Vol. 29, No. 7JOURNAL OF CLINICAL MICROBIOLOGY, JUlY 1991, P. 1498-15030095-1137/91/071498-06$02.00/0Copyright © 1991, American Society for Microbiology

Localization of Endogenous Activity of PhospholipasesA and C in Ureaplasma urealyticumNIHAL S. DE SILVA"3 AND PATRICIA A. QUINN' 2,4*

Department of Microbiology, The Hospital for Sick Children, Toronto M5G 1X8,1* and Departments of Microbiology,2Clinical Biochemistry,3 and Obstetrics and Gynaecology,4 University of Toronto, Toronto M55 IAH, Canada

Received 26 November 1990/Accepted 16 April 1991

Endogenous activities of phospholipases A and C in Ureaplasma urealyticum were assayed in cellularfractions of exponential-phase cells. Enzymatic studies indicated that ATPase activity was localized in theplasma membrane fraction and NADH and NADPH dehydrogenase activities were localized in the cytosolfraction. Studies with purified ureaplasma membranes demonstrated that, of three serovars tested, endogenousphospholipase A1, A2, and C activities were localized in the plasma membrane. Very low levels of activity wereobserved in the cytosol fractions. Phospholipase A2 activity in the plasma membrane was 3- to 5-fold higherthan the activity in the lysates and 60- to 300-fold higher than the activity of phospholipase A1. PhospholipaseC was localized mainly in the plasma membrane, with 20% found in the cytosol fraction. The levels of activitywere comparable among the three serovars. There was a significantly lower level of activity in cells from thestationary growth phase than in the exponential phase. Significant differences were observed in thephospholipase A activities among the U. urealyticum serovars 3, 4, and 8. Phospholipase A2 activity was twofoldhigher in serovar 8 membranes, and phospholipase A1 activity was twofold higher in serovar 3 membranes.These results demonstrate that endogenous activities of phospholipase A and C are localized primarily in theplasma membrane fraction of U. urealyticum. The specific activities in the membranes of the phospholipasesvaried among the three serovars. Phospholipase enzymes may function as virulence factors in U. urealyticumand may vary among the serovars.

Ureaplasma urealyticum is associated with a variety ofreproductive failures ranging from infertility, stillbirth, andpremature delivery to fetal or neonatal lung disease in asubset of colonized patients (for a review, see reference 16).The ability of Ureaplasma diversum to cause infertility,spontaneous pregnancy loss, premature delivery, and lungdisease is well documented by experimental infection in thebovine model (16). Infection of the placenta leads to chorio-amnionitis, with an increased risk for stillbirth and prematu-rity. One hypothesis in delineating the role of placentalinfection in premature delivery is that bacterial phospholi-pases may affect prostaglandin biosynthesis and thus indi-rectly cause premature labor (2). The activation of microbialphospholipases could hydrolyze placental or membranephospholipids to produce free arachidonic acid which couldstimulate a variety of eicosinoid biosynthetic pathways,leading to prematurity or stillbirth.

Phospholipases are important catabolic enzymes in phos-pholipid metabolism, but the mechanisms of their action inthe infectious process are not well understood (5). Phospho-lipases A1 and A2 hydrolyze phospholipids to producelysophospholipid and fatty acids. Phospholipase C is aphosphoryl hydrolase which liberates 1.2-diglyceride andphosphorylester (5, 28). In addition to its function in phos-pholipid turnover, phospholipase A2 is involved in theresynthesis of phospholipids via deacylation-reacylation (9)and in the production of prostaglandin precursors (28).Phospholipase C activity has been reported as secretedphosphorylhydrolase in many bacterial species (1, 5). Thesecretion of phospholipase C may play a role in the physio-logical effects of infection, but the function of phospholi-

* Corresponding author.

pases as secreted toxins in perinatal infection is not yetdelineated.

