Roles of Adenylate Cyclases in Ciliary Responses of Paramecium to Mechanical Stimulation.

Paramecium shows rapid forward swimming due to increased beat frequency of cilia in normal (forward swimming) direction in response to various kinds of stimuli applied to the cell surface that cause K+ -outflow accompanied by a membrane hyperpolarization. Some adenylate cyclases are known to be functional K+ channels in the membrane. Using gene-specific knockdown methods, we examined nine paralogues of adenylate cyclases in P. tetraurelia to ascertain whether and how they are involved in the mechanical stimulus-induced hyperpolarization-coupled acceleration of forward swimming. Results demonstrated that knockdown of the adenylate cyclase 1 (ac1)-gene and 2 (ac2)-gene inhibited the acceleration of forward swimming in response to mechanical stimulation of the cell, whereas that spared the acceleration response to external application of 8-Br-cAMP and dilution of extracellular [K+ ] induced hyperpolarization. Electrophysiological examination of the knockdown cells revealed that the hyperpolarization-activated inward K+ current is smaller than that of a normal cell. Our results suggest that AC1 and AC2 are involved in the mechanical stimulus-induced acceleration of ciliary beat in Paramecium.

Evaluation of phenol-induced ecotoxicity in two model ciliate species: Population growth dynamics and antioxidant enzyme activity.

The application of identical exposure dosages in different species generally leads to a limited understanding of dose-response patterns because of species-specific factors. To evaluate phenol-induced ecotoxicity, antioxidant enzyme activity and population growth dynamics were compared in two model ciliates, the marine species Euplotes vannus and the freshwater species Paramecium multimicronucleatum. Dosage ranges of phenol exposure were based on tolerance limits of test ciliates as determined by their carrying capacity (K) and growth rate (r). When the exposure duration of phenol increased from 48 h to 96 h, the median effective dose (ED50) for P. multimicronucleatum decreased faster than that for E. vannus, and the ratio of the former to the latter declined from 2.75 to 0.30. When E. vannus was exposed to increasing concentrations of phenol (0-140 mg l-1), r rose initially and then dropped significantly at concentrations higher than 40 mg l-1, whereas K decreased linearly over the entire range. For P. multimicronucleatum, both r and K declined gradually over the range 0-200 mg l-1 phenol. Dose-response patterns of activities of three individual antioxidant enzymes, and the integrative index of the three enzymes, presented a biphasic (inverse U-shaped) curve at each of four durations of exposure, i.e. 12 h, 24 h, 36 h and 48 h. Cluster analyses and multidimensional scaling analyses of antioxidant enzyme activities revealed differences in the temporal succession of physiological states between the two model ciliates. In brief, combining ED50 with growth dynamic parameters is helpful for designing exposure dosages of toxicants in ecotoxicity tests.

Retrieval of lysosomal membrane and acid phosphatase from phagolysosomes of Paramecium caudatum.

Little is known about the fate of lysosomal membrane in phagocytic cells. Because the age of the digestive vacuoles in Paramecium caudatum can be easily determined, we have been able to study the dynamic membrane events in the older vacuoles. Late in the phagolysosomal stage (DV-III) the vacuole membrane undergoes a burst of tubule formation. The tubules expand into vesicles which have characteristics resembling lysosomes in both thin sections and freeze-fracture replicas. The tubules also contain acid phosphatase activity when they arise from acid phosphatase-reactive vacuoles. We conclude that after active digestion lysosomal membrane is retrieved in whole or in part along with some membrane-associated hydrolases. A logical extension of these results is that the lysosome-like vesicles, after being recharged with hydrolases by fusing with primary lysosomes, are recycled back to DV-II for reuse.

Vesicle transport along microtubular ribbons and isolation of cytoplasmic dynein from Paramecium.

