relationship between the host cell mitochondria and the parasitophorous vacuole in cells infected...

I Eukuryor. Miudid. , S I ( I ). 2004 pp. 8 1-87 D 2004 hy the Society of Prnloi.t~nlngisls

Relationship between the Host Cell Mitochondria and the Parasitophorous Vacuole in Cells Infected with Encephalitozoon Microsporidia

MARY SCANLON,u GORDON J. LEITCH," GOVINDA S. VISVESVARAh and ANDREW P. SHAW "Department of Physiology, Morehouse School of Medicine, Atlanta, Georgia 3031 0, USA, and

"Division of Parasitic Diseases, National Center for Infectious Diseases, Centers for Disease Control, Atlanta. Georgiu 30341. USA

ABSTRACT. Encephalitozoon microsporidia proliferate and differentiate within a parasitophorous vacuole. Using the fluorescent probe, calcein, and the mitochondrial probe, MitoTracker-CMXRos, a vital method was developed that confirmed ultrastructural reports that the host cell mitochondria frequently lie in immediate proximity to the parasitophorous vacuole. Morphometry failed to demonstrate any infection-induced increase in host cell mitochondria as there was no correlation between the mitochondrial volume and the extent of infection as judged by the parasitophorous vacuole volume. The total ATP concentration of infected cells did not differ from that of uninfected cells in spite of the increased metabolic demands of the infection. Treatment with M albendazole, more than ten times the antiparasitic IC,,, dose, and demecolcine had no subjective effect on the proximity of mitochondria to the parasitophorous vacuole membrane when studied by either transmission electron microscopy or by confocal microscopy even though these drug concentrations affected microtubule structure. Thus, once the association between mitochondria and the parasitophorous vacuole has been established, host cell microtubule integrity is probably not required for its maintenance. It is unlikely that the antimicrosporidial action of albendazole involves physically uncoupling developing parasite stages from host cell organelle metabolic support. Key Words. Albendazole, Encephalitozoon, demecolcine, meront, microsporidia, mitochondria, parasitophorous vacuole, sporont,

spore.

NCEPHALITOZOON microsporidia were initially believed E to be protozoan parasites but now are recognized as fungi (Katinka et al. 2001; Van de Peer, Ben, and Meyer 2000). Mi- crosporidia proliferate and differentiate within a parasitophorous vacuole (PV) in the host cell (Didier, Snowden, and Shad- duck 1998). The three species of Encephalitozoon, Encephali- tozoon cuniculi, Encephalitozoon hellem, and Encephalitozoon intestinalis, are of clinical interest because of the extensive morbidity and mortality caused by infections in immunocom- promised patients (Garcia 2001; Weber et al. 1994). They are of biomedical interest because, in spite of a huge parasite bur- den, infection is sustained for several days without inducing apoptosis or necrosis of the host cell. We have previously reported on changes induced within the host cell by the parasite, including cytoskeletal rearrangement (Leitch et al. 1999) and inhibition of host cell cycling (Scanlon et al. 2000). Both of these effects are presumably necessary to create favorable en- vironmental conditions for the growth and differentiation of the parasite.

Ultras~ructural observations suggest that reorientation of host cell mitochondria to the membrane of the PV may represent another change induced within the host cell that is beneficial to the parasite (Canning and Hollister 1992; Shadduck and Pakes 1971). There is evidence with some microsporidial infections that host cell oxidative metabolism is significantly increased, particularly early in the infection, placing an increased demand on the mitochondria to support parasite proliferation, parasite differentiation, and host cell repair (Weidner et al. 1999). The realignment of mitochondria would optimize energy transfer to the developing stages of the parasite, which themselves lack mitochondria. Reorganization of host cell mitochondria and endoplasmic reticulum around PVs has been reported in infections with both protozoan parasites and bacteria (Sinai, Webster, and Joiner 1997), suggesting a common technique for providing metabolic support from the host cell to invading microorganisms while they develop within a protected intracellular niche.

Therefore, in order to test the hypothesis that microsporidial infection induces a realignment of host mitochondria to areas near the parasite or near the PV membrane in order to provide energy sources for the developing stages, we developed an in bitu technique using fluorescent dyes and confocal microscopy. MitoTracker-CMXRos (MitoTracker), a red fluorescent probe,

Corresponding Author: M. Scanlon-Telephone number: (404) 752- 1683; FAX number: (404) 752-1045; E-mail: [email protected]

stained the host cell mitochondria; calcein, a green fluorescent probe, outlined the developing stages and mature spores within the PV since they do not load with calcein (Leitch et al. 1997). Cells were imaged using confocal microscopy. This method has the advantage of ensuring that the cells under study are viable since dead cells do not cleave calcein/AM, the permeable pre- cursor to calcein. This in situ method, relying upon fluorescent probes and confocal microscopy, will test the hypothesis that mitochondria realign themselves in Encephalitozoon-infected cells. The importance of microtubules in the realignment of mitochondria in infected cells will also be examined.

