Effects of Metabolites Present During Growth of Tetrahyrnena pyriforrnis on the Subsequent Secretion of Lysosomal Hydrolases J. J . BLUM Department of Physiology and Pharmacology, Duke University Medical Center, D u r h a m , North Carolina 2771 0

ABSTRACT Tetrahymena were grown in proteose-peptone medium supplemented with glucose, mannose, fructose, galactose, acetate, succinate, or pyruvate and then washed and resuspended in a non-nutrient salt solution and the amounts of 7 acid hydrolases secreted into the medium in a one hour incubation were measured. Cells that had been grown in the presence of glucose secreted about half the amounts of acid phosphatase, p-N-acetylglucosaminidase and acid protease as did control cells grown i n unsupplemented medium. Pyruvate was about as effective as glucose and both were slightly more effective than acetate or fructose. Succinate had little effect. Similar experiments showed that a-mannosidase, p-fucosidase, and p-galactosidase are secreted into the salt solution and that secretion is reduced by prior growth of the cells in medium supplemented with glucose or mannose but not galactose. Except for a-mannosidase, these reductions in amounts of hydrolase secreted were not accompanied by appreciable changes in intracellular activity, and therefore demonstrate a persistent effect of growth in the presence of certain metabolites on the subsequent secretion of lysosoma1 hydrolases. Since the inhibition of subsequent secretion depended on both the individual metabolite and the particular hydrolase examined, it appears that the effect of metabolites is not limited to a general inhibition of secretion but may differentially alter some properties of lysosomal subpopulations. A preliminary characterization of the secreted acid protease of Tetrahymena suggests that there may be two acid proteases released, since up to 25% of the activity was not inhibited by high concentrations of pepstatin, leupeptin, or chymostatin.

The ciliate Tetrahymena pyriforrnis releases several lysosomal acid hydrolases into the extracellular medium both during growth in proteose-peptone medium and when placed in a non-nutrient medium (Muller, '70, '72). It seemed likely to Muller ('72) and to Rothstein and Blum ('73) that the release of these hydrolases was a secretory process. Strong supportive evidence for this view comes from recent findings (Rothstein and Blum, '72a) that the egestion of pre-ingested inert hydrocarbon particles and the release of several acid hydrolases were increased by the addition of the catecholamine antagonists dichloroisoproterenol and desmethylimipramine. Rothstein and Blum ('73) found that when Tetrahymena are placed in a dilute salt solution supplemented with glucose, acetate, or pyruvate, there were significant differences in the amounts of acid phosphaJ. CELL. PHYSIOL.,86: 131-142.

tase, a-glucosidase, and ribonuclease released compared to the amounts released by control cells incubated in unsupplemented dilute salt solution. The pattern of release depended on which hydrolase was examined and which metabolite was present during the incubation, indicating that there were at least two populations of lysosomes in Tetrahymena (Muller, '70, '72; Rothstein and Blum, '73, '74b). These findings suggest that Tetrahymena might be a useful model for studying the influence of metabolism on the secretion of lysosomal hydrolases, a problem which has received remarkably little attention. In the experiments of Rothstein and Blum ('73), the effects of metabolites added to the dilute salt solution on hydrolase release were small. It seemed possible that Received June 26, '74. Accepted Nov. 29,'74.

131

132

J. J. BLUM

the addition of metabolites to the cells during growth might have a more profound effect on the subsequent secretion of lysosomal hydrolases, and preliminary experiments indicated that this was the case. It was also found that T e t r a h y m e n a contained a-mannosidase, p-fucosidase, and p-galactosidase activities, all with optima at acidic pH values, and that these activities were secreted into the medium. Since these enzymes had not previously been described for T e t r a h y m e n a , it was of interest to examine the effects of metabolites during growth on the intracellular content of these acid glycosidases and on their rates of secretion into a dilute salt solution. The results to be presented below will therefore analyze the pattern of variation of six lysosomal hydrolases (both with respect to intracellular content and to the subsequent rates of secretion) in response to growth in proteose-peptone medium supplemented with various metabolites. It was also noted during the course of these experiments that the acid protease activity that was secreted probably consisted of two different proteases as judged by the effects of proteolytic inhibitors, and some of the properties of these asid proteases and of their responses to the presence of substrates in the growth medium will be described. An abstract of part of this work has been published (Blum, '74). MATERIALS AND METHODS

Growth a n d haruesting of cells T e t r a h y m e n a pyriformis, strain HSM, were grown at 25" in a medium of 1% proteose peptone and 0.05% liver extract in 20 mM potassium phosphate adjusted to pH 6.5 with NaOH. Cells were grown in a gyrotary shaker in 500 ml flasks containing 130 ml total volume, supplemented if desired with subtrates dissolved in distilled water and sterilized by filtration through ultrafine sintered glass filters. At most 2 ml of the substrate was used in the 130 ml total volume, and the same amount of water was added to the control cells. Cultures were inoculated from log phase shaken cultures at approximately 9 AM, i.e., after about 17 hours of growth. Initial and final cell densities (Ni and Nf, respectively) were determined for each experiment. Typical values for Ni ranged from 10,000-40,000 per ml, and typical

values of Nf/Ni ranged from 10-25, depending on Ni. There was no effect of any of the substrates used here on the ratio Nf/ Ni except for pyruvate, which caused a slight reduction in Nf/Fi. Cells were collected a t room temperature by centrifugation for three minutes at 200 X g . The pellets were washed twice (200 X g for 3 min; 200 g for 2 min) with a 1:100 dilution of the salt solution described by Wagner ('56) and resuspended to a density of about 2 X 1 0 6 cellslml. Twelve ml portions were added to 1 liter Erlenmeyer flasks containing 4 ml of the dilute salt solution and the flasks were incubated without shaking in a n incubator at 25" for one hour. After the one hour incubation the cells were centrifuged at 0 " for 3 min at 200 g and the supernatants were collected and recentrifuged. The cells were resuspended in 10 ml of dilute salt solution, chilled in ice, and treated with ultrasound twice for 30 seconds separated by a 30 second lapse, using a Branson Model LS-75 ultrasonic generator a t a setting of "5". For experiments in which it was desired to check on the intracellular content of the various hydrolases before the one hour incubation, the identical procedures were followed except that the cells were incubated for 30 seconds in the incubator instead of one hour and then collected by centrifugation. The supernatant was discarded and the pellet was treated with ultrasound as above. Assay procedures Cell counts were performed with a Coulter counter (Coulter Electronics, Inc., Hialeah, Fla.), using a 100 pm aperture. All assays were performed as described in table 1. For the pN02-phenylated substrates, four samples of 0.5 ml were taken at intervals from about 10-40 minutes and pipetted into 0.5 ml of ice cold 0.5 M tris (hydroxymethy1)-aminomethane and the absorbance measured at 410 nm. The assays were linear with time and enzyme concentrations. Activities are expressed as pmoles of substrate hydrolyzed/106 cells. min. For the acid protease assay, duplicate samples of 0.5 ml were taken shortly after the addition of the enzyme to the hemoglobin substrate and added to 0.5 ml of ice cold 5% (w/v) trichloroacetic acid. DupIicate samples were also taken after 1-2

METABOLITE EFFECTS ON HYDROLASE SECRETION

133

TABLE 1

Assay conditions Enzyme

Substrate ~~

Acid phosphatase

Buffer ~

0.1 M acetate, pH 4.5 0.1 M acetate, pH 4 . 5

Acid protease a-mannosid ase

p nitrophenyl phosphate p-nitrophenyl P-N acetyl glucosaminide denatured hemoglobin p-nitrophenyl a-D-mannoside

p-fucosidase

p-nitrophenyl p D-fucoside

0.1 M lactate, pH 3 . 4 0.1 M cacodylate, 0.1 M acetate, pH 4.0 0.1 M cacodylate, 0.1 M ace-

p-galactosidase

p-nitrophenyl 8-D-galactopyranoside

0.1 M cacodylate, 0 . 1 M acetate, pH 4.0

p-N acetylhexoseaminidase

tate, pH 4 . 0

All assays were carried out at 25'C in a total volume of 4.1 ml of which 0 . 1 ml was 2% Triton X-100. p-Nitrophenylated substrates were made up to 10 mM in the buffers indicated. Denatured hemoglobin was made up to 20 mglml in 0.1 M lactate, pH 3.4, and dialysed overnight against lactate buffer. Reactions were started by adding 2 ml of supernatant or homogenate, suitably diluted with the dilute salt solution, to 2 ml of reagent (plus Triton X-100).

hours of incubation with the hemoglobin. The precipitated hemoglobin was removed by centrifugation and 0.5 ml of the clear supernatant was added to 1.0 ml of 0.2 N NaOH and analyzed for released peptides by the procedure of Lowry et al. ('51). Protease activity is expressed as pg protein equivalents hydrolysed/l O6 cells.min, using bovine serum albumin as the standard.

minutes. The tubes were then centrifuged three minutes at 200 g and the pellets washed twice with about 8 ml of ice cold 95% ethanol and frozen. Glycogen content (determined in triplicate) was assayed by the glucose oxidase method after hydrolysis with H2S04as described by Blum ('72).

Reagents All p-nitro-phenylated substrates were pH-curves purchased from Sigma. Hemoglobin stanThe pH curves of a-mannosidase, p-fuco- dardized for protease assay was purchased sidase, and p-galactosidase were done from Nutritional Biochemical Co. Antipain, using 1 ml of supernatant enzymes from leupeptin, chymostatin, and pepstatin were cells that had been allowed to secrete for the generous gifts of Dr. H. Umezawa of one hour, 1 ml of 20 mM substrate in water, the Microbial Chemistry Research Foundaand 2 ml of a buffer consisting of 0.1 M so- tion, Tokyo. All other chemicals were redium acetate plus 0.1 M sodium cacodylate agent grade. adjusted to the required pH with NaOH or RESULTS HC1. Release 0.f a-mannosidase, p-galactosidase, and p-fucoszaase Proteolytic inhibitors In view of the ubiquitous occurrence of Antipain, leupeptin, and chymostatin were dissolved in 0.1 M acetate, pH 3.4 at glycosidases with acid pH optima in mam1 mglml and diluted as required in lactate malian lysosomes,it seemed likely that some buffer. Pepstatin (5 mg) was dissolved in of these might be present in Tetrahymena. 1 ml of methanol and then the volume A preliminary survey indicated no hybrought to 5.0 ml by the addition of lactate drolysis of pnitrophenyl p-D-xylopyranobuffer. In experiments where the effect side, p-nitrophenyl p-D-glucuronide, p-niof pepstatin on protease activity was tested, trophenyl a-D-fucoside, or p-nitrophenyl the same amounts of methanol were added a-D-galactopyranoside by homogenates of to control tubes. Tetrahymena, but hydrolysis was observed with p-nitrophenyl p-D-galactopyranoside, Assay of glycogen content p-nitrophenyl a-D-mannoside, and p-nitroAliquots of cells to be analyzed for glyco- phenyl p-D-fucoside. It was then found that gen were pipetted into centrifuge tubes p-galactosidase, p-fucosidase, and a-mancontaining 8 ml of ice cold 95% ethanol, nosidase activities were released into dilute and allowed to stand in ice for about 20 salt solution during a 1 hour incubation.

134

J. J. BLUM

The pH versus activity profiles of these three hydrolases show optima at about 3.7 for p-galactosidase and a-mannosidase and about 4.1 for p-fucosidase (fig. 1). For convenience, all future assays of these three hydrolases were made at pH 4.0 in a cacodylate-acetate buffer each at final concentrations of 0.05 M . Experiments to be published in detail elsewhere yield data which pertain to the question as to whether these three glycosidase activities could be due to a single enzyme. First, cells were allowed to secrete

1.1

for one hour and the dilute salt solution was subjected to column chromatography. The pooled peak of a-mannosidase activity had one-fifth as much p-fucosidase activity and one-tenth as much p-galactosidase activity as a-mannosidase activity, whereas the amounts of p-fucosidase and p-galactosidase activities secreted by the cell are roughly equal to each other and to the mannosidase activity secreted. Furthermore, when cultures are grown in proteose peptone for two days (i.e., well into stationary phase) and then centrifuged and al(Y-

\ 0-galactosldase

I.0-

-

.9

-

v)

s.0(D

0 .-

.7-

E \ $ .6-

2

\ d-mannosidase

2 5-

\

v -

h

c .-

.? c

.4-

2

2.3

2.7

3.1

3.5 3.9

4.3 4.7

5.1

5.5 5.9 € 3

PH Fig. 1 pH versus activity curves for p-galactosidase, a-mannosidase, and p-fucosidase. Cells were grown for 17 hours i n unsupplemented proteose-peptone medium and then washed, resuspended in dilute salt solution, and allowed to release hydrolases for one hour as described i n the section on MATERIALS AND METHODS. Two experiments were performed, one at pH values ranging from 2.3-4.1 and the other from pH 3.5 to pH 6.0. For each enzyme the values obtained i n the first experiment were multiplied by a constant so that the activity at pH 3.5 became identical to that obtained in the second experiment, i.e., the curves were matched at pH 3.5.

135

METABOLITE EFFECTS O N HYDROLASE SECRETION

lowed to secrete for one hour into a dilute salt solution, the amount of a-mannosidase activity secreted goes up over 10-fold compared to cells grown for 17 hours, whereas the amount of p-fucosidase secreted is only doubled and the amount of p-galactosidase activity is hardly changed. These observations, plus the slight difference in pH optimum between the p-galactosidase and the p-fucosidase, render it highly unlikely that a single unspecific glycosidase is responsible for the observed activities. Sensitivity of secreted acid proteases to proteolytic inhibitors Dickie and Liener ('62) reported that there were three proteases in Tetrahymena: an intracellular protease with a pH optimum of 5.5, which was not released into the medium; and two extracellular proteases, one released in the absence of glucose and having a pH optimum of 6-7, the other released during growth in media supplemented with glucose and having a pH optimum of 7-8. They reported that although the protease released in the presence of glucose had a higher specific activity than the one released in the absence of

$401

glucose, less total activity was released in the presence of glucose than in its absence. Muller et al. ('66), however, found that the pH optimum of the released acid protease was about 3.5, and we have also been unable to detect any released activity at pH values near 7, although we confirm the observation of Dickie and Liener ('62) that glucose reduces the protease activity in the medium. Recently Levy and Sisskin ('74) found that when cells grown under relatively anaerobic conditions are subjected to a step-up in oxygen tension (i.e., changed from deep static to shallow shaken conditions) a protease with a pH optimum near neutral appears in cell homogenates. It seemed worthwhile, therefore, to attempt to further characterize the acid protease activity secreted by strain HSM growing under the present conditions. For this purpose, the proteolytic inhibitors antipain (Suda et al., '72), leupeptin (Aoyagi et al., '69), chymostatin (Umezawa et al., '70) and pepstatin (Aoyagi et al., '71) were used. These inhibitors, isolated from the growth media of actinomycetes cultures, are very potent proteolytic inhibitors which have been used to discriminate between types of

/

Antipain

/ -/ , ' x ,A

Leupeptin

Pepsta t ii

II

I

.2

I

I

.4

1

I

I

Ill

.6 .8 I

I I I 1 I 1 I l l 2 4 6 810 Inhibitor (&g/ml)

1

20

It1 1 40 li

Fig. 2 Effect of proteolytic inhibitors on secreted acid protease activity. Cells were grown for 17 hours in unsupplemented proteose-peptone medium, collected, washed, resuspended in dilute salt solution, and allowed to secrete hydrolases for one hour a s described in the section on MATERIALS AND METHODS. Aliquots of the cell-free supernatant were then assayed at pH 3.5 in the presence of the indicated concentrations of chymostatin, leupeptin, antipain, and pepstatin.

136

J. J. BLUM

proteases both in Tetrahymena (Levy, personal commun.) and many other species. Figure 2 shows that the secreted acid protease of Tetrahymena is insensitive to pepstatin but sensitive to antipain, leupeptin, and chymostatin, the latter being the most potent inhibitor. It can be seen that even at relatively enormous concentrations of antipain, leupeptin, or chymostatin, only three-fourths of the protease activity was inhibited. This suggests that there are two acid proteases secreted by Tetrahymena, one of which, accounting for three-fourths of the secreted activity, is inhibited 50% by about 0.3 pg/ml of chymostatin, by 0.5 pg/ml of leupeptin, and by 3.8 pg/ml of antipain, and the other, accounting for

Acid prolease

-

one-fourth of the secreted activity, not inhibited by any of these proteolytic inhibitors.

Effect of growth in media supplemented with metabolizable substrates on the subsequent release of six acid hydrolases Figures 3a and 3b show the results obtained when cells grown €or 17 hours in unsupplemented proteose-peptone medium or in media supplemented with various substrates were collected, washed and allowed to secrete lysosomal hydrolases into a dilute salt solution for one hour. There was little

n

100

-

0 80 +

5 60 r

8 40 20 n

I

d-n-annosidase

Fig. 3 Effect of growth i n medium supplemented with sumtrates on the intracellular activity and subsequent release of acid hydrolases. Cells were grown for 17 hours in unsupplemented proteose-peptone medium (controls) or in medium supplemented with glucose (15.4 mM), fructose (15.4 mM), acetate (7.7 mM), pyruvate (7.7 mM), succinate (7.7 mM), mannose (7.7 mM), or galactose (7.7 mM). Cells were then collected by centrifugation, washed, resuspended in dilute salt colution, and allowed to secrete acid hydrolases for 1 hour as described in the section on MATERIALS AND METHODS. In each experiment, the activity remaining in the (sonicated) cell pellet, i.e., the intracellular activity (open bars) and the activity released into the cell-free supernatant, i.e., the secreted activity (hatched bars) was measured. The ordinates are the percent of the intracellular activity and of the secreted activity relative to control values taken as 100%. Figure 3a shows data from two experiments for acid protease, acid phosphatase, and P-N-acetylglucosaminidase. Typical control values for these hydrolases are given i n table 3. Figure 3b presents data from three experiments for a-mannosidase, P-fucosidase. and P-galactosidase; the control values for these glycosidases are: 14.0, 3.42, and 2.94 pmoles/hr.lOs cells, respectively, for the intracellular activities and 0.75, 0.56, and 0.70 pmoleslhr.106 cells for the released activities. Vertical bars indicate one standard deviation.

137

METABOLITE EFFECTS ON HYDROLASE SECRETION

effect on the intracellular activity of acid phosphatase, P-N-acetylglucosaminidase, or acid protease in cells grown with any of these substrates as compared to cells grown in unsupplemented medium (fig. 3a). The amounts of these three hydrolases secreted in the one hour incubation, however, were reduced by half or more in cells grown with glucose and about one-third in cells grown in fructose. Acetate also reduced the secretion of these three hydrolases but because of the variability in the results obtained in these two experiments, it cannot be decided from the data in figure 3a whether acetate was more effective than fructose or whether acetate affected the secretion of acid protease more than it affected the secretion of acid phosphatase. F'yruvate inhibited the secretion of acid protease and of p-N-acetylglucosaminidase about 50% , but scarcely inhibited the secretion of acid phosphatase. Growth in the presence of succinate caused only a slight reduction of the subsequent secretion of either of these three hydrolases. Although there were no appreciable effects Of growth in media with glucose, pyruvate, acetate, or fructose on the intracellular activity (assumed proportional to content) of acid phosphatase, PN-acetylglucosaminidase or acid protease when measured at the end of the one hour incubation period, i t Was nevertheless pOSsible that there were reductions in intracellular content at the beginning of the one hour incubation in dilute salt solution which accounted for the reduced rate of secretion but which were no longer apparent at the end of the hour incubation. To check on this possibility, experiments were done in which the cells were grown for 17 hours and collected and resuspended in dilute salt solution but allowed to incubate for 30 sec instead of one hour. Table 2 shows that exposure to glucose, pyruvate, acetate, or fructose during the 17 hour growth period caused only a slight reduction in the intracellular activity of the three hydrolases. Thus the reduction in amount of hydrolase activity released is not caused by a reduction in the amount of enzyme present in the cell, but must reflect a reduced rate of secretion. The data in figure 3a showed that growth in glucose caused a stronger inhibition of subsequent secretion than growth in fruc-

TABLE 2 m e c t of growth i n the presence of substTates the intracellular activities of some lysosomal hydrolases

on

Percent of control activity A.

Acid protease

0-N-acetyl glucosaminidase

~l~~~~~

86 85

96 83

84 97

Fructose

90 88

96 81

103 92

Pyruvate

78 95

84 92

107 114

Acetate

85 90

96 80

91 80

a-Mannosidase

0-Fucosidase

P-Galactosidase

Glucose

67 74

105 115

104 110

M ann ose

72 81

93 101

97 107

B.

Acid phosphatase

Cells were grown overnight as described in the section on METHODS in unsupplemented proteose-peptone medium (controls) or supplemented with the indicated substrates at an initial concentration of 15.4 mM for glucose and 7.7 mM for any of the other substrates. After the 17 hour growth period (during which cell density increased at least 12-fold) the cells were collected, washed, resuspended, and sonicated after being allowed to release hydrolases into the medium for only one minute. TWO experiments are shown for each substrate and enzyme. The numbers in the table are the intracellular activities in percent of the control value for each experiment.

tose, but were inadequate to establish the relative potencies of fructose versus acetate or pyruvate. To examine this, experiments were done in which hydrolase secretion from cells that had been grown with one substrate present was compared to secretion from the same stock of cells that had been grown with another substrate (table 3). In two such experiments in which glucose was compared to fructose, the cells that had been grown with fructose secreted slightly more of each hydrolase than the cells that had been grown with glucose, as expected from figure 3a. Cells that had been grown with acetate secreted more of each hydrolase than cells that had been grown with pyruvate, indicating that pyruvate caused a larger inhibition of secretion than acetate. Paired experiments in which

138

J. J. BLUM

acetate was compared to fructose indicated that these two substrates were approximately equal in their effectiveness as inhibitors of the secretion of these hydrolases. These experiments indicate that the order of potency for these substrates is glucosepyruvate > fructose rv acetate. It must be emFhasized that the differences in potency between glucose or pyruvate on the one hand and fructose or acetate on the other hand are small, and are of interest primarily because they demonstrate subtle differences in the effects of metabolites on hydrolase secretion rates. Similarly, the small inhibition of acid phosphatase release relative to that of acid protease by pyruvate demonstrates that a given substrate may affect the secretion rate of two hydrolases in different ways. It should be noted that there were only small differences between the intracellular activities of each hydrolase in these paired experiments (table 3), as expected from the data already presented in figure l a and table 2. Figure 3b shows that the presence of glucose or mannose caused a reduction by about one-fourth in the intracellular activity of a-mannosidase but essentially no change in the intracellular activity of either p-fucosidase or p-galactosidase. This reduction in intracellular a-mannosidase activity is manifest before the one hour incubation (table 3) and is thus a consequence of growth in the presence of glucose

or mannose. Growth in media supplemented with glucose caused a reduction in the amounts of hydrolases secreted in a one hour incubation to 62,48, and 61 % of control values for a-mannosidase, p-fucosidase, and p-galactosidase, respectively. Similar results were obtained from cells that had been grown in medium supplemented with mannose (fig. 3b). It seems likely that glucose and mannose caused less of a reduction in the secretion of p-galactosidase, but more experiments would have to be done to establish this point with certainty. In contrast to glucose or mannose, the presence of acetate during growth caused no significant change in the intracellular activity of any of these three glycosidases and reduced their subsequent secretion to a lesser extent that did mannose or glucose. Galactose caused only a slight reduction in intracellular activity or in the amounts of each glycosidase released. The failure of galactose to alter the intracellular activity or amount of the three glycosidases released (in contrast to glucose or mannose) raised the question as to whether galactose was metabolized by Tettrahymena. Previous investigations (summarized by Holz, '64) had left this issue unresolved, although i t was established that mannose could be fermented by Tetrahymena. The data in table 4 show that whereas mannose was approximately as effective as glucose in increasing the glyco-

TABLE 3

meet of

growth in medium supplemented w i t h one substrate as compared to another substrate on the subsequent release of acid hydrolases Acid protease

N-acetyl-p-Dglucosaminidase

Acid phosphatase

Substrate Intracell

Released

Intracell

Released

Intracell

Released

Glucose Fructose

mg proteinimin.106 cells 1.4,2.0 0.16,0.20 1.5,1.8 0.22,0.24

pmoksimin.106 cells 14.9,20.8 3.2,3.9 15.5,21.5 4.2,4.1

pmoksimin.lO6 cells 293,526 21.4,38.3 308,530 26.9,38.5

Pyruvate Acetate

2.3,2.1 2.0,1.9

0.13,0.14 0.24,0.25

19.9,20.5 15.1,20.8

3.3,3.5 4.4,4.7

414,471 343,386

26.2,33.9 38.0,44.4

Acetate Fructose

2.6,2.6 2.3,3.0

0.31,0.29 0.34,0.32

13.1,15.7 14.4,17.3

4.9,4.6 4.7,4.4

381,378 334,399

40.1,45.1 43.8,31.6

Aliquots of a stock culture of Tetrahymena were inoculated into identical volumes of proteose peptone supplemented with glucose or fructose (15.4 mM each), acetate or pyruvate (7.7 mM each), or acetate or fructose (7.7 mM each) and grown for 17 hours. The cultures were then collected, washed, resuspended in dilute salt solution, and placed in the incubator for one hour. The cells were then collected by centrifugation, the supernatant assayed for secreted hydrolase activity and the pellets assayed (after sonication) for intracellular activity. Two experiments are shown for each pair of substrates. In the first experiment with glucose and fructose, for example, the intracellular protease activities were 1.4 and 1.5 units, respectively, while the secreted protease activities were 0.16 and 0.22 units, respectively.

139

METABOLITE EFFECTS ON HYDROLASE SECRETION TABLE 4

Glycogen c o n t e n t of cells g r o w n w i t h glucose, f r u c t o s e , m a n n o s e , or galactose Experiment I Sugar added

Nf/Ni

Glycogen

Experiment I1 Percent of (control)

Nf/Ni

(100) (298) (356) (179)

5.4 5.3 4.4 5.2

pg1106 cells -

Glucose Mannose Galactose

7.8 7.9 7.0 7.7

182 541 646 326

Glycogen

Percent of (control)

pg/106 cells

144 583 315 226

(100) (404) (219) (1%)

Fifty ml of cells were grown for 17 hours in 300 ml flasks with shaking in proteose-peptone supplemented with 7.9 mM glucose, mannose, or galactose as indicated. N i , the cell density at the beginning of the growth period, was 89,900 cellslml for experiment I and 124,000 cells/ml fo: experiment 11. At the end of the growth period, samples were taken in triplicate from each flask and assayed for glycogen as described in the section on METHODS. TABLE 5

m e e t of c h y m o s t a t i n on released acid protease activity Substrate

No inhibitor

10 p d m ? chymostatin

m g protein/min.106 cells

Control Glucose (Percent of control)

0.29 2 0.09 0.13 2 0.05 (44 & 6)

0.096 It 0.037 0.027 t 0.023

Control Mannose (Percent of control)

0.41 2 0.05 0.22 t 0.03 ( 5 6 2 14)

0.093 c 0.015 o,048 o,oo9

Control Acetate (Percent of control)

0.39&0.08 0.25 t 0.08 (61 & 9)

o.122 o.041 0.066 t 0.025

Control Galactose (Percent of control)

0.50 2 0.16 0.492 0.14 (99 & 3)

Cells were grown a s described in the legend to figure 3. The values shown are the mean protease activity f standard deviation for three experiments for the control and for each substrate.

gen content of Tetrahymena during a 17 hour growth period, galactose was considerably less effective as a glycogen precursor. It must be emphasized, however, that growth in the presence of galactose caused about a 1.6-fold increase in glycogen content, and it is, therefore, metabolized at an appreciable rate by Tetrahymena.

Effect of chymostatin o n released protease activity Figure 3a showed that the presence of glucose, fructose, and acetate during growth caused a reduction in the subsequent release of acid protease activity when the cells were incubated in a dilute salt solution. Since the experiments with proteo-

lytic inhibitors (fig. 2) suggested that two protease actjvities may be secreted, it was of interest to ascertain whether metabolites cculd be preferentially altering the release of one of the two putative proteases. Experiments were therefore performed in which cultures were supplemented with glucose, mannose, or acetate, and the sensitivity of the protease activity secreted into a dilute salt solution to chymostatin was determined. Table 5 shows that mannose was roughly as effective as glucose or acetate (or fructose; cf. fig. la) in reducing the secretion of acid protease activity, and that there was no indication of any change in the sensitivity of the secreted protease activity to chymostatin. DISCUSSION

It is well established both for Tetrahym e n u (Levy and Wasmuth, '70; Hogg and Kornberg, '63; Whitlow et al., '72; Diesterhaft et al., '72; Voichick et al., '73) and for other cells that growth in the presence of glucose may cause alteration of the enzyme complement of the cell, usually referred to as glucose repression, although in Tetrahymena growth in the presence of glucose can cause an increase in the activity of several enzymes of intermediary metabolism (Porter and Blum, '73). This framework does not apply to lysosomal hydrolases in Tetrahymena except for a-mannosidase which will be discussed separately below. For all the other hydrolases examined, growth in media supplemented with glucose (or fructose, mannose, acetate, or pyruvate) caused an appreciable reduction in the amount of activity released during a one hour incubation in dilute salt solution

140

J. J . BLUM

without any appreciable reduction in the intracellular activity of these hydrolases or in the high rate of growth (except for a slight reduction in growth rate caused by pyruvate). There are essentially two mechanisms which could account for the reduced rate of acid hydrolase release. First, the rate of egestion per se (e.g., cytoproct activity) might be inhibited by prior exposure to the effective substrates. If this were the case, one would expect the same percentage reduction in the secretion of all the hydrolases examined. Pyruvate, however, reduced the rate of release of acid protease and pN-acetylglucosaminidase more than that of acid phosphatase, glucose reduced the release of acid protease more than it reduced that of p-galactosidase, and mannose reduced the rate of release of p-fucosidase more than it reduced that of p-galactosidase.These observations do not rule out the possibility that growth in medium supplemented with these substrates reduced the subsequent rate of egestion by the cytoproct, but would require that even if that occurred, a change in selectivity for which kinds of lysosomes were egested would also have occurred. Alternatively - and more parsimoniously - one could suppose that prior exposure to the effective substrates caused a modification in the properties of the lysosomal membranes which reduced the probability of their capture and extrusion by the cytoproct. Since it is well established that there are at least two populations of lysosomes in Tetrahymena (Muller, '70, '72; Rothstein and Blum, '73, '74b), one might expect that the two or more populations would be modified differently and this would provide a simple explanation for the subtle differential effect on secretion rate noted above. Zahlten et al. ('72) reported that glycogen changed the amount of 3*P incorporated in the proteins and lipids of lysosomal membranes, and it seems possible that changes in lysosomal membranes might occur in Tetrahymena in response to the presence of certain substrates in the growth medium. Such a mechanism would also account for the persistence of the effect for at least 1.5 hours after the cells are washed free of all substrates (about 0.5 hours for washing the cells and resuspending plus the 1 hour incubation). Although pyruvate and acetate were fairly strong inhibitors of hydrolase release,

succinate was not. This could indicate that Krebs cycle intermediates are relatively ineffective in suppressing lysosomal hydrolase release, but we have found (Raugi, Liang and Blum, unpublished) that oxidation of added succinate is very slow in Tetrahymena, and the failure of succinate to inhibit hydrolase release may simply indicate that it is not metabolized at an appreciable rate. As far as we have been able to ascertain, only an acid protease activity is released into the medium by strain HSM of Tetrahymenu in agreement with Muller's ('72) finding for syngen I, mating type 11, but not with that of Dickie and Liener ('62) working with strain W. The latter workers observed that addition of glucose to the proteose-peptone medium reduced the protease activity of the cell-free culture fluid. We find that this effect of glucose is persistent, i.e., the rate of secretion of protease from cells grown with glucose is lower than that of control cells grown in unsupplemented medium even when the cells are washed and placed in a non-nutrient salt solution. We further find that the protease activity released into the medium is not completely inhibited by even very large concentrations of chymostatin, leupeptin, antipain, or pepstatin. This suggests that the released protease activity may consist of two proteases. If so the major constituent, which accounts for three-fourths or more of the secreted protease activity does not resemble any of the 14 proteases tested by Suda et al. ('72) for their sensitivity to these four proteolytic inhibitors, although this could be due in part to the use of substrates other than hemoglobin in most of their assays. Muller et al. ('66) noted that Tetrahymenu had virtually no p-glucuronidase activity or arylsulfatase activity. We confirmed the absence of p-glucuronidase, and failed to demonstrate the presence of p-xylosidase, a-fucosidase, or a-galactosidase. The finding that Tetrahymena contain a-mannosidase, p-fucosidase, and p-galactosidase in addition to amylase, a-glucosidase, pglucosidase, and p-N-acetylaminoglucosidase, which were already known to be present (Muller, '70), indicates that this protozoan secretes at least 4 glycosidases into the medium and that these, as well as proteases, nucleases, and acid phosphatases may play a role in the initiation of the

METABOLITE EFFECTS ON HYDROLASE SECRETION

digestive process in the external environment. In contrast to the other six hydrolases studied, the intracellular content of a-mannosidase was reduced by growth in the presence of glucose and mannose, but not galactose. Both glucose and mannose enter metabolism by phosphorylation to glucose6 - phosphate or mannose - 6 - phosphate, which is converted to glucose-6-phosphate by a phosphomannose isomerase. Galactose, however, generally enters as galactose-1-phosphate by the action of galactokinase and then may be converted to uridine diphosphate galactose and then to UDPG (e.g., White et al., '67), thus serving as a precursor for glycogen synthesis without building up appreciable levels of glucose-6-phosphate, fructose-6-phosphate, or related metabolites. This may suggest that the effects of glucose and fructose in causing the reduction in intracellular a-mannosidase activity are mediated via glucose6-phosphate andlor related metabolites rather than via UDPG. Whatever the reason why galactose is less effective than mannose or glucose in causing the reduction in intracellular activity of a-mannosidase, it is of interest that certain substrates - i.e., glucose and mannose -can preferentially reduce the content of a particular lysosomal hydrolase. This indicates that the enzyme complement of lysosomes is potentially variable by suitable nutritional regimens and may provide a useful tool for further analysis of lysosomal function in certain diseases. ACKNOWLEDGMENT

This work was supported by Grant 5 R01 HD01269 from the National Institutes of Health. I am grateful to Dr. H. Umezawa for generously supplying me with the protease inhibitors, to Ms. Carolyn Edwards for excellent technical assistance, and to Dr. Michael Levy for permitting me to see his work on the neutral proteases of Tetrahymena prior to publication. LITERATURE CITED Aoyagi, T., T. Takeuchi, A. Matsuzaki, K. Kawamura, S . Kondo, M. Hamada, K. Maeda and H. Umezawa 1969 Leupeptins, new protease inhibitors from actinomycetes. J. Antibiotics, 2 2 : 283-286. Aoyagi, T., S. Junimoto, H. Morishima, T. Takeuchi and H. Umezawa 1971 Effect of pepstatin on acid proteases. J. Antibiotics, 24: 687-694.

141

Blum, J. J. 1972 Effect of AMP and related compounds on glycogen content of T e t r a h y m e n a . J. Cell Physiol., 80: 4 4 3 4 5 2 . 1974 Effects of prior exposure to metabolites on secretion of lysosomal acid hydrolases in Tetruhymena pyriformis, strain HSM. J. Protozool., 21 : 439. Dickie, N., and I. E. Liener 1962 A study of the proteolytic system of T e t r a h y m e n a pyriformis W. Biochim. Biophys. Acta, 64: 41-51. Diesterhaft, M. D., H. Hsieh, C. Elson, H. J . Sallach and E. Shrago 1972 Enzymatic regulation of the metabolism of phosphoenolpyruvate in T e t r a h y m e n a pyriformis. J. Biol. Chem., 247: 2755-2762. Hogg, J. F., and H. L. Kornberg 1963 The metabolism of Cn-compounds i n micro-organisms 9. Role of the glyoxylate cycle in protozoal glyconeogenesis. Biochem. J., 86: 46-68. Holz, G. G . , Jr. 1964 Nutrition and metabolism of ciliates. In: Biochemistry and Physiology of Protozoa. Vol. 111. S. H. Hutner, ed. Academic Press, New York, pp. 199-242. Levy, M. R., and J. J. Wasmuth 1970 Effects of carbohydrate on glycolytic and peroxisomal enzymes in T e t r a h y m e n a . Biochim. Biophys. Acta, 201: 205-214. Levy, M., and E. E. Sisskin 1974 Neutral protease activity in T e t r a h y m e n a during a period of enzymic adjustment. Submitted for publication. 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: 265275. Muller, M. 1970 Release of hydrolases by Tetruh y m e n a pyriformis. J. Protozool., 17 (Suppl.): 13. 1972 Secretion of acid hydrolases and its intracellular source in T e t r u h y m e n a pyriform i s . J . Cell Biol., 52: 478-487. Muller, M., P. Baudhuin and C . de Duve 1966 Lysosomes in T e t r a h y m e n a pyriformis. I. Some properties and lysosomal localization of acid hydrolases. J. Cell Physiol., 68: 165-176. Porter, P., and J. J. Blum 1973 On the regulation of tyrosine transaminase, glutamic dehydrogenase and aspartic transaminase in T e t r a h y m e n a . Exptl. Cell Res., 77: 335-345. Rothstein, T. L., and J. J. Blum 1973 Effect of glucose, acetate, pyruvate, and carmine on intracellular content and extracellular release of three acid hydrolases. J. Cell Biol., 57: 63-41, 1974a Lysosomal physiology in Tetrah y m e n a . 111. Pharmacological studies on acid hydrolase release and the ingestion and egestion of dimethylbenzanthracene particles. J. Cell Biol., 6 2 : 844-859. 1974b Lysosomal physiology in Tetruhym e n a . IV. Effect of dichloroisoproterenol on the intracellular source of released acid hydrolases. Exptl. Cell Res., 87: 168-174. Suda, H., T. Aoyagi, M. Hamada, T. Takeuchi and H. Umezawa 1972 Antipain, a new protease inhibitor isolated from actinomycetes. J. Antibiotics, 25: 263-266. Umezawa, H., T. Aoyagi, H. Morishima, S. Kunimoto, M. Matsuzaki, M. Hamada and T. Takeuchi 1970 Chymostatin, a new chymotrypsin inhibitor produced by actinomycetes. J. Antibiotics, 23: 4 2 5 4 2 7 .

142

J. J. BLUM

Voichick, J., C. Elson, D. Granner and E. Shrago 1973 Relationship of adenosine 3’,5’-monophosphate to growth and metabolism of T e t r a h y m e n u pyriformis. J. Bact., 115: 68-72. Wagner, C. 1956 The glycogen metabolism of T e t r a h y m a a p y r i f o m i s . Ph.D. Thesis, University of Michigan, Ann Arbor. White, A,, P. Handler and E. L. Smith 1964 Principles of Biochemistry. McGraw-Hill, N.Y.

Whitlow, K. J., A. DIorio and C. Mavrides 1972 Regulation of the enzymes of tyrosine catabolism i n T e h a h y m e n a p y r i f o m i s . Biochim. Biophys. Acta, 264: 4 4 M 4 9 . Zahlten, R. N., A. A. Hochberg, F. W. Stratman and H. A. Lardy 1972 Glucagon-stimulated phosphorylation of mitochondria1 and lysosoma1 membranes of rat liver in viva Proc. Nat. Acad. Sci. (U.S.A.), 69 : 800-804.