, 25); and
chloroquine, a drug likely to affect nonmitochondrial target(s)
(26-29). Representative fluorescence histograms of infected red blood
cells in the presence and absence of antimalarial compounds are
presented in Fig. 6. It is clear that only atovaquone
caused the depolarization of plasmodial mitochondria and dissipated the dye accumulation as observed by a change in flow cytometric profile, while the other two antimalarials chloroquine (Fig. 6B) and
tetracycline (Fig. 6C) had no effect on membrane potential.
In order to ascertain that atovaquone specifically affects malarial
mitochondria, we also tested the effect of atovaquone on mouse
lymphocyte mitochondria. No effect of atovaquone was detected on
membrane potential of lymphocyte mitochondria (Fig. 6D). It
is known that atovaquone does not inhibit mammalian mitochondrial
bc1 complex (24). However, in lymphocytes other
respiratory inhibitors also fail to collapse
m (see
"Discussion"). Concentration dependence of the effect of the
antimalarials tested on P. yoelii
m is presented in Fig. 7A. Atovaquone had an
EC50 of 2.6 × 10
7 M, with
the maximum inhibition observed at ~2 × 10
6
M. Neither chloroquine nor tetracycline had any effect on
malarial
m even at millimolar concentrations. We also
tested the effect of atovaquone on the rate of O2
consumption by parasitized red blood cells as shown in Fig.
7B. Atovaquone inhibited P. yoelii respiration at
an EC50 of 7 × 10
8 M. At
2 × 10
5 M and above, atovaquone
inhibited 90% of the O2 consumption by infected red blood
cells.
It has become apparent that mitochondria are quite heterogeneous with regard to the various functions they serve in different cell types; while ATP synthesis may be a common critical function for mitochondria in most eukaryotic cells, additional crucial functions may be determined by the differentiation status of the cell or the organism in which these organelles reside. Our knowledge of mitochondrial physiology, however, is based largely upon classical studies done with easily accessible systems derived from tissues such as the liver and the heart (30, 31). To broaden our view it is necessary to study mitochondrial physiology in various cell types and organisms.
In this report we describe a flow cytometric assay employing a cationic
lipophilic fluorescent probe to monitor m in a rodent
malarial parasite. Fluorescence-based demonstration of mitochondrial
membrane potential has been used for many years with fluorochromes such
as rhodamine 123 (32-35). Most of these studies, however, involved
fluorescence microscopy and employed micromolar concentration of the
probe. Because the rhodamine dyes have adverse effects on mitochondrial
respiration (36, 37) and do not appear to be strictly dependent on
m for intramitochondrial
accumulation,2 their use in careful studies on
mitochondrial physiology is problematic. A cationic lipophilic
fluorochrome, DiOC6(3) originally used at high
concentrations to monitor intracellular membranes (39, 40), appears to
be a better alternative to assay
m when used at low
concentrations. We empirically determined the optimal concentration of
the probe for
m assay in intact malarial parasites
as the concentration at which probe accumulation was essentially
dependent on membrane potential and maximally dissipated by the
protonophore CCCP. When used at the optimal concentration (2 nM), the probe did not dissipate significantly from the
intact parasitized erythrocytes for up to 2 h. We also determined
the effect of the probe on respiration by intact parasites, observing
that at 10 nM and higher concentrations
DiOC6(3) significantly affected respiration rate. We
suggest that such parameters need to be established prior to
utilization of this fluorochrome for studying mitochondrial physiology.
The concentration of DiOC6(3) used for studying
mitochondrial membrane potential in pre-apoptotic lymphocytes (40 nM) is 20-fold higher than what we have used in assaying
m in malarial parasites (41). This concentration completely inhibited mitochondrial respiration in lymphocytes and the
fluorescence level was highly quenched.2 The optimal
concentration for the determination of
m in lymphocytes is
0.2 nM.2
Using this assay we determined the effects of various conventional
mitochondrial inhibitors on m of intact malarial parasites and compared them for their effects on respiration. Rotenone
did not affect either
m or respiration, consistent with
the previous studies that failed to observe NADH-ubiquinone oxidoreductase (site I) of the respiratory chain in malarial parasites (17). Two inhibitors of ubiquinol-cytochrome c
oxidoreductase (site II), myxothiazole and antimycin, as well as
cyanide (a cytochrome c oxidase and site III inhibitor)
collapsed
m and inhibited respiration. These components
of the respiratory chain have been observed in malarial parasites. It
is interesting to note that the inhibition of the respiratory chain
seems to result in rapid dissipation of
m. This is in
sharp contrast to what we observed in lymphocytes.2 For
technical reasons, it was not possible to determine the effects of
electron transport inhibitors on
m and respiration at the
same inhibitor/cell ratios. It is possible that higher inhibitor/cell
ratios are necessary to collapse
m than those required
for respiration inhibition. Nevertheless, our results suggest that
m in the parasite cannot be supported by ATP synthase,
which is in agreement with the notion that erythrocytic stages of
malarial parasite lack the machinery for oxidative phosphorylation. A
puzzling observation was the lack of respiration stimulation following
CCCP treatment. This is contrary to the general observation of
increased oxygen consumption in mitochondria with collapsed membrane
potential. This suggests that the electron transport in the malarial
parasite is not controlled by the mitochondrial
H+.
Atovaquone, a hydroxynaphthoquinone, is a drug initially developed as
an antimalarial but now known to be also effective against several
other eukaryotic microbial parasites such as toxoplasma and
pneumocystis (42). Using cholate-lysed mitochondria from Plasmodium falciparum and P. yoelii, and
externally provided heterologous (horse) cytochrome c at 100 micromolar concentration, Fry and Pudney (24) observed that atovaquone
inhibited cytochrome c reductase activity. Based upon these
results, it was concluded that atovaquone acts at the cytochrome
bc1 complex of the malarial respiratory chain.
Unique structural features of the parasite cytochrome b a
were speculated as being responsible for the therapeutic value of some
of the hydroxynaphthoquinones as antimalarials (16). Because malarial
mitochondria do not seem to contribute much to the ATP pool, it has
long been suggested that the main purpose for these organelles was to
dispose of electrons generated by dihydroorotate dehydrogenase (7-9),
an essential enzyme in pyrimidine biosynthesis. Since the parasites are
unable to salvage pyrimidines, inhibition of dihydroorotate
dehydrogenase has been suggested as the reason for antimalarial
activity of compounds such as atovaquone (43). Our observation that
atovaquone rapidly dissipates m provides another possible
mechanism for the antimalarial activity of this drug. It is not
entirely clear whether the collapse of
m is a direct
result of respiration inhibiton by atovaquone.
Recent studies on programmed cell death (PCD) pathways in metazoan
organisms have implicated mitochondria to be involved at earlier stages
of the PCD cascade (41, 44-49). PCD has also been observed in
unicellular organisms such as Trypanosoma (50, 51), Dictyostelium (52), and Tetrahymena (53) although
underlying mechanisms are not known at present. We suggest that the
inhibition of mitochondrial electron transport chain in malarial
parasites also acts to initiate a cascade of events similar to PCD. It
would be important to investigate the possibility that the PCD-like response is elicited by collapsed m in malarial
parasites. Nevertheless, maintenance of
m is likely to be
critical for all stages of the malaria parasite's life cycle, and
drugs that affect
m are likely to act on all such stages. In accordance with this proposal atovaquone has been shown also to act
against liver (54, 55) and insect stages of malarial parasite (38).
Further investigations on parasite mitochondrial physiology will be
illuminating in our attempts to understand mechanisms of drug action
and the development of newer strategies for chemotherapy of
malaria.
We thank Glaxo Wellcome for providing the antimalarial atovaquone and Joanne Morrisey for technical assistance.