MATERIALS AND METHODS

Cell and parasite culture. African green monkey kidney (E6) cells were infected with Encephalitozoon hellem, Enceph- alitozoon intestinalis or Encephalitozoon cuniculi as previously described (Visvesvara et al. 1999). Infected and uninfected cells were maintained in Dulbecco's modified Eagle's medium sup- plemented with 10% heat-inactivated fetal bovine serum, 10 pg/ ml gentamicin, 16 pg/rnl tylosin and 0.5 pg/ml amphotericin B at 37 "C in a 5% CO, incubator.

Electron microscopy. Electron microscopy was performed as previously described (Visvesvara et al. 199 I) .

Visualization of host cell mitochondria. Infected and uninfected E6 cells were plated on No. 1 glass coverslips that had been silicon-glued to the bottom of 35-mm (diam.) tissue culture dishes whose bottoms had been removed to accommodate the glass coverslips. The cells were incubated overnight and then treated with M verapamil for 45 min at 37 "C in a 5% CO, incubator. Pre-incubation with verapamil enhanced the retention of fluorochromes within the host cell by inhibiting P- glycoprotein on the surface of the E6 cells (Leitch et al. 2001). Cells were then incubated with media containing M verapamil, 150 nM MitoTracker-CMXRos (MitoTracker) (Molec- ular Probes, Eugene, OR), and 1-5 pM calcein/AM (Molecular Probes, Eugene, OR) for 15 min. CalceinIAM is converted to calcein, a fluorescent probe that distributes within the cytoplasm and nucleus of the host cell as well as in the parasite-free volume of the PV but not within the parasite (Leitch et al 1997). The parasites are thus negatively stained and appear as dark shapes against a green background. After washing the cells to remove extracellular dye, the cells were maintained at 37 "C in a HEPES-buffered solution (20 mM HEPES, 135 mM NaCl, 5 mM KCl, 5 mM NaHCO,, 1.2 mM KH,PO,, 1.2 mM CaCI,, 1.2 mM MgSO,, 5.5 mM glucose, 1 mg/ml bovine serum al-

81

82 J. EUKARYOT. MICROBIOL., VOL. 51, NO. 1, JANUARY-FEBRUARY 2004

Table 1. Comparison of volume measurements in Encephalitozoon hellem-infected E6 cells treated with DMSO or 10 " M Albendazole.

Mean volume measurements 2 S.E. in E. hellem-infected cells

DMSO carrier 10." M Albendazole control (n = 36) (n = 21)

Cellular volume, pm' Nuclear volume, pm' PV volume, pm' Mitochondrial volume, p,m' % Cytopla\m occupied by mitochondria

3,436 2 293 459 ? 36

1,241 ? 180 199 2 18 16.0 ? 2.4

4,545 t 537 579 2 64

1,603 * 208 282* -t 31 14.2 i- 1.9

* Volume in albendazole-treated, E. hellem-infected cells significantly different from volume of infected cells treated with DMSO, f-value, -2.31; p < 0.05.

bumin, pH 7.4) (Leitch et al. 1997) and visualized by confocal microscopy.

In some cases, after the cells had plated, they were treated for 24 h with albendazole M) or demecolcine (lo-' M), inhibitors of microtubules (both agents were administered in DMSO). Carrier control cells were treated with the same amount of DMSO (0.01 %). Albendazole inhibits the polymer- ization of microtubules in vitro and slows assembly in vivo (Solana et al. 1998). Demecolcine depolymerizes microtubules in living cells (Hagiwara and Takata 2002).

Measurement of cellular volumes. MitoTracker- and calcein-loaded, E. hellem-infected E6 cells were optically sec- tioned at 0.5-pm intervals throughout the total volume of the cell using confocal microscopy. Background fluorescence was removed from the sections by thresholding. Volumes were measured using 1mageSpace software (Molecular Dynamics, Sun- nyvale, CA) run on a Silicon Graphics (Mountain View, CA) workstation. Cellular, nuclear and PV volumes were measured by manual delimitation of the perimeters of the cell, the nucleus and the PV, respectively, in the slice of the cell that was 0.5 pm removed from the glass coverslip at the bottom of the dish. The software then counted, within the perimeter, the number of pixels containing calcein in the slice. This process was repeated with 0.5-pm sections throughout the cell. Pixels in all slices of the cell were summed to arrive at a value for each volume. Mitochondrial volume was measured in a similar manner with the exception that pixels containing MitoTracker, rather than calcein. were measured in each slice and summed throughout the entire cellular volume. Cytoplasmic volume was obtained by subtracting the nuclear and PV volumes from the cellular volume. Measurements were performed on carrier control (DMS0)-treated and lo-" M albendazole-treated cells infected with E. hellem.