Cytoplasmic microtubule-based motility in Paramecium was investigated using video-enhanced contrast microscopy, the quick-freeze, deep-etch technique, and biochemical isolations. Three distinct vesicle populations were found to be transported unidirectionally along the cytopharyngeal microtubular ribbons. This minus-end-directed movement exhibited unique in vivo features in that the vesicle transport was nonsaltatory, rapid, and predominantly along one side of the microtubular ribbons. To identify candidate motor proteins which may participate in vesicle transport, we prepared cytosolic extracts of Paramecium and used bovine brain microtubules as an affinity matrix. These preparations were found to contain a microtubule-stimulated ATPase which supported microtubule gliding in vitro. This protein was verified as a cytoplasmic dynein based upon its relative molecular mass, sedimentation coefficient of 16S, susceptibility to vanadate photocleavage, elevated CTPase/ATPase ratio, and its typical two-headed dynein morphology. This dynein was directly compared with the axonemal dyneins from Paramecium and found to differ by five criteria: morphology, sedimentation coefficient, CTPase/ATPase ratio, vanadate cleavage patterns, and polypeptide composition. The cytoplasmic dynein is therefore not an axonemal dynein precursor, but rather it represents a candidate for supporting the microtubule-based vesicle transport which proceeds along the microtubular ribbons.

Regulation of adenylyl cyclase from Paramecium by an intrinsic potassium conductance.

Hyperpolarization of the cell membrane of Paramecium stimulates adenosine 3',5'-monophosphate (cAMP) formation. Manipulations of the K+ resting conductance of the ciliate by adaptation in different buffers affected excitability of the cAMP generating system. Blockade of K+ channels inhibited hyperpolarization-stimulated cAMP formation. A mutant of Paramecium that is unable to control its K+ resting conductance had a defect in cAMP formation. Purified adenylyl cyclase, when incorporated into an artificial lipid bilayer membrane, revealed properties of a voltage-independent K+ channel. This indicates that the adenylyl cyclase of Paramecium has a secondary function as carrier of the K+ resting conductance. A hyperpolarization-activated K+ efflux appears to directly regulate adenylyl cyclase activity in vivo.

Kinetic Analyses of the Substrate Inhibition of Paramecium Arginine Kinase.

Paramecium tetraurelia expresses four types of arginine kinase (AK1-AK4). In a previous study, we showed that AK3 is characterized by typical arginine substrate inhibition, where enzymatic activity markedly decreases near a concentration of 1 mM of arginine substrate. This is in sharp contrast to the three other AK types, which obey the Michaelis-Menten reaction curve. Since cellular arginine concentration in another ciliate Tetrahymena is estimated to be 3-15 mM in vivo, Paramecium AK3 likely functions in conditions that are strongly affected by substrate inhibition. The purpose of this work is to find some novel aspect on the kinetic mechanism of the substrate inhibition of Paramecium AK3 enzyme. Substrate inhibition kinetics for AK3 were analyzed using three models and their validity were evaluated with three static parameters (R2, AICc, and Sy.x). The most accurate model indicated that not only ES but also the SES complex reacts to form products, the latter being the complex with two substrates in the active center. The maximum reaction rate for the SES complex, VmaxSES = 30.4 µmol Pi/min/mg protein, was one-eighth of the ES complex, VmaxES = 241.7. The dissociation constant for the SES complex (KiSES: 0.34 mM) was two times smaller than that of the ES complex (KsES: 0.61 mM), suggesting that after the primary binding of the arginine substrate (ES complex formation), the binding of a second arginine to the secondarily induced inhibitory site is accelerated to form an SES complex with a lower VmaxSES. The same kinetics were used for the S79A, S80A, and V81A mutants. The results indicate that the S79 residue is significantly involved in the process of binding the second arginine substrate. Herein, the KiSES value was ten times (3.62 mM) the value for the wild-type (0.34 mM), weakening substrate inhibition. In contrast, VmaxES and VmaxSES values for the mutants decreased by one-third, except for the VmaxSES of the S79A mutant, which had a value that was comparable with the value for the wild-type.

Receptor guanylyl cyclases.

Three different guanylyl cyclase cell receptors are known, but others will likely be discovered within the next few years. The general function of these receptors appear to relate to the regulation of fluid volume or fluid movement. New receptors, or possibly the currently known receptors, therefore, may be discovered in areas of the body where fluid volume regulation is important. Such fluids whose volume or composition might be regulated by guanylyl cyclase receptors include synovial fluid, uterine/oviductal luminal fluid, follicular fluid, aqueous humor, cerebral spinal fluid, seminiferous tubule luminal fluid, epididymal luminal fluid, seminal plasma, and airway luminal fluid. The function of the heterodimeric forms of guanylyl cyclase appear to relate to a primary regulation of nitric oxide (or similar molecules) concentrations, which are in turn regulated by a Ca2+/calmodulin-sensitive nitric oxide synthase.

