JOURNAL OF CLINICAL MICROBIOLOGY, May 1988, p. 965-9700095-1137/88/050965-06$02.00/0Copyright C 1988, American Society for Microbiology

Biochemical and Ultrastructural Study of Blastocystis hominisCHARLES H. ZIERDT,l* CALEB T. DONNOLLEY,' JACQUELINE MULLER,2

AND GEORGE CONSTANTOPOULOS3

Microbiology Service, Clinical Pathology Department, National Institutes of Health,' Center for Biologics Evaluation andResearch, Food and Drug Administration,2 and Developmental and Metabolic Neurology Branch, National Institute of

Neurological Communicative Disorders and Stroke,3 Bethesda, Maryland 20892

Received 3 November 1987/Accepted 24 January 1988

This study was prompted by the paradox of strong presence of mitochondria in an anaerobic protozoan,recently reclassified from the yeasts. Stemming from publication in 1911 to 1912, Blastocystis hominis has beengenerally accepted as a harmless intestinal yeast of humans, with short standardized textbook (parasitology)descriptions, even to the present day. Reports since 1967 have changed the classification of B. hominis fromyeast to protozoan (Sarcodina), and this has been followed by interest in B. hominis-caused disease, resultingin documentation of disease in humans and other primates. In this study of B. hominis, the basic ultrastructureof the mitochondria was shown by thin-section electron microscopy to be identical to that of an archetypicalmitochondrion. There were hundreds of them in large B. hominis cells (100 to 200 ,im in diameter).Mitochondria were confined to a peripheral ring of cytoplasm bounded by the outer cell membrane (there is nocell wall) and the membrane of the large, spherical, organelle-free central body that constitutes 75% of thecell's volume. Mitochondria tended to surround the cell's usual two to four nuclei. Rhodamine 123 stained themitochondria selectively, visualized by fluorescence microscopy. The cell was devoid of cytochromes. Additionof 0.1% cytochrome c to the growth medium increased utilization of glucose by 34% and that of lactate by 17%.Furthermore, it markedly increased the number of mitochondrion-filled cells. At higher concentrations,cytochrome c inhibited the growth of the cells. Despite the presence of large numbers of mitochondria, activitiesof the mitochondrial enzymes pyruvate dehydrogenase complex, a-ketoglutarate dehydrogenase complex,isocitrate dehydrogenase, glutamate dehydrogenase, and cytochrome c oxidase were absent. Thus, the functionof the mitochondria in B. hominis remains unknown. Considerable activities of aspartate aminotransferase andalanine aminotransferase were found. Aldolase activity was prominent. Pyruvate decarboxylase was present.Diaphorase and lactate dehydrogenase were detectable but in suspect quantities. Other missing enzymes were

gamma glutamyl transpeptidase, alkaline phosphatase (a lysosomal marker), and creatine kinase isoenzymes.

Blastocystis hominis is an intestinal protozoan parasiterestricted to humans and other primates. It is minimallyinvasive and has no cyst form and no life cycle outside of itsprimate host. Although Brumpt (3) is credited with itsclassification in 1912, Alexieff (1) created and published thegenus in 1911 as Blastocystis enterocola. It was described as

a yeast, and the yeast designation became firmly entrenchedin textbooks and in the minds of parasitologists. Sporadicpapers over the decades puzzled over the classification (2)but made no positive recommendation for change. Research-ers were particularly bemused by the organism's amazingmultiplicity of forms. Failure to find intracellular or tissue-invasive B. hominis, a life cycle, a cyst form, or intermediatehosts contributed to the continuing enigma. In 1967, over a

dozen physiological and morphological B. hominis charac-teristics were reported (9) that fit a protozoan classification,with no strong characteristic suggesting a yeast classifica-tion. Since then, other papers (10; C. H. Zierdt, Clin.Microbiol. Newsl. 5:57-59, 1983) corroborated and extendedthis work, and the organism is now recognized as an emerg-ing protozoan parasite of humans, causing intestinal diseasewith potentially disabling symptoms (Zierdt, Clin. Micro-biol. Newsl. 5:57-59, 1983). It is now classified with theamebae, in kingdom Protista, subkingdom Protozoa, phy-lum Sarcomastigophora, subphylum Sarcodina, superclass

* Corresponding author.

Rhizopoda, class Lobosea, subclass Gymnamoeba, orderAmoebida, new suborder Blastocystina.A strict anaerobe (7), B. hominis yet may include hun-

dreds of mitochondria in each cell. These mitochondria werereported to be free of cytochromes (7). The present report isof a multiple approach: the search for mitochondrial en-

zymes in B. hominis, an ultrastructural study of B. hominismitochondria, and increased uptake of carbohydrates in thepresence of added cytochrome c.

MATERIALS AND METHODS

B. hominis cultures. Strains 1, 2, 3, 4, and 5, obtained fromhuman infection, were rendered bacteria-free (axenic) withthe use of ampicillin, streptomycin, colistin, and amphoter-icin (9). The medium formula is 1 part beaten whole egg to3.6 parts Locke's solution (NaCl, 8 g; CaCl2, 0.2 g; KCI, 0.2g; MgCl2, 0.01 g; Na2HPO4, 2.0 g; NaHCO3, 0.4 g; KH2PO4,0.3 g; H20, 1,000 ml). After being filtered through gauze,medium (4 ml) was pipetted into screw-cap tubes (18 by 125mm) and coagulated in flowing steam for 10 min at a 300slant, followed by the addition of 4.5 ml of Locke's solutionand autoclaving at 15 lbs/in2 steam for 15 min. After additionof 2 ml of horse serum, the medium was reduced in anaero-bic jars for at least 2 days before use. B. hominis grew as a

glutinous sediment at the slant base, and transfers (one-thirdof sediment) were done at 3- and 4-day intervals, at whichtime the cell mass was approximately 5 x 108 B. hominiscells per tube.

965

Vol. 26, No. 5

966 ZIERDT ET AL.

Cytochrome c effect on glucose and lactate uptake. Cyto-chrome c from horse heart, type VI (C-7752, lot no. 76F-7045; Sigma Chemical Co., St. Louis, Mo.), was used. Freshstock (2%, wt/vol) was prepared daily in neutral (0.85%NaCi) saline and filtered through a 0.2-ptm membrane filter.Concentrations of 0.001, 0.01, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,and 1.0% (wt/vol) were added to the growth medium todetermine (i) toxicity to B. hominis, effect on (ii) numbers ofmitochondrial cells and (iii) cell growth, and (iv) effect onuptake of glucose and lactate. Population counting was donewith Neubauer hemacytometer chambers. Two counts weredone on each sample, after a 10-fold dilution. The percentageof mitochondrial cells was estimated by counting 500 cells.

Toxicity of cytochrome c to bacteria. One percent cyto-chrome c was added in nutrient agar pour plates, and thesewere inoculated with stock cultures of Klebsiella pneumo-niae, Listeria monocytogenes, Streptococcus pneumoniae,and Escherichia coli. These plates were incubated at 36°Cfor 48 h and observed for inhibitory effect.

Glucose was made up in Locke's solution as 50% (wt/vol)stock, and 50 ,ul was added to each culture tube, resulting ina ca. 400-mg/dl final concentration. Sodium lactate (60%syrup; L-1375, Sigma) was made up in Locke's solution at25% (vol/vol), and 0.25 ml ofthis stock solution was added toeach tube, resulting in about 17 meq/liter (range, 10.0 to 24.9meq/liter), depending on the individual experiment.

Cell homogenates of B. hominis and normal fibroblastswere examined for activities of pyruvate dehydrogenasecomplex, aoketoglutarate dehydrogenase complex, isocitratedehydrogenase, glutamate dehydrogenase, cytochrome coxidase, lactate dehydrogenase, aspartate aminotransferase,alanine aminotransferase, and other enzymes.

