Lysosomes as photochemical targets

Int. J. Cancer: 59,814-822 (1994) 0 1994 Wiley-Liss, Inc.

Publication of the International Union Against Cancer Publication de I'Union lnternationale Contre le Cancel

LYSOSOMES AS PHOTOCHEMICAL TARGETS Kristian BERG'and Johan MOAN Institute for Cancer Research, Department of Biophysics, Montebello, N-0310 Oslo 3 Noway. Sulfonated tetraphenyl porphines (TPPS, ) are photosensitizing dyes that localize in lysosomes of NHIK 3025 cells. In order to elucidate the mechanismsof cell inactivation by photochemical treatment with TPPS,, lysosomal enzyme inactivation and release of Iysosomal contents were examined after treatment. In cells treated with TPPS4 and light, the lysosomal enzymes P-N-acetyl-D-glucosaminidase (fl-AGA) and cathepsin(L 6 ) were almost completely inactivated and no enzyme activities were released from the lysosomes. In contrast, a maximum of 30 and 50% of the initial P-AGA activity was released from lysosomes after treatment with TPPS, and TPPSa, respectively. Forty per cent of the initial P-AGA activity was released after treatment with TPPSk and a non-cytotoxic dose of light. After such a treatment only approximately I W o of the initial cathepsin activity was found in the cytosol fraction and in all other cases no cathepsin activity was recovered in the cytosol fraction after photochemicaltreatment. It was found that the constituents of the cytosol partly inhibited cathepsin activity. T h i s inhibitory effect was not influenced by the photochemical treatment, neitherwas the colony-formingability of photochemically treated cells influenced by pre-treatment with the cathepsin inhibitor EM. The present results indicate that NHlK 3025 cells are not killed by lysosomal disruption after photochemical treatment. This is partly due to photochemical inactivation of the lysosomal enzymes and to the action of cytosolic cysteine cathepsin inhibitors. The present results also indicate that cells can survive a partial lysosomal disruption.

+

o 1994 Wiley-Liss,Inc. Photochemotherapy of cancer [PCT, also named photodynamic therapy (PDT)] is an experimental treatment modality which is presently undergoing clinical trials (Henderson and Dougherty, 1992; Moan and Berg, 1992). The cytotoxic effects are thought to be mainly mediated through the formation of singlet oxygen. However, the main cellular targets for the cytotoxic effects of PCT have still not been identified. Several photosensitizers with wide differences in lipophilicity, including TPPS, (tetraphenylporphine with n sulfonate groups), AlPcS, (aluminium phthalocyanine with n sulfonate groups), chlorin e6, Nile blue derivatives, a-terthienyl, tin etiobenzochlorine, hematoporphyrin (Hp), HpD, uroporphyrin I, protoporphyrin and phylloerythrin have been shown to be localized in extranuclear granules or to mediate the destruction of lysosomes in light-exposed cells (Moan and Berg, 1992). Upon irradiation, several of these dyes, such as TPPS, and AIPcS,, penetrate the lysosomal membranes and relocate to other parts of the cytoplasm and/or the nucleoplasm (Moan et al., 1989; Berg et al., 1991). It is not clear, however, what consequences the treatment with these sensitizers may have on the integrity and function of the lysosomes, and whether lysosomal damage contributes to the inactivation of the cells. In the present study, the activities of 2 lysosomal enzymes, P-AGA (P-N-acetyl-D-glucosaminidase) and cathepsin(L + B), have been measured after PCT. These activities have also been used to quantify lysosomal intactness and relocation of enzyme activities by isolating cytosol and cytosol-free cells containing intact vesicles. The results have been used to evaluate the importance of release of hydrolytic enzymes from the lysosoma1 compartment for the cytotoxic effects of PCT. MATERIAL AND METHODS

Cell cultivation The established cell line NHIK 3025, derived from a carcinoma in situ of the cervix, was used. The cells were

subcultured twice a week in Medium E2a containing 10% human serum and 10% horse serum.

