- Email: [email protected]
Phosphohydrolytic activity in Paramecium caudatum at neutral pH
Acta histochem. 100,395-408 (1998) © Gustav Fischer Verlag
A~la
laisl.tte.iea
Phosphohydrolytic activity in Paramecium caudatum at neutral pH Kensuke Chikamori l , Tsutomu Arak? and Katsuyuki Sat0 3 I Department of Chemistry, Naruto University of Education, Takashima, Narutoelcho, Naruto-shi, Tokushima, 772-8502, 2 Department of Systems and Human Science, Graduate School of Engineering Science, Osaka University, Toyonaka, Osaka, 560-8531 and 3 Department of Biology, Naruto University of Education, Takashima, Naruto-cho, Naruto-shi, Tokushima, 772-8502, Japan
Accepted 12 July 1998
Summary Phosphohydrolytic activity was cytochemically characterized in Paramecium caudatum, a ciliated protozoa, at neutral pH. We stained cells in the presence of several mononucleotides as substrates, namely adenosine 5'-monophosphate (5'-AMP), adenosine 2'-monophosphate, guanosine 5'-monophosphate (5'-GMP) and p-glycerophosphate (p-GLP) using a lead capture method at 37°C. Cells were also incubated in the presence of 5'-AMP with the inhibitor for alkaline phosphatase, tetramisole. In all cases, varying amounts of final reaction product, lead sulfide, was observed in Paramecium cytoplasm. Tetramisole did not have any effect on Paramecium 5'-AMP hydrolytic activity. The phosphohydrolytic activity was measured as the increase in total absorbance of "test minus control" reactions at 440 nm per unit time after 20 min of incubation using a microphotometric system for image analysis that has been developed by us. From the relationship between the concentrations of 5'-AMP and activity, an apparent Km value was estimated to be 0.20 mM. These results suggest that mononucleotides and phosphate monoesters are hydrolyzed by one or more enzymes with wide substrate specificity in P caudatum. All the activity distribution patterns in Paramecium cultures, that were tested, were monomodal. The mean activity for 5'-AMP hydrolysis widely varied in these cultures. To investigate substrate affinity, distribution patterns and mean activity with 5'-AMP as substrate were compared with those in the presence of 2 other substrates, 5'GMP and p-GLP. Affinity of the enzyme(s) was similar for 5'-AMP and 5'GMP and lower for p-GLP. Correspondence to: K. Chikamori
396
K. Chikamori et al.
Key words: Paramecium caudatum - phosphohydrolytic activity - 5'-AMP enzyme histochemistry - image analysis - quantitative
Introduction In 1980s, Fok and coworkers investigated hydrolytic enzyme activities in relation to dynamics in intracellular digestion in Paramecium with biochemical and cytochemical means (Fok, 1983; Fok and Paeste, 1982; Fok et al., 1982, 1984). In a biochemical study, it was found that the activities depended on the age of the cultures (Fok and Paeste, 1982). Data were presented supporting the idea that Paramecium cells contain a distinct lysosomal acid phosphatase and hydrolysis of adenosine 5'-monophosphate (5'-AMP) is mainly performed by this enzyme even at neutral pH in Paramecium caudatum (Fok, 1983). Cytochemically, acid phosphatase activity was demonstrated in a specific stage of the digestive vacuole (DV), and was also found in endoplasmic reticulum (ER), some Golgi vesicles, and small vesicles which may represent primary lysosomes (Fok et al., 1984). From late 1980s to early 1990s, Van Noorden, Frederiks and coworkers intensively studied phosphohydrolytic enzymes in relation to diurnal variations in 5'-nucleotidase activity (Frederiks et al., 1987), changes in rat liver under experimentally induced pathological conditions in alkaline phosphatase and 5'-nucleotidase activity after partial hepatectomy (Van Noorden et al., 1988), acid phosphatase, alkaline phosphatase and 5'-nucleotidase activity during cholestasis (Frederiks et al., 1990), and 5'-nucleotidase activity in ischaemia (Frederiks et al., 1988). In the series of studies, the reliability of quantitative cytochemical analysis was indicated for the demonstration and measurement of enzymatic 5'-AMP hydrolysis at neutral pH. Van Noorden and Jonges (1992) indicated the possibility of a direct comparison of histochemical and biochemical data using the molar extinction coefficient of lead sulfide (PbS), the final reaction product of the histochemical phosphatase reaction. The aim of the present study was the characterization of Paramecium 5'AMP hydrolytic activity at neutral pH (pH 7.2) and the analysis of variation of the activity in individual cells in populations of P. caudatum. On the basis of the studies mentioned above, we investigated phosphohydrolytic activity at neutral pH (pH 7.2) using several mononucleotides and a phosphate monoester as natural substrates, because thus far, 5'-AMP hydrolysis in P. caudatum has not been studied cytochemically at neutral pH. For this purpose, we combined the cytochemical lead capture method according to Frederiks et al. (1987) with a cytophotometric method using a microphotometric system for image analysis that has been developed by ourselves (Chikamori et al., 1991). At first, we qualitatively characterized Paramecium phosphohydrolytic activity by its substrate affinity using 5'-AMP, adenosine 2'-monophosphate (2'-AMP), guanosine 5'-monophosphate (5'-GMP) and ftglycerophospahate (p-GLP) as natural substrates, and by the behavior of the enzyme(s) in the presence of tetramisole, a potent inhibitor of alkaline phosphatase in various tissues (Borgers, 1973). Then we investigated quantita-
Phosphohydrolytic activity in Paramecium
397
tively 5'-AMP hydrolytic activity in individual Paramecium cells to characterize its enzymatic properties and to reveal individual differences in activity in Paramecium populations. The density images (ODI) of single cells were obtained using monochromatic light of 440 nm to determine the amount of lead sulfide in the cells. Activity was expressed as total absorbance at 440 nm (TA). Using the relationship between TA of "test minus control" reactions (~TA) and incubation time, we expressed 5'-AMP hydrolytic activity as ~TA per unit time (~TA/min). In the presence of various 5'-AMP concentrations, we investigated 5'-AMP hydrolytic activity in different Paramecium populations and also determined its apparent K m value for 5'-AMP. To investigate substrate affinity of the phosphohydrolytic activity and to compare the data with those for 5'-AMP as substrate, cells were also incubated in the presence of 5'-GMP, p-GLP and detected quantitatively their hydrolysis in the same way as for 5' -AMP.
Materials and Methods Preparation of specimens and activity staining. P. caudatum cells were cultured and immobilized on the surface of cover slips as described previously (Chikamori et aI., 1998). Before the immobilization procedure, cover slips were washed thoroughly after immersing in a detergent dissolved in hot water for at least half a day and then kept in an ethanol-ether (1: 1) mixture before use. Cells, immobilized on the surface of cover slips, were dried in air with a blower for 1 min, then they were fixed in 4% formaldehyde dissolved in 0.1 M Tris-maleate buffer (pH 7.2) at 4°C for 5 min. After washing once in 0.1 M Tris-maleate buffer (pH 7.2) at room temperature, cells were dried in air with a blower for 1 min. This was essential to prevent cells from detaching from the cover slips during the staining procedures. Then, after washing twice in the same buffer, cells were incubated in a reaction mixture at 37°C for a specified period of time. Phosphohydrolytic activity was detected according to the method of Frederiks et al. (1987). The reaction mixture consisted of 0.1 M Tris-maleate buffer (pH 7.2), 1 mM 5'-AMP, 10 mM magnesium chloride and 3.6 mM lead nitrate. The reaction was stopped by immersing cover slips in ice-cold 4% formaldehyde dissolved in the same buffer (pH 7.2). After 30 min, cells were washed in 3 changes of distilled water each for 3 min and dehydrated in a graded series of ethanol. After coating with 2% celluloidin to prevent destruction of the cells, cells were hydrated and treated in 1% ammonium sulfide for 1 min. After rinsing 3 times in distilled water, the cover slips with cells were mounted in non-fluorescent glycerin. As controls, 5' -AMP and the other substrates were omitted from the reaction mixture. 2'-AMP, 5'-GMP and p-GLP were used instead of 5' -AMP for the test of substrate affinity at the final concentration of 1 mM. 1 mM tetramisole was added to the reaction mixture in the inhibition test. Mononucleotides and inhibitor were purchased from Sigma Chemical Co. (St Louis, MO, USA). p-GLP (disodium salt) was obtained from Tokyo Kasei Organic Chemicals (Tokyo, Japan). The other reagents were obtained from Wako Purechemical Industries (Osaka, Japan). They were used without further purification. Measurement of activity. Staining intensity was measured at 440 nm, the absorption maximum of the final reaction product in the range between 410 nm and 680 nm, by the microphotometric system for image analysis designed by ourselves. The instrumentation has been described previously (Chikamori et aI., 1991). From
398
K. Chikamori et a1.
the digitized ODI of stained cells (256' 256 pixels with 8 bit density), total absorbance (TA) values were estimated using the ODI of adjacent areas of the cells to correct for background density. Both determinations of absorbance of pixel and TA value were previously described (Chikamori, 1998). The commercially available application software (Delta graph 3.5; Nippon Polaroid, Tokyo, Japan) was used to fit the experimental data in an equation, when necessary. Statistical analysis. Differences in mean activity against different substrates were tested with one-way analysis of variance (ANOVA) and with the Scheffe test for multiple comparison; p = 0.05 was taken as the level of significance.