Mollicutes are unique microorganisms with no cell wallsand limited metabolic capabilities. They adhere to andcolonize epithelial surfaces in the respiratory and urogenitaltracts by attachment to host plasma membranes (21). Theymay affect host cells by the production of toxic metabolitessuch as 02 and H202 (21), or the mycoplasma membraneenzymes may interact directly with required substratespresent in the host cell membrane (11).We have reported the presence of phospholipases A1, A2,

and C in U. urealyticum and that phospholipase A2 activityvaries among the serovars studied (6). It is, therefore,possible that the phospholipases of U. urealyticum, Myco-plasma hominis (23), and other species (3, 22, 29) function asvirulence factors in infection as has been proposed forbacteria (2, 5). In this study, we demonstrate that theureaplasma phospholipases are localized in the membrane,an important factor in delineating the mechanism of patho-genesis.

MATERIALS AND METHODS

Materials. 1-Palmitoyl-2[2-'4C]oleoylphosphatidylcholinewith a specific activity of 57 mCi/mmol, L-a-dipalmitoyl-phosphatidylcholine-[choline-methyl-3H] with a specific ac-tivity of 40.5 Ci/mmol, and L-a-[dipalmitoyl-1,2-14C]phos-phatidylcholine with a specific activity of 112 mCi/mol werepurchased from New England Nuclear (Boston, Mass.).Phospholipase A2 (from Naja naja venom), phospholipase C(from Clostridium welchii), and egg phosphatidylcholinewere purchased from Sigma Chemical Co. (St. Louis, Mo.).The purity of substrates was checked by thin-layer chroma-tography in silica gel by using chloroform-methanol-water(65:35:4, vol/vol/vol) as the developing solvent. All lipid

1498

on June 25, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

PHOSPHOLIPASE ACTIVITY LOCALIZATION IN U. UREALYTICUM 1499

standards and thin-layer chromatography plates were ob-tained from Supelco Chromatography Supplies (Bellefonte,Pa.) and Mandel Scientific Co. (Rockwood, Ontario, Cana-da). All other chemicals were of reagent grade and werepurchased from Fisher Scientific Co. (Toronto, Ontario,Canada).Organism and growth. U. urealyticum serovar 3 (strain

27), serovar 4 (strain 58), and serovar 8 (strain T960-CX3)were originally obtained from R. Purcell at the NationalInstitutes of Health, Bethesda, Md., in 1967. As with othermicroorganisms, it is possible that certain U. urealyticumserovars may be more virulent than others. The most prev-alent ureaplasma serovar in our locale is serovar 3 (17), yet,in antibody studies, U. urealyticum serovar 3 did not inducea significantly elevated antibody response in neonates withrespiratory disease (18, 19). Thus, serovar 3 was chosen as apossibly nonvirulent strain. U. urealyticum serovar 8 wasassociated with a case of fatal ureaplasma pneumonia in aneonate, even though the mother was also colonized withserovar 3 (18). Elevated antibody responses to serovars 4, 7,and 8 were observed in neonates with respiratory disease(18, 19). Serovar 4 was first reported as the most prevalentserovar in nongonococcal urethritis (25) and as the serovarwhich had the greatest inhibitory effect on the fertilization ofhamster ova by human sperm exposed to U. urealyticumserovars (4). For these reasons, serovars 4 and 8 werechosen as potentially pathogenic since virulent and avirulentserovars have not yet been identified in humans. The urea-plasmas were cultured in lean trypticase soy medium (BBLMicrobiology Systems, Cockeysville, Md.) supplementedwith 5% fetal calf serum (Bocknek Ltd., Rexdale, Ontario,Canada) and 5% fresh yeast extract (25%, wt/vol) (TSmedium) (6). Stocks were stored at -70°C in polypropylenecryotubes (Nunc, GIBCO Canada, Burlington, Ontario,Canada).To ensure that the phospholipase activities were measured

in actively metabolizing organisms, standard growth curvesfor the specific pools of each of the three serovars weredetermined at 37°C by using 100-ml volumes of lean TSmedium. Estimates of the incubation times to exponentialand stationary growth phases were made on the basis of theobserved growth curve as previously described (6). Cultureswere then inoculated at the same concentration and from thesame pool and grown to exponential and stationary phases.Ureaplasma cells were harvested in late-exponential phaseby centrifugation (35,000 x g for 45 min) at 15 h for serovar8, 24 h for serovar 4, and 40 h for serovar 3 as predeter-mined. The ureaplasmas were harvested in the stationaryphase at 28 h for serovar 8, 35 h for serovar 4, and 47 h forserovar 3. The pHs of the culture media were monitored, andthe viability and titer of the cultures were determined at thetime of harvest. The growth rates were reduced because ofthe lower concentrations of serum and yeast extract in leanTS medium. The use of lean medium enabled better demar-cation between the exponential and stationary phases.

Preparation of membrane and cytosol fractions. The cellpellet was resuspended in 2 ml of 0.25 M NaCl and usedimmediately for the preparation of cellular fractions byosmotic lysis (6, 20). The membrane fraction was washedthree times consecutively in distilled water and resuspendedin 1 ml of P buffer (20). Membrane and cytosol fractions werestored at -70°C until assayed. The medium control wasprepared by incubating uninoculated lean TS medium for 24h, harvesting by centrifugation, and processing as describedabove for the ureaplasmas. Protein was determined by themethod of Lowry et al. (12).

Characterization of membrane and cytosol fractions.(ii) Electron microscopy. An aliquot (100 RI) of the plasmamembrane fraction was examined by negative staining andtransmission electron microscopy. For negative staining, themembrane fraction was stained with 2% phosphotungsticacid and directly examined with an electron microscope. Fortransmission electron microscopy, the membrane fractionwas fixed in 2.5% glutaraldehyde and postfixed in 1% os-

mium tetroxide in 0.1 M phosphate buffer (pH 7.2). Thepellets were embedded in Epon-Araldite. Thin sections were

cut and double stained with uranyl acetate-lead citrate.Stained sections were viewed by use of a Philips electronmicroscope.

(ii) Enzyme assays. ATPase activity was determined bymeasuring the release of Pi from ATP (23). NADH andNADPH dehydrogenase activities were assayed spectropho-tometrically (13).

Phospholipase A. The assays for phospholipases A1 and A2were performed as previously described by using lecithinlabeled specifically in the 2 position, i.e., 1-palmitoyl-2-[2-'4C]oleoylphosphatidylcholine, and prepared as unsonicatedliposomes for the substrate (6). The reaction mixture (1.05ml) contained 18.3 nmol of labeled substrate diluted with eggphosphatidylcholine, 30 to 50 ,ug of ureaplasma cell fractionprotein, and 0.1 M Tris buffer (pH 7.4). The phospholipidswere extracted by a modification of the Bligh and Dyermethod (7). Phospholipase A1 activity was calculated fromthe yield of labeled lysophosphatidylcholine, and phospho-lipase A2 activity was calculated from the yield of labeledfree fatty acid. The labeled fatty acids were isolated bythin-layer chromatography using chloroform-methanol-wa-ter (65:35:5, vol/vol/vol) and measured directly by use ofliquid scintillation spectrometry. The positional specificity of14C in lecithin was determined by using purified phospholi-pase A2 by the above-described techniques. More than 98%of the counts were recovered in the lysophosphatidylcholinefraction, and 1 to 2% were recovered in the fatty acidfraction. Three controls were assayed: inactivated urea-

plasma cell fractions (boiled enzyme at 100°C for 5 min);substrate only (at time zero); and a medium control preparedfrom incubated lean TS medium, processed in the same

manner as the ureaplasma lysate. To calculate the urea-

plasma enzyme activity, the activity values found in thecontrols were subtracted from the activity values found inthe ureaplasma membrane preparations.

Phospholipase C assay. Phospholipase C assays were per-formed as previously described by using lecithin, labeledspecifically in the choline moiety as L-a-dipalmitoylphos-phatidylcholine-[choline-methyl-3H]. Lecithin was preparedas unsonicated liposomes to use as substrate (6). The reac-

tion mixture (1.2 ml) contained 0.18 to 0.36 nmol of labeledsubstrate, 30 to 70 ,ug of enzyme protein (ureaplasma cellfraction), 10 mM CaCl2, and 0.1 M Tris (pH 7.4). Theenzymatic activities in ureaplasma cellular fractions were

monitored by the cleavage of the choline base to liberateradiolabeled phosphorylcholine and unlabeled diglyceride(5). Phosphorylcholine (Fig. 1, spot B) had an Rf of 0.45 andwas identified by the Hanes-Ischerwood spray for the iden-tification of organic phosphorous on the chromatogram (6).Spot A was unidentified, and spot C was identified as Pi.

