Purification and Characterization of a Ca2l-Dependent Membrane ...

JOURNAL OF BACTERIOLOGY, Apr. 1987, p. 1626-1631

Vol. 169, No. 4

0021-9193/87/041626-06$02.00/0 Copyright C 1987, American Society for Microbiology

Purification and Characterization of a Ca2l-Dependent Membrane Peptidase Involved in the Signaling of Mating Pheromone in Rhodosporidium toruloides TOKICHI MIYAKAWA,* MIHOKO KAJI, YONG KEE JEONG, EIKO TSUCHIYA, AND SAKUZO FUKUI

Department of Fermentation Technology, Faculty of Engineering, Hiroshima University, Saijo, Higashi-Hiroshima, 724 Japan Received 12 June 1986/Accepted 12 January 1987

A mating-type-specific, membrane thiol peptidase (referred to as trigger peptidase) that seems to play a key role in the transmembrane signaling of the lipopeptidyl mating pheromone rhodotorucine A at the cell surface of mating type a cells of Rhodosporidium toruloides (T. Miyakawa, M. Kaji, T. Yasutake, Y. K. Jeong, E. Tsuchiya, and S. Fukui, J. Bacteriol. 162:294-299, 1985) was purified to homogeneity and characterized. The following lines of evidence support the contention that the enzyme we purified was the trigger peptidase: (i) the identical specificity of hydrolysis at the Arg-Asn sequence of rhodotorucine A and the sensitivity of the reaction to sulfhydryl-blocking reagents; (ii) the identical specificity for the substrate, with a strict requirement for the presence of the lipid moiety; and (iii) the absence of the corresponding activity in the pheromone-producing strain (mating type A) and in a sterile mutant strain, M-39 (type a), that lacks trigger peptidase activity in vivo. The apparent molecular weight of trigger peptidase was estimated to be 68,000 by Sepharose 6B gel filtration in the presence of octylglucoside and 63,000 by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Trigger peptidase alone was inactive but exhibited enzymatic activity with the simultaneous addition of Ca2', membrane phospholipids, and a nonionic detergent such as octylglucoside. The concentration of Ca2' required for maximum activation was approximately 1 mM. Only Mn2+ could replace Ca2' at comparable concentrations. Among the phospholipids tested, only phosphatidylserine and phosphatidylethanolamine supported trigger peptidase activation. Solubilized trigger peptidase was strongly inhibited by antipain and phosphoramidon.

Intercellular communication by means of diffusible signaling molecules plays an important role in the control of cellular functions in eucaryotic organisms. In the life cycle of the heterobasidiomycetous yeast Rhodosporidium toruloides, the alteration from the sexual to asexual mode of growth is achieved by conjugation of two yeast-form haploid cells with compatible mating types, a and A. The sexual cell interaction is mediated by mating-type-specific mating pheromones secreted by the haploid cells (1). The haploid vegetative cells differentiate to gamete cells responding to the mating pheromone of the opposite mating type. The sexual differentiation is assumed to be a preparatory process for mating: the G1-arrested yeast-form vegetative cells are converted to gamete cells which elongate a mating tube toward the mating partner cell, resulting in cell fusion at the apex of the tube (1). The resulting conjugant grows as dicaryotic cells with hyphal morphology. Rhodotorucine A, the mating pheromone produced by the A cells, elicits sexual differentiation specifically in the a cells. Rhodotorucine A has been shown to be a lipopeptide, a polyisoprenylated undecapeptide of the following sequence: H-Tyr-Pro-GluIle-Ser-Trp-Thr-Arg-Asn-Gly-Cys (S-farnesyl)-OH (9). Both the farnesyl and peptidyl moieties are indispensable for the biological activity of the pheromone (22). The unique lipopeptidyl structure is a common feature of all three characterized mating pheromones of the heterobasidiomycetous yeast species (9, 20, 21). Accordingly, it was expected that a unique signaling mechanism might be involved in the action of these mating pheromones. During our studies on the signaling mechanism of rhodotorucine A, we demon*