Results Light microscopic observation of phosphohydrolytic activity in the presence of various substrates and inhibitor. Final reaction product, the brownish lead sulfide, when using 5'-AMP as substrate was mainly deposited in food vacuoles in Paramecium cytoplasm (Fig. 1 A). Besides activity in the round food vacuole, dot-like or fibrous deposition of final reaction product was found in the cytoplasm. In the presence of 2'-AMP (Fig. 1 C) and of 5'-GMP (Fig. 1 D), final reaction product was also deposited in Paramecium cytoplasm in a similar manner as in the presence of 5'-AMP (Fig. 1 B). Staining intensity appeared to be somewhat lower when using 5'-GMP as substrate than when using 5'-AMP. Incubation in the presence of fi-GLP resulted in lower staining intensity than when using 5'-AMP (Fig. 1 E). Tetramisole did not have any effect on Paramecium 5'-AMP hydrolytic activity (Fig. 1 F). In control experiments, almost no final reaction product was found when incubations were performed in the absence of substrates. Data are not shown here, but will be presented as quantitative data in the following section. Measurement ofphosphohydrolytic activity using 5' -AMP as substrate. The absorption spectrum of the final reaction product was obtained from measurement of a final reaction product-containing food vacuole using a measuring spot of 10 11m in diameter at wavelengths between 410 nm and 680 nm taking steps of 30 nm, as shown in Fig. 2. The absorption maximum of final reaction product was at 440 nm in the visible region irrespective of incubation period. Fig. 3 shows the relationships between TA at 440 nm and incubation time in test and control experiments. In controls, TA was nearly constant and independent of time. In both test experiments, TA at 440 nm increased with time, but its relationships with time was nonlinear during the first 20 min,
Fig. 1. Photomicrographs of Paramecium cells stained for phosphohydrolytic activity in the presence of various substrates and of an inhibitor. A: Paramecium cell stained in the presence of 5'-AMP observed at high magnification. Arrows indicate round food vacuoles in Paramecium cytoplasm. Brownish final reaction product of lead sulfide indeposited in the vacuole. Paramecium cells stained in the presence of 1 mM of 5'-AMP (B), 2'-AMP (C), 5'-GMP (D), p-GLP (E), and 5'-AMP in the presence of 1 mM tetramisole (F) are shown at low magnification. Scale bar: 100 !-lm.
Phosphohydrolytic activity in Paramecium
A
B
: -.
.'
c
..,.
..
:
F
399
400
K. Chikamori et al.
1.5
Ql U C
ttl
-e ~
Ll c(
0.5
~
8
8
8
8
8
~
0'----'--....L.--.l...-----1.---'--...I-J 400 450 500 550 600 650 700 Wavelength (nm)
25 ~20
'"o
x
I
E15 c
o
'"~10 ttl
~
5
o
i
.. '
..... ~
I
. .0
X .. _... ~ ~: O. L...-_ _--el':.:.:....-_ _L...-_ _--'-..J •.........
o
20
40
Incubation time (min)
60
Fig. 2. Absorbance spectrum of brownish final reaction product deposited in a food vacuole in P caudatum. .: incubation lasted 60 min; 0: incubation lasted 15 min.
Fig. 3. Relationships between mean _total absorbance at 440 nm and incubation time. .: test-1; .: test-2; .&.: control; 0: test-1 minus control; 0: test-2 minus control. Vertical bars indicate standard deviation (n = 20). Dotted lines are regression lines for tests-1 and test-2 minus control. Test-1 minus control: y =30.4x-487, R2 = 0.998; test-2 minus control: y =9.18x-162, R2 = 0.970.
caused by a lag period. Plots of TA values of "test minus control" reactions (~TA) and time after 20 min showed linear relationships for both tests (test-1 minus control: y = 30.4x-487, R 2 = 0.998; test-2 minus control: y = 9.18x-162, R 2 = 0.970). Linear regression lines representing the reaction between 20 min and 60 min showed similar x-intercepts at 16.0 min for test-1 minus control and 17.6 min for test-2 minus control. Only by measuring TA at 60 min of the test reaction, we were able to calculate 5' -AMP hydrolytic activity as defined as the increase in ~TA per min (~TA/min), by equation (1). .. (ATA/ . ) _ TA at 60 min of test - 437 A ctlVlty Ll mm 60.0 _ 16.8
(1)
Where 437 is the mean value of TA in controls (S.D. = 61.3, n = 80) and 16.8 is the average of x-intercepts of the regression lines in Fig. 3. To investigate the enzymatic properties of Paramecium 5'-AMP hrdrolytic activity, cells from 2 cultures were stained in the presence of various con-
Phosphohydrolytic activity in Paramecium
401
25,------------------------,
0 0.05m 0 O.lmM
20
~ 0.15m
0 0.2m
~ 15
~ O.4mM
u
• •
'0 ... Q)
.J:J
§1O
z
0.6mM 1 mM
5
0-5
510
1015
15-20
20-25
AC\lVlly ( T
25-30
3035
3540
40-45
min)
Fig. 4. 5'-AMP hydrolytic activity distribution patterns in Paramecium cell populations at 7 different substrate concentrations. 5'-AMP concentrations ranging from 0.05 mM to 1 mM are indicated in the figure with respective symbols. Thirty cells were measured at each concentration. centrations of 5'-AMP, 0.05 mM, 0.1 mM, 0.15 mM, 0.2 mM, 0.4 mM, 0.6 mM and 1 mM, for 60 min and 30 cells were measured at each concentration. To correct the difference in activity between cultures, cells from both cultures were stained in the presence of 0.2 mM 5'-AMP and their mean activities were measured. As almost no difference in mean activity was detected for one culture (12.3 ± 8.4; mean ± S.D., n = 30) and for the other (12.7 ± 6.9; mean ± S.D., n = 30), we made activity distribution patterns in Paramecium populations in the presence of various substrate concentrations without any correction, as shown in Fig. 4. As expected, with increasing substrate concentrations in the incubation medium, the top of the histograms shifted to the right, meaning that higher amounts of final reaction products per cell were formed with increasing substrate concentrations. The distribution patterns did not change significantly when varying the concentration of substrate. Monomodal distribution profiles were observed when cells were incubated in the presence of more than 0.2 mM 5'-AMP. Mean activity of each group (n = 30) was plotted against the substrate concentration (Fig. 5). The experimental data fitted well in the MichaelisMenten equation (y = 23.0x/(0.201 + x), R 2 = 0.912). Wilkinson plots (Wilkinson, 1961) provided linear regression lines of 5'-AMP concentration in mM versus 5'-AMP concentration/5'-AMP hydrolytic activity in arbitrary unit of ~TAlmin (y = 0.0458x + 0.0091, R 2 = 0.944, p = 0.002), as shown in Fig. 6. From this, V max and apparent K m values were estimated at 21.8 arbitrary units of 5'-AMP hydrolytic activity and 0.20 mM, respectively.
402
K. Chikamori et al. 30 25
Fig. 5. Relationship between 5'-AMP concentration and Paramecium 5'-AMP hydrolytic activity. Mean activities with standard deviations (n =30) are plotted against substrate concentrations. Fitting curve: y = 23.0x/(0.201 + x); R 2 = 0.912.