Chemicals TPPS,s were provided by Porphyrin Products (Logan, UT). Each of the TPPS,s showed one peak on HPLC (Berg et al., 1991). Photofrin was kindly supplied by Lederle Parenterals (Carolina, Puerto Rico). The enzyme inhibitor L-transepoxysuccinyl-leucylamido(4-guanidino)butane (E-64) was purchased from Sigma (St. Louis, MO). Labelling with potphyrins, irradiation and measurements of survival The cells were inoculated in plastic tissue-culture dishes (Nunclon, Roskilde, Denmark) containing E2a medium with 20% serum, and left at 37°C for 4-5 hr for proper attachment to the substratum. Subsequently they were exposed to TPPS, or Photofrin in E2a medium with 3% serum (2% human serum and 1% horse serum) for 18 hr. The following concentrations of sensitizers were used in all the experiments: TPPS4: 75 kg/ml; TPPS2,: 19 kg/ml; TPPS2,: 3.2 kg/ml; TPPSI: 3.1 kg/ml; Photofrin: 6.0 pg/ml. The concentrations of TPPS, (Fig. 1) were adjusted so that the cells contained similar intracellular amounts of fluorescing, supposedly photoactive dyes (Moan and Sommer, 1981). After 18 hr incubation with porphyrins the cells were washed with E2a (20% serum) and incubated for 1 hr in sensitizer-free E2a containing 20% serum before exposure to the light from a bank of 4 fluorescent tubes (mod. 3026, Applied Photophysics, London, UK). The light intensity reaching the cells was 36 W/m2, with the highest fluence rate around 405 nm. Cell survival was measured by means of a colony-forming test as previously described (Berg et al., 1991). In these experiments 2,000 cells were seeded out in 25-cm2plastic tissue-culture dishes (Nunclon). Similar survival curves were obtained by seeding out the same number of cells (per cm2) as in the electrodisruption experiments (see below). The concentrations of photosensitizers applied in the present work did not reduce the colony-forming ability of the cells. Cell disruption and separation of sedimentable cell components from cytosol Five million cells were seeded out in 175-cm2plastic tissueculture flasks and treated as described above except that after 18 hr of incubation with the sensitizers followed by 1 hr in sensitizer-free medium the cells were trypsinized off the substratum, pelleted and suspended in E2a medium containing 20% serum. One ml of the cell suspension was distributed all over the bottom of 10-cm2 dishes (Falcon 1008, Oxnard, CA). The cells were allowed to sediment for 5 min before exposure to light. Immediately after light exposure, the cells were incubated at 37°C for 20-25 min to allow intracellular relocalization of lysosomal components to be completed (Berg et al., 1991). The cells were then collected, pelleted and suspended in 400 PI 10% sucrose. This cell suspension was subjected to electrodisruption by a single pulse from a commercial electroporation apparatus (BioRad, Richmond, CA). The cells were treated in a cuvette (BioRad) with 2 mm distance between the electrodes and exposed to 300 V and 125 kF. 'To whom correspondence and reprint requests should be sent. Fax: 47 22 93 42 70.

Received: June 7,1994 and in revised form August 22,1994.

815

LYSOSOMES AS PHOTOCHEMICAL TARGETS

TABLE I - p-AGA ACTIVITY IN NHIK 3025 CELLS AFTER TREATMENT WITH PHOTOSENSITIZERSIN THE ABSENCE OF LIGHT. THE ENZYMATIC ACTIVITIES ARE DESCRIBED AS PERCENTAGE OF UNTREATED CONTROL VALUES

R4

TPPS, : R1-4=S0,

TPPSl : R,

=SO,

-

.R,-,=H

FIGURE1- Structural formula for sulfonated tetraphenylporphines (TPPS,).

Separation of sedimentable cell components from cytosol was performed essentially as described by Kopitz et al. (1990). After electrodisruption, 200 ~1 phosphate-buffered sucrose [lo0 mM potassium phosphate, pH 7.5, 2 mM dithiothreitol (DTT), 2 mM ethylenediaminetetraacetic acid (EDTA), 0.05 mM sucrose] were used to wash the cuvette and added to the suspension of disrupted cells. This suspension was layered on top of a 600-p1 ice-cold density cushion of metrizamide (8% metrizamide, 50 mM potassium phosphate, pH 7.5, 1 mM DTT, 1 mM EDTA, 64 mM sucrose) in 1.8-ml Eppendorf tubes. After centrifugation at 4°C for 30 min at 15,000 rpm (17.000 x g) (Heraeus Sepatech Biofuge 17RS, Osterode, Germany) the pellet, containing all the intact lysosomes and other sedimentable cell components, was dissolved in phosphate-buffered sucrose (50 mM potassium phosphate, pH 7.5, 1 mM DTT, 1 mM EDTA) and used for determination of enzyme activities. In all cases 25 pl were used for the p-AGA and cathepsin measurements. The cytosolic fraction was carefully isolated from the top of the centrifuge tube.

Enzyme assays Lactate dehydrogenase (LDH) was assayed spectrophotometrically by measuring the oxidation of NADH at 340 nm as described by Reeves and Fimognari (1966). The method measures the sum of all isoenzymes. P-AGA was measured as described by Beaufay et al. (1974). The method is based on the formation of p-nitrophenol (from the substrate p-nitrophenylN-acetyl-D-glucosaminide)which can be measured spectrophotometrically at 420 nm. Cathepsins were measured mainly as described by Barrett and Kirschke (1981). The substrate, N-CBZ-Phe-Arg 7-amido 4-methylcoumarin, used at the concentration of 20 p M , is a substrate for both cathepsin Land B (Khalfan, 1991). The product formation was measured fluorometrically (hex= 380 nm; ,A = 440 nm) at 30°C by means of a Perkin-Elmer (Norwalk, CT) LS 5 spectrofluorimeter equipped with a temperature-controlled cuvette. RESULTS

p-AGA and cathepsin activity in sensitizer-treated cells After treatment of NHIK 3025 cells with TPPS, or Photofrin in the absence of light, P-AGA and cathepsin activities were measured in the cells. In all cases, except after treatment with TPPS4, the P-AGA activity was not significantly changed by exposure of the cells to these sensitizers (Table I). TPPS4 was found to reduce the total 6-AGA activity in the cells by approximately 25% of the activity in untreated cells. Photofrin, the drug most frequently used in the clinic, is a mixture of various porphyrins. It was included in the present study since

Sensitizer

!3-AGA activity

TPPS4 TPPSZo TPPS2, TPPSl Photofrin

73 t 10' (512 109 2 4.1 3 311 +- 6.5 4 102 t 7.8 4 102 t 2.3 (3)

ii

'Mean 2 SD.-ZFigures in parentheses indicate number of experiments. TABLE I1 - CATHEPSIN ACTIVITY IN NHlK 3025 CELLS AFTER TREATMENT WITH PHOTOSENSITIZERSIN THE ABSENCE OF LIGHT THE ENZYMATIC ACTIVITIES ARE DESCRIBED AS PERCENTAGE OF UNTREATED CONTROL VALUES ~~

Sensitizer

Cathepsin activity

TPPS4 TPPSb TPPS2, TPPSl Photofrin

102 & 4.4' (4)2 123 2 4.0 (4) 109 2 2.0 5 109 2 4.0 [4] 106 +- 11 (4)

'Mean -t SD.-ZThe figures in parentheses indicate number of experiments. previous work had shown that it is not localized in lysosomes in NHIK 3025 cells (Berg et al., 1990a, 1991). Cathepsin activity was not influenced by treatment of the cells with TPPS, or Photofrin in the absence of light (Table 11).

p-AGA and cathepsin sensitiviy tophotochemical treatment Treatment of NHIK 3025 cells with TPPS, or Photofrin at the concentrations used in the present study is not cytotoxic in the absence of light (Berg et al., 1991). However, exposure of treated cells to light of wavelengths absorbed by the porphyrins induces cytotoxic effects as measured by the colony-forming assay. The total enzyme activities in the cells were measured immediately after photochemical treatment (Figs. 2, 3). The P-AGA activity in the cells was attenuated by treatment with TPPS, and light, but was not affected by treatment with Photofrin and light (Fig. 2). The reduction in the f3-AGA activity correlates with the hydrophobicity of the sensitizer, TPPSl being the least efficient one. The sensitivity of cathepsins to the different sensitizers and light showed the same pattern as for P-AGA, ie., increasing sensitivity with increasing hydrophilicity (Figs. 2 and 3 ) . However, cathepsin activity seemed in all cases to be more sensitive to PCT than P-AGA activity. Additionally, a slight reduction in cathepsin activity was found after treatment with Photofrin and light. In the case of TPPS4 (and possibly also TPPS2J the photochemically induced inactivation of P-AGA activity was bimodal and 20-30% of the activity in TPPS4-treated cells not exposed to light was resistant to PCT. For TPPS4 this fraction of P-AGA activity has recently been found to be located in vesicles with density similar to that of lysosomes but not containing TPPS4 and in Golgi vesicles (3-5% of the total activity) (Berg et al., 1993). Some of the P-AGA may therefore reside in telolysosomes (ie., lysosomes not active in fusion with endocytic vesicles; Holtzman, 1991) or in vesicles transporting lysosomal enzymes from trans-Golgi network to the lysosomes (Brown et al., 1986). Previous reports indicate that 4 hs of cycloheximide treatment (L of 1mM) is sufficient to empty the compartments involved in synthesis and transport of P-AGA to the lysosomes (Brown et aL, 1986). Pre-treatment of the cells with cycloheximide reduced the PCT-resistant fraction of P-AGA by only 50% (Fig. 4). It is therefore likely that as much

816

BERG AND MOAN

1 .o

1 0.8

a

0.6

0

< I

0.4

1

0.4

1

o.2f-\\*

0.2

0.0

\

0 0.0 1

0.1

0.01

1

2

3

4

Time (hours)

Surviving fraction

F~GURE 4 -Effect of cycloheximide on the photochemically resistant pool of P-AGA activity in TPPS,-treated NHIK 3025 FIGURE 2 - Relative P-AGA activity in cells immediately after cells. The cells were incubated for 22 hr in the presence of TPPS4, treatment with TPPS, or Photofrin and light vs. surviving fraction the last 0-4hr in the presence of 1 mM cycloheximide. The cells of cells. The enzymatic activity as presented is relative to cells were irradiated in the presence of a drug-free medium applied treated neither with photosensitizer nor with light. The cells were immediately before light exposure. P-AGA activity was measured exposed to light before detachment from the substratum (see from cells unexposed to light (0)or exposed to 210 sec of light (0) text). The results are from 3 se arate ex eriments. Symbols: 0, (D5)as a function of time after addition of cycloheximide. The TPPS4;0,TPPS,; V,TPPS2,; TPPS,; ! ! I Photofrin. , Bars, SD. measurements at 0 hr were performed on cells not exposed to cycloheximide. Bars, range from 2 experiments.

$

1.0

L

-I

>r

-

f 0.8 > c 0

0

c

I

n

0 .6

a 0

f

o

-?

0.4

Q

0:

0.2

0.0 1

0.1

Surviving fraction

FIGURE3 - Relative cathepsin (L + B) activity in cells immediately after treatment with TPPS, or Photofrin and light vs. surviving fraction of cells. The enzymatic activity as presented is relative to cells treated neither with photosensitizer nor with light. The cells were exposed to light before detachment from the substratum (see text). The results are from 3 separate experiments. Symbols: 0, TPPS4; 0, TPPSz,; V, TPPS,; V, Photofrin. Bars, SD.

as 10% of the total 6-AGA activity in TPPS4-treated cells resides in telolysosomes (Fig. 4 ) .

p-AGA and cathepsin activity in electrodisrupted cells Porphyrins have been shown to escape compartmentalization in lysosomes and endosomes within 20 min after exposure to light (Berg et al., 1991). Therefore, to reveal the possibility

that functionally intact lysosomal enzymes also relocalize to extralysosomal compartments, cytosol-free cells were isolated 2&25 min after exposure to light. This was performed by electrodisruption of the cells and separation of sedimentable organelles from cytosol by centrifugation through metrizamide/ sucrose density cushions (Kopitz et aL, 1990) as described above. Nearly all of the cells (> 99%) were disrupted by the electrical treatment, as revealed by the trypan-blue exclusion method. The separation of cytosol and sedimentable organelles was controlled in each experiment by measuring lactate dehydrogenase (LDH) activity in the pellet after metrizamide centrifugation. Generally, less than 5% of the LDH activity in the cells was associated with the cell pellets, due to nonspecificbinding to sedimentable material and LDH in autophagic vacuoles lacking proteolytic enzymes (Kopitz et al., 1990). In all cases LDH activity was unaffected by photochemical treatment of the cells (data not shown). The P-AGA activity was measured as a function of light dose from the whole disruptate, from the cell pellet (intact vesicles) and from the non-sedimentable activity, i.e. cytosol, remaining on top of the metrizamide cushion (Fig. 5). The sum of the pelletable and non-pelletable enzymatic activities was similar, to within

.-

.-> 2

.-.-x> c

0.2

.-C Y)

n ID

f

0.1

0.4

IX

0

40

20

60

80

100

120

140

Light exposure (sec) 1.6

I

I

i

0.6

-6 I

0.2

t

0.0

‘

0

I

I

I

I

1

2

3

4

[CytosoI]/[CytosoI-free

I

cells]

F~GURE I - Cytosol inhibition of cathepsin activity. Cytosol and cytosol-free cells were isolated as described in the text. Relative cathepsin activity was measured in mixtures of cytosol and cytosol-free cells, the ratio being obtained by varying the amount of added cytosol. The results were normalized to the 1:l mixture of cytosol and cytosol-free cells. The buffers used during electrodisruption and isolation of cytosol and cytosol-free cells were adjusted to pH 5.5.

14

Dose/D

I

I

I

10

FIGURE 6 - Fluence-response curves for inactivation of cells (a) and cathepsin (L + B) activity (b) in NHIK 3025 cells treated with TPPS2,. The cells used for cathepsin (L + B) activity measurements were electrodisrupted 20-25 min after light exposure as described in the text. Cathepsin (L + B) activity was measured in the sedimentfrom the cell disruptate before centrifugation (0); and in the non-sedimentable able fraction (cytosol-free cells) (7); fraction (cytosol) (0). Bars, SD from 3 experiments.

0.5 2.

5.-

0.3

. I -

0

0

.0

. I -

0

determined. In all cases, except for cells treated with TPPSB and light, no significant effects of E64 on the survival of the cells were found (Fig. 10). The presence of E64 in TPPS4treated cells slightly reduced the photosensitivity of these cells. This was not due to a significantly reduced uptake of TPPS4 in the presence of E64 (data not shown). Lysosomes in cells treated with E64 for 18 hr are probably enlarged and filled with undigested proteins. Such a change in lysosomal morphology may influence the distribution of TPPS4 in the cells. The cathepsin activity was therefore tested after 1 hr incubation with different concentrations of E64. It was found that treatment for 1 hr with 100 pg/rnl E64 inhibited cathepsin activity by 90% (88 2 2%). Treatment of TPPS4-loaded cells with 100 pg/ml E64 for 1 hr prior to light exposure did not reduce their photosensitivity. In no cases did E64 show any cytotoxic effects. E64 did not affect the release of @-AGAfrom the lysosomes and the inactivation of P-AGA was slightly reduced in cells treated with TPPS2, and light (Table 111).

5 :

0.1

ID

-ID m

0.05

0.03 ~

0

50

100

Light exposure

150

200

(5)

F~GURE 8 - P-AGA ( 0 , O )and cathepsin (L + B) (0)activities in cytosol-free cells after treatment of NHIK 3025 cells with TPPS4 and light. The total activities in cytosol-free cells are presented (0, 0 )as well as P-AGA activity after subtraction of the photochemical resistant fraction of p-AGA (0,22%) (Berg et al., 1993).

LYSOSOMES AS PHOTOCHEMICALTARGETS I

1.2

I

I

I

I

819

exposure to light (Moan et al., 1989; Berg et al., 1991; Berg et aZ., 1993). Porphyrins are relatively small molecules (MW

/

1.0

T

0.8 0.6

0.4

0.2 0.0

0

20

40

60

80

100

p g / m l E64

FIGURE9 -Effects of E64 on cathepsin (L + B) activity. The cells were treated with E64 for 18 hr in 3% E2a medium and then for 1 hr in drug-free medium before cellular extracts were prepared for determination of total protein (V), P-AGA activity (O), and cathepsin (L + B) (0)activities. Bars, range from 2 experiments. DISCUSSION

Singlet oxygen (lo2) is thought to be the main cytotoxic product formed during photochemical treatment of cells when porphyrins are used as sensitizers (Moan and Berg, 1992). This reactive oxygen species has a very short lifetime in cells ( < 0.04 ks), corresponding to a diffusion length of 0.014.02 pm (Moan and Berg, 1991). The primary cytotoxic effect of PCT is therefore executed during light exposure and very close to the formation. Significant amounts of several porphysites of *02 rins, such as TPPS4, TPPS2,, TPPS2, and, to a lesser extent, TPPSl are found in fluorescing spots in NHIK 3025 cells (Berg et ab, 1990~).This indicated a lysosomal localization of these dyes (Berg et al., 1990~).Similar localization has previously been suggested for TPPS4 in other cell lines (Wessels et al., 1992). In the present study, we found that the lysosomal marker enzymes P-AGA and cathepsin (B L) were inactivated by treatment with TPPS4, TPPS20,TPPS2, and to some extent with TPPSI, in combination with light (Figs. 2,3). These results provide further evidence for a lysosomal localization of sulfonated tetraphenyl porphins in NHIK 3025 cells. The lack of lysosomal enzyme inactivation by treatment with Photofrin and light confirms the previous suggestion that Photofrin is not located in lysosomes in NHIK 3025 cells (Figs. 2,3) (Berg et al., 1991). Photofrin is therefore probably not taken up to a large extent by endocytosis in NHIK 3025 cells through its partial binding to LDL particles, as may occur in other cell lines (Geze et al., 1993). In the case of TPPS4, a small but significant reduction in P-AGA activity was found after treatment of the cells in the absence of light (Table I). This reduction in P-AGA activity may be due either to TPPS4being an inhibitor of the enzymatic activity or influencing the rate of protein synthesis, or compartmentalization/exocytosis of P-AGA. However, TPPS4does not inhibit protein synthesis in general since the activity of other lysosomal enzymes (cathepsins) are not changed by the treatment with TPPS4 (Table 11) and since cell growth is not reduced by the treatment (Berg et a!., 1992). Lysosomally located TPPS,s, as well as sulfonated AlPcs, have previously been found to relocate intracellularly upon

+

600-1200) that seem to be released through damaged lysosomal membranes. Such damaged membranes might conceivably retain larger molecules, such as proteins. Light exposure of the cells treated with TPPS4 resulted in a rapid reduction in the total P-AGA and cathepsin activities and only insignificant amounts of active enzymes were relocated to the cytosol (Fig. 5). It is not clear from the present results whether lysosomal proteins, e.g. photoinactivated enzymes, are released at all after this treatment. In contrast, substantial amounts of P-AGA activity are released after treatment with TPPS2, or TPPSl and light (Fig. 5 ) . These results show that PCT can induce relocation of lysosomal enzymes into the cytosolic compartment. Similar observations have previously been reported after treatment with Hp and light (Santus et al., 1983) and after treatment with derivatives of the lysosomotropic photosentizer Nile blue (Lin et al., 1993). Agents that disrupt lysosomes are widely believed to kill cells by allowing leakage of lysosomal hydrolases into the cytoplasm (“suicide sac” hypothesis) (de Duve et al., 1974). In fibroblast cell killing by the lysosomotropic agent N-dodecylimidazole (CI2-Im),evidence is provided for a central role of the release of enzymes from the lysosomal compartments (Wilson et al., 1987). Amino-acid methyl esters, like L-leucine methyl ester and L-leucyl-L-leucine methyl ester, and g1ycyl-Dphenylalanine-2-naphthylamideare also considered to execute their cytotoxic effect through disruption of lