Giardia lamblia: Characterization of proteinase activity in trophozoites

EXPERIMENTAL

PARASITOLOGY

Giardia

68, 168-175

lamblia:

(1989)

Characterization of Proteinase in Trophozoites

Activity

DAWN F. HARE, EDWARD L. JARROLL,ANDDONALDG.LINDMARK Department of Biology, Cleveland State University, 1983 E. 24th Street, Cleveland, Ohio 44115, U.S.A. HARE, D. F., JARROLL, E. L., AND LINDMARK, D. G. 1989. Giardia lamblia: Characterization of proteinase activity in trophozoites. Experimental Parasitology 68, 168-175. The proteinase activity of Giardia lamblia trophozoites, Portland 1 strain, was characterized with respect to substrate specificities and inhibitor sensitivities. Proteinase activity with urea-denatured hemoglobin (UDH), a-N-benzoyl-DL-arginine-2-naphthylamide (BANA), and a-N-benzoyl-argininamide (BAA) as substrates exhibited pH optima of 5.8,3.8, and 5.0, respectively. For BANA, the apparent K, was 0.20 n&f and the V,,,, was 2.56 pi&f. For BAA, the apparent K,,, was 4.0 mM and the V,,,, was 8.69 PM. Dithiothreitol (DTT, 5 mM) enhanced proteinase activity threefold for UDH, fourfold for BAA, and fivefold for BANA. Iodoacetamide, L-tosylamide-2-phenylethyl chloromethyl ketone (TPCK), and N-a-p-tosyl+lysine chloromethyl ketone (TLCK), each at 1 mM, inhibited proteinase activity by greater than 90% with BANA and BAA. Iodoacetamide inhibited proteinase activity by 35% with UDH; TPCK and TLCK inhibited activity greater than 70% with UDH. Activity on BAA was inhibited by 91% with Zn*+ and activity on UDH was inhibited by 30% with Cu*+. Virtually complete inhibition of proteinase activity on BANA and BAA was obtained with leupeptin and chymostatin at 1 &ml. Pepstatin A, chelators, and other heavy metals had no apparent effect on proteinase activity. Two polypeptide bands (ca. 10.5and 40 kDa) indicative of proteinase activity were visualized by sodium dodecyl sulfate-gelatin polyacrylamide gel electrophoresis. The 105 kDa band was visible over the pH range of 4 to 7, but with greater intensity from pH 5 to 7. The 40 kDa band, while present at pH 5, was most intense at pH 6 and 7. The proteinase activity of these two bands on gelatin was D’lT dependent and was inhibited by 1 mM iodoacetamide, 1 mM TPCK, and 1 n&f TLCK, but not by up to 1.2 mM phenylmethylsulphonyl fluoride. These results indicate that the proteinase activities in G. lamblia are primarily of the cysteine type. 8 1989 Academic &SS, II-L INDEX DESCRIPTORS AND ABBREVIATIONS: Giardia lamblia; Protozoa, parasitic; Proteinases, characterization; Phosphate-buffered saline (PBS); Dimethyl sulfoxide (DMSO); Urea-denatured hemoglobin (UDH); a-N-benzoyl-DL-arginine-2-naphthylamide (BANA); a-N-benzoyl-argininamide (BAA); L-tosylamide-2-phenylethyl chloromethyl ketone (TPCK): N-a-p-tosyl-L-lysine chloromethyl ketone (TLCK); phenylmethylsulfonyl fluoride (PMSF); Sodium dodecyl sulfate (SDS); Polyacrylamide gel electrophoresis (PAGE); SDSgelatin PAGE (GEL-PAGE); Ethylenediaminetetraacetic acid (EDTA); Dithiothreitol W-U.

have been shown to function as cellular regulators (Bond and Butler 1987; Chertow 1981; Holzer and Tschensche 1979), and differentiation (Holzer and Tschensche 1979; North and Coombs 1981) and pathogenesis factors (Holzer and Heimich 1980; Gadasi and Kessler 1983; Munoz et al. 1982). However, little is known of the proteinase of Giurdiu. In fact, to date, only one study of proteinase activity in Giurdiu using UDH and BANA as substrates has been reported. Lindmark (1988), in a study of G.