Control human fibroblast line IMR-90 was obtained fromthe Human Genetic Mutant Cell Repository, Coriell Institutefor Medical Research, Camden, N.J. The cells were grownin McCoy SA medium with 2 mmol of glutamine per liter,supplemented with 10% fetal calf serum, 100 U of mycosta-tin per ml, and 50 ptg of gentamicin per ml. The cells wereharvested at confluency by trypsinization and washed twicewith phosphate-buffered saline (137 mmol ofNaCI, 2.7 mmolof KCI, 8 mmol of Na2HPO4, and 1.5 mmol of KH2PO4 perliter, pH 7.3). Cell pellets were suspended in 0.2 to 0.5 ml ofa mixture of 40 ml of glycerol with 60 ml of 10 mmol Trishydrochloride buffer, pH 7, per liter subjected to threecycles of freeze-thawing, and briefly sonicated in a celldisruptor (model W140; Heat Systems-Ultrasonics, Plain-view, N.Y.), using the microprobe, output position 4, for 10s at 0°C.Enzyme assays. Pyruvate dehydrogenase complex and

a-ketoglutarate dehydrogenase complex were assayed radio-chemically as described previously (4). The activities ofdihydrolipoamide dehydrogenase, isocitrate dehydrogenase,glutamate dehydrogenase, cytochrome c oxidase, and lac-tate dehydrogenase were measured spectrophotometrically(4). To confirm the presence or absence of glutamate dehy-drogenase, aspartate aminotransferase, and alanine amino-transferase in B. hominis, specific radiochemical assayswere developed to enable us to identify the product of theenzymatic reaction. Conditions for these assays, as deter-mined by us, were optimal for the assays of the mammalianenzymes. Concentrations in the assay mixture of glutamate-dehydrogenase were as follows (in millimoles per liter): Trishydrochloride buffer as Tris, 50; ammonium acetate, 100;EDTA, 2.6; NADH, 0.3; at-[1-'4C]ketoglutarate, 0.27 (about200,000 cpm); and when indicated, ADP, 1. The pH of thereaction was 7.6 and the final volume was 200 ,u. Incubation

time varied from 15 to 60 min. Concentrations in the assaysystem of aspartate aminotransferase were as follows (inmillimoles per liter): Tris hydrochloride buffer (pH 7.8) as

Tris, 80; aspartate, 240; cx-[1-'4C]ketoglutarate, 1 (500,000cpm); and pyridoxal phosphate, 0.1. Concentrations in theassay system of alanine aminotransferase were as follows (inmillimoles per liter): Tris hydrochloride buffer (pH 7.3) as

Tris, 100; L-alanine, 500; a-[1-'4C]ketoglutarate, 1; andpyridoxal phosphate, 0.1. In both assays, the final volumewas 200 ,u and incubation time at 37°C varied from 5 to 30min. At the end of the incubation, the assay mixtures of allthree enzymes were acidified with 1 drop (20 pil) of 3 N HCl,diluted to 2.0 ml with distilled water, and chromatographedon small columns (6-mm inside diameter by 50 mm; in5.75-in. [14.6-cm] Pasteur pipettes of Dowex 50W hydrogenform, 8% cross-linked, 50-100-mesh ion-exchange resin, indistilled water). The unchanged radioactive substrate a-[1-'4C]ketoglutarate was eluted with two washings (5.0 mleach) of distilled water (including the void volume). Theradioactive product, [1-'4C]glutamate, was retained on theresin and was subsequently eluted with two 5.0-ml washingsof 2 M ammonia solution. Radioactivity was assessed bycounting samples of the fractions. The remainder of thefractions was lyophilized. Residues were dissolved in 0.5 Msodium acetate buffer (pH 5.0), and 100 ,ul was assayed forglutamic acid with glutamic decarboxylase. The reactionmixture, in tubes also containing plastic capsules with paperwicks (impregnated with 50 pl of 2.5 M NaOH for thecollection of the evolved C02), contained 30 ,umol of NaCIand 0.1 to 0.25 U of glutamic acid decarboxylase in a totalvolume of 200 ,ul. At the end of the incubation (30 to 120min), the reaction mixture was acidified with 2 drops of 3 MHCI, and C02 collection was completed by equilibration at37°C for 1 h. The radioactivity was assessed in an LS-230liquid scintillation system (Beckman Instruments, Inc., Ful-lerton, Calif.). Protein content was 'determined by themethod of Lowry et al. (6), using bovine crystalline serumalbumin as a standard.The activities of the enzymes alkaline phosphatase,

gamma glutamyl transpeptidase, aldolase, and creatine ki-nase isoenzymes were determined by the Clinical ChemistryService of the Clinical Center of this institute.