The values for cytoplasmic volume and for the percentage of the cytoplasm occupied by mitochondria were calculated for each cell and then mean values were determined (Table 1). Intuitively, one would expect to generate the same mean values whether one used data from individual cells or one divided the mean mitochondria1 volume by the mean cytoplasmic volume. However, this is not the case. When means are generated from individual data points exhibiting large variation, not all values are equally weighted. This explains the discrepancy between the percentages of cytoplasm occupied by mitochondria in Ta- ble I and those numbers obtained by dividing the mean mitochondrial volume in Table 1 by the cytoplasmic volume that one generates from the means for cellular, nuclear and PV volumes in Table 1.

Immunocytochemistry. E. hellem-infected E6 cells were plated in 8-well chamber-slides and treated overnight with DMSO, albendazole (10 or lo-' M) or demecolcine (lo-' M).

They were fixed and permeabilized with ice-cold methanol for 10 minutes on ice. Following several washes with Tris-buffered saline, the cells were incubated with mouse anti-a tubulin (Sig- ma, St. Louis, MO) (1/500 dilution) for 1 hr at 37 "C. The cells were washed with Tris-buffered saline three times and then incubated with biotinylated goat anti-mouse 1gG (Jackson ImmunoResearch, West Grove, PA) (1/300 dilution) for I h at 37 "C. The cells were washed again and then incubated with streptavidin Oregon Green 488 (Molecular Probes, Eugene, OR) ( 1/300 dilution) for 45 min at room temperature. The scaf- folding of the chamber-slide was removed and the slide mount- ed with N-propyl gallate- and glycerol-containing phosphate- buffered saline. The cells were viewed with confocal micros- COPY.

Measurement of ATP. Uninfected and infected cells plated in 75-mm flasks were rinsed with phosphate-buffered saline and were lysed in a buffer containing 0.1 M KH2P04, 0.2% Triton X-100, 1 mM dithiothreitol, 100 pM leupeptin, 1 pg/ml pep- statin A, 1 mM benzamidine, and 1 mM AEBSF at pH 7.8. The lysed cells were detached from the flask with a cell scraper. The cell lysates were transferred to a microcentrifuge tube and centrifuged for 2 min. The extracts were transferred to a fresh tube and either stored at -70 "C or used immediately. The ATP Determination Kit (Molecular Probes, Eugene, OR) was used to calculate the ATP concentration of cellular lysates, whose volumes were identical (Karamohamed et al. 2001; Ronner et al. 1999). This kit relies on ATP to drive the enzyme-substrate reaction of luciferase and luciferin to evolve light, the amount of which is linearly correlated with the amount of ATP present. The luminescences of three replicates of each sample were read on an LS 6500 Multipurpose Scintillation Counter (Beckman Coulter, Fullerton, CA) and converted to ATP concentrations by comparison to a standard curve of known ATP standards (Molecular Probes, Eugene, OR). Protein in the extracts was measured using the Bio-Rad DC protein assay kit (Bio-Rad, Hercules, CA) as previously described (Scanlon et al. 1999). The ATP values were standardized to the total protein of each sample.

Statistical analysis. In studies involving more than two groups, a one-way analysis of variance was used followed by a post-hoc Tukey's protected t-test. In studies involving only two groups, the Student's t-test was used to determine the significance of differences between mean values. Differences were considered significant when the p value was < 0.05.

RESULTS CalceidAM is a nun-fluorescent, membrane-permeable

probe that is converted to impermeable, fluorescent calcein free acid by intracellular esterases. The resulting green fluorescence is visible throughout the cell cytoplasm. In infected cells (Fig.

SCANLON ET AL.-HOST CELL MITOCHONDRIA IN MICROSPORIDIAL INFECTION 83

Fig. 1. (upper panel) Confocal images of (a) an uninfected E6 cell and (b and C ) Encephalitozoon hellem-infected E6 cells probed with calcein (green fluorescence) and MitoTracker (orange/yellow fluorescence). The host cell accumulates calcein but intracellular parasite stages do not; larger developing stages and smaller mature spores are visible by negative staining. Mitochondria appear yellow or orange and are more evenly distributed around the central nucleus (N) in the uninfected cell (a). In contrast, they accumulate around the parasitophorous vacuole (PV) in the infected cell (b). Rarely, dramatic elongation of a few mitochondria is observed in infected cells (c). Bar = 2 pm.

Confocal images of E6 cells infected with Encephalifozoon hellem and incubated with (a) DMSO (carrier control), (b) 10-* M albendazole, or (c) lo-’ M demecolcine for 24 h showing that microtubule disruption did not affect mitochondria1 proximity to the parasitophorous vacuole (PV). (a) Two uninfected cells (arrowheads) exhibit broader distributions of mitochondria than infected cells. (b) Two uninfected cells (arrowheads) stained with MitoTracker whose color appears orange rather than yellow due to weaker loading of calcein by these cells. (c) A chain of four developing stages (arrowhead) suggests that this cell is at an early stage of infection. The clustering of mitochondria around these developing stages suggests that redistribution of mitochondria to the vicinity of the parasite occurs very soon after infection. N. Nucleus; Bar = 2 pn.

Fig. 2. (lower panel)

Ib and Ic), the intracellular parasite stages appear dark and unstained because calcein cannot cross the parasite cell membrane. The smaller mature spores can be distinguished from the larger meronts and sporonts by this method. MitoTracker has a high affinity for mitochondria and emits a red fluorescence. When images of cells loaded with both calcein and MitoTracker are superimposed, the resultant color of the mitochondria can vary from orange to yellow dependent upon the level of calcein loading in the cell. When calcein loading is high (Fig. 2a, arrowheads). the color of the mitochondria tends toward yellow; when calcein loading is lower, the resultant color is more orange (Fig. 2b, arrowheads). When cells are loaded with

MitoTracker alone, the red color is the same in uninfected and infected cells (data not shown).

In uninfected, viable cells, the distribution of mitochondria is fairly uniform around the nucleus and throughout the cytoplasm (Fig. la), while in infected cells, there is a distinct ori- entation around the PV membrane (Fig. 1 b). In some cases, the mitochondria seem to be interspersed among the rnicrosporidia but they are not. These mitochondria are located between the PV and the plasma membrane of the host cell. Their small size in comparison with the thickness of the optical section makes them appear to be in the same plane as the parasites when in reality they are abutting the membrane of the PV.

84 J . EUKARYOT. MICROBIOL., VOL. 51, NO. I , JANUARY-FEBRUARY 2004

The reorientation of the mitochondria around the PV does not occur because of passive compression of the mitochondria as the PV grows but appears to be a more active process. This point is demonstrated by a cell at a very early stage of infection. Even at this early stage of infection, mitochondria have reori- ented to the vicinity of the developing stages (Fig. 2c, arrowhead).

Qualitatively similar observations of reorientation of mitochondria to the PV were made with E. cuniculi and E. intestin- ah-infected E6 cells (data not shown). Very rarely an infected cell was encountered in which host mitochondria exhibited ab- errant morphology. For example, a very long mitochondrion (or several mitochondria aligned perfectly in series) was seen to cross approximately half the length of the host cell (Fig. Ic). Similar, apparently elongated mitochondria were also occasionally observed in E. cuniculi and E. intestinalis-infected E6 cells. Thus, Encephalitozoon infection results in host cell mitochondrial reorientation around the PV and may occasionally result in significant structural changes within mitochondria.

The reorientation of the mitochondria in infected cells was not affected by agents that interfere with microtubule integrity, at least after infection has been established within host cells. While both confocal and transmission electron microscopy suggest that the mitochondria are most frequently seen in close apposition to PV membrane areas with associated meronts, there was considerable variation within and between infected cells. Mitochondria clustered around the PV in a DMSO carrier control preparation (Fig. 2a). Clustering was retained in infected cells after 24 h exposure to M albendazole or to lo-’ M demecolcine (Fig. 2b, 2c, respectively). The concentration of albendazole used in these studies was more than ten times the antimicrosporidial ICso.

Since the mitochondrial clustering was unaffected by microtubule disrupters, we examined the microtubule distribution in E. hellem-infected cells treated with DMSO, albendazole or demecolcine to ensure that albendazole and demecolcine were, in fact, causing microtubule disarray. Carrier control cells (Fig. 3a) exhibited abundant, bundled microtubules spanning throughout all regions of the cell except those near the nucleus and the PV. The microtubules of infected cells treated with albendazole at M (Fig. 3b) demonstrated slightly less bundling and were more likely to lie at angles to each other in a cross-hatched pattern than carrier control cells. These effects were exacerbated in infected cells after exposure to M albendazole (Fig. 3c) or lo-’ M demecolcine (Fig. 3d). At lo-’ M demecolcine (Fig. 3d), depolymerization of tubulin was observed as evidenced by the granular material in parts of the infected cells. Thus, at concentrations of albendazole and demecolcine that disrupted microtubule architecture, the rearrangement of mitochondria to the PV in infected cells was not affected. We have demonstrated this point not only with im- munochemical techniques but also with transmission electron microscopy. Transmission electron micrographs of infected cells illustrated the close apposition of the mitochondria to the membrane of the PV in the presence of M albendazole- treated (Fig. 4a) and lo-’ M demecolcine-treated (Fig. 4b).