A single telomerase RNA is sufficient for the synthesis of variable telomeric DNA repeats in ciliates of the genus Paramecium.

Paramecium telomeric DNA consists largely of a random distribution of TTGGGG and TTTGGG repeats. Given the precise nature of other ciliate telomerases, it has been postulated that there are two distinct types of the Paramecium enzyme, each synthesizing perfect telomeric repeats: one with a template RNA that specifies the addition of TTTGGG and the second dictating the synthesis of TTGGGG repeats. We have cloned and sequenced telomerase RNA genes from Paramecium tetraurelia, P. primaurelia, P. multimicronucleatum, and P. caudatum. Surprisingly, a single gene encodes telomerase RNA in all four species, although an apparently nontranscribed pseudogene is also present in the genome of P. primaurelia. The overall lengths of the telomerase RNAs range between 202 and 209 nucleotides, and they can be folded into a conserved secondary structure similar to that derived for other ciliate RNAs. All Paramecium telomerase RNAs examined include a template specific for the synthesis of TTGGGG telomeric repeats, which has not been posttranscriptionally edited to account for the conventional synthesis of TTTGGG repeats. On the basis of these results, possible mechanisms for the synthesis of variable telomeric repeats by Paramecium telomerase are discussed.

Early stages of functional diversification in the Rab GTPase gene family revealed by genomic and localization studies in Paramecium species.

New gene functions arise within existing gene families as a result of gene duplication and subsequent diversification. To gain insight into the steps that led to the functional diversification of paralogues, we tracked duplicate retention patterns, expression-level divergence, and subcellular markers of functional diversification in the Rab GTPase gene family in three Paramecium aurelia species. After whole-genome duplication, Rab GTPase duplicates are more highly retained than other genes in the genome but appear to be diverging more rapidly in expression levels, consistent with early steps in functional diversification. However, by localizing specific Rab proteins in Paramecium cells, we found that paralogues from the two most recent whole-genome duplications had virtually identical localization patterns, and that less closely related paralogues showed evidence of both conservation and diversification. The functionally conserved paralogues appear to target to compartments associated with both endocytic and phagocytic recycling functions, confirming evolutionary and functional links between the two pathways in a divergent eukaryotic lineage. Because the functionally diversifying paralogues are still closely related to and derived from a clade of functionally conserved Rab11 genes, we were able to pinpoint three specific amino acid residues that may be driving the change in the localization and thus the function in these proteins.

Detection of NADPH-diaphorase activity in Paramecium primaurelia.

Recently, we showed that Paramecium primaurelia synthesizes molecules functionally related to the cholinergic system and involved in modulating cell-cell interactions leading to the sexual process of conjugation. It is known that nitric oxide (NO) plays a role in regulating the release of transmitter molecules, such as acetylcholine, and that the NO biosynthetic enzyme, nitric oxide synthase (NOS), shows nicotinamide adenine dinucleotide phosphate-diaphorase (NADPH-d) activity. In this work, we detected the presence of NADPH-d activity in P. primaurelia. We characterized this activity histochemically by examining its specificity for beta-NADPH and alpha-NADH co-substrates, and sensitivity both to variations in chemico-physical parameters and to inhibitors of enzymes showing NADPH-d activity. Molecules immunologically related to NOS were recognized by the anti-rat brain NOS (bNOS) antibody. Moreover, bNOS immunoreactivity and NADPH-d activity sites were found to be co-localized. The non-denaturing electrophoresis, followed by exposure to beta-NADPH or alpha-NADH co-substrates, revealed the presence of a band of apparent molecular mass of about 124 kDa or a band of apparent molecular mass of about 175 kDa, respectively. In immunoblot experiments, the bNOS antibody recognized a single band of apparent molecular mass of about 123 kDa.

Site-directed mutagenesis, in vivo electroporation and mass spectrometry in search for determinants of the subcellular targeting of Rab7b paralogue in the model eukaryote Paramecium octaurelia.