Chemicals. [1-'4C]pyruvic acid (sodium salt; 5 to 20mCi/mmol), ao-[1-'4C]ketoglutaric acid (sodium salt; 40 to 60mCi/mmol), and L-[1'4C]glutamic acid (49.6 mCi/mmol) werepurchased from New England Nuclear Corp., Boston, Mass.Nonradioactive sodium pyruvate, sodium a.-ketoglutarate,coenzyme A, thiamine PP1, NAD', NADH, nicotinamide,lipoamide, pyridoxal 5'-phosphate, L-aspartic acid, L-ala-nine, glutamic acid decarboxylase, protein (27 U/mg), andDowex 50W hydrogen form, 8% cross-liniked, 50-100-meshion-exchange resin were obtained from Sigma Chemical Co.

Electron microscopy. Cell sediments from 72-h B. hominiscultures were fixed with 4% glutaraldehyde for 2 h andtransferred to phosphate-buffered saline for storage at 4°C.The cells were treated with 1% aqueous tannic acid for 30min at 4°C and postfixed in Dalton chrome-osmium (1%OS04, 1% K2Cr207 in 0.8% NaCI) for 2 h at 4°C. Stepwiseethanol dehydration was followed by embedding in epoxyresin. Thin sections were stained by routine methods withuranyl acetate and lead citrate.Rhodamine 123 staining. Rhodamine 123 (Polysciences,

Inc., Warrington, Pa.), one of about six closely relatedfluorescing compounds derived from rhodamine B that stainsmitochondria in living cells brightly and specifically (5), withno toxicity to the cells, was added to the cultures at 0.01, 0.1,

J. CLIN. MICROBIOL.

BLASTOCYSTIS HOMINIS 967

TABLE 1. Effect of cytochrome c levels on B. hominis cellgrowth and percentage of mitochondrial cells

% (wt/vol) ClconpemlApproximate % ofcytochrome c Celi count per mi mitochrondrion-filled celis

0 6.6 x 107 200.001 6.2 x 107 200.01 6.4 x 107 400.05 6.5 x 107 600.1 6.1 x 107 800.2 4.4 x 107 900.3 1.4 x 107 950.4 6.0 x 106 950.5 1.9 x 106b 201.0 1.8 x 106b 20

a Inoculum, 2.0 x 106 B. hominis cells per ml. Counts were done after 3days of growth.

b Inhibited; no multiplication.

1.0, 10.0, and 100 ,ug/ml. Staining was checked by epifluo-rescence optics at 546 ,um on a Zeiss photomicroscope withZeiss Planapochromat 63 x où immersion objective. The filmused was Kodak Tri-X (ASA 400; Eastman Kodak Co.,Rochester, N.Y.). Wet mounts of the living cells wereviewed. Cultures were reincubated to determine toxicitylevels of rhodamine 123 to B. hominis.

RESULTS

In 3-day-old cultures, there occurred a fraction of the cellpopulation in which the peripheral cytoplasmic band waspacked with mitochondria, totally obscuring the centralbody. We called these mitochondrial cells. The rest of thecells were dominated by the central body and had relativelyfew mitochondria. The evidence supporting this was fromthin-section electron microscopy, examination of living cellsvia Nomarski interference contrast optics, and fluorescencemicroscopy after rhodamine 123 staining.