INTRODUCTION Giardia lamblia parasitizes the upper small intestine of humans, and it is the most common cause of waterborne intestinal disease in the United States. This intestinal flagellate exists in two morphologically distinct forms: a trophozoite and a cyst. However, little is known of the mechanisms by which this flagellate differentiates (i.e., encysts and excysts) or by which it causes pathogenesis. In other systems, proteinases 168 0014-4894/89 Copyright Au ri@s

$3.00

0 1989 by Academic Press, Inc. of reproduction in any form reserved.

Giardia

lamblia:

trophozoite hydrolases, showed that proteinase activity with either UDH or BANA as substrates is latent and tightly bound to a lysosome-like particle population with an equilibrium density of 1.15 in sucrose. Since proteinases have been shown, in other systems, to function in various aspects of cellular regulation, differentiation, and pathogenesis, this study was designed to begin the characterization of the proteinase activities in G. lamblia (Portland 1 strain) trophozoites. Such information will serve as a reference for the characterization of proteinases from encysting and excysting trophozoites, and from cysts. lamblia

MATERIALS

AND METHODS

Parasite cultivation. Giardia lamblia (Portland 1 strain) trophozoites were grown axe&ally in TYIS-33 medium (Keister 1983) moditied by omitting bile. Cells in late log phase were harvested after 96 hr of growth at 37 C by centrifugation at room temperature, washed twice, and resuspended in ice-cold 0.25 M sucrose for enzymatic studies and for electrophoresis. Enzyme assays. Proteinase activity (EC 3.4.1) was measured by (1) release of peptides, determined with fluorescamine, from UDH, (2) release of 2naphthylamine, determined by fluorescence, from BANA, and (3) release of NH,+, determined by fluorescence, from BAA (McLaughlin and Milller 1979). The UDH and BAA assays were performed in the presence of 0.1% Triton X-100, and the BANA assay contained 12.5% DMSO (McLaughlin and Miiller 1979). All enzymes were assayed at 30 C for 30 min, and pH optima were determined in 0.1 M acetateborate-citrate buffer. Enzyme units are defined as the amount of enzyme necessary to degrade 1 umole of substrate/mm under the assay conditions stated. Protein was measured by the method of Bradford (1976). Activators and inhibitors. Activators and inhibitors were prepared: In deionized water, dithiothreitol (DTT), ethylenediaminetetraacetic acid (EDTA), HgCl,, Z&O,, CuSO,, MgCl,, CaC12, leupeptin, iodoacetamide, and iodoacetate; in 100% methanol, (PMSF, TLCK, TPCK, and pepstatin A; and in 100% DMSO, chymostatin. Preincubation of inhibitor with enzyme was performed only when no inhibition was detected without preincubation. Electrophoresis. Prior to electrophoresis, intact trophozoites in sucrose were lysed by vigorously mixing them in a pipette with an equal amount of sample loading buffer (0.75 g Tris, 2 g SDS, 2.5 ml B-mercaptoethanol, 5 ml glycerol, and 50 ml water) (Lockwood

169

PROTEINASES

et al. 1987). Proteinases were visualized by SDSPAGE using gels which contained 10% acrylamide and 0.2% (w/v) copolymerized gelatin GEL-PAGE as described by Lockwood et al. (1987) or 0.2% copolymerized casein. Gels were developed at 37 C in buffers at pH 4 (0.1 M acetate), 5 (0.1 M acetate), 6 (0.1 M citrate phosphate), and 7 (0.1 M citrate phosphate) with and without 5 mM DTT for 4,8, and 24 hr. Tritrichomonas foetus was used as a control organism prior to using G. lamblia in this system. Our results with T. foetus were similar to those obtained by Lockwood et al. (1987). RESULTS

The pH optima for proteinase activity with BANA as a substrate was 3.8 (Fig. 1). However, proteinase activity of up to 50% maximum activity was observed over a pH range of from 3 to 8. The optimal proteinase activity with UDH as a substrate was at pH 5.8, with up to 50% maximum activity over a range from pH 5 to 7.2 (Fig. 2). With BAA, optimal proteinase activity was detected at pH 5 while a range of up to 50% maximum activity was exhibited from pH 4 to 7 (Fig. 3). The specific activity with respect to each substrate is presented in Table I. The highest specific activity (ca. 124.8 mU/mg protein) was exhibited with UDH as the substrate followed by the specific activity (ca.

700 -

fj

30-

.I

o-

? P

4020 t

FIG. 1. The effect of pH on Giardia lamblia trophozoite proteinase activity on a-N-benzoyl-Dr.- arginine2-naphthylamide. Assays were performed at 30 C for 30 min in 0.1 it4 acetate-boratiitrate buffer. Vertical bars represent the standard error of the plotted means. Maximum activity (100%) = 1.7 + 0.4 mU/mg protein.

170

HARE, JARROLL, AND LINDMARK

120

-“D”

100

-

TABLE I Specific Activities of Proteinases

Activity mU/mg protein + SD (No. of determinations)

B $

90-

!

90-

t J!

40-

20

0

Proteinase (BANA) F’roteinase (BAA) Eroteinase (UDH)

1.7 + 0.4 (3) 3.6 f 0.5 (3) 124.8 f 29 (3)

Note. All assays were performed at 30 C for 30 min using trophozoite preparations as described under Materials and Methods. Substrates used were BANA, BAA, and UDH. SD, standard deviation.

-

4

5

6

7

a

9

P”

FIG. 2. The effect of pH on G. lamblia trophozoite proteinase activity on urea-denatured hemoglobin. Assays were performed at 30 C for 30 min in 0.1 M acetateborate-citrate buffer. Vertical bars represent the standard error of the plotted means. Maximum activity (100%) = 124 2 29 mU/mg protein.

3.5 mU/mg protein) with BAA. The lowest specific activity (ca. 1.7 mU/mg protein) was observed when BANA was the substrate. Table II presents the results of the proteinase kinetic studies. For BANA, the apparent K, was 0.2 mM and the V,, of 2.56

l&; the apparent Km for BAA was higher at 4.0 mM and the V,, was 8.69 t&f. The effect of activators on Giardiu proteinase activity is summarized in Table III. With BANA as the substrate, proteinase activity was increased fivefold above control levels with DTT at a concentration of 5 mM. A 30% increase above control levels was observed with EDTA at a 5 mM concentration. With UDH as the substrate, proteinase activity was increased threefold by the addition of 5 mM DTT, and by 50% with 5 mM EDTA. Proteinase activity with BAA was increased fourfold by the addition of 5 mM DTT, but no increase in this

MA

120

t

TABLE II Michaelis-Menton Kinetics of Giurdiu lambliu Proteinase Activity

loo -

6 B p

of Giardia lamblia

go-

Apparent K,,,

5

.6

go-

t

40 -

W4

B

Proteinase (BANA) Proteinase (BAA)

20 -

PH

FIG. 3. The effect of pH on G. lamblia trophozoite proteinase activity on a-N-benzoyl-rxirgininamide. Assays were performed at 30 C for 30 min in 0.1 M acetateborate-citrate buffer. Vertical bars represent the standard error of the plotted means. Maximum activity (100%) = 3.6 f 0.5 mU/mg protein.

0.2 4.0

cvh?

2.56 8.69

Note. All assays were performed at 30 C for 30 min using trophozoite preparations as described under Materials and Methods. K,,, and V_ were derived from Lineweaver-Burke plots using determinations from at least three sets of replicates. BANA substrate varied in concentration from 0.025 to 0.675 p&f; BAA substrate varied in concentration from 0.5 to 25 )LM. While the amount of protein used per tube for each data set was constant, the amount used per assay tube varied among data sets from 0.35 to 0.41 mg for BANA, and 1.5 to 2.2 mg for BAA.

Giardia

lamblia:

TABLE III Effect of Proteinase Activators on Giardia lamblia Proteinase Activity

171

PROTEINASES

TABLE IV Effect of Proteinase Inhibitors on Giardia lamblia Proteinase Activity Percentage of maximum activity

Percentage of maximum activity Activator

BANA

BAA

UDH

Control activity” Dithiothreitol (DTT)

100

100

100

518

405

300

136 565

99 400

320

EDTA DTT + EDTA”

150

Note. Activator final concentrations for results reported here were 5 mM. Substrates were BANA, BAA, and UDH. Values represent the mean of at least two sets of replicates reported in the percentage of maximum activity relative to the control which was 100%. 0 Control specific activities were 0.3 + 0.05 mU/mg protein (BANA), 0.9 k 0.08 mU/mg protein (BAA), and 43.2 k 4.2 mU/mg protein (UDH). ’ Ethylenediaminetetraacetic acid.

activity was observed with the addition of EDTA up to 10 mM. The results of inhibitor studies are summarized in Table IV. ZnSO, inhibited proteinase activity 91% with BAA as the substrate, and C&O, inhibited activity by 30% with UDH as the substrate. Significant inhibition of the substrates by other heavy metals was not detected. Iodoacetamide inhibited activity greater than 90% with BANA and BAA, and by 35% with UDH. TPCK and TLCK inhibited proteinase activity by greater than 90% with BANA and BAA, and by greater than 70% with UDH. PMSF inhibited proteinase activity by 75% with BAA as a substrate, but not with BANA or UDH. Leupeptin and chymostatin at a concentration of 1 l&ml of assaymixture inhibited proteinase activity with BANA by 89 and 97%, respectively (Fig. 4). With BAA, leupeptin at 0.5 l&ml of assay mixture and chymostatin at 1 t&ml of assay mixture gave 100% inhibition (Fig. 5). Pepstatin A did not inhibit proteinase activity with ei-

BANA

UDH

Control activity”

loo

loo

100

Heavy metals WA znso, cuso, MEI, C&I,

% 103 % 105 95

84 84 70 85 92

101 9 100 101 85

7:

65 85

1 94

81* 96 9 106

% 32 28 91

25 2

Inhibitor

Sulthydral reagents Iodoacetamide Iodoacetate

Other inhibitors PMSF (in methanol) TPCK(inmethanol) TLCK (in methanol) Methanol

BAA

9:

Note. The final concentration of all inhibitors was 1 mM. Substrates were BANA, UDH,and BAA. Values represent a mean of at least two sets of replicates reported in the percentage of maximum activity relative to the control which was 100%. 0 Control activities were 1.7 it 0.4 mU/mg protein (BANA), 124.8 ? 29 mU/mg protein (UDH), and 3.6 f 0.5 mU/mg protein (BAA). Assays were performed at 30 C for 30 min. b Inhibition required preincubation at 30 C for 5 min. Lack of inhibition by other inhibitors was not affected by preincubation.

ther BANA or BAA at concentrations of up to 1 @ml of assay mixture. The proteinase activity band pattern obtained using GEL-PAGE is presented in Fig. 6 and Table V. Even though gels were incubated for up to 24 hr in buffer, no additional bands were detected after 8 hr. l Chymostatin c

so

0

20

50

loo 200 Inhibitor (w/ml)

500

1000

FIG. 4. The effect of proteinase inhibitors of microbial origin on the hydrolysis of a-N-benzoylDL-arginine-2- naphthylamide by G. lamblia trophozoite proteinases. Assays were performed at 30 C for 30 min. Data points represent the means of at least two sets of replicates.

172

HARE,

JARROLL,

A Leuwptin l Chymostatin

hmbitw

hghlo

FIG. 5. The effect of proteinase inhibitors of microbial origin on the hydrolysis of a-iV-benzoylL-argininamide by G. lambliu trophozoite proteinases. Assays were performed at 30 C for 30 min. Data points represent the means of at least of two sets of replicates.

Bands were only faintly visible when the gels were developed in the absence of DTT. At pH 4 through 7 (Fig. 6), one polypeptide with an apparent M, value of 105 kDa appeared consistently but with less intensity at pH 4 than at pH 5 and above. Another polypeptide band with an apparent M, value of 40 kDa appeared consistently at

FIG. 6. Proteinase activity bands of G. lamblia trophozoites appearing in a representative sodium dodecyl sulfate-gelatin polyacrylamide gel. The samples were electrophoresed from cathode (top) to anode (bottom). The molecular weight markers, in ascending order, are carbonic anhydrase, egg albumin, bovine albumin, phosphorylase b, f3- galactosidase, and myosin. All lanes contained 0.2 mg protein. The M, values are given in kilodaltons. Lanes represent development at 37 C for 8 hr in (1) pH 4 (acetate buffer), (2) pH 5 (acetate buffer), (3) pH 6 (citrate phosphate buffer), and (4) pH 7 (citrate phosphate buffer). Gels contained 10% polyacrylamide and 0.2% gelatin.