All protein estimations and, hence, the specific enzymeactivities of ureaplasma cell fractions were corrected forresidual culture medium components as described for phos-pholipase A. All radioactivities in the controls (boiled con-

trol, medium control, and zero time control) were subtracted

VOL. 29, 1991

on June 25, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

1500 DE SILVA AND QUINN

2000 B

1000 c

4 8 12 16 20 24 28

Distance from origin, cm.

FIG. 1. Fractionation of upper phase of lipid extract by paperchromatography to detect [14C]phosphorylcholine. The solvent sys-tem consisted of butanol-acetic acid-water (25:5:5, vol/vol/vol).

from the radioactivities of experimental samples beforecalculation of total and specific enzymatic activities.

RESULTSCharacterization of the membrane and cytosol fractions.

Cellular characteristics and purity of isolated plasma mem-branes from U. urealyticum were examined by negativestain, transmission electron microscopy, and enzymatic as-says. The bilayered structure of plasma membrane ghostswas evident (Fig. 1). ATPase and NADH and NADPHdehydrogenase activities of ureaplasma cellular fractions arepresented in Table 1. The relative specific activity of ATPaseassociated with the plasma membrane of the serovars, whencompared with that of whole-cell lysates, varied between 1.2and 9.1. No ATPase activity was detected in the cytosolfraction. However, both NADH and NADPH dehydroge-nase activities were highest in the cytosol fraction. Theseresults indicated that ATPase activity was localized primar-ily in the plasma membrane fraction of U. urealyticum, andNADH and NADPH dehydrogenase activities were local-ized mainly in the cytosol fraction. There also appear to bedifferences between the serovars in the levels of activity ofthese enzymes.

Phospholipase A activity in cellular fractions. Phospholi-pase A activity was assayed in lysates, plasma membrane,and cytosol fractions of ureaplasma cells. Phospholipase A1was estimated by the amount of radioactivity in the liberatedlysophospholipid. Of the radioactivity added, 85 to 90% wasrecovered as products and unreacted substrate after incuba-

TABLE 1. Enzymatic characterization of cellular fractions ofU. urealyticum serovars

Sp act of enzymeCellularSerovar fraction ATPasea NADH NADPH

dehydrogenaseb dehydrogenaseb3 Whole-cell lysate 12.8 6.7 89.2

Plasma membrane 116.9 21.3 <0.1CCytosol <0.1 61.2 1,955.0

4 Whole-cell lysate 31.9 34.0 28.3Plasma membrane 53.5 33.0 24.4Cytosol <0.1 111.5 1,104.2

8 Whole-cell lysate 27.6 12.2 154.2Plasma membrane 33.8 4.8 <0.1Cytosol <0.1 53.8 5,054.4

a ATPase activity is expressed as nanomoles of Pi released per minute.b NADH and NADPH dehydrogenase activities are expressed as decreases

in A340 per milligram of protein per minute (10-3).c <0.1, no activity detected at the lower limit of the assay.

tion, extraction of reaction products, thin-layer chromatog-raphy, etc. Of the recovered radioactivity, only a minimalamount (1 to 2%) was associated with the glycerophospho-rylcholine fraction. In the zero time control, about 1% of theadded radioactivity from substrate was recovered in eitherfatty acid or lysophosphatidylcholine. These experimentsdemonstrate that phospholipase A1 activity is localized inthe plasma membrane fraction of the three serovars exam-ined (Table 2). Very low activity of phospholipase A1 wasobserved in the cytosol fractions. Phospholipase A2 inureaplasma lysates, plasma membrane, and cytosol fractionswere assayed by using the same substrate labeled specifi-cally in the fatty acid moiety in the sn-2 position only. Theexperiments demonstrated that A2 activity was confined tothe plasma membrane fraction of the three ureaplasmaserovars (Table 3). The specific activity of the membranefraction of serovar 8 was the highest. The A2 activitiesshowed a three- to fivefold enrichment over the activity inthe lysate. The cytosol fraction showed a low but detectablelevel of A2 activity in all three serovars.

Phospholipase C activity in cellular fractions. The activityof phospholipase C was estimated as the amount of radioac-tivity liberated as phosphorylcholine by using the substratespecifically labeled in the choline moiety (choline-methyl-

kA

tv @ x- 2SX.i

FIG. 2. Transmission electron micrograph of membrane ghosts of serovar 8 of U. urealyticum cells fixed in 2.5% glutaraldehyde andpostfixed in 1% osmium tetroxide. The gold sections were stained with uranyl acetate-lead citrate. Note the triple-layered structure of theplasma membrane (M). Bar = 0.2 ,um.

J. CLIN. MICROBIOL.

,; .. jbk.. .:. vP" raw'.%,

on June 25, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

PHOSPHOLIPASE ACTIVITY LOCALIZATION IN U. UREALYTICUM 1501

TABLE 2. Localization of phospholipase A1 activity in cellularfractions of exponential-phase cells of U. urealyticum serovars

Lysophosphatidylcholine liberatedCellular (pmol/mg of protein per min) by serovar"fraction

3 4 8

Cell lysate 2.5 ± 2.4 7.7 ± 1.2 1.2 ± 0.4Plasma membrane 24.2 ± 2.6 9.4 ± 1.5 11.1 ± 2.9Cytosol 0.01 ± 0.05 0.01 ± 0.06 0.01 ± 0.05

a The values are means ± standard deviations of three separate experi-ments.

3H) (Fig. 1). Furthermore, no radioactivity was detected inthe diglyceride, isolated in the lipid phase, which indicatedthat phospholipase C activity was present in the lysates.Experiments were done to determine the time course ofhydrolysis and the effect of protein concentration, and theresults showed that the hydrolytic action of the enzyme waslinear up to 120 min of incubation and up to 100 mg of lysateprotein. These experiments indicated that phospholipase Cactivity was localized predominantly in the plasma mem-brane fraction of U. urealyticum, with 20% in the cytosolfraction (Table 4).

Phospholipase C activity in stationary-phase cells of U.urealyticum. The growth curve of U. urealyticum is typical ofmost mycoplasma species in that there is an exponentialphase, a short stationary phase, and a rapid decline phase.To control for the stage of growth, experiments were carriedout on the basis of carefully standardized growth curves foreach serovar pool. If one hypothesized that the phospholi-pases were providing lipid metabolites to the growing organ-ism, it was important to assay the cells at the stage of growthin which the enzymes were active. The specific activities ofphospholipase C during the exponential and stationaryphases of growth were therefore compared. As shown inTable 5, there was a marked decline in specific activity ofphospholipase C from the exponential phase (Table 4) to thestationary phase in all cellular fractions. The phospholipaseC activity was again confined to the membrane fraction,although membrane degradation could have taken place bythis stage of growth. The data therefore indicate that thephospholipases are most active during the exponentialgrowth phase.

DISCUSSION

This study demonstrates that the activities of endogenousphospholipases A1, A2, and C in U. urealyticum lysates arelocalized primarily in the plasma membrane of exponential-phase cells. These findings are in general agreement with

TABLE 3. Localization of phospholipase A2 activity in cellularfractions of exponential-phase cells of U. urealyticum serovars

Fatty acid liberated (pmol/mg of protein per min)Cellular by serovarlfraction

3 4 8

Cell lysate 344.0 ± 12.9 465.0 ± 14.5 1,138.2 ± 21.6Plasma mem- 1,666.7 + 6.8 1,686.0 ± 10.1 3,484.8 ± 28.6

braneCytosol 0.04 ± 0.01 0.04 ± 0.02 0.05 ± 0.03

a The values are means ± standard deviations of three separate expen-ments.

TABLE 4. Localization of phospholipase C activity in cellularfractions of exponential-phase cells of U. urealyticum serovars

Phosphorylcholine liberatedCellular (pmol/mg of protein per min) by serovar'fraction

3 4 8

Cell lysate 8.2 ± 2.6 8.6 ± 2.1 9.2 ± 2.4Plasma membrane 14.2 ± 3.2 12.8 ± 3.1 16.6 ± 3.6Cytosol 3.5 + 1.4 1.5 ± 2.9 4.5 ± 1.6

a The values are means + standard deviation of three separate experiments.

similar observations on other mycoplasmas such as M.hominis (23), Acholeplasma laidlawii (29), and Mycoplasmamycoides (3). In M. hominis and A. laidlawii, the phospho-lipase C activity and lysophospholipase activity, respec-tively, were localized in the plasma membrane fractions.

Transmission electron microscopy, negative staining, andenzyme assays indicated that the membrane ghosts preparedby our technique were of sufficient purity. The ureaplasmaghosts exhibited the typical bilayer membrane of Mollicutesspecies. Since Na+K+-ATPase has been generally recog-nized as a marker enzyme for the mycoplasma membrane(24), enzymatic assays to check the purity of the membranepreparation were done. U. urealyticum had ATPase activitywhich was confined mainly to the plasma membrane frac-tion. Membrane-bound ATPases have been previously re-ported in A. laidlawii B (10) and in Mycoplasma gallisepti-cum (26). NADH and NADPH dehydrogenase activitieswere found in U. urealyticum and were markedly elevated inthe cytosol fractions. These observations are in agreementwith studies showing these enzymes to be markers for thecytosol fraction. Thus, U. urealyticum appears to be com-parable to other mycoplasmas in terms of localization ofATPase in the membrane. NADH and NADPH dehydroge-nase activities are localized in the cytoplasm.

Specifically labeled substrates of phosphatidylcholinewere used to directly assay for phospholipase A and Cactivities in cellular fractions of U. urealyticum. Since thehydrolysis of the substrate is influenced by physiochemicalstates (5, 27) and by the transition temperature of the lipid(14), all assays were done with unsonicated liposomes of thesubstrate in the presence of Ca2" ions for phospholipase Aand at near-transition temperature (37°C). Thus, interpreta-tion of enzyme activity is subject to these limitations.The specific activity of phospholipase A1 was localized in

the plasma membrane of exponential-phase cells of the threeserovars. The enrichment factor of the activity in the mem-brane in comparison with that of the lysate varied from 1.2 inserovar 4 to about 9 in serovars 3 and 8. These values mayreflect the purity of the preparation of the plasma membranefraction or basic differences between the serovars. The very

TABLE 5. Localization of phospholipase C activity in cellularfractions of stationary-phase cells of U. urealyticum serovars

Phosphorylcholine liberatedCellular (pmollmg of protein per min) by serovar'fraction

3 4 8

Cell lysate 0.03 0.06 0.04Plasma membrane 1.00 0.16 0.20Cytosol <0.01 <0.01 <0.01

a The values are the means of two separate experiments.

VOL. 29, 1991

on June 25, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

1502 DE SILVA AND QUINN

low activities of phospholipases A1 and A2 in the cytosolindicated that these enzymes were not liberated duringpreparation of the plasma membranes.The specific activity of phospholipase A2 was also found

to be localized exclusively in the plasma membrane ofexponential-phase cells in the three serovars. The enrich-ment factors of activity in the membrane compared to that inthe cell lysate of the three serovars were similar, rangingfrom 3.1 to 4.8. However, the ratio of phospholipase A2activity to phospholipase A1 activity in the membrane wassignificantly higher and varied from 69:1 in serovar 3 to 179:1in serovar 4 and to 317:1 in serovar 8. Since the isolatedmembrane protein is only about 25% of the total cell lysateprotein (unpublished observations), these observations ofhigh specific activity of phospholipases in the membrane,greater than that in the whole-cell lysates, support theconcept that these enzymes are localized in the plasmamembrane of U. urealyticum.The specific activity of phospholipase C was also elevated

in the plasma membrane in comparison with its activity inthe whole-cell lysate. The enrichment factor of activity in themembrane over that of the lysates is about 2. This suggeststhat phospholipase C is membrane bound in U. urealyticum,which differs from the case in bacteria, where phospholipaseC is secreted into the culture medium (1, 5). Since thecytosol fraction has phospholipase C activity, it may be thatthe enzyme was loosely bound to the plasma membrane ordistributed through the membrane. Thus, in processing, acertain amount of activity was recovered in the cytosolfraction because of a difference in stereochemistry within theureaplasma membrane. These results are in contrast to thelocalization studies of phospholipase C activity in M. hom-inis, where the enzyme has been found exclusively in theplasma membrane (23). This may be a reflection of a differ-ence in lipid metabolism between the ureaplasmas and otherMycoplasma species. It has been proposed that lipids maydiffer in U. urealyticum to maintain membrane function overa wide pH range (pH 5.0 to 9.0). Ureaplasmas are reported todiffer from other mollicutes in a number of enzyme systems(15). This study also demonstrates differences among thethree serovars in the levels of activity for ATPase andNADH and NADPH dehydrogenase.The decline in phospholipase C activity in the stationary

phase may reflect a general reduction in phospholipid turn-over as the cell metabolism decreases. In aging Mycoplasmacapricolum cultures, there was a decrease in the phospho-lipid-to-protein ratio as cells moved from the exponential tothe stationary phase (8). These observations emphasize that,for comparisons between Mollicutes species and serovars,enzyme assays should be done at the same stage of thegrowth cycle.The phospholipase enzymes of U. urealyticum may be

important as virulence factors in pathogenesis. Mycoplas-mas adhere to host cells, and their enzymes can interact withthe cell membrane (11). The utilization of cell membranephospholipids such as phosphatidylcholine for enzyme sub-strates and the production of metabolites such as fatty acids(arachidonic acid) and lysophospholipids may be significantin delineating the pathogenic effects of ureaplasmas. Bythese mechanisms, the ureaplasmas could exert an effect onhost cell biosynthesis, membrane function, or immune func-tion. Thus, ureaplasma phospholipases may act as virulencefactors in infection of fetal membranes, the placenta, andneonatal lungs. Variation in levels of specific activity orunique enzyme characteristics may account for differencesin virulence among the ureaplasma serovars.

ACKNOWLEDGMENTSThis work was supported by grant no. 6656-2140-53 from Health

and Welfare Canada and grants from the Easter Seals ResearchFoundation (grant no. 22-16-03-87-1) and the Ontario ThoracicSociety. N. S. DeSilva held a Career Development Award from theEaster Seals Research Foundation, Toronto, Ontario, Canada.

REFERENCES1. Baine, W. B., J. K. Rasheed, D. C. Mackel, C. A. Bopp, J. G.

Wells, and A. F. Kaufmann. 1979. Exotoxin activity associatedwith the Legionnaires disease bacterium. J. Clin. Microbiol.9:453-456.

2. Bejar, R., V. Curbelo, C. Dairs, and L. Gluck. 1981. Prematurelabour. Obstet. Gynecol. 57:479-482.

3. Bhandari, S., and P. J. Asnani. 1989. Characterization ofphospholipase A2 of mycoplasma species. Folia Microbiol.34:294-301.

4. Busolo, F., and R. Zanchetta. 1985. The effect of Mycoplasmahominis and Ureaplasma urealyticum on hamster egg in vitropenetration by human spermatozoa. Fertil. Steril. 43:110-114.

5. Dennis, E. A. 1983. Phospholipases, p. 307-353. In P. D. Boyer(ed.), The enzymes. Academic Press, Inc., New York.

6. De Silva, N. S., and P. A. Quinn. 1986. Endogenous activity ofphospholipases A and C in Ureaplasma urealyticum. J. Clin.Microbiol. 23:354-359.

7. De Silva, N. S., and C.-H. Siu. 1980. Preferential incorporationof phospholipids to plasma membrane during cell aggregation ofDictyostelium discoideum. J. Biol. Chem. 255:8489-8496.

8. Gross, Z., and S. Rottem. 1986. Lipid interconversions in agingMycoplasma capricolum cultures. J. Bacteriol. 167:986-991.

9. Lands, W. E. M. 1960. Metabolism of glycerolipids. II. Theenzymatic acylation of lysolecithin. J. Biol. Chem. 235:2233-2237.

10. Lewis, R. N., and R. N. McElhaney. 1983. Purification andcharacterization of the membrane (Na+ +Mg2+)-ATPase fromAcholeplasma laidlawii B. Biochim. Biophys. Acta 735:113-122.

11. Lingwood, C., S. Schramayr, and P. A. Quinn. 1990. The malegerm cell specific sulfogalactoglycerolipid is recognized anddegraded by mycoplasms allociated with male infertility. J. Cell.Physiol. 142:170-176.

12. Lowry, 0. 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.

13. Ne'eman, Z., and S. Razin. 1975. Characterization of the myco-plasma membranes. Biochim. Biophys. Acta 375:54-68.

14. Op den Kamp, J. A. F., J. de Gier, and L. L. M. Van Deenen.1974. Hydrolysis of phosphatidylcholine liposomes by pancre-atic phospholipase A2 at the transition temperature. Biochim.Biophys. Acta 345:253-256.

15. Pollack, J. D. 1986. Metabolic distinctiveness of ureaplasmas.Pediatr. Infect. Dis. J. 5(Suppl. 6):S305-S307.

16. Quinn, P. A. 1988. Mycoplasma infection of the fetus andnewborn, p. 107-151. In D. G. Scarpelli and G. Migaki (ed.),Transplacental effects on fetal health. Alan R. Liss, Inc., NewYork.

17. Quinn, P. A. and L. U. Arshoff. 1986. Serotypes of Ureaplasmaurealyticum found in fertile and infertile couples. Pediatr. In-fect. Dis. 5:S352.

18. Quinn, P. A., J. E. Gillan, T. Markestad, et al. 1985. Intrauterineinfection with Ureaplasma urealyticum as a cause of fatalneonatal pneumonia. Pediatr. Infect. Dis. J. 4:538-543.

19. Quinn, P. A., S. Rubin, H. C. S. Li, D. M. Nocilla, S. E. Read,and M. Chipman. 1983. Serological evidence of Ureaplasmaurealyticum infection in neonatal respiratory disease. Yale J.Biol. Med. 56:565-572.

20. Razin, S. 1983. Cell lysis and isolation of membranes, p.225-243. In S. Razin and J. G. Tully (ed.), Methods in my-coplasmology. Academic Press, Inc., New York.

21. Razin, S. 1985. Mycoplasma adherence, p. 161-202. In S. Razinand M. F. Barile (ed.), The mycoplasmas. Academic Press,Inc., New York.

22. Rottem, S., L. Adar, Z. Gross, Z. Ne'eman, and P. J. Davis.

J. CLIN. MICROBIOL.

on June 25, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

PHOSPHOLIPASE ACTIVITY LOCALIZATION IN U. UREALYTICUM 1503

1986. Incorporation and modification of exogenous phosphati-dylcholines by mycoplasmas. J. Bacteriol. 167:299-304.

23. Rottem, S., M. Hasin, and S. Razin. 1973. Differences insusceptibility to phospholipase C of free and membrane-boundphospholipids of Mycoplasma hominis. Biochim. Biophys. Acta323:520-531.

24. Rottem, S., and S. Razin. 1966. Adenosine triphosphate activityof mycoplasma membranes. J. Bacteriol. 92:714-722.

25. Shepard, M. C., and C. D. Lunceford. 1978. Serological typingof Ureaplasma urealyticum isolates from urethritis patients byan agar growth inhibition method. J. Clin. Microbiol. 8:566-574.

26. Shirvan, M. H., S. Schuldiner, and S. Rottem. 1987. Control of

sodium fluxes in Mycoplasma gallisepticum. Isr. J. Med. Sci.23:384-388.

27. Thuren, T., P. Vainio, J. A. Virtanen, P. Somerharjie, K.Blomquist, and P. K. J. Kunnunen. 1984. Evidence for thecontrol of the action of phospholipases A by the physical stateof the substrate. Biochemistry 23:5129-5134.

28. Van den Bosch, H. 1980. Intracellular phospholipases A. Bio-chim. Biophys. Acta 604:191-246.

29. Van Golde, L. M. G., R. N. McElhaney, and L. L. M. V. Deenen.1971. A membrane-bound lysophospholipase from Mycoplasmalaidlawii strain B. Biochim. Biophys. Acta 231:245-249.

VOL. 29, 1991

on June 25, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from