TABLE 3. B. hominis lactate utilization withand without cytochrome c

Lactate (meq/liter)

B. hominids + B. hominis +B.homlmcdium+Strain Expt medium + Mcdium +

lactatc (% lactatc + 0.1% lactatcof control cytochrome c (control)utilized) (% of control

utilized)1 1 15.5 (15) 11.6 (37) 18.3

2 12.3 (14) 10.9 (24) 14.33 24.4 (2) 20.9 (16) 24.94 16.7 (9) 14.3 (22) 18.35 20.2 (0) 15.0 (24) 19.76 19.3 (9) 17.9 (16) 21.27 11.9 (2) 8.6 (29) 12.18 6.6 (43) 4.2 (63) 11.59 17.1 (13) 15.6 (20) 19.610 10.1 (35) 8.6 (45) 15.5il 9.5 (10) 5.5 (48) 10.612 20.4 (9) 18.0 (19) 22.313 15.1 (0) 10.3 (31) 14.914 12.7 (14) 9.9 (33) 14.715 13.2 (11) 10.2 (32) 14.9

2 1 14.7 (20) 12.1 (34) 18.32 12.4 (13) 10.9 (24) 14.33 6.7 (33) 5.6 (44) 10.04 23.1 (7) 18.7 (25) 24.95 16.8 (8) 15.4 (16) 18.36 18.0 (9) 16.7 (15) 19.7

3 1 18.3 (14) 13.1 (38) 21.22 11.2 (7) 9.6 (21) 12.13 7.2 (37) 5.6 (51) 11.54 16.1 (18) 11.0 (44) 19.65 11.2 (28) 9.6 (38) 15.5

4 1 19.4 (13) 14.2 (36) 22.32 13.6 (10) 9.8 (35) 15.1

5 1 13.8 (7) 8.3 (44) 14.92 14.8 (3) 9.5 (38) 15.3

Total lactate 63.5 (13) 154.2 (30) 505.8utilized

TABLE 2. B. hominis glucose utiliZation withand without cytochrome c

Glucose (mg/dl)

B. hominids + B. hominis +Strain Expt medium + medium with Medium +medium + glucose + 0.1% glucose

glucose (% of cytochrome c (% (control)control utilized) of control utilized)

1 1 351 (17) 156 (63) 4212 366 (15) 188 (56) 4323 210 (38) 119 (65) 3394 186 (48) 142 (60) 3565 192 (45) 113 (67) 3476 242 (10) 197 (27) 269

2 1 331 (20) 122 (70) 4132 376 (18) 147 (68) 4573 336 (16) 273 (32) 4024 399 (6) 146 (65) 4235 413 (3) 155 (63) 4246 374 (4) 237 (39) 391

3 1 376 (14) 158 (62) 4212 382 (12) 172 (60) 4323 205 (40) 187 (45) 3394 172 (52) 130 (63) 3565 168 (52) 146 (58) 347

4 1 240 (48) 130 (72) 460Total glucose 5,319 (24) 2,918 (58) 7,029

utilized

The effect of cytochrome c additions to the cultures is seenin Table 1. There was no evident increase in growth, but thepercentage of mitochondrion-filled cells increased from 20 to40% at the 0.01% cytochrome concentration and to 95% atthe 0.3% level. Growth inhibition was strong at 0.4% cyto-chrome, and there was evident cessation of growth at 0.5%cytochrome. None of the bacterial species was inhibited by1% cytochrome.

B. hominis cultures used, on average, 34% more glucoseover a 3-day growth period when 0.1% cytochrome c wasincluded in the medium (Table 2), or 58% utilization withcytochrome and 24% utilization without cytochrome.When sodium lactate was added, there was an average

30% utilization when cytochrome c was present and 13%utilization in its absence (Table 3).Rhodamine 123 blocked division of B. hominis at 100

,ug/ml, was partially inhibitory at 10 ,ug/ml, and showed noinhibition at 1 p.g/ml or lower. Staining of B. hominismitochondria was bright at 0.1 ,ug/ml or higher, but dimin-ished at 0.01 ,ug/ml. Staining of organelles other than mito-chondria was not observed. Rhodamine 123 permitted accu-

rate visualization of distribution, numbers, and shape ofmitochondria in living B. hominis cells (Fig. 1).Photomicrographs from earlier work with B. hominis

mitochondria (Zierdt, Clin. Microbiol. Newsl. 5:57-59, 1983)were not clear in detailing the peripheral opening of the

VOL. 26, 1988

J. CLIN. MICROBIOL.968 ZIERDT ET AL.

FIG. 1. B. hominis. (Top) Rhodamine 123 stain showing fluores-cence of mitochondria. Inset shows "mitochondria" cell. Centralbody seen as darker sphere occupying center of cell. (Bottom) Largedark cell is degenerate giant cell, with membrane bound centralbody remaining. Magnification, x 1,750.

goblet-shaped cristae. There was critical conjecture that thenecks of the cristae opened to the cytoplasm. An additionalfixation with tannic acid resulted in the present photomicro-graphs (Fig. 2), showing cristae opening to the common

peripheral space between the outer and inner membranes.Thus, the ultrastructure of B. hominis mitochondria is typi-cal of mitochondria in general.

Negative staining with phosphotungstate did not revealthe pedunculated spheres lining the matrix surface of theinner membrane that incorporate ATPase.

Activities of the pyruvate dehydrogenase complex, a-

ketoglutarate dehydrogenase complex, isocitrate dehydroge-nase, glutamate dehydrogenase, and cytochrome c oxidase,enzymes commonly associated with mitochondrial metabo-lism, were absent from B. hominis (Table 4). Mixed experi-ments with fibroblast homogenates revealed the presence ofan endogenous inhibitor of isocitrate dehydrogenase in ho-mogenates of B. hominis but ruled out the existence of suchan inhibitor(s) of the other enzymes. Considerable activitiesof aspartate aminotransferase and alanine aminotransferasewere found in B. hominis, and the product, [1-14C]glutamate,was identified and characterized. Also present were diapho-rase, lactate dehydrogenase, and aldolase. Other enzymestested for, but with negative results, were gamma glutamyltranspeptidase, alkaline phosphatase (a lysosomal marker),and creatine kinase isoenzymes.A notable characteristic of B. hominis is accumulation of

free fat globules in the cytoplasm of older cells (Fig. 3). Theglobules increase in size as they coalesce with adjacentglobules. The final appearance before cell membrane ruptureis sometimes that of a single fat globule filling most of thespace enclosed by the membrane. Work toward elucidationof lipid synthesis in B. hominis is in progress.

DISCUSSION

Addition of 0.1% cytochrome c increased utilization ofglucose by 34% and that of lactate by 17%. The mechanismby which this occurs is not understood. It could be postu-lated that B. hominis mitochondria were derived from aero-bic bacteria which gradually lost their cytochromes andoxidative enzymes as the host cell became anaerobic. B.hominis strain 1 utilized 28% of supplied glucose (six exper-iments), while strain 2 utilized 45% (six experiments), strain3 utilized 24% (five experiments), and strain 4 utilized 24%(one experiment). Thus, strain 2 for this series of experi-ments utilized much more glucose. However, B. hominisstock strains may fluctuate over time in growth rate andmetabolic activity. The increased glucose utilization bystrain 2 may reflect a temporary rather than a permanentdifference from the other strains.The possibility exists that residual substances from the

commercial preparation of cytochrome c could inhibit B.hominis. Inhibitory chemicals residual to the preparationtechnique might be expected to inhibit bacteria also, but this

TABLE 4. Enzymatic activities in B. hominis

nmol min` mg of protein-îEnzyme

B. hominis Fibroblasts

Pyruvate dehydrogenase complex NDb ' 1.45a-Ketoglutarate dehydrogenase <0.004 0.91complex

Isocitrate dehydrogenase NADP+ ND 36.90Glutamate dehydrogenase NADH ND 2,554.00Cytochrome c oxidase ND 10.45Diaphorase (dihydro-lipoamide 14.9 58.60dehydrogenase)"

Lactate dehydrogenase 4.9 699.00Aspartate aminotransferase 37.0 0.72Alanine aminotransferase 41.0 0.30

Values are averages of at least two experiments.">ND, Nondetectable.'Activity of pyruvate decarboxylase, 578 pmol min- mg of protein'-., was

found, but no requirement for thiamine pp,, coenzyme A, and NAD+ could bedemonstrated.

" The rate of oxidation of NADH in the presence of lipoamide increasedwith time.