Protein products of the paralogous genes resulting from the whole genome duplication may acquire new function. The role of post-translational modifications (PTM) in proper targeting of Paramecium Rab7b paralogue - distinct from that of Rab7a directly involved in phagocytosis - was studied using point mutagenesis, proteomic analysis and double immunofluorescence after in vivo electroporation of the mutagenized protein. Here we show that substitution of Thr200 by Ala200 resulted in diminished incorporation of [P32] by 37.4% and of 32 [C14-]UDP-glucose by 24%, respectively, into recombinant Rab7b_200 in comparison to the non-mutagenized control. Double confocal imaging revealed that Rab7b_200 was mistargeted upon electroporation into living cells contrary to non- mutagenized recombinant Rab7b correctly incorporated in the cytostome area. We identified the peptide ion at m/z=677.63+ characteristic for the glycan group attached to Thr200 in Rab7b using nano LC-MS/MS and comparing the peptide map of this protein with that after deglycosylation with the mixture of five enzymes of different specificity. Based on the mass of this peptide ion and quantitative radioactive assays with [P32]and  [C14-]UDP- glucose, the suggested composition of the adduct attached to Thr200 might be (Hex)1(HexNAc)1(Phos)3 or (HexNAc)1 (Deoxyhexose)1 (Phos)1 (HexA)1. These data indicate that PTM of Thr200 located in the hypervariable C-region of Rab7b in Paramecium is crucial for the proper localization/function of this protein. Moreover, these proteins differ also in other PTM: the number of phosphorylated amino acids in Rab7b is much higher than in Rab7a.

Cyclic AMP-dependent protein kinases of Paramecium. I. Chromatographic and physical properties of the enzymes from cilia.

The cAMP-dependent protein kinases of the cilia of the protozoan Paramecium tetraurelia were resolved and characterized. Two cAMP-dependent activities were present in cilia; the two ciliary kinases resemble types I and II from vertebrate tissues. Part of the ciliary kinase activity (primarily type II) was released by freeze-thawing, but a significant amount remained particulate. Both kinases were found as aggregates of about 220 kDa and of about 70 kDa. A portion of the cAMP-binding activity in ciliary extracts separated from kinase activity, and eluted at 36 kDa during gel filtration. Photoaffinity labeling with 8-azido-cAMP identified cAMP-binding proteins of 45-52 kDa in type II kinase from cilia, and of 43-46 kDa in type I kinase. The type II kinase was apparently autophosphorylated, causing a decrease in mobility during sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

Cyclic AMP-dependent protein kinases of Paramecium. II. Catalytic and regulatory properties of type II kinase from cilia.

The type II cAMP-dependent protein kinase (cAMP-PK-II) from cilia of Paramecium, purified free of type I cAMP-PK (cAMP-PK-I) and of cGMP-dependent protein kinase (cGMP-PK), phosphorylated several basic proteins and a heptapeptide containing serine (Kemptide). The enzyme was partially inhibited by the protein kinase inhibitor (Walsh inhibitor), but only at relatively high inhibitor concentrations. Half-maximal activation of cAMP-PK-II occurred at 15-25 nM cAMP. Several cAMP analogs were tested for ability to bind and activate the enzyme. 8-bromo-cGMP, a potent activator of Paramecium cGMP-PK, was a poor activator of Paramecium cAMP-PK-II. Activation of cAMP-PK-II was influenced by the phosphorylation assay buffer. Phosphate buffers provided increased activation by cAMP but decreased total activity relative to that measured in Mops-Tris buffer. The kinase was cAMP-independent when the pH of the assay buffer was high. Preincubation of cAMP-PK-II with histones also activated the enzyme in the absence of cAMP. The cAMP-PK-II bound cAMP with a Kd of 23 nM, and bound cAMP was released with a biphasic time course, suggesting two non-identical binding sites. The properties of the cAMP-PK of this ciliated protozoan appear to be closely similar to those of vertebrates.

Characterization of Ca2+- or Mg2+-ATPase of the excitable ciliary membrane from Paramecium tetraurelia: comparison with a soluble Ca2+-dependent ATPase.

We have characterized divalent-cation-stimulated nucleoside triphosphate hydrolase activity of the excitable ciliary membrane and compared it with a soluble Ca2+-ATPase released upon deciliation of Paramecium. The membrane-bound activity is strongly dependent on a divalent cation; calcium stimulates the basal activity of this enzyme at least 10-fold; magnesium and manganese stimulate less well, and strontium and barium, although less effective, also give measurable stimulation. This membrane-bound activity prefers ATP and GTP as substrates but also hydrolyzes UTP and CTP at measurable rates. The maximum velocity at saturating ATP concentrations and optimal calcium concentrations is 0.3 mumol/min per mg. The pH optimum for the membrane-bound activity is broad and centers around pH 7. From the temperature dependence of ATP hydrolysis, we calculate activation energies of 14 and 11 kcal/mol for the Ca2+- and Mg2+-stimulated activities, respectively. The Arrhenius plot is linear over the temperature range of 4 to 25 degrees C. The membrane ATPase is relatively insensitive to ouabain, oligomycin, N,N'-dicyclohexylcarbodiimide, vanadate, Ruthenium red and two calmodulin antagonists. Polyclonal antisera raised against the purified soluble ATPase from the deciliation supernatant show low reactivity with the membrane-bound ATPase. We conclude from the comparison of properties of the two activities that the ciliary membrane-bound ATPase is distinct from the soluble ATPase released by deciliation.

Transfer RNA methyltransferase activity in paramecium aurelia.

The tRNA methyltransferases from Paramecium aurelia were investigated. The effects of varying the Mg2+ and NH4+ concentrations, pH, and temperature on the methylation of Escherichia coli B tRNA using extracts from P. aurelia were determined. Optimum tRNA methyltransferase activity was observed at pH 7.8 and 37 degrees C. The Mg2+ optimum occurred at 0.66 mM in the absence of NH4+ while the NH4+ optimum occurred at 100 mM in the absence of Mg2+. Analysis of the bases methylated in (E. coli B) tRNA by extracts of P. aurelia showed the presence of 1-methyladenine, 1-methylguanine, N2-methylguanine, N2,N2-dimethylguanine and methylated pyrimidine nucleotides. In comparison, an analysis of the in vivo methylation of tRNA from P. aurelia showed the presence of 1-methyladenine, 6-methyladenine, 6,6-dimethyladenine, 1-methylguanine, N2-methylguanine, N2,N2-dimethylguanine, 7-methylguanine, and methylated pyrimidine nucleotides. The pattern of methylation of tRNA in P. aurelia is similar to that observed in other eukaryotes.

DNA-dependent DNA polymerase activities from Paramecia macronuclei.

Three DNA-dependent DNA polymerase activities, A, B, and C, have been detected from the macronucleus of Paramecium. The enzymes were separated and partially purified by DEAE-cellulose and phosphocellulose chromatography. These three enzymes can be further distinguished from each other by the effect of KCl, araCTP and different DNA templates on their activities. All three enzymes have the same molecular weight (900 000-110 000), are dependent on Mg2+ and added DNA template and are inhibited by sodium pyrophosphate. The relative amounts of the three enzyme activities vary with the growth stage of the Paramecia from which they are isolated. In balanced growth, three polymerase activities (A, B AND C) are observed, whereas in late log or stationary phase polymerase C is absent, while the level of polymerase B decreases by a larger factor than polymerase A. These results have been related to the possible functional roles of the three activities and their relationship to activities in other organisms studies is discussed.

Alkaline phosphatase from Paramecium cilia and cell bodies: purification and characterization.

A soluble alkaline phosphatase was purified 10 000-fold in an overall yield of 8% from both of the cilia and cell bodies of the protozoan Paramecium tetraurelia. The concentration in cilia (1.7 microM) was 6-fold higher than in cell bodies, although the latter contained most of the activity due to their much greater volume. The purified protein showed a single (36 kDa) protein staining band on SDS-PAGE. This value, in conjunction with the apparent molecular mass of 66 kDa for the native enzyme (gel filtration) suggests a dimeric structure. The specific activity of the purified phosphatase ranged from 10 to 70 mumols.min-1.mg-1 at the pH-optimum of 8.0 and the Km for p-nitrophenyl phosphate was 81 microM. Basal enzyme activity was inhibited by metal chelators and stimulated up to 12-fold by addition of divalent cations. Mg2+ acted as a non-essential mixed-type activator with a half-maximal effect at 7 microM. Ca2+ was inhibitory, the extent of inhibition was dependent on the concentration of Mg2+ in the assay. Furthermore, the kinetics of inhibition by Ca2+ varied with the Mg2+ concentration. Phosphate, pyrophosphate, and SH-group blocking agents also strongly inhibited. The enzyme did not dephosphorylate Tyr- or Ser-/Thr-phosphoproteins. The Paramecium enzyme is not of lysosomal origin and its properties are quite different from all known phosphatases. It is a novel type of phosphatase since it (i) shows F(-)-inhibition like Ser/Thr-phosphatases but (ii) is inhibited by vanadate and molybdate like Tyr-phosphatases, and (iii) inhibition by Ca2+ has not been reported for any other phosphatase.

Characterization of a putative Ca2(+)-transporting Ca2(+)-ATPase in the pellicles of Paramecium tetraurelia.

In Paramecium, no Ca2(+)-ATPases with the properties of Ca2+ pumps have been identified. Here we report a pellicle associated Ca2(+)-ATPase activity and a corresponding phosphoprotein intermediate characteristic of a pump. The Ca2(+)-ATPase activity requires 3 mM Mg for optimal Ca2+ stimulation (KCa = 90 nM) and is specific for ATP as substrate (Km = 75 microM). Vanadate and calmidazolium inhibit Ca2(+)-stimulated activity with an EC50 of about 2 microM and 0.5 microM, respectively. Likewise, 10 microM trifluoperazine inhibits 80% of Ca2(+)-ATPase activity, but bovine calmodulin fails to stimulate. The Ca2(+)-ATPase is not inhibited by sodium azide (10 mM), oligomycin (10 micrograms/ml) or ouabain (0.2 mM). Incubation of pellicles with [gamma-32P]ATP specifically labels a 133 kDa protein in a Ca2(+)-dependent, hydroxylamine-sensitive manner, and the level of phosphorylation is increased by 100 microM La3+. Phosphorylation of an endoplasmic reticulum-enriched fraction labels a Ca2(+)-dependent protein different from the pellicle protein, being lower in molecular mass and unaffected by La3+. Ca2+ uptake by the alveolar sacs, integral components of the pellicle membrane complex, is poorly coupled to Ca2(+)-stimulated ATP hydrolysis (Ca2+ transported/ATP hydrolysed less than 0.2) and is much less sensitive to vanadate inhibition (EC50 approx. 20 microM) compared to the total Ca2(+)-ATPase activity. Therefore, the majority of the Ca2(+)-ATPase activity is likely to be plasma membrane associated.

Biochemical studies of the excitable membrane of Paramecium. IV. Protein kinase activities of cilia and ciliary membrane.

Two protein kinases (ATP: protein phosphotransferase, EC 2.7.1.37) were detected in disrupted cilia of Paramecium tetraurelia. One of the enzymes exhibited maximum activity at pH 6.0, required 4 mM Mg2+ for its maximum activity and was stimulated by cyclic AMP and cyclic GMP. Histone was a good exogenous protein substrate for this enzyme, but protamine sulfate was not. The other protein kinase showed a peak of activity at pH 8.0, required 10 mM Mg2+ for its maximum activity and was slightly inhibited by cyclic AMP and cyclic GMP. Protamine sulfate was a good exogenous substrate for this enzyme. The pH 8.0 activity partitioned preferentially with the axonemes, but the pH 6.0 activity was divided almost equally between the axonemes and the membranes. We also found indirect evidence for the presence in cilia of phosphoprotein phosphatase (phosphoprotein phosphohydrolase, EC 3.1.3.16) and adenyl cyclase (ATP pyrophosphate-lyase (cyclizing), EC 4.6.1.1) activity.

A Ca2+-activated ATPase specifically released by Ca2+ shock from Paramecium tetraurelia.

Deciliation of Paramecium tetraurelia by a Ca2+ shock procedure releases a discrete set of proteins which represent about 1% of the total cell protein. Marker enzymes for cytoplasm (hexokinase), endoplasmic reticulum (glucose-6-phosphatase), peroxisomes (catalase), and lysosomes (acid phosphatase) were not released by this treatment. Among the proteins selectively released is a Ca2+-dependent ATPase. This enzyme has a broad substrate specificity which includes GTP, ATP, and UTP, and it can be activated by Ca2+, Sr2+, or Ba2+, but not by Mg2+ or by monovalent cations. The crude enzyme has a specific activity of 2-3 mumol/min per mg; the optimal pH for activity is 7.5. ATPase, GTPase, and UTPase all reside in the same protein, which is inhibited by ruthenium red, is irreversibly denatured at 50 degrees C, and which has a sedimentation coefficient of 8-10 S. This enzyme is compared with other surface-derived ATPases of ciliated protozoans, and its possible roles are discussed.