5
0.2
0.4
0.6
0.8
5 '-AMP (mM)
g
6
c:
"E ~
E
4
'"",....0 X
o
'--_....I..-_----L.._---'_ _.l..-_....I..-....l
o
0.2
0.4
0.6
0.8
5'-AMP(mM)
25
A
20
•
5'-AMP
o
5'-GMP
~0 15 ~
§10
z
5 0
0-5
5·10
10-15 15-20 20-25 25-30 30-35 Activity (ATAlmin)
Fig. 6. Wilkinson plot of the relationship between 5'-AMP concentration and mean Paramecium 5'-AMP hydrolytic activity. Solid line indicates the first linear regression line; y = 0.0458x + 0.0091 (R2 = 0.944, p = 0.002).
Fig. 7. Activity distribution patterns in the presence of 1 mM 5'-GMP (A) and 1 mM fi-GLP (B). Activity distribution pattern for 5'-AMP is indicated in each pattern as a reference. This reference pattern was obtained from cells of the same culture as those used for 5'-GMP or fi-GLP.
Phosphohydrolytic activity in Paramecium
403
Substrate affinity of phosphohydrolytic activity in Paramecium. The activity distribution profiles were determined in the presence of different substrates (Fig. 7). The profiles and mean activities were compared in cells of identical cultures. In both cultures, 5'-AMP hydrolytic activity was distributed monomodally, yet one profile was considerably broader than the other. In the presence of 5' -GMP, the peak shifted to the left without any change in the mode of distribution. Mean activity (11.4 ± 6.0; mean ± S.D., n = 30) was lower than that in the presence of 5'-AMP (15.4 ± 5.1; mean ± S.D., n = 30). However, this difference is not statistically significant (p = 0.12). In the presence of p-GLP, approx. 75% of the cells tested showed low activity, yet some cells (approx. 7%) contained rather high activity against p-GLP. This resulted in a lower mean activity with a high standard deviation (4.5 ± 5.8; mean ± S.D., n = 30) than in the presence of 5'-AMP (16.8 ± 8.6; mean ± S.D., n = 30). This difference was statistically significant (p < 0.0001).
Discussion We have characterized phosphohydrolytic activity in P caudatum using natural substrates, several mononucleotides and a phosphate monoester, and an inhibitor at neutral pH (pH 7.2). Since Paramecium is a unicellular ciliated free organism, it has to be immobilized on the surface of cover slips by centrifugation as was described for bacterial cells by Barer (1991). In the present study, we found that the use of clean cover slips and 1 min air drying with a blower after pre-fixation were essential to prevent cells from detaching from cover slips during the preparation of specimens. Coating of the cells with 2 % celluloidin was required to protect cells from destruction in the ammonium sulfide solution. The combination of these procedures with pre- and post-fixation in 4% buffered formaldehyde (pH 7.2) resulted in a clear and valid demonstration of phosphohydrolytic activity in the presence of various substrates and an inhibitor (Fig. 1). Frederiks et al. (1987) described that short periods of fixation reduce 5'-nucleotidase activity, a 5'AMP-specific hydrolytic enzyme, by 20% only whereas a good localization of final reaction product was observed. Since we did not measure 5'-AMP hydrolytic activity without pre-fixation in the present work, we have no data about the effect of the fixative on the activity. 5'-AMP hydrolytic activity at neutral pH was mainly present in round food vacuoles in Paramecium cytoplasm. The other mononucleotides tested as substrate with the phosphate group at a different position or a different base moiety and the phosphate monoester also yielded a dot-like deposition of final reaction product in Paramecium, although staining intensities were generally lower. Tetramisole, a potent inhibitor of alkaline phosphatase (Borgers, 1973), did not have any effects on 5' -AMP hydrolysis. These results indicate that the Paramecium phosphohydrolytic activity at neutral pH is not specific for the position of phosphate group or the base moiety but
404
K. Chikamori et al.
specific only for the phosphate ester bond. Furthermore it is not attributable to alkaline phosphatase. Biochemical data on 5'-AMP hydrolytic activity at neutral pH are limited. Fok (1983) measured hydrolytic activities using three different substrates, 5'-AMp, p-GLP and glucose-6-phosphate as substrates in P. caudatum homogenates at various pH. All these activities showed a maximum at pH 5.0 with comparable specific activity and their major portions were found in lysosomal fractions. Hydrolysis of AMP at pH 7.5 was approx. 50% of the maximal activity at pH 5.0 and its major portion was found in lysosomal fractions of P. caudatum. p-GLP hydrolysis was also detected at pH 7.5, but its activity ratio to maximal activity (approx. 8%) was much lower than that of AMP hydrolytic activity. This is in agreement with our findings that staining intensities of cells in the presence of 5'AMP were higher than in the presence of p-GLP. Food vacuoles in P. caudatum were cytologically grouped into four orderly stages of digestion by Fok et al. (1982) as DV-I, DV-II, DV-III and DV-IY. The presence of acid phosphatase activity in DV-III has been verified using the Gomori technique in Paramecium (Muller and Toro, 1962; Fok et al. 1984). Combining our microscopic findings in Fig. 1 with their biochemical and cytological findings indicate that we demonstrated acid phosphatase activity in food vacuoles corresponding to DV-III at neutral pH. Fibrous or dot-like deposition of final reaction product near the macronucleus may be due to acid phosphatase activity in cytoplasmic organelles such as ER, some Golgi vesicles and primary lysosome-like vesicles as described by Fok et al. (1984). Unfortunately, TA at 440 urn showed a nonlinear concave increase with time with a lag period of approx. 20 min (Fig. 2), although we used the same staining method as Frederiks et al. (1987) and measured at 440 nm, nearly the same wavelength they used (450 nm). They indicated a linear relationship between mean integrated absorbance at 450 nm and reaction time up to 60 min in the study of rat liver 5'-nucleotidase. Van Noorden and longes (1992) also indicated a linear relationship between absorbance at 450 nm and the concentration of lead sulfide in solidified gelatin solutions. As Frederiks et al. (1987) used cryostat sections of 8 ~m thickness, this discrepancy appears to be attributable to the penetration of substrate and other constituents in the reaction mixture across the Paramecium cell membrane; one of the obstacles in cytochemistry of intact cells. Lund-Hansen et al. (1984) described that air-drying was an effective means of rendering cultured human fibroblasts permeable to an indoxyl substrate. Barer (1991) proposed the possibility of disruption of relatively hydrophilic barriers by air-drying. However, it is questionable whether their results can be directly compared with our results because of the differences in chemical nature of the substrates used. Another explanation for the lag phase could be the time needed for building up sufficient liberated phosphate in the cells to obtain precipitate. Van Duijn and co-workers have studied this phenomenon in great detail especially for enzyme reactions visualized with lead precipitation methods (Van Duijn, 1992). The lag phase is dependent on the rate of production of phosphate and this could explain that we found a lag phase, whereas others
Phosphohydrolytic activity in Paramecium
405
did not. However, ~TA, TA values of "test minus control" reactions were linearly after 20 min of incubation in both tests (Fig. 2). Furthermore, the fact that "test minus control" reactions showed Michaelis-Menten's kinetics in relation to the substrate concentration also indicated the validity of the rate of reaction after approx. 20 min in the presence of various substrate concentrations. As all the substrates that were used in the present study were 5'-AMP related compounds in their chemical structures and we stained cells for rather long periods (60 min), we have assumed that all the reactions proceeded in a similar manner as in the presence of 5'-AMP and that tetramisole also penetrated into the cells. The mean value of 5'-AMP hydrolytic activity widely varied in each culture, in particular, in 2 of 5 cultures tested. Hydrolytic activity in P. caudatum depends on the age of culture (Fok and Paeste, 1982). We used cell in an early stationary phase in the present study. In this phase, a relatively steep decrease in acid phosphatase activity was found using p-GLP as substrate. The number of acid phosphatase activity-positive DVs changed with time after a pulse of polystyrene latex spheres (Fok et al. 1984). Therefore, it is likely that the timing of cell-harvesting greatly influences activity as well as minor changes in culture condition. Thus when we compared activity and activity distribution patterns, we used cells that were harvested at the same time from identical cultures (Fig. 7). If not possible, we stained cells in the presence of similar substrate concentrations and measured in 2 cultures to correct data as mentioned in the experiment to analyze the relationship between activity and 5'-AMP concentration (Figs. 4, 5 and 6). Irrespective of the 5'-AMP concentration and mean activity, activity distribution patterns were monomodal (Figs. 4 and 7). We proposed previously that it was bimodal in an early stationary phase of P. caudatum for succinate dehydrogenase activity (Chikamori et al., 1998). This suggest that the activity distribution profile may be different for each enzyme even in an identical culture. The mean activity well fitted to the Michaelis-Menten equation in relation to the 5'-AMP concentration. From the linear Wilkinson plot, we could estimate an apparent Km value for 5'-AMP of 0.20 mM at pH 7.2. Although direct comparison is not possible because of the use of different pH, Fok (1983) reported a Km value of 0.42 mM at pH 5.0 for AMP hydrolysis by lysosomal acid phosphatase in P. caudatum. To investigate substrate affinity, the activity and distribution patterns of 5'-AMP hydrolysis were compared with those of two other substrates at the same concentration of 1 mM. At this concentration, the affinity was similar for 5'-AMP and 5'-GMP and lower for p-GLP. On the basis of microscopic observations and the measurements, the affinity for 5'-GMP was expected to be lower than that for 5'-AMP (Figs. 1 D and 7 A). However, the difference was not statistically significant. We found some cells showing rather high p-GLP hydrolytic activity (Figs. 1 D and 7 B). These findings indicate that it is necessary to estimate kinetic constants of phosphohydrolytic enzyme activity for various substrates to confirm its substrate affinity at neutral pH.
406
K. Chikamori et al.
We started the present study by investigating whether true adenosine 5'monophosphatase (5'-nucleotidase) could be detected in P. caudatum or not. Our cytochemical results from both microscopic observations and measurements were nearly consistent with the biochemical data of Fok (1983) that 5'AMP hydrolytic activity at neutral pH is attributable to acid phosphatase with a broad substrate specificity and 5'-nucleotidase is lacking in P. caudatum. Recently, reports have been published about the application of whole living organisms or cells as biosensors or test organisms for environmental monitoring based on cytochemical methods. Kinetic parameters of glucose6-phosphate dehydrogenase in livers of North Sea flat fish were established (Van Noorden et aI., 1997; Kohler and Van Noorden, 1998) and the marine dinoflagellate Prorocentrum micans was used as test organism to measure parathion toxicity using fluorescence induction kinetic measurements (Prevot et aI., 1993). Araki et al. (1995) also proposed the use of Paramecium cells as a single cell biosensor for the detection of pH, herbicide and Ni+ contents in a microenvironment on the basis of their movement and fluorescence intensity of rhodamine 123, an indicator of mitochondrial membrane potential. In the previous study of P. caudatum, we have investigated succinate dehydrogenase activity as a measure of its energy metabolism (Chikamori, 1998). Our results presented here have provided one more parameter of the physiological state of P. caudatum under contaminant-free conditions. It will contribute to the application of Paramecium cell as single cell biological sensors in environmental monitoring.
Acknowledgement We gratefully thank Dr. Kazuyuki Mikami, Miyagi University of Education for providing kindly a strain of P. caudatum, and it is a pleasure for us to acknowledge the skilful technical assistance of Yoko Ueyama, Department of Chemistry, Naruto University of Education and Kei Horie, Department of Mechanical Engineering, University of Tokushima (present address: Nikon Co., Tokyo, Japan) during the experiments, and Dr. Shinichi Kudo, Department of Biology, Naruto University of Education, for the statistical treatment of the data. We gratefully thank Dr. C. 1. F. Van Noorden, Laboratory of Cell Biology and Histology, Academic Medical Center, University of Amsterdam, for his critical reading of the manuscript and valuable suggestions. This work was partly supported by Grant-in-Aid for Exploratory Research (09875070) and for Scientific Research (10558134) from The Ministry of Education in Japan.
References Araki T, Misawa H, Chikamori K, and Tohno Y (1995) A biological-sensor utilizing a living cell for environment monitoring in micro space. Proceedings of the International Symposium on Microsystem, Intelligent Materials and Robots, September 27-29,1995, Sendai, Japan, pp 301-304
Phosphohydrolytic activity in Paramecium
407
Barer MR (1991) New possibilities for bacterial cytochemistry: light microscopic demonstration of p-galactosidase in unfixed immobilized bacteria. Histochem J 23: 529-533 Borgers M (1973) The cytochemical application of new potent inhibitors of alkaline phosphatases. J Histochem Cytochem 21: 812-824 Chikamori K, Yamamoto A, and Araki T (1991) A microphotometer for multispectral image analysis. Instrumentation and application to multicolor-stained specimen. Acta Histochem Cytochem 24: 395-403 Chikamori K, Araki T, and Sato K (1998) Succinate dehydrogenase (SDH) activity in single Paramecium caudatum cells. Acta histochem 100: 25-36 Fok AK, Lee Y, and Allen RD (1982) The correlation of digestive vacuole pH and size with the digestive cycle in Paramecium caudatum. J Protozool29: 409-414 Fok AK, and Paeste RM (1982) Lysosomal enzymes of Paramecium caudatum and Paramecium tetraurelia. Exp Cell Res 139: 159-169 Fok AK (1983) An inhibition and kinetic study of acid phosphatase in Paramecium caudatum and Paramecium tetraurelia. J Protozool 30: 14-20 Fok AK, Muraoka JH, and Allen RD (1984) Acid phosphatase in the digestive vacuole and lysosomes of Paramecium caudatum: A timed study. J Protozool 31: 216-220 Frederiks WM, Marx F, Bosch KS and Van Noorden CJF (1987) Diurnal variation in 5'-nucleotidase activity in rat liver. A quantitative histochemical study. Histochemistry 87: 439-443 Frederiks WM, Marx F, and Myagkaya GL (1988) A quantitative histochemical study of 5'-nucleotidase activity in rat liver after ischaemia. J Phathol 154: 277286 Frederiks WM, Van Noorden CJF, Aronson DC, Marx F, Bosch KS, Jonges GN, Vogels IMC, and James J (1990) Quantitative changes in acid phosphatase, alkaline phosphatase and 5'-nucleotidase activity in rat liver after experimentally induced cholestasis. Liver 10: 158-166 Kohler A, and Van Noorden CJF (1998) Initial velocities in situ of G6PDH and PGDH and expression of proliferating cell nuclear antigen (PCNA): sensitive diagnostic markers of environmentally induced hepatocellular carcinogenesis in a marine flatfish (Platichthys flesus L.). Aquat Toxicol 40: 233-252 Lund-Hansen T, Hoyer PE, and Andersen H (1984) A quantitative cytochemical assay of p-galactosidase in single cultured human skin fibroblast. Histochemistry 81: 321-330 Muller M, and Ton) (1962) Studies on feeding and digestion in protozoa. III. Acid phosphatase activity in food vacuoles Paramecium multimicronucleatum. J Protozool9: 98-102 Prevot P, Perret E, Jupin H, and Soyer-Gobillard MO (1993) Fluorescence induction used to measure parathion toxicity in the marine dinoflagellate Prorocentrum micans: A possible model for biodetection. Ecotoxicol Environment Safety 25: 360371
Van Duijn P (1992) Model systems. Principles and practice of the use of matrix immobilized enzymes for the study of the fundamental aspects of cytochemical enzyme methods. In: Stoward P and Pearse AGE (Eds) Histochemistry. Theoretical and applied vol 3, 4th ed. Churchill Livingstone, Edinburgh, pp 433-472 Van Noorden CJF, Vogels IMC, and Houtkooper JM (1988) Cytophotometric analysis of alkaline phosphatase and 5'-nucleotidase activity in regenerating rat liver after partial hepatectomy. Cell Biochem Function 6: 53-60
408
K. Chikamori et al.
Van Noorden CJF, and Jonges GN (1992) Molecular extinction coefficients of lead sulfide and polymerized diaminobenzidine as final reaction products of histochemical phosphatase reactions. Cytometry 13: 644-648 Van Noorden CJF, Bahns S, and Kohler A (1997) Adaptational changes in kinetic parameters of G6PDH but not of PGDH during contamination-induced carcinogenesis in livers of North Sea flatfish. Biochim Biophys Acta 1342: 141-148 Wilkinson GN (1961) Statistical estimations in enzyme kinetics. Biochem J 80: 324332
A~la
laisl.tte.iea
Phosphohydrolytic activity in Paramecium caudatum at neutral pH Kensuke Chikamori l , Tsutomu Arak? and Katsuyuki Sat0 3 I Department of Chemistry, Naruto University of Education, Takashima, Narutoelcho, Naruto-shi, Tokushima, 772-8502, 2 Department of Systems and Human Science, Graduate School of Engineering Science, Osaka University, Toyonaka, Osaka, 560-8531 and 3 Department of Biology, Naruto University of Education, Takashima, Naruto-cho, Naruto-shi, Tokushima, 772-8502, Japan
Accepted 12 July 1998
Summary Phosphohydrolytic activity was cytochemically characterized in Paramecium caudatum, a ciliated protozoa, at neutral pH. We stained cells in the presence of several mononucleotides as substrates, namely adenosine 5'-monophosphate (5'-AMP), adenosine 2'-monophosphate, guanosine 5'-monophosphate (5'-GMP) and p-glycerophosphate (p-GLP) using a lead capture method at 37°C. Cells were also incubated in the presence of 5'-AMP with the inhibitor for alkaline phosphatase, tetramisole. In all cases, varying amounts of final reaction product, lead sulfide, was observed in Paramecium cytoplasm. Tetramisole did not have any effect on Paramecium 5'-AMP hydrolytic activity. The phosphohydrolytic activity was measured as the increase in total absorbance of "test minus control" reactions at 440 nm per unit time after 20 min of incubation using a microphotometric system for image analysis that has been developed by us. From the relationship between the concentrations of 5'-AMP and activity, an apparent Km value was estimated to be 0.20 mM. These results suggest that mononucleotides and phosphate monoesters are hydrolyzed by one or more enzymes with wide substrate specificity in P caudatum. All the activity distribution patterns in Paramecium cultures, that were tested, were monomodal. The mean activity for 5'-AMP hydrolysis widely varied in these cultures. To investigate substrate affinity, distribution patterns and mean activity with 5'-AMP as substrate were compared with those in the presence of 2 other substrates, 5'GMP and p-GLP. Affinity of the enzyme(s) was similar for 5'-AMP and 5'GMP and lower for p-GLP. Correspondence to: K. Chikamori
396
K. Chikamori et al.
Key words: Paramecium caudatum - phosphohydrolytic activity - 5'-AMP enzyme histochemistry - image analysis - quantitative
Introduction In 1980s, Fok and coworkers investigated hydrolytic enzyme activities in relation to dynamics in intracellular digestion in Paramecium with biochemical and cytochemical means (Fok, 1983; Fok and Paeste, 1982; Fok et al., 1982, 1984). In a biochemical study, it was found that the activities depended on the age of the cultures (Fok and Paeste, 1982). Data were presented supporting the idea that Paramecium cells contain a distinct lysosomal acid phosphatase and hydrolysis of adenosine 5'-monophosphate (5'-AMP) is mainly performed by this enzyme even at neutral pH in Paramecium caudatum (Fok, 1983). Cytochemically, acid phosphatase activity was demonstrated in a specific stage of the digestive vacuole (DV), and was also found in endoplasmic reticulum (ER), some Golgi vesicles, and small vesicles which may represent primary lysosomes (Fok et al., 1984). From late 1980s to early 1990s, Van Noorden, Frederiks and coworkers intensively studied phosphohydrolytic enzymes in relation to diurnal variations in 5'-nucleotidase activity (Frederiks et al., 1987), changes in rat liver under experimentally induced pathological conditions in alkaline phosphatase and 5'-nucleotidase activity after partial hepatectomy (Van Noorden et al., 1988), acid phosphatase, alkaline phosphatase and 5'-nucleotidase activity during cholestasis (Frederiks et al., 1990), and 5'-nucleotidase activity in ischaemia (Frederiks et al., 1988). In the series of studies, the reliability of quantitative cytochemical analysis was indicated for the demonstration and measurement of enzymatic 5'-AMP hydrolysis at neutral pH. Van Noorden and Jonges (1992) indicated the possibility of a direct comparison of histochemical and biochemical data using the molar extinction coefficient of lead sulfide (PbS), the final reaction product of the histochemical phosphatase reaction. The aim of the present study was the characterization of Paramecium 5'AMP hydrolytic activity at neutral pH (pH 7.2) and the analysis of variation of the activity in individual cells in populations of P. caudatum. On the basis of the studies mentioned above, we investigated phosphohydrolytic activity at neutral pH (pH 7.2) using several mononucleotides and a phosphate monoester as natural substrates, because thus far, 5'-AMP hydrolysis in P. caudatum has not been studied cytochemically at neutral pH. For this purpose, we combined the cytochemical lead capture method according to Frederiks et al. (1987) with a cytophotometric method using a microphotometric system for image analysis that has been developed by ourselves (Chikamori et al., 1991). At first, we qualitatively characterized Paramecium phosphohydrolytic activity by its substrate affinity using 5'-AMP, adenosine 2'-monophosphate (2'-AMP), guanosine 5'-monophosphate (5'-GMP) and ftglycerophospahate (p-GLP) as natural substrates, and by the behavior of the enzyme(s) in the presence of tetramisole, a potent inhibitor of alkaline phosphatase in various tissues (Borgers, 1973). Then we investigated quantita-
Phosphohydrolytic activity in Paramecium
397
tively 5'-AMP hydrolytic activity in individual Paramecium cells to characterize its enzymatic properties and to reveal individual differences in activity in Paramecium populations. The density images (ODI) of single cells were obtained using monochromatic light of 440 nm to determine the amount of lead sulfide in the cells. Activity was expressed as total absorbance at 440 nm (TA). Using the relationship between TA of "test minus control" reactions (~TA) and incubation time, we expressed 5'-AMP hydrolytic activity as ~TA per unit time (~TA/min). In the presence of various 5'-AMP concentrations, we investigated 5'-AMP hydrolytic activity in different Paramecium populations and also determined its apparent K m value for 5'-AMP. To investigate substrate affinity of the phosphohydrolytic activity and to compare the data with those for 5'-AMP as substrate, cells were also incubated in the presence of 5'-GMP, p-GLP and detected quantitatively their hydrolysis in the same way as for 5' -AMP.
Materials and Methods Preparation of specimens and activity staining. P. caudatum cells were cultured and immobilized on the surface of cover slips as described previously (Chikamori et aI., 1998). Before the immobilization procedure, cover slips were washed thoroughly after immersing in a detergent dissolved in hot water for at least half a day and then kept in an ethanol-ether (1: 1) mixture before use. Cells, immobilized on the surface of cover slips, were dried in air with a blower for 1 min, then they were fixed in 4% formaldehyde dissolved in 0.1 M Tris-maleate buffer (pH 7.2) at 4°C for 5 min. After washing once in 0.1 M Tris-maleate buffer (pH 7.2) at room temperature, cells were dried in air with a blower for 1 min. This was essential to prevent cells from detaching from the cover slips during the staining procedures. Then, after washing twice in the same buffer, cells were incubated in a reaction mixture at 37°C for a specified period of time. Phosphohydrolytic activity was detected according to the method of Frederiks et al. (1987). The reaction mixture consisted of 0.1 M Tris-maleate buffer (pH 7.2), 1 mM 5'-AMP, 10 mM magnesium chloride and 3.6 mM lead nitrate. The reaction was stopped by immersing cover slips in ice-cold 4% formaldehyde dissolved in the same buffer (pH 7.2). After 30 min, cells were washed in 3 changes of distilled water each for 3 min and dehydrated in a graded series of ethanol. After coating with 2% celluloidin to prevent destruction of the cells, cells were hydrated and treated in 1% ammonium sulfide for 1 min. After rinsing 3 times in distilled water, the cover slips with cells were mounted in non-fluorescent glycerin. As controls, 5' -AMP and the other substrates were omitted from the reaction mixture. 2'-AMP, 5'-GMP and p-GLP were used instead of 5' -AMP for the test of substrate affinity at the final concentration of 1 mM. 1 mM tetramisole was added to the reaction mixture in the inhibition test. Mononucleotides and inhibitor were purchased from Sigma Chemical Co. (St Louis, MO, USA). p-GLP (disodium salt) was obtained from Tokyo Kasei Organic Chemicals (Tokyo, Japan). The other reagents were obtained from Wako Purechemical Industries (Osaka, Japan). They were used without further purification. Measurement of activity. Staining intensity was measured at 440 nm, the absorption maximum of the final reaction product in the range between 410 nm and 680 nm, by the microphotometric system for image analysis designed by ourselves. The instrumentation has been described previously (Chikamori et aI., 1991). From
398
K. Chikamori et a1.
the digitized ODI of stained cells (256' 256 pixels with 8 bit density), total absorbance (TA) values were estimated using the ODI of adjacent areas of the cells to correct for background density. Both determinations of absorbance of pixel and TA value were previously described (Chikamori, 1998). The commercially available application software (Delta graph 3.5; Nippon Polaroid, Tokyo, Japan) was used to fit the experimental data in an equation, when necessary. Statistical analysis. Differences in mean activity against different substrates were tested with one-way analysis of variance (ANOVA) and with the Scheffe test for multiple comparison; p = 0.05 was taken as the level of significance.