BLASTOCYSTIS HOMINIS 969

FIG. 2. B. hominis. Transmission electron microscopy. Saccatecristae are dominant, with filamentous branched cristae alsopresent. Both photomicrographs show cristae opening into theintracristal space. B. hominis mitochondria elongate from saccate orglobular to filamentous as the cell matures. They commonly formrosettes around the nuclei. Magnification: top, x82,500; bottom,x 137,500.

did not occur. Therefore, when partial inhibition of B.hominis occurred, at the 0.3 to 0.4% level, it probably wasdue to cytochrome c toxicity.We have been unable to demonstrate the presence of

enzymes commonly associated with mitochondria. The func-tional metabolic pathways, if any, are unknown. Evidencethat B. hominis mitochondria may be functional follows. (i)It seems unlikely that a vestigial, useless organelle would beretained by the cell. (ii) Mitochondria normally surround theB. hominis nucleus in the young cell as small spheres (0.2 to0.5 ,um in diameter) and then elongate as tubular, sometimesbranched, mitochondria up to 5 ,im in length and migratefrom the nucleus as the cell ages. (iii) They stain brightly and

FIG. 3. B. hominis cells from 5-day culture, showing intracellu-lare lipid making up much of the volume of the cells. Nomarskiinterference contrast. Magnification, x 1,600.

VOL. 26, 1988

970 ZIERDT ET AL.

specifically with Janus green or rhodamine 123, indicating atypical, physiologically active outer membrane. (iv) Mito-chondrial numbers per cell are not fixed and vary from twoto four in B. hominis cells from the human intestine or inrapidly dividing B. hominis cells to hundreds in old largercells and giant cells, reaching 100 ,um or more in diameter.(v) B. hominis synthesizes and stores lipid in quantitiessufficient to displace most of the volume of the cytoplasm.There is a possibility that B. hominis mitochondria synthe-size the stored lipid. This is based on microscopic evidence(in culture chambers), seen as lipid extrusion from massiveaggregations of mitochondria that escaped from rupturedcells (8). (vi) We saw increased numbers of mitochondria andincreased utilization of carbohydrates in the presence ofadded cytochrome c.

ACKNOWLEDGMENTS

We thank Francis Waterhouse for performing the cytochrome cexperiments and Pamula Simms for manuscript preparation.

LITERATURE CITED1. Alexieff, A. 1911. Sur la nature des formations dites "Kystes de

"Trichomonas intestinalis." C. R. Soc. Biol. 71:296-298.

2. Bach, F. W., and K. H. Kiefer. 1922. Untersuchungen uberBlastocystis. Zentralbl. Bakteriol. Parasitenkd. lnfektionskr.Hyg. Abt. 1 Orig. 89:72-98.

3. Brumpt, E. 1912. Colite à Tetranmitiis mesnili (Wenyon, 1910) etcolite à Trichomonas intestinalis (Leuckart, 1879); Blastocystishominis n. sp. et voisines formes. Bull. Soc. Pathol. Exot. 5:725-730.

4. Constantopoulos, G., M. A. Greenwood, and S. H. Sorrell. 1986.Mitochondrial abnormalities in fibroblast line GM 3093 defec-tive in oxidative metabolism. Experientia 42:315-318.

5. Johnson, L. V., M. L. Walsh, and L. B. Chen. 1980. Localizationof mitochondria in living cells with rhodamine 123. Proc. Natl.Acad. Sci. USA 77:990-994.

6. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall.1951. Protein measurement with the Folin phenol reagent. J.Biol. Chem. 193:265-275.

7. Zierdt, C. H. 1986. Cytochrome-free mitochondria of an anaer-obic protozoan-Blastocystis hoininis. J. Protozool. 33:67-69.

8. Zierdt, C. H. 1988. Blastocystis homninis, a long-misunderstoodintestinal parasite. Parasitol. Today 4:15-17.

9. Zierdt, C. H., W. S. Rude, and B. S. Bull. 1967. Protozoancharacteristics of Blastocystis hominis. Am. J. Clin. Pathol. 48:495-501.

10. Zierdt, C. H., and R. L. Williams. 1974. Bl/astocystis hoininis:axenic cultivation. Exp. Parasitol. 36:233-243.

J. CI-IN. MICROBIOL.