strated that the mating pheromone is hydrolyzed at the surface of the target cell (mating type a) by a mating-typespecific endopeptidase and that the reaction is necessary for the development of the pheromone responses in the recipient cell (17). We also showed that the enzyme responsible for pheromone metabolism is a cell surface thiol peptidase highly specific for rhodotorucine A. The external disposition of the active site of the enzyme with respect to the membrane has been demonstrated by using a membraneimpermeable inhibitor of the peptidase (15). Furthermore, we have showed in an in vitro system that phosphorylation of proteins of the particulate fraction occurs in response to the hydrolysis of the pheromone (15, 16). After pheromone metabolism, a very rapid Ca2+ influx into the cell occurs as an immediate physiological response of the a cell to the pheromone (17a, 18). Since the peptidase seems to play a key role at the surface of a cells in the trigger reaction of the signaling, we refer to this enzyme as trigger peptidase. We describe in this report the solubilization, purification, and characterization of the trigger peptidase. MATERIALS AND METHODS

Materials. Antipain, chymostatin, E-64, leupeptin, pepstatin, and phosphoramidon were purchased from Peptide Institute, Inc., Osaka, Japan. Phospholipids were obtained from Pharmacia P-L Biochemicals. Iodobeads were purchased from Pierce Chemical Co. N-Octyl-p-D-glucopyranoside (OG) was obtained from Wako Pure Chemical Industries, Osaka, Japan. Phenylmethylsulfonyl fluoride (PMSF), dithiothreitol (DTT), concanavalin A (ConA), 5,5'dithiobis-(2-nitrobenzoic acid) (DTNB), and N-tosyl-Lphenylalanine chloromethyl ketone-trypsin were from Sigma

Corresponding author. 1626

VOL. 169, 1987

Ca2+-DEPENDENT MEMBRANE PEPTIDASE OF R. TORULOIDES

Chemical Co. All other reagents were of the highest purity commercially available. Na125I was purchased from Amersham Corp. Microorganisms and growth conditions. Haploid strains of R. toruloides IFO 0559-M919 (mating type A), IFO 0880M1057 (mating type a), and M-39, a sterile tnutant derived from the latter strain, and the conditions for cell growth were previously described (17). Enzyme assay. Trigger peptidase was determined biologically by its ability to decrease the pheromonal effect of rhodotorucine A, since no proper chromogenic substrates for the enzyme have been found because of its very high specificity for rhodotorucine A. Trigger peptidase was routinely assayed by measuring the ability of DTNB-sensitive peptidase to decrease the biological effect of rhodotorucine A in the absence of DTNB. The reaction mixture (0.2 ml) contained 10 mM Tris hydrochloride (pH 7.3), 1 mM CaCl2, 0.1 mM DTT, 0.3 mM OG, 0.07 mM phospholipids extracted from a cells, 8 U of rhodotorucine A, and the enzyme preparation to be assayed. After incubation for 60 min at 280C, the reaction was stopped by heating the mixture in a boiling water bath for 1 min. The remaining pheromonal activity was determined biologically by a serial dilution method as previously described (15). The enzyme assay was performed in both the presence and absence of 1 mM DTNB. One unit of the peptidase was defined as that amount of enzyme which inactivated 1 U of rhodotorucine A in 1 min. The difference in the activities determined in the absence and presence of DTNB was presumed to be the trigger peptidase activity. During this assay, the reaction proceeded linearly with time, and the activity was proportional to the amount of enzyme used.

khodotorucine A and its analogs. Partially purified rhodotorucine A prepared as described previously (17) was used as the substrate for biological detetmination of trigger peptidase. Pure rhodotorucine A prepared by the procedure described previously (17) was used for 1251 iabeling of the pheromone. Chemically synthesized desfarnesyl rhodotorucine A was a gift from Y. Kamiya. 1251 labeling of rhodotorucine A and desfarnesyl rhodotorucine A and their purification were performed as described previously (17). Solubilization of trigger peptidase. Cells (mating type a) were grown at 280C in five 5-liter flasks that contained 2 liters each of YPG medium (17) on a rotary shaker (GlO Gyrotory shaker; New Brunswick Scientific Co., Inc.) at 200 rpm. The cells were harvested at the late logarithmic phase of growth (3 x 107 cells per ml) by centrifugation (6,000 x g for 10 min at 40C) and washed twice by suspension in 200 ml of 10 mM Tris hydrochloride (pH 7.3) (buffer A) followed by centrifugation. All of the following operations were performed at 0 to 4°C. The washed cells were suspended in 200 ml of buffer A containing 1 mM each PMSF, DTT, and EDTA and disrupted by passage through a French press (20,000 lb/in2), and the disrupted cells were pelleted by centrifugation (11,000 x g for 20 min). The pellets were washed three times with 300 ml each of supplemented buffer A (described above) suspended in 100 ml of buffer A containing 4% Nonidet P-40. Proteins were extracted by sonication for 1 mnin. The resulting suspension was left at 0°C for 1 h and centrifuged at 11,000 x g for 60 min, and the clear supernatant was obtained. Proteins in the pellet fraction were extracted again by repeating the procedure described above. The supernatant solutions thus obtained were combined for further purification. IDE52 ion-exchange column chromatography. The protein extract was adjusted to pH 8.2 with NaOH before applying it