AND

LINDMARK

pH 5 and above. A group of one to three faintly visible bands occasionally appeared at the higher pH values with apparent M, values ranging from 80 to 95 kDa. However, these bands were not considered major activity bands since they did not appear consistently. The use of casein as a substrate for proteinase activity in GEL-PAGE revealed no additional bands. The addition of TLCK, TPCK, or iodoacetamide at final concentrations of 1 mJI4in both the sample and developing buffers resulted in complete inhibition of all activity bands. PMSF did not noticeably inhibit the activity bands even at a concentration of 1.2 mM. DISCUSSION

Perez-Montford et al. (1987) reported that three of the four major proteinases of Entamoeba histolytica were of the cysteine type. McLaughlin and Mtiller (1979) and Lockwood et al. (1984, 1987) reported that proteinases of several trichomonads were cysteine proteinases. Our studies indicate that, like those in the other anaerobic protozoan parasites, the major proteinases of G. lambliu trophozoites are also cysteine type. This conclusion is based on the observation that: (1) Giurdia proteinase activity was optimal in the range of pH 3.8 to 5.8 which is consistent with the range for cysteine proteinases. Lindmark (1988) observed pH optima of 3.7 for BANA and 5.8 for UDH with Giardiu. These values more closely resemble those for the mammalian cysteine proteinase, cathepsin B which exhibits pH optima of from 3.5 to 5 (Otto 1971). (2) Proteinase activity was enhanced by DTT and inhibited by iodoacetamide, leupeptin, chymostatin, TLCK, TPCK, and with some substrates by Zn*+ and Cu*+. These substancesare known to inhibit cysteine type proteinases. Similar results were reported by Lindmark (1988) for iodoacetamide, Zn*+, and Cu*+ who also reported inhibition by p-chloromercuribenzoate. (3) Proteinase activity was not inhibited by

Giardia lamblia: PROTEINASES

173

TABLE V Apparent Molecular Weights of Giurdia lambliu Proteinases Detected by Sodium Dodecyl Sulfate-Gelatin Polyacrylamide Gel Electrophoresis pH4 103IIZ2(3)

PH 5 40 + 3 (3) 106+ 2 (3)

PH6 41 + 3 (3) 106+ 4 (3)

pH7 42 f 2 (3) 105+ 3 (3)

Note. Gels contained 10% polyacrylamide and 0.2% gelatin. M, values are in kDa 2 SD (No. of determinations). This table does not include the occasionally appearing polypeptides with M, values between 80 and 95 kDa.

pepstatin A or by chelating agents which are known inhibitors of aspartic and metalloproteinases, respectively. The fact that PMSF, formerly thought to be specific for serine type proteinases, inhibited activity with BAA may be explained by the fact that PMSF has been reported to inhibit certain cysteine type proteinases (Barrett 1977; Bond and Butler 1987). PMSF did not inhibit Giardia proteinase activity with the other substrates tested. The specific activities of the G. lamblia proteinases with UDH, BAA, and BANA as substrates followed those observed by McLaughlin and Mtiller (1979) for T. foetus, i.e., the descending order of specific activities was UDH > BAA > BANA. Lindmark (1988) observed virtually identical specific activities to those reported in this study for UDH and BANA. BAA was not used as a substrate in the Lindmark (1988) study. While the specific activity was somewhat greater for BAA than BANA, the lower K, for BANA suggests that, at least in this crude extract, there was a greater proteinase affinity for BANA than BAA. Further kinetic studies, using purified proteinases, will be required to confirm these data. The proteinase activity in G. lamblia trophozoites, at least of the Portland 1 strain, observed by Lindmark (1988) and detected in the present study, appears to be attributable to two major proteinases with approximate A4,values of 105 and 40 kDa. At least with the gelatin substrate, the 40 kDa proteinase seems most active in the pH range

from 5 to 7 while the activity of the 105 kDa proteinase appears over the entire range from pH 4 to 7. When present, the polypeptides with M, values between 80 and 95 appear at the higher pH values, and their failure to appear may be related to their concentration in the cells. The GEL-PAGE technique for observing proteinase activity assumes that proteinases are not irreversibly denatured by the SDS treatment, but that activity is restored following the incubation in Triton X-100 which removes the SDS (Lockwood et al. 1987). While this assumption may not necessarily be valid in all cases, Perez-Montford et al. (1987) noted that when they estimated the number of proteinases from E. histolytica by both gelatin gels and the o-Zmacroglobulin methods, similar results were observed. The Giardia trophozoites used in this study exhibited fewer proteinase bands in GEL-PAGE than either E. histolytica or the trichomonads. Perez-Montford et al. (1987) observed four major bands in E. histolytica, a protozoan residing in the large intestines. Lockwood et al. (1987) detected four bands for the intestinal trichomonads, Pentatrichomonas hominis, also found in the large intestines, and 8 and 11 bands for the urogenital species T. foe&s and Trichomonas vaginalis, respectively. Lockwood et al. (1987) speculated that in the trichomonads the number of proteinases varies depending on the parasite’s location in the host. If this is true for protozoan parasites in general, then it could explain why Giardia, the only small intestine dwelling proto-

174

HARE,

JARROLL,

zoan, only exhibits two major proteinases. By extension of the Lockwood et al. (1987) idea, it is possible that because of its habitat in the small intestine where host proteases are already present, even fewer endogenous proteinases may be required by Giardia than are required for large intestinal and urogenital protozoa. However, other explanations could be envisioned: (1) The length of time in culture for a particular strain might affect the number of proteinases exhibited, (2) strain differences might account for the number exhibited, (3) the substrates which are copolymerized with the polyacrylamide may affect the number visualized, although in this case the substitution of the casein for gelatin made no detectable difference, or as suggested earlier, (4) irreversible SDS sensitivity of some proteinases but not others may alter the number detected. These possibilities remain to be examined in Giardia. Cysteine appears to be quite important to the biology of Giardia. The requirement of cysteine for growth in vitro has been demonstrated (Gillin and Diamond 1981)as well as the expression of thiol groups (Gillin et al. 1984) and antigenically variable cysteine-rich proteins on the surface of Giardia trophozoites (Adam et al. 1988). The observation, in this paper, that the major proteinases of Giardia trophozoites are of the cysteine type represents yet another facet of Giardia physiology in which cysteine is important. The invasiveness and virulence of E. histolytica has been ascribed, at least in part, to one or more proteinases which exhibit collagenase activity (Gadasi and Kessler 1983; Munoz et al. 1982; Munoz et al. 1984). Whether or not the proteinases of Giardia function as pathogenesis factors remains to be determined. It seems unlikely that the Giardia proteinases which are tightly bound to lysosome-like particles (Lindmark 1988)are acting as pathogenesis factors; however, we have not yet determined if growing cultures are producing ex-

AND

LINDMARK

tracellular proteinases that may be acting as such factors. In the flagellate protozoan, Leishmania mexicana, North and Coombs (1981) suggested that the increased level of cysteine type proteinase in the amastigote stage was a stage-specific function of the organism. They speculated that the higher proteinase activity in the amastigote may be a factor in the transformation of a promastigote to an amastigote. Whether or not proteinases are involved in the differentiation of Giardia from cyst to trophozoite (excystation) or from trophozoite to cyst (encystment) is currently under investigation. This study should be facilitated by our ability to cultivate trophozoites in vitro, to induce in vitro excystation, and recently to induce in vitro encystment (Gillin et al. 1987; Schupp et al. 1988). ACKNOWLEDGMENTS

The authors acknowledge the Cleveland State University College of Graduate Studies grant to Dawn Hare and the Ohio Board of Regents Academic Challenge Program for support of this project. REFERENCES

ADAM, R., AGGARWAL, A., LAL, A., DE LA CRUZ, V., MCCUTCHAN, T., AND NASH, T. 1988. Antigenic variation of a cysteine-rich protein in Giardia lamblia. Journal of Experimental Medicine 167, 109118.

BARRETT, A. J. 1977. Proteinases in mammalian cells and tissues. In “Research Monographs in Cell and Tissue Physiology” (J. T. Dingle, Ed.), pp. 10-18. North-Holland, New York. BOND, J., AND BUTLER, P. 1987. Intracellular proteases. Annual Review of Biochemistry 56,333-364. BRADFORD, M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72, 248-254. CHERTOW, B. S. 1981. The role of lysosomes and proteases in hormone secretion and degradation. Endocrinology Review 2, 137-165. GADASI, H., AND KESSLER, E. 1983. Correlation of virulence and collagenolytic activity in Entamoeba histolytica. Infection and Immunity 39, 529-531. GILLIN, F., AND DIAMOND, L. 1981. Entamoeba histolytica and Giardia lamblia: Effects of cysteine and oxygen tension on trophozoite attachment to glass

HARE,

JARROLL,

and survival in culture media. Experimental Parasitology 52, 9-17. GILLIN, F., REINER, D., LEVY, R., AND HENKART, P. 1984. Thiol groups on the surface of anaerobic protozoa. Molecular and Biochemical Parasitology 13, 1.

GILLIN, F., REINER, D., GAULT, M., DOUGLAS, H., WUNDERLICH, A., AND SAUCH, J. 1987. Encystation and expression of cyst antigens by Giardia lamblia in vitro. Science 235, 104&1043. HOLZER, H., AND HEINRICH, P. C. 1980. Control of proteolysis. Annual Review of Biochemistry 49, 6391. HOLZER, H., AND TSCHENSCHER, H. 1979. “Biological Function of Proteinases,” pp. W-115. SpringerVerlag, Berlin. KEISTER, D. B. 1983. Axenic culture of Giardia lamblia in TYI-S-33 medium supplemented with bile. Transactions of the Royal Society of Tropical Medicine and Hygiene 11, 487-488. LINDMARK, D. G. 1988. Giardia lamblia: Localization of hydrolase activity in lysosome-like organelles in trophozoites. Experimental Parasitology 65, 141147. LOCKWOOD, B., NORTH, M., AND COOMBS, G. 1984. Trichomonas vaginalis, Tritrichomonas foetus, and Trichomitus batrachorum: Comparative proteolytic activity. Experimental Parasitology 58, 245-253. LOCKWOOD, B., NORTH, M., SCOTT, K., BREMNER, A., AND COOMBS, G. 1987. The use of a highly sensitive electrophoretic method to compare the proteinases of trichomonads. Molecular and Biochemical Parasitology 24, 8%95.

AND

LINDMARK

175

MCLAUGHLIN, J., AND MULLER, M. 1979. Purification and characterization of a low molecular weight thiol proteinase from the flagellate protozoan Tritrichomonas foetus. Journal of Biological Chemistry 254, 1526-1533. MUNOZ, M., CALDERON, J., AND ROJKIND, M. 1982. The collagenase of Entamoeba histolytica. Journal of Experimental Medicine 155, 42-51. MUNOZ, M., ROJKIND, M., CALDERON, J., TANIMOTO, M., ARIAS-NEGRETE, S., AND MARTINEZPALOMO, A. 1984. Entamoeba histolytica: Collagenolytic activity and virulence. Journal of Protozoology 31, 468-470. NORTH, M., AND COOMBS, G.1981. Proteinases of Leishmania mexicana amastigotes and promastigates: Analysis by gel electrophoresis. Molecular and Biochemical Parasitology 3, 293-300. Oreo, K. 1971. In “Tissue Proteinases” (A. J. Barrett and J. Dingle, Eds.), pp. l-28. American Elsevier, New York. PEREZ-M• NTFORD, R., OSTOA-SALOMA, P., VELASQUEZ-MEDINA, L., MONTORT, I., AND BECKER, I. 1987. Catalytic classes of proteinases of Entamoeba histolytica. Molecular and Biochemical Parasitology 26, 87-98. SCHUPP, D., JANUSCHKA, M., SHERLOCK, L., ERLANDSEN, S., MEYER E., BEMRICK, W., AND STIBBS, H. 1988. Production of viable Giardia cysts in vitro: Determination by fluorogenic dye staining, excystation, and animal infectivity in the mouse and Mongolian gerbil. Gastroenterology 95, I-10. Received 5 July 1988; accepted with revision 28 Sep tember 1988