Results Light microscopic observation of phosphohydrolytic activity in the presence of various substrates and inhibitor. Final reaction product, the brownish lead sulfide, when using 5'-AMP as substrate was mainly deposited in food vacuoles in Paramecium cytoplasm (Fig. 1 A). Besides activity in the round food vacuole, dot-like or fibrous deposition of final reaction product was found in the cytoplasm. In the presence of 2'-AMP (Fig. 1 C) and of 5'-GMP (Fig. 1 D), final reaction product was also deposited in Paramecium cytoplasm in a similar manner as in the presence of 5'-AMP (Fig. 1 B). Staining intensity appeared to be somewhat lower when using 5'-GMP as substrate than when using 5'-AMP. Incubation in the presence of fi-GLP resulted in lower staining intensity than when using 5'-AMP (Fig. 1 E). Tetramisole did not have any effect on Paramecium 5'-AMP hydrolytic activity (Fig. 1 F). In control experiments, almost no final reaction product was found when incubations were performed in the absence of substrates. Data are not shown here, but will be presented as quantitative data in the following section. Measurement ofphosphohydrolytic activity using 5' -AMP as substrate. The absorption spectrum of the final reaction product was obtained from measurement of a final reaction product-containing food vacuole using a measuring spot of 10 11m in diameter at wavelengths between 410 nm and 680 nm taking steps of 30 nm, as shown in Fig. 2. The absorption maximum of final reaction product was at 440 nm in the visible region irrespective of incubation period. Fig. 3 shows the relationships between TA at 440 nm and incubation time in test and control experiments. In controls, TA was nearly constant and independent of time. In both test experiments, TA at 440 nm increased with time, but its relationships with time was nonlinear during the first 20 min,
Fig. 1. Photomicrographs of Paramecium cells stained for phosphohydrolytic activity in the presence of various substrates and of an inhibitor. A: Paramecium cell stained in the presence of 5'-AMP observed at high magnification. Arrows indicate round food vacuoles in Paramecium cytoplasm. Brownish final reaction product of lead sulfide indeposited in the vacuole. Paramecium cells stained in the presence of 1 mM of 5'-AMP (B), 2'-AMP (C), 5'-GMP (D), p-GLP (E), and 5'-AMP in the presence of 1 mM tetramisole (F) are shown at low magnification. Scale bar: 100 !-lm.
Phosphohydrolytic activity in Paramecium
A
B
: -.
.'
c
..,.
..
:
F
399
400
K. Chikamori et al.
1.5
Ql U C
ttl
-e ~
Ll c(
0.5
~
8
8
8
8
8
~
0'----'--....L.--.l...-----1.---'--...I-J 400 450 500 550 600 650 700 Wavelength (nm)
25 ~20
'"o
x
I
E15 c
o
'"~10 ttl
~
5
o
i
.. '
..... ~
I
. .0
X .. _... ~ ~: O. L...-_ _--el':.:.:....-_ _L...-_ _--'-..J •.........
o
20
40
Incubation time (min)
60
Fig. 2. Absorbance spectrum of brownish final reaction product deposited in a food vacuole in P caudatum. .: incubation lasted 60 min; 0: incubation lasted 15 min.
Fig. 3. Relationships between mean _total absorbance at 440 nm and incubation time. .: test-1; .: test-2; .&.: control; 0: test-1 minus control; 0: test-2 minus control. Vertical bars indicate standard deviation (n = 20). Dotted lines are regression lines for tests-1 and test-2 minus control. Test-1 minus control: y =30.4x-487, R2 = 0.998; test-2 minus control: y =9.18x-162, R2 = 0.970.
caused by a lag period. Plots of TA values of "test minus control" reactions (~TA) and time after 20 min showed linear relationships for both tests (test-1 minus control: y = 30.4x-487, R 2 = 0.998; test-2 minus control: y = 9.18x-162, R 2 = 0.970). Linear regression lines representing the reaction between 20 min and 60 min showed similar x-intercepts at 16.0 min for test-1 minus control and 17.6 min for test-2 minus control. Only by measuring TA at 60 min of the test reaction, we were able to calculate 5' -AMP hydrolytic activity as defined as the increase in ~TA per min (~TA/min), by equation (1). .. (ATA/ . ) _ TA at 60 min of test - 437 A ctlVlty Ll mm 60.0 _ 16.8
(1)
Where 437 is the mean value of TA in controls (S.D. = 61.3, n = 80) and 16.8 is the average of x-intercepts of the regression lines in Fig. 3. To investigate the enzymatic properties of Paramecium 5'-AMP hrdrolytic activity, cells from 2 cultures were stained in the presence of various con-
Phosphohydrolytic activity in Paramecium
401
25,------------------------,
0 0.05m 0 O.lmM
20
~ 0.15m
0 0.2m
~ 15
~ O.4mM
u
• •
'0 ... Q)
.J:J
§1O
z
0.6mM 1 mM
5
0-5
510
1015
15-20
20-25
AC\lVlly ( T
25-30
3035
3540
40-45
min)
Fig. 4. 5'-AMP hydrolytic activity distribution patterns in Paramecium cell populations at 7 different substrate concentrations. 5'-AMP concentrations ranging from 0.05 mM to 1 mM are indicated in the figure with respective symbols. Thirty cells were measured at each concentration. centrations of 5'-AMP, 0.05 mM, 0.1 mM, 0.15 mM, 0.2 mM, 0.4 mM, 0.6 mM and 1 mM, for 60 min and 30 cells were measured at each concentration. To correct the difference in activity between cultures, cells from both cultures were stained in the presence of 0.2 mM 5'-AMP and their mean activities were measured. As almost no difference in mean activity was detected for one culture (12.3 ± 8.4; mean ± S.D., n = 30) and for the other (12.7 ± 6.9; mean ± S.D., n = 30), we made activity distribution patterns in Paramecium populations in the presence of various substrate concentrations without any correction, as shown in Fig. 4. As expected, with increasing substrate concentrations in the incubation medium, the top of the histograms shifted to the right, meaning that higher amounts of final reaction products per cell were formed with increasing substrate concentrations. The distribution patterns did not change significantly when varying the concentration of substrate. Monomodal distribution profiles were observed when cells were incubated in the presence of more than 0.2 mM 5'-AMP. Mean activity of each group (n = 30) was plotted against the substrate concentration (Fig. 5). The experimental data fitted well in the MichaelisMenten equation (y = 23.0x/(0.201 + x), R 2 = 0.912). Wilkinson plots (Wilkinson, 1961) provided linear regression lines of 5'-AMP concentration in mM versus 5'-AMP concentration/5'-AMP hydrolytic activity in arbitrary unit of ~TAlmin (y = 0.0458x + 0.0091, R 2 = 0.944, p = 0.002), as shown in Fig. 6. From this, V max and apparent K m values were estimated at 21.8 arbitrary units of 5'-AMP hydrolytic activity and 0.20 mM, respectively.
402
K. Chikamori et al. 30 25
Fig. 5. Relationship between 5'-AMP concentration and Paramecium 5'-AMP hydrolytic activity. Mean activities with standard deviations (n =30) are plotted against substrate concentrations. Fitting curve: y = 23.0x/(0.201 + x); R 2 = 0.912.
5
0.2
0.4
0.6
0.8
5 '-AMP (mM)
g
6
c:
"E ~
E
4
'"",....0 X
o
'--_....I..-_----L.._---'_ _.l..-_....I..-....l
o
0.2
0.4
0.6
0.8
5'-AMP(mM)
25
A
20
•
5'-AMP
o
5'-GMP
~0 15 ~
§10
z
5 0
0-5
5·10
10-15 15-20 20-25 25-30 30-35 Activity (ATAlmin)
Fig. 6. Wilkinson plot of the relationship between 5'-AMP concentration and mean Paramecium 5'-AMP hydrolytic activity. Solid line indicates the first linear regression line; y = 0.0458x + 0.0091 (R2 = 0.944, p = 0.002).
Fig. 7. Activity distribution patterns in the presence of 1 mM 5'-GMP (A) and 1 mM fi-GLP (B). Activity distribution pattern for 5'-AMP is indicated in each pattern as a reference. This reference pattern was obtained from cells of the same culture as those used for 5'-GMP or fi-GLP.
Phosphohydrolytic activity in Paramecium
403
Substrate affinity of phosphohydrolytic activity in Paramecium. The activity distribution profiles were determined in the presence of different substrates (Fig. 7). The profiles and mean activities were compared in cells of identical cultures. In both cultures, 5'-AMP hydrolytic activity was distributed monomodally, yet one profile was considerably broader than the other. In the presence of 5' -GMP, the peak shifted to the left without any change in the mode of distribution. Mean activity (11.4 ± 6.0; mean ± S.D., n = 30) was lower than that in the presence of 5'-AMP (15.4 ± 5.1; mean ± S.D., n = 30). However, this difference is not statistically significant (p = 0.12). In the presence of p-GLP, approx. 75% of the cells tested showed low activity, yet some cells (approx. 7%) contained rather high activity against p-GLP. This resulted in a lower mean activity with a high standard deviation (4.5 ± 5.8; mean ± S.D., n = 30) than in the presence of 5'-AMP (16.8 ± 8.6; mean ± S.D., n = 30). This difference was statistically significant (p < 0.0001).
Discussion We have characterized phosphohydrolytic activity in P caudatum using natural substrates, several mononucleotides and a phosphate monoester, and an inhibitor at neutral pH (pH 7.2). Since Paramecium is a unicellular ciliated free organism, it has to be immobilized on the surface of cover slips by centrifugation as was described for bacterial cells by Barer (1991). In the present study, we found that the use of clean cover slips and 1 min air drying with a blower after pre-fixation were essential to prevent cells from detaching from cover slips during the preparation of specimens. Coating of the cells with 2 % celluloidin was required to protect cells from destruction in the ammonium sulfide solution. The combination of these procedures with pre- and post-fixation in 4% buffered formaldehyde (pH 7.2) resulted in a clear and valid demonstration of phosphohydrolytic activity in the presence of various substrates and an inhibitor (Fig. 1). Frederiks et al. (1987) described that short periods of fixation reduce 5'-nucleotidase activity, a 5'AMP-specific hydrolytic enzyme, by 20% only whereas a good localization of final reaction product was observed. Since we did not measure 5'-AMP hydrolytic activity without pre-fixation in the present work, we have no data about the effect of the fixative on the activity. 5'-AMP hydrolytic activity at neutral pH was mainly present in round food vacuoles in Paramecium cytoplasm. The other mononucleotides tested as substrate with the phosphate group at a different position or a different base moiety and the phosphate monoester also yielded a dot-like deposition of final reaction product in Paramecium, although staining intensities were generally lower. Tetramisole, a potent inhibitor of alkaline phosphatase (Borgers, 1973), did not have any effects on 5' -AMP hydrolysis. These results indicate that the Paramecium phosphohydrolytic activity at neutral pH is not specific for the position of phosphate group or the base moiety but
404
K. Chikamori et al.
specific only for the phosphate ester bond. Furthermore it is not attributable to alkaline phosphatase. Biochemical data on 5'-AMP hydrolytic activity at neutral pH are limited. Fok (1983) measured hydrolytic activities using three different substrates, 5'-AMp, p-GLP and glucose-6-phosphate as substrates in P. caudatum homogenates at various pH. All these activities showed a maximum at pH 5.0 with comparable specific activity and their major portions were found in lysosomal fractions. Hydrolysis of AMP at pH 7.5 was approx. 50% of the maximal activity at pH 5.0 and its major portion was found in lysosomal fractions of P. caudatum. p-GLP hydrolysis was also detected at pH 7.5, but its activity ratio to maximal activity (approx. 8%) was much lower than that of AMP hydrolytic activity. This is in agreement with our findings that staining intensities of cells in the presence of 5'AMP were higher than in the presence of p-GLP. Food vacuoles in P. caudatum were cytologically grouped into four orderly stages of digestion by Fok et al. (1982) as DV-I, DV-II, DV-III and DV-IY. The presence of acid phosphatase activity in DV-III has been verified using the Gomori technique in Paramecium (Muller and Toro, 1962; Fok et al. 1984). Combining our microscopic findings in Fig. 1 with their biochemical and cytological findings indicate that we demonstrated acid phosphatase activity in food vacuoles corresponding to DV-III at neutral pH. Fibrous or dot-like deposition of final reaction product near the macronucleus may be due to acid phosphatase activity in cytoplasmic organelles such as ER, some Golgi vesicles and primary lysosome-like vesicles as described by Fok et al. (1984). Unfortunately, TA at 440 urn showed a nonlinear concave increase with time with a lag period of approx. 20 min (Fig. 2), although we used the same staining method as Frederiks et al. (1987) and measured at 440 nm, nearly the same wavelength they used (450 nm). They indicated a linear relationship between mean integrated absorbance at 450 nm and reaction time up to 60 min in the study of rat liver 5'-nucleotidase. Van Noorden and longes (1992) also indicated a linear relationship between absorbance at 450 nm and the concentration of lead sulfide in solidified gelatin solutions. As Frederiks et al. (1987) used cryostat sections of 8 ~m thickness, this discrepancy appears to be attributable to the penetration of substrate and other constituents in the reaction mixture across the Paramecium cell membrane; one of the obstacles in cytochemistry of intact cells. Lund-Hansen et al. (1984) described that air-drying was an effective means of rendering cultured human fibroblasts permeable to an indoxyl substrate. Barer (1991) proposed the possibility of disruption of relatively hydrophilic barriers by air-drying. However, it is questionable whether their results can be directly compared with our results because of the differences in chemical nature of the substrates used. Another explanation for the lag phase could be the time needed for building up sufficient liberated phosphate in the cells to obtain precipitate. Van Duijn and co-workers have studied this phenomenon in great detail especially for enzyme reactions visualized with lead precipitation methods (Van Duijn, 1992). The lag phase is dependent on the rate of production of phosphate and this could explain that we found a lag phase, whereas others
Phosphohydrolytic activity in Paramecium
405
did not. However, ~TA, TA values of "test minus control" reactions were linearly after 20 min of incubation in both tests (Fig. 2). Furthermore, the fact that "test minus control" reactions showed Michaelis-Menten's kinetics in relation to the substrate concentration also indicated the validity of the rate of reaction after approx. 20 min in the presence of various substrate concentrations. As all the substrates that were used in the present study were 5'-AMP related compounds in their chemical structures and we stained cells for rather long periods (60 min), we have assumed that all the reactions proceeded in a similar manner as in the presence of 5'-AMP and that tetramisole also penetrated into the cells. The mean value of 5'-AMP hydrolytic activity widely varied in each culture, in particular, in 2 of 5 cultures tested. Hydrolytic activity in P. caudatum depends on the age of culture (Fok and Paeste, 1982). We used cell in an early stationary phase in the present study. In this phase, a relatively steep decrease in acid phosphatase activity was found using p-GLP as substrate. The number of acid phosphatase activity-positive DVs changed with time after a pulse of polystyrene latex spheres (Fok et al. 1984). Therefore, it is likely that the timing of cell-harvesting greatly influences activity as well as minor changes in culture condition. Thus when we compared activity and activity distribution patterns, we used cells that were harvested at the same time from identical cultures (Fig. 7). If not possible, we stained cells in the presence of similar substrate concentrations and measured in 2 cultures to correct data as mentioned in the experiment to analyze the relationship between activity and 5'-AMP concentration (Figs. 4, 5 and 6). Irrespective of the 5'-AMP concentration and mean activity, activity distribution patterns were monomodal (Figs. 4 and 7). We proposed previously that it was bimodal in an early stationary phase of P. caudatum for succinate dehydrogenase activity (Chikamori et al., 1998). This suggest that the activity distribution profile may be different for each enzyme even in an identical culture. The mean activity well fitted to the Michaelis-Menten equation in relation to the 5'-AMP concentration. From the linear Wilkinson plot, we could estimate an apparent Km value for 5'-AMP of 0.20 mM at pH 7.2. Although direct comparison is not possible because of the use of different pH, Fok (1983) reported a Km value of 0.42 mM at pH 5.0 for AMP hydrolysis by lysosomal acid phosphatase in P. caudatum. To investigate substrate affinity, the activity and distribution patterns of 5'-AMP hydrolysis were compared with those of two other substrates at the same concentration of 1 mM. At this concentration, the affinity was similar for 5'-AMP and 5'-GMP and lower for p-GLP. On the basis of microscopic observations and the measurements, the affinity for 5'-GMP was expected to be lower than that for 5'-AMP (Figs. 1 D and 7 A). However, the difference was not statistically significant. We found some cells showing rather high p-GLP hydrolytic activity (Figs. 1 D and 7 B). These findings indicate that it is necessary to estimate kinetic constants of phosphohydrolytic enzyme activity for various substrates to confirm its substrate affinity at neutral pH.
406
K. Chikamori et al.
We started the present study by investigating whether true adenosine 5'monophosphatase (5'-nucleotidase) could be detected in P. caudatum or not. Our cytochemical results from both microscopic observations and measurements were nearly consistent with the biochemical data of Fok (1983) that 5'AMP hydrolytic activity at neutral pH is attributable to acid phosphatase with a broad substrate specificity and 5'-nucleotidase is lacking in P. caudatum. Recently, reports have been published about the application of whole living organisms or cells as biosensors or test organisms for environmental monitoring based on cytochemical methods. Kinetic parameters of glucose6-phosphate dehydrogenase in livers of North Sea flat fish were established (Van Noorden et aI., 1997; Kohler and Van Noorden, 1998) and the marine dinoflagellate Prorocentrum micans was used as test organism to measure parathion toxicity using fluorescence induction kinetic measurements (Prevot et aI., 1993). Araki et al. (1995) also proposed the use of Paramecium cells as a single cell biosensor for the detection of pH, herbicide and Ni+ contents in a microenvironment on the basis of their movement and fluorescence intensity of rhodamine 123, an indicator of mitochondrial membrane potential. In the previous study of P. caudatum, we have investigated succinate dehydrogenase activity as a measure of its energy metabolism (Chikamori, 1998). Our results presented here have provided one more parameter of the physiological state of P. caudatum under contaminant-free conditions. It will contribute to the application of Paramecium cell as single cell biological sensors in environmental monitoring.
Acknowledgement We gratefully thank Dr. Kazuyuki Mikami, Miyagi University of Education for providing kindly a strain of P. caudatum, and it is a pleasure for us to acknowledge the skilful technical assistance of Yoko Ueyama, Department of Chemistry, Naruto University of Education and Kei Horie, Department of Mechanical Engineering, University of Tokushima (present address: Nikon Co., Tokyo, Japan) during the experiments, and Dr. Shinichi Kudo, Department of Biology, Naruto University of Education, for the statistical treatment of the data. We gratefully thank Dr. C. 1. F. Van Noorden, Laboratory of Cell Biology and Histology, Academic Medical Center, University of Amsterdam, for his critical reading of the manuscript and valuable suggestions. This work was partly supported by Grant-in-Aid for Exploratory Research (09875070) and for Scientific Research (10558134) from The Ministry of Education in Japan.
References Araki T, Misawa H, Chikamori K, and Tohno Y (1995) A biological-sensor utilizing a living cell for environment monitoring in micro space. Proceedings of the International Symposium on Microsystem, Intelligent Materials and Robots, September 27-29,1995, Sendai, Japan, pp 301-304
Phosphohydrolytic activity in Paramecium
407
Barer MR (1991) New possibilities for bacterial cytochemistry: light microscopic demonstration of p-galactosidase in unfixed immobilized bacteria. Histochem J 23: 529-533 Borgers M (1973) The cytochemical application of new potent inhibitors of alkaline phosphatases. J Histochem Cytochem 21: 812-824 Chikamori K, Yamamoto A, and Araki T (1991) A microphotometer for multispectral image analysis. Instrumentation and application to multicolor-stained specimen. Acta Histochem Cytochem 24: 395-403 Chikamori K, Araki T, and Sato K (1998) Succinate dehydrogenase (SDH) activity in single Paramecium caudatum cells. Acta histochem 100: 25-36 Fok AK, Lee Y, and Allen RD (1982) The correlation of digestive vacuole pH and size with the digestive cycle in Paramecium caudatum. J Protozool29: 409-414 Fok AK, and Paeste RM (1982) Lysosomal enzymes of Paramecium caudatum and Paramecium tetraurelia. Exp Cell Res 139: 159-169 Fok AK (1983) An inhibition and kinetic study of acid phosphatase in Paramecium caudatum and Paramecium tetraurelia. J Protozool 30: 14-20 Fok AK, Muraoka JH, and Allen RD (1984) Acid phosphatase in the digestive vacuole and lysosomes of Paramecium caudatum: A timed study. J Protozool 31: 216-220 Frederiks WM, Marx F, Bosch KS and Van Noorden CJF (1987) Diurnal variation in 5'-nucleotidase activity in rat liver. A quantitative histochemical study. Histochemistry 87: 439-443 Frederiks WM, Marx F, and Myagkaya GL (1988) A quantitative histochemical study of 5'-nucleotidase activity in rat liver after ischaemia. J Phathol 154: 277286 Frederiks WM, Van Noorden CJF, Aronson DC, Marx F, Bosch KS, Jonges GN, Vogels IMC, and James J (1990) Quantitative changes in acid phosphatase, alkaline phosphatase and 5'-nucleotidase activity in rat liver after experimentally induced cholestasis. Liver 10: 158-166 Kohler A, and Van Noorden CJF (1998) Initial velocities in situ of G6PDH and PGDH and expression of proliferating cell nuclear antigen (PCNA): sensitive diagnostic markers of environmentally induced hepatocellular carcinogenesis in a marine flatfish (Platichthys flesus L.). Aquat Toxicol 40: 233-252 Lund-Hansen T, Hoyer PE, and Andersen H (1984) A quantitative cytochemical assay of p-galactosidase in single cultured human skin fibroblast. Histochemistry 81: 321-330 Muller M, and Ton) (1962) Studies on feeding and digestion in protozoa. III. Acid phosphatase activity in food vacuoles Paramecium multimicronucleatum. J Protozool9: 98-102 Prevot P, Perret E, Jupin H, and Soyer-Gobillard MO (1993) Fluorescence induction used to measure parathion toxicity in the marine dinoflagellate Prorocentrum micans: A possible model for biodetection. Ecotoxicol Environment Safety 25: 360371
Van Duijn P (1992) Model systems. Principles and practice of the use of matrix immobilized enzymes for the study of the fundamental aspects of cytochemical enzyme methods. In: Stoward P and Pearse AGE (Eds) Histochemistry. Theoretical and applied vol 3, 4th ed. Churchill Livingstone, Edinburgh, pp 433-472 Van Noorden CJF, Vogels IMC, and Houtkooper JM (1988) Cytophotometric analysis of alkaline phosphatase and 5'-nucleotidase activity in regenerating rat liver after partial hepatectomy. Cell Biochem Function 6: 53-60
408
K. Chikamori et al.
Van Noorden CJF, and Jonges GN (1992) Molecular extinction coefficients of lead sulfide and polymerized diaminobenzidine as final reaction products of histochemical phosphatase reactions. Cytometry 13: 644-648 Van Noorden CJF, Bahns S, and Kohler A (1997) Adaptational changes in kinetic parameters of G6PDH but not of PGDH during contamination-induced carcinogenesis in livers of North Sea flatfish. Biochim Biophys Acta 1342: 141-148 Wilkinson GN (1961) Statistical estimations in enzyme kinetics. Biochem J 80: 324332