1627

to a column of DE52 (1.9 by 40 cm) equilibrated with Tris hydrochloride (pH 8.2) containing 1 mM each PMSF and EDTA, 0.1 mM DTT, and 0.3 mM OG. After the column was washed with the same buffer until the optical density of the washing solution at 280 nm decreased to below 0.1 (ca. 1,500 ml), elution was performed with 400 ml of a linear gradient of NaCl (0 to 0.5 M) in the same buffer. Fractions of 4 ml each were collected. ConA-Sepharose 4B affinity chromatography. The active fractions from DE52 chromatography were combined and adjusted to pH 7.3 with HCl. CaCl2 was added to a final concentration of 2 mM. The protein sample was applied to a column of ConA-Sepharose 4B (1.5 by 22 cm) which was

equilibrated with buffer A containing 0.1I mM DTT, 0.3 mM OG, and 0.5 mM CaCl2. After the column was washed with the same buffer, elution was performed with 0.5 M methyla-D-mannopyranoside in the same buffer. ConA:-Sepharose 4B (0.5 mg of ConA per ml of gel) was prepared by the procedure described previously (11). Sepharose 6B gel filtration. The active fractions from Con A-Sepharose affinity chromatography were pooled and dialyzed against buffer A containing 0.5 mM CaC12, 0.1 mM DTT, 100 mM NaCl, and 0.3 mM OG. Proteins were concentrated to about 1/10 of the original volume by subsequent dialysis against a dry polyethylene glycol 4000 powder. The concentrated sample was then applied to a column of Sepharose 6B (1.9 by 74 cm) equilibrated with the same buffer. The molecular weight markers used were blue dextran (voided volume), 3-galactosidase (Mr = 540;000), catalase (Mr = 260,000), glucose oxidase (Mr = 160,000), lactoperoxidase (Mr = 78,000), and cytochrome c (Mr = 11,500). Extraction of phospholipids. Membrane phospholipids were extracted from the particulate fraction of French pressdisrupted a cells with chloroform-methanol (2:1) and partially purified by extraction to ether by the procedure described previously (4). Phospholipids were quantitated by determination of Pi in a perchloric acid digest of the lipids

(4).

Electrophoretic analysis in polyacrylamide gel. Purified trigger peptidase was labeled with 1251 by using lodo-beads (chloramine T-derivatized polystyrene beads) by the procedure described by Markwell (13). After being labeled the sample was dialyzed extensively against deionized water, lyophilized, and subjected to sodium dodecyl sulfatepolyacrylamide gel electrophoresis (12.5% acrylamide) by the system of Laemmli as previously described (14). Protein determination. Protein concentrations were measured by the method of Lowry et al. (12). RESULTS Purification of trigger peptidase. Trigger peptidase is localized in the particulate fraction of a cells (17). For purification of the enzyme, usually ca. 3 x 1011 cells of mating type a harvested at the logarithmic phase of growth (3 x 107 cells per ml) in a 10-liter culture were used as the starting material. Membranes in the particulate fraction of French press-disrupted cells were solubilized with 4% Nonidet P-40. The enzyme activity in this fraction was usually unmeasurable or very low in comparison with that of intact cells, possibly because of unknown inhibitory activity present in this fraction. As described below, purified trigger peptidase required simultaneous addition of Ca2 , phospholipids, and nonionic detergents (e.g., OG) for activity. Therefore, these components were included in all the reaction mixtures for enzyme assay.

1628

MIYAKAWA ET AL.

J. BACTERIOL. TABLE 1. Purification of trigger peptidasea

Purification step

Intact cell Cell envelope Nonidet P-40 extract DE52-cellulose chromatography ConA-Sepharose chromatography

Total protein

(mg)

7,200 1,100 500 0.48 0.15

Total

Sp act (U/mg of

activity

(U) 1,400 1,400 b 750 800

Purification

protein) 0.19 1.20

(fold) 1.0 6.3

1,600 5,300

8,400 28,000

a A total of 3 x 1011 celis of mating type a were used as the starting material. b_, Determination of the activity was not possible due to the presence of an unknown inhibitor in the extract.

The results of purification of trigger peptidase are given in Table 1. DE52 ion-exchange column chromatography proved to be very effective for the purification of trigger peptidase. Trigger peptidase was eluted from the column very early by a concentration gradient of sodium chloride, separating from most other proteins. Under the conditions used, only 0.1% of the proteins applied to the column were recovered in the active fraction, with an activity recovery of at least 50%. Further purification was achieved by ConA affinity chromatography. Occasionally, a portion of a DTNB-insensitive proteolytic activity which was eluted immediately after the trigger peptidase fractions from DE52 column chromatography contaminated the active fractions of trigger peptidase. However, the contaminating activity was completely eliminated by the next step, ConA affinity chromatography. After the affinity chromatography, purification of trigger peptidase was achieved 28,000-fold relative to the specific enzyme activity of the cell. Trigger peptidase is a mating-type-specific enzyme that is missing in A cells in vivo (17). Also, a sterile mutant (M-39) of an a strain lacking trigger peptidase activity in vivo has been identified (17). To confirm that the enzyme we purified was trigger peptidase, the activity in partially purified preparations (ConA affinity chromatography fraction) identically prepared from these strains as from a cells by the procedure shown in Table 1 was measured. As expected, DTNBsensitive peptidase activity was not present at any detectable levels in the preparations from these strains (data not shown). Pattern of hydrolysis of rhodotorucine A by trigger peptidase. Intact a cells hydrolyze rhodotorucine A by the trypsin-type specificity at the Arg-Asn sequence of the pheromone, producing two segments, an N-terminal octapeptide and a C-terminal farnesyl tripeptide (8, 17). The pattern of hydrolysis of rhodotorucine A by purified trigger peptidase was examined. The electrophoretic pattern of the hydrolysis products of ['25I]rhodotorucine A labeled at the N-terminal tyrosine residue, obtained after incubation with the enzyme (ConA affinity chromatography fraction), is shown in Fig. 1A. In agreement with the in vivo result, a labeled peptide fragment with an electrophoretic mobility identical to that of the trypsin hydrolysate was produced by trigger peptidase. The conformity of the two products was verified by thin-layer chromatography (17) (data not shown). The hydrolysis of rhodotorucine A was completely inhibited by the addition of 1 mM DTNB, an inhibitor of trigger peptidase in vivo (15). Other sulfhydryl-blocking reagents such as N-ethylmaleimide also inhibited the peptidase (data not shown). Trigger peptidase has a very high substrate specificity for rhodotorucine A in vivo: desfarnesyl rhodotorucine A, a biologically inactive analog of rhodotorucine A that has the complete amino acid sequence but lacks the farnesyl residue, does not serve as the substrate for trigger peptidase

(17). Hydrolysis of the analog by the purified enzyme was examined. The peptide fragment corresponding to the Nterminal octapeptide of rhodotorucine A was not produced from this' substrate, even with solubilized enzyme (Fig. 1B), suggesting that farnesyl residue of the pheromone is involved in substrate recognition by the peptidase. The results presented above strongly suggest that the purified enzyme was the trigger peptidase. Molecular weight of trigger peptidase. The molecular weight of trigger peptidase was determined under nondenaturing and denaturing conditions. The activity of trigger peptidase was eluted in a fraction corresponding to a molecular size of approximately 68,000 daltons by gel filtration through Sepharose 613 in the presence of 0.3 mM OG (Fig. 2A). The Sepharose 6B fractions containing the enzyme were pooled and further analyzed by sodium dodecyl sulfatepolyacrylamide gel electrophoresis. For detection of small amounts of proteins, the proteins of the pooled fraction were labeled with 1251I before electrophoresis. The autoradiograph shown in Fig. 2B demonstrated that the trigger peptidase obtained was essentially homogeneous and that the enzyme possibly consists of a single polypeptide chain with a molecular weight of 63,000. The heavy band at the electrophoretic front appeared to be the result of an iodination artifact, since a similar band was present even in a sample iodinated in the absence of protein. Properties of trigger peptidase. On the basis of the data