The mitochondrial volume was compared in uninfected and infected cells to assess whether infection increased or decreased the number or volume of mitochondria in a cell. In a represen- tative experiment, the mitochondrial volume expressed as a percentage of the total cytoplasmic volume was 15.60 & 0.02% in uninfected cells and 14.68 & 0.04% in E. hellem-infected cells. This difference was not statistically significant.

Because albendazole is the drug of choice for the treatment of Encephalitozoon microsporidiosis (Conteas et al. 2000), further morphometric studies were conducted using this agent. To

determine if either infection per se or 24 h of albendazole treatment affected the host cell microtubule organizing center (MTOC), the depth of the MTOC was measured. We chose to count the number of 0S+m optical slices in which the MTOC was visible as an estimate of MTOC depth rather than measure MTOC area or volume. We did so because of the subjectivity associated with the selection of the background setting and/or the determination of where the MTOC ended and the microtubule bundles began. The mean estimates of MTOC depth (+ S.E.) in the various treatment groups were as follows: uninfected, DMSO carrier control, 1.6 ? 0.2 p,m; uninfected, M albendazole-treated, 1.6 2 0.1 pm; E. hellem-infected, DMSO carrier control, 1.6 2 0.1 p,m; E. hellem-infected, 1 O+ M albendazole-treated. 1.7 & 0.2 pm. No statistically significant differences were observed in MOTC depth as a result of infection or of albendazole treatment. Thus, at concentrations of albendazole that cause disorganization of microtubules (Fig. 3c), the size of the MTOC was unaffected.

To further determine the effects of albendazole on morphological parameters in infected viable cells, morphometric analyses were performed on E. hellem-infected E6 cells treated for 24 h with M albendazole or DMSO (camer control). While cell volumes, nuclear volumes, and PV volumes were larger in albendazole-treated cells, these increased volumes were not statistically significant (Table 1). The mean mitochondrial volume of infected cells was significantly increased by albendazole treatment. However, when the mitochondrial volume was nor- malized to the cytoplasmic volume of each cell, there was no significant difference between the groups. Thus, morphometry failed to show any statistically significant effect of albendazole on the relative volume of the cytoplasm occupied by the mitochondria.

There was considerable variation in all measured volumes, in part because of the cell-to-cell variance in the parasite load as suggested by the broad range of PV volumes, 43 to 4,890 pn3. The range of PV volumes reflects a continuum from early- to late-stage infection. To determine if variability in host mitochondrial volume could be accounted for or correlated with parasite load, regression analyses were performed using PV volume as the independent variable and mitochondrial volume as the dependent variable. The assumption made in this analysis is that PV volume correlates with parasite load. Two separate experiments were analyzed yielding non-significant correlation coefficients (r) for linear regression equations of 0.016 and 0.095. Thus, the host cell mitochondrial volume did not cor- relate with the PV volume or parasite load.

To determine whether a biochemical change in energy content accompanied mitochondrial reorientation in infected cells, we compared cellular concentrations of ATP in uninfected and E. hellem-infected cells. The ATP content of non-confluent uninfected cells was 46.8 2 18.9 nM/p,g protein. This value was not statistically different from the ATP content of infected cell cultures, 43.4 & 10.5 nM/pg protein, in which at least 50% of the cells in the population were infected with E. hellem.

DISCUSSION It is not unusual for microorganisms occupying an intracel-

lular niche within a vacuole to restructure many of the host cell organelles around that vacuole (Sinai, Webster, and Joiner 1997). This organelle restructuring may only be limited to one species in a genus, such as in Chlamydia where it only occurs in Chlamydia psittaci (Matsumoto et al. 1991). One of the most thoroughly studied parasites in this regard is Toxoplasma gondii. With this species there is restructuring and attachment of host cell mitochondria and endoplasmic reticulum to the PV, which, after having been established, is retained when host cell

SCANLON ET AL-HOST CELL MITOCHONDRIA IN MICROSPORIDIAL INFECTION 85

Fig. 3. Imrnunocytochemical images of E6 cells infected with Encephalitozoon hellem, stained for tubulin and visualized with confocal microscopy. Microtubule organization is illustrated in (a) an infected, carrier control-treated E6 cell and infected cells treated for 24 h with (b)

M albendazole, and (d) lo-' M demecolcine. Compared to (a) DMSO-treated cells, there was less microtubule bundling following treatment with (b) M albendazole. The microtubules in cells treated with (b) lo-' M albendazole were a150 more likely to lie at an angle to each other than in (a) DMSO-treated cells. These effects were exacerbated by (c) 10-" M albendazole or by (d) M demecolcine. Addition of (d) lo-' M demecolcine also produced microtubule depolymerization, evidenced by the granular material in the infected cell. Nu, nucleus; PV, parasitophorous vacuole. Bar = 2 pm.

M albendazole, (c)

microtubules are disrupted (Sinai, Webster, and Joiner 1997). The parasite-derived linker protein involved in this attachment has been identified (Sinai and Joiner 2001).

The PV membrane, while providing protective advantages to an intracellular pathogen, may act as a permeability barrier to host nutrients. In the case of microorganisms, such as Chla- mydia, which are dependent on the host for all high energy nutrients, the replicating forms are found in close proximity to the inclusion membrane. Several chlamydia1 proteins have been identified that insert into the membrane of the inclusion body

and serve to regulate traffic across the membrane into the inclusion body (Fields and Hackstadt 2002). The proliferative meront stages of Encephalitozoon microsporidia also abut the PV membrane. However, in the case of Encephalitozoon, as with T. gondii (Schwab, Beckers, and Joiner 1994), the PV membrane is highly permeable and probably does not pose a significant barrier to solute acquisition by the parasite (bitch et al. 1995). The small genome of the Encephalitozoon microsporidia with the apparent loss of many genes encoding proteins in biosynthetic pathways (Katinka et al. 2001) and the lack of

86 J. EUKARYOT. MICROBIOL., VOL. 51, NO. 1, JANUARY-FEBRUARY 2004

Fig. 4. Transmission electron micrographs of E6 cells infected with Encephalirozoon hellem and incubated for 24 h with (a) M albendazole or (b) 10.’ M demecolcine. In both cases, mitochondria (M) are seen in close proximity to the membrane of the parasitophorous vacuole (PV) (arrowheads). S, sporont; Sp, mature spore. Bar = 200 nm.

parasitic mitochondria indicate that this parasite is highly dependent on host metabolism. The identification of four genes for ADP/ATP carrier proteins in the genome of E. cuniculi, that are homologous to ADP/ATP translocases in Rickettsia and Chlumydiu, suggests that microsporidia are able to import host cell ATP (Katinka et al. 2001). It would not be surprising if the PV contents acted as an energy sink for such solutes as ATP. This assumption has been supported morphologically by electron micrographic images illustrating host cell mitochondria closely abutting the PV (Canning and Hollister 1992; Shadduck and Pakes I97 I ). However, drawing physiological and biochemical conclusions from ultrastructural studies is hazardous; interpretations of electron micrographs are subject to sampling bias as well as to errors caused by fixation and embedding artifacts.

Our results, obtained from confocal microscopic visualization of living, infected cells stained with calcein and MitoTracker, confirm the ultrastructural reports of an association between a population of host cell mitochondria and the Encephalitozoon PV. A significant number of mitochondria in E. hellem-infected cells localize to the membrane of the PV, often very close to the developing stages of the parasite. As with T. gondii, the maintenance of this realignment of the mitochondria is not in- hibited by microtubule inhibitors, such as demecolcine or al- benddzole. In the case of albendazole, we used a concentration of the drug that was more than tenfold the antimicrosporidial ICs,, and it was still without detectable effect on the mainte-

nance of the realignment of the mitochondria. This concentration was, however, high enough to cause disorganization of microtubules in infected E6 cells. Transmission electron microscopic observations demonstrate that the host cell mitochondria often are within 10 nm of the PV membrane. Thus, Encepha- lituzoon microsporidia may dock host cell organelles on the PV membrane. Although microtubule integrity may be necessary for the initial organelle reorganization, it is not necessary to sustain the association with the PV. The biological significance of the grossly elongated mitochondria, or possibly several, perfectly aligned mitochondria in series, occasionally seen in infected cells has yet to be determined.

Whatever role host cell mitochondria play in the proliferation and differentiation of Encephalitozoon microsporidia, our results suggest that it does not appear to require a change in mitochondrial number and/or mass. We also compared the effect of albendazole on various cellular volumes. The mean cellular, nuclear, and PV volumes were statistically indistinguish- able between carrier control and albendazole-treated, E. hellem- infected cells, as was the mitochondrial volume but only when expressed as a percentage of the cytoplasmic volume.

It has been suggested that microsporidia harvest energy from the host cell, perhaps by a method similar to the ATP/ADP translocases seen with other parasites (Weidner et al. 1999). In the case of the Encephalitozoon microsporidia the realignment of mitochondria to the PV would provide those developing stages of the parasite abutting the PV membrane with an adjacent energy source for their growth and differentiation. Interestingly, we did not detect a difference in ATP content between uninfected and E. hellem-infected cells suggesting that cellular levels of ATP are sufficient for the growth and development of the parasite. We also found that there was no correlation between the mitochondrial volume and the PV volume measured over a range of PV volumes from 43 to 4,890 p,mR indicating that an increase in parasite load was not accompanied by an increase in host mitochondria. Thus, the existing host cell mitochondria appear capable of meeting the energy requirements of an increasing parasitic load while maintaining cellular ATP levels. This situation may be ameliorated by the limited use of ATP by the host cell. We have previously shown that the cell cycle of infected cells is stopped at several checkpoints (Scan- Ion et al. 2000), and this would be expected to reduce the host cell energy demands during infection. In addition, as the parasite load rises with time, most of the increase is due to an increased number of mature spores. Mature spores would be expected to impose less of an energy drain than the proliferating and differentiating stages. Thus, the host cell may be able to shunt energy, normally used by itself, to the developing stages without having to ramp up ATP production. Alternatively, the rate of ATP production may be increased in infected cells; however, if so, the rate of ATP usage must be proportionally increased to keep the ATP content similar between infected and uninfected cells.

The studies described here exhibit parallels with studies conducted on T. gondii. Sinai, Webster, and Joiner ( I 997) have described mitochondrial association with T. gondii vacuoles in infected host cells. The association can be disrupted by treatment with nocodazole prior to infection suggesting that microtubules are involved in the reorientation of the mitochondria. Sinai and Joiner (2001) have identified parasitic rhoptry proteins as potential candidates for inducing and maintaining the association between mitochondria and parasite. We do not know if there are similar microsporidial proteins involved in promoting and/or maintaining the realignment of the host cell mitochondria to the PV. Once established, the extremely close association between the mitochondria and the PV is unaffected

SCANLON ET AL.-HOST CELL MITOCHONDRIA IN MICROSPORIDIAL INFECTION 87

by microtubule inhibitors suggesting that the relationship between mitochondria and the PV membrane is maintained by (a) linker protein(s).

ACKNOWLEDGMENT Supported in part by U.S. Public Health Service grant

RR03034.

LITERATURE CITED Canning, E. U. & Hollister, W. S. 1992. Human infections with mi-

crosporidia. Rev. Med. Microhid., 3:35-42. Conteas, C. N., Berlin, 0. G., Ash, L. R. & Pruthi, J. S. 2000. Therapy

for hurnan gastrointestinal microsporidiosis. Am. J. Trop. Med. Hyg.,

Didier, E. S., Snowden, K. F. & Shadduck, J. A. 1998. Biology of microsporidian species infecting mammals. Adv. Purasitr,l., 40:283- 320.

Fields, K. A. & Hackstadt, T. 2002. The chlamydial inclusion: escape from the endocytic pathway. Annu. Rev. Cell Dev. Biol., 18:221-245.

Garcia, L. S. 200 I . Laboratory identification of the microsporidia. J. Clin. Microbiol., 40: 1892-1901.

Hagiwara, H. & Takata, K. 2002. Depolymerization of microtubules by colcemid induces the formation of elongated and centriole-nonasso- ciated striated rootlets in PtK(2) cells. Cell Tiss. Res.. 309:287-292.

Karamohumed, S. & Guidotti, G. 2001. Biolurninometric method for real-time detection of ATPase activity. Biorechniques, 3L420-425.

Katinka, M. D., Duprat, S., Cornillot, E., Metenier, G., Thomarat, F., Premier, G., Barbe. V., Peyretaillade, E., Brottier, P., Wincker, P., Delbac. F., Alaoui, H. E.. Peyret, P., Saurin, W., Gouy, M., Weis- senbach, J . & Vivares, C. P. 2001. Genome sequence and gene com- paction of the eukaryote parasite Encephulitozoon cuniculi, Nature, 414:45(-53.

Leitch, G. J., Scanlon. M., Shaw, A. P. & Visvesvara, G. S. 1999. Modification of the cytoskeleton in Encephalirozoon-infected cells. J. Eiihrvot. Microhiol.. 463363-37s.

Leitch, G. J.. Scanlon, M., Shaw, A. & Visvesvara, G. S. 2001. Role of P glycoprotein in the course and treatment of Encephalitozoon microsporidiosis. Antimicroh. Agents Chemotherup., 45:73-78.

Leitch, 0. J., Scanlon, M., Visvesvara, G. S. & Wallace, S. 1995. Cal- cium and hydrogen ion concentrations in the parasitophorous vacuole of epithelial cells infected with the microsporidial, Encephalitozoon hellem. J. E u k u r y ~ t . Microbiol.. 421445-45 1.

Leitch, G. J., Scanlon, M., Shaw, A., Visvesvara, G. S. & Wallace, S. 1997. LJse of a fluorescent probe to assess the activities of candidate agents against intracellular forms of Encephalitozoon microsporidia. Antimicroh. A,genr.s Chernother.. 41:337-344.

Matsumoto, A.. Bessho, H., Uehira, K. & Suda, T. 1991. Morphological

63: I2 I - 127.

studies of the association of mitochondria with chlamydial inclusions and the fusion of chlamydial inclusions. J. E/CCI)TJII Micros[.. (Ilbkyo), 40:356-363.

Ronner, P., Friel, E., Czerniawski, K. & Frankle. S. 1999. Lurninornetric assays of ATP, phosphocreatine, and creatine for estimation of free ADP and free AMP. And. Biochenz., 279208-2 16.

Scanlon, M., Shaw. A. P., Visvesvara. G. S. & Leitch. G. J. 2000. Infection by microsporidia disrupts the host cell cycle. J. Eiiktryt. Microbiol. 47525-53 I .

Scanlon, M., Leitch, G. J., Shaw. A. P.. Moura. H. & Visvesvara. G. S. 1999. Susceptibility to apoptosis is reduced in the microsporidia- infected host cell. J. Euhryor. Microhiol., 46:348-353.

Schwab, J. C., Beckers, C. J. M. & Joiner. K. A.. 1994. The parasitophorous vacuole membrane surrounding intracellular Toxoplctsmu gondii functions as a molecular sieve. Proc. Nut. A w d . Sci . USA. 91: 509-5 13.

Shadduck, J. A. & Pakes, S. P. I97 I . E n ~ ~ ~ / J h U ~ ~ f ~ l ~ f i f J i f ~ J . ~ i . s (nosemato- sis) and toxoplasmosis. Am. J. Pu/hol., 64:657-67 I .

Sinai, A. P. & Joiner, K. A. 200 I . The Toxo/7Iu.sn~n Konrlii protein ROP2 mediates host organelle association with the parasitophorous vacuole membrane. J. Cell Biol.. 154:95-108.

Sinai, A. P., Webster, P. & Joiner, K. A. 1997. Association of host cell mitochondria and endoplasmic reticulum with the Toxopltr.smrr gondii parasitophorous vacuole membrane-a high affinity interaction. J. Cell. Sci., 110:2117-2128.

Solana, H. D., Teruel, M. T., Najle, R., Lanusse. C. E. & Rodriguez, J . A. 1998. The antihclminthic albendazole affects in vivo thc dynamics and detyrosination-tyrosination cycle of rat brain microtubules. Actcr Ph-ysiol. Phar~nacol. Ther. Lut.. 48: I9Y-205.

lating molecular evidence that a group of amitochondriate and sus- pectedly primitive eukaryotes are just curious fungi. Genr, 246: 1-8.

Visvesvara, G. S., Moura, H., Leitch, G. J. & Schwartz, D. A. 1999. Culture and propagation of microsporidia. In: Wittner, M. & Weiss, L.M. (ed.), The Microsporidia and Microsporidiosis. ASM Press, Washington, D.C. p. 363-392.

Visvesvara, G. S., Leitch, G. J., Moura, H.. Wallace. S., Wehcr, R. & Bryan, R. T. 199 1. Culture, electron microscopy. and immunoblot studies on a microsporidian parasite isolated from thc urine of a pa- tient with AIDS. J. Prorozool., 38: 105s-I l l S.

Weber, R., Bryan, R. T., Schwartz, D. A. & Owen, R. L. 1994. Human microsporidial infections. Clin. Micmhiol. Rw. , 7:426-46 I .

Weidner, E., Findley, A. M., Dolgikh. V. & Sokolova, J. 1999. Micro- sporidian biochemistry and physiology. In: Wittner. M. & Weiss, L. M. (ed.), The Microsporidia and Microsporidiosis. ASM Press, Wash- ington, D.C. p 172-195.

Van de Peer, Y., Ben, A. & Meyer, A. 2000. Microsporidi

Received 11/06/02; ucceptecl 10//4/03

relationship between the host cell mitochondria and the parasitophorous vacuole in cells infected...

Documents