A

B

_ 3 a

b

c

d

*

e

f

g

FIG. 1. Pattern of hydrolysis of rhodotorucine A and desfarnesyl rhodotorucine A by trigger peptidase. [125I]rhodotorucine A or [1251]desfarnesyl rhodotorucine A (ca. 20,000 cpm each) labeled at the N-terminal tyrosine was incubated with ConA affinity chromatography-purified trigger peptidase under the conditions described in Materials and Methods for the biological assay of trigger peptidase. The samples were analyzed by thin-layer electrophoresis as described previously (14). (A) [1251]rhodotorucine A as the substrate. Lanes: a, no addition; b, with trypsin; c, with trigger peptidase; d, with trigger peptidase and 1 mM DTNB. (B) [tIII]desfarnesyl rhodotorucine A as the substrate. Lanes: e, no addition; f, with trypsin; g, with trigger peptidase. Arrows 1, 2, and 3 indicate [125I]rhodotorucine A, [I251]desfarnesyl rhodotorucine A, and the N-terminal octapeptides of [251I]rhodotorucine A and [1251I]desfarnesyl rhodotorucine A, respectively.

_ .I


VOL. 169, 1987

I1

E1

A 600 4001

200*

0

- loC

-cn 3:

5a U 4)

-5

20l

0k

.

a

TABLE 2. Activatiop of trigger peptidase

Mr (1O3)

Enzyme activity

Reaction system

(U/ml)

Enzyme alone ......................................

0

Completea ...........................................

11.0 0 2.3 1.5

-Ca2.

.............................................

-Phospholipids ...................................

2

-Nonidet P-40 ......................

The complete reaction mixture contained 10 mM Tris hydrochloride (pH 7.3), 8 U of rhodotorucine A, 0.1 mM DTT, 1 mM CaC12, 0.3 mM OG, 50 ,g of phospholipids extracted from a cells per ml, and ConA affinity chromatography-purified enzyme preparation. a

3

-463

4

Aft

..a I 100

1629

200

Elution volume, ml FIG. 2. Determination of molecular weight of trigger peptidase. (A) The active fraction from ConA affinity chromatography was analyzed by Sepharose 6B gel filtration in the presence of 0.3 mM OG. The molecular weight markers used were as follows: 1, 3-galactosidase (Mr = 540,000); 2, catalase (M, = 260,000); 3, glucose oxidase (Mr = 160,000); 4, lactoperoxidase (Mr = 78,000); 5, cytochrome c (Mr = 11,500). The arrow indicates the position of trigger peptidase. (B) Proteins from the active fraction from Sepharose 6B gel filtration were analyzed by sodium dodecyl sulfatepolyacrylamide gel electrophoresis. Proteins in the active fraction (ca. 4 ml) from Sepharose 6B gel filtration'were labeled by incubation with 200 ,uCi of Na125I and two Iodo-beads for 30 min at room temperature. The labeled protein sample was dialyzed extensively against deionized water, lyophilized, and then subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (12.5% acrylamide). The radioactive bands on the dried gel were detected by autoradiography.

Sr2+ and Mn2+ stimulated enzyme activity (50 and 113% as active as Ca2+, respectively). In the experiments described thus far, a mixture of phospholipids extracted from a cells was used in the reaction mixture for enzyme assay. The specificity of phospholipids in trigger peptidase activation was studied (Fig. 4). As indicated by the concentration dependence of trigger peptidase activity on the phospholipids, only phosphatidylserine and phosphatidylethanolamine were effective in trigger peptidase activation. The maximum levels of activation were similar for the two phospholipids. However, phosphatidylethanolamine was more effective than phosphatidylserine at lower concentrations. The effect of various microbial protease inhibitors on trigger peptidase was tested. Antipain and phosphor4midon (each at 100 ,ug/ml), thiol protease and metalloprotease inhibitors, respectively, inhibited trigger peptidase almost completely (Table 3). In contrast, leupeptin and pepstatin, inhibitors of carboxyl protease and serine-thiol protease, respectively, had no inhibitory effect at all at the same I~~~~~~~~~,

I

I

60 0.

g-

0

~

~

~

~

40

_

_r~~~~

I

described thus far, the ConA affinity chromatography fraction containing the enzyme was judged to be free of other proteolytic enzymes. Thus, the ConA affinity chromatography enzyme fraction was used for the studies of the properties of trigger peptidase. Trigger peptidase was normally inactive after purification but was reactivated by the simultaneous addition of Ca2+, membrane phospholipids, and a nonionic detergent such as OG (Table 2). The dependence of trigger peptidase activity on Ca2+ concentration is shown in Fig. 3. The profile shows that trigger peptidase absolutely required Ca2+ for enzyme activity. At Ca2+ concentrations below 0.5 mM, the activity increased, depending on the Ca21 concentration. The enzyme was fully activated by 1 mM Ca2+. The Km of the enzyme for Ca2+ was approximately 0.25 mM. Of various divalent cations (chloride form, 1 mM each) tested, Cu2+, Zn2+, Co2+, and Fe2+ had little or no stimulatory effect (0 to 10% as active as Ca2+). Mg2+ had a weak stimulatory effect (21% as active as Ca2+). In contrast,

a

a

E 20 c wU

0

0.5

1

Ca2+

"

3

mM

FIG. 3. Dependence of trigger peptidase activity on calcium concentration. Trigger peptidase activity was determined by incubation of ConA affinity chromatography-purified enzyme in the presence of 10 ,ug of phospholipids and various concentrations of Ca2". Other conditions were as described in Materials and Methods.

1630

J. BACTERIOL.

MIYAKAWA ET AL.

concentration. The peptidase activity was suppressed to about 50% of that of the control in the presence of E-64 or chymostatin, both thiol protease inhibitors. Other protease inhibitors tested, such as PMSF (1 mM) and soybean trypsin inhibitor (100 ,ug/ml), had no effect at all (data not shown). Surprisingly, none of the inhibitors of trigger peptidase in vitro (antipain, pepstatin, E-64, and phosphoramidon) inhibited in vivo hydrolysis of the mating pheromone by trigger peptidase (data not shown). DISCUSSION In this study, we purified a mating-type-specific membrane thiol peptidase (trigger peptidase) which appears to be responsible for the signaling of rhodotorucine A at the membrane of target cells. The following lines of evidence support the contention that the enzyme we purified was trigger peptidase: (i) the identical pattern of hydrolysis of rhodotorucine A at the Arg-Asn sequence and the sensitivity of the reaction to sulfhydryl-blocking reagents; (ii) the identical specificity for the substrate, with a requirement for the lipid moiety in the substrate structure; and (iii) the absence of the corresponding activity in both the pheromone-producing strain (mating type A) and in a sterile mutant strain, M-39 (type a), that lacks trigger peptidase activity in vivo. Yeast cells have various proteolytic enzymes that play fundamental roles in the regulation of cellular events (2). There are apparently no reports of yeast proteolytic enzymes that are required for pheromone action in the target cell. In regard to the proteolytic enzymes that act on mating pheromone a factor of Saccharomyces cerevisiae, involvement of the enzymes in maturation to yield the active mating pheromone a factor from its precursor polypeptide (6, 7) and in cleavage of factor at the surface of the target cells (3) has been reported. However, in the latter reaction, the proteolytic enzyme is reported not to be required in the signaling reaction but is required for recovery of the cell from the pheromonal effect (3). Solubilized trigger peptidase activity was absolutely dependent on the presence of Ca2+, and enzyme activity also required the simultaneous presence of phospholipids and a

15 0

x 10

._

.;

5

0

10

15

P_ospholipids, pq FIG. 4. Effect of phospholipids on activation of trigger peptidase. ConA affinity chromatography-purified enzyme was preincubated with various amounts of each phospholipid in the presence of 1 mM Ca2l for 1 h at 0°C, and the enzyme activity was assayed as described in Materials and Methods. Symbols: 0, phosphatidylethanolamine; 0, phosphatidylserine; *, phosphatidylcholine; 0, phosphatidylinositol. In the presence of more than 5 F.g of phosphatidylinositol, determination of enzyme activity was not possible due to loss of the mating pheronmone by adsorption to the lipid.

TABLE 3. Effect of various protease inhibitors on trigger

peptidase activitya Inhibitor

Inhibition

E-64 ...........

53.3

Chymostatin ..........

64.7

Phosphoramidon .......... Antipain ..........

93.3 100

Leupeptin ..........

0.0

Pepstatin ...........

0.0

a The trigger peptidase activity of partially purified enzyme (CohA affinity chromatography fraction) in the presence of various protease inhibitors (100 p.g of each per ml) was determined, and the result was expressed as the percent activity remaining compared with that of the control (without inhibitor).

nonionic detergents. The activation of trigger peptidase was achieved by specific phospholipids. Only phosphatidylethanolamine and phosphatidylserine supported trigger peptidase activation. Ca2+-requiring neutral proteases have recently attracted considerable interest because of their apparent involvement in a variety of cellular processes (19). On the basis of concentration dependence on Ca2+, trigger peptidase appears to resemble Ca2+ protease I (10) or mCANP (5) of mammalian cells. However, trigger peptidase is different from these Ca2+-dependent proteases in its requirement for phospholipids, very high substrate specificity for the mating pheromone, and effective inhibitors. A striking feature of trigger peptidase is its requirement for the farnesyl residue in the substrate structure both in vivo (17) and in vitro (Fig. 1). The lipophilic moiety may play a role in substrate recognition. Supporting this possibility, desfarnesyl rhodotorucine A had no effect on the biological activity of rhodotorucine A (17) or on the rate of hydrolysis of rhodotorucine A by the purified enzyme (data not shown). Inhibitors effective against solubilized trigger peptidase (antipain, phosphoramidon, E-64, and chymostatin) were tested for their effect on trigger peptidase activity in vivo. However, none of these inhibitors was effective against in vivo hydrolysis of rhodotorucine A. This was not due to inactivation of the inhibitors by the cell during incubation. Since there is substantial evidence indicating that trigger peptidase is an externally disposed membrane enzynie (15, 17), it seems most likely that the active site of the peptidase is inaccessible to these low-molecular-weight peptide inhibitors. The lipophilic moiety of rhodotorucine A may be necessary for the access of the pheromone to the active site, which is possibly located in a hydrophobic environment. The molecular weights of trigger peptidase determined under nondenaturing and denaturing conditions were 68,000 and 63,000, respectively (Fig. 2 and 3). The small difference in the molecular weights determined by the two procedures may be accounted for by the presence of a nonionic detergent bound to the enzyme during gel filtration. Thus, trigger peptidase seems to consist of a single polypeptide chain. Based on specific binding to ConA-Sepharose, the enzyme appears to be a glycoprotein that contains mannose or glucose or both. The purified enzyme is readily incorporated into phospholipid liposomes prepared from a cells (M. Kaji, unpublished data). Thus, the enzyme seems to have a hydrophobic domain that interacts with the membrane lipids. The properties of trigger peptidase determined in these studies will facilitate further examination of the role of trigger peptidase in transmembrane signaling. For this purpose, we are studying the correlation of the hydrolysis

VOL. 169, 1987


reaction with protein phosphorylation in a membrane reconstitution system using phospholipid liposomes, trigger peptidase, and other membrane components. ACKNOWLEDGMENTS We thank K. Imai for performing some of the early experiments involved in this study. This work was supported in part by a grant-in-aid for scientific research from the Ministry of Education, Science and Culture of Japan. LITERATURE CITED 1. Abe, K., I. Kusaka, and S. Fukui. 1975. Morphological change in the early stages of the mating process of Rhodosporidium toruloides. J. Bacteriol. 122:710-718. 2. Achstetter, T., and D. H. Wolf. 1985. Proteinases, proteolysis and biological control in the yeast Saccharomyces cerevisiae. Yeast 1:139-157. 3. Ciejek, E., and J. Thorner. 1979. Recovery of S. cerevisiae a cells from G1 arrest by a-factor pheromone requires endopeptidase action. Cell 18:623-635. 4. Dittmer, J. C., and M. A. Wells. 1969. Quantitative and qualitative analysis of lipids and lipid components. Methods Enzymol. 14:482-530. 5. Ishiura, S., H. Murofushi, K. Suzuki, and K. Imahori. 1978. Studies of a calcium-activated neutral protease from chicken skeletal muscle. Purification and characterization. J. Biochem. 84:225-230. 6. Julius, D., L. Blair, A. Brake, G. Sprague, and J. Thorner. 1983. Yeast a factor is processed from a larger precursor polypeptide: the essential role of a membrane-bound dipeptidyl aminopeptidase. Cell 32:839-852. 7. Julius, D., A. Brake, L. Blair, R. Kunisawa, and J. Thorner. 1984. Isolation of the putative structural gene for the lysinearginine-cleaving endopeptidase required for processing of yeast prepro a-factor. Cell 37:1075-1089. 8. Kamiya, Y., A. Sakurai, and N. Takahashi. 1980. Metabolites of mating pheromone, rhodotorucine A, by a cells of Rhodosporidium toruloides. Biochem. Biophys. Res. Commun. 94:855860. 9. Kamiya, Y., A. Sakurai, S. Tamura, E. Tsuchiya, K. Abe, and S. Fukui. 1979. Structure of rhodotorucine A, a peptidyl factor inducing mating tube formation in Rhodosporidium toruloides. Agric. Biol. Chem. 43:363-369. 10. Kishimoto, A., N. Kajikawa, H. Tabuchi, M. Shiota, and Y. Nishizuka. 1981. Calcium-dependent neutral protease, widespread occurrence of a species of protease active at lower concentrations of calcium. J. Biochem. 90:889-892.

1631

11. Lloid, K. 0. 1970. The preparation of two insoluble forms of the phytohemagglutinin, concanavalin A and their interactions with polysaccharides and glycoproteins. Arch. Biochem. Biophys. 137:460-468. 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. Markwell, M. A. K. 1982. A new solid-state reagent to iodinate proteins. I. Conditions for the efficient labeling of antiserum. Anal. Biochem. 125:427-432. 14. Miyakawa, T., T. Kadota, Y. Okubo, T. Hatano, E. Tsuchiya, and S. Fukui. 1984. Mating pheromone-induced alteration of cell surface proteins in the heterobasidiomycetous yeast, Tremella mesenterica. J. Bacteriol. 158:814-819. 15. Miyakawa, T., M. Kaji, T. Yasutake, Y. K. Jeong, E. Tsuchiya, and S. Fukui. 1985. Involvement of protein sulfhydryls in the trigger reaction of rhodotorucine A, a farnesyl peptide mating pheromone of Rhodosporidium toruloides. J. Bacteriol. 162: 294-299. 16. Miyakawa, T., H. Maruyama, E. Tsuchiya, and S. Fukui. 1982. In vivo and in vitro stimulatory effect of mating pheromone on the endogenous protein phosphorylation in Rhodosporidium toruloides. Agric. Biol. Chem. 46:2179-2181. 17. Miyakawa, T., M. Nishihara, E. Tsuchiya, and S. Fukui. 1982. Role of metabolism of the mating pheromone in sexual differentiation of the heterobasidiomycete Rhodosporidium toruloides. J. Bacteriol. 151:1184-1194. 17a.Miyakawa, T., T. Tachikawa, R. Akada, E. Tsuchiya, and S. Fukui. 1986. Involvement of Ca2+/calmodulin in sexual differentiation induced by mating pheromone rhodotorucine A in Rhodosporidium toruloides. J. Gen. Microbiol. 132:1453-1457. 18. Miyakawa, T., T. Tachikawa, Y. K. Jeong, E. Tsuchiya, and S. Fukui. 1985. Transient increase of Ca2' uptake as a signal for mating pheromone-induced differentiation in the heterobasidiomycetous yeast Rhodosporidium toruloides. J. Bacteriol. 162: 1304-1306. 19. Murachi, T. 1983. Calpain and calpastatin. Trends Biochem. Sci. 8:167-169. 20. Sakagami, Y., A. Isogai, S. Suzuki, C. Kitada, and M. Fujino. 1979. Structure of tremerogen A-10, peptidal pheromone inducing conjugation tube formation in Tremella mesenterica. Agric. Biol. Chem. 43:2643-2645. 21. Sakagami, Y., M. Yoshida, A. Isogai, and A. Suzuki. 1981. Structure of tremerogen a-13, a peptidal sex hormone of Tremella mesenterica. Agric. Biol. Chem. 45:1045-1047. 22. Tsuchiya, E., S. Fukui, Y. Kamiya, Y. Sakagami, and M. Fujino. 1978. Requirement of chemical structure of lipopeptidyl factors inducing sexual differentiation on vegetative cells of heterobasidiomycetous yeast. Biochem. Biophys. Res. Commun. 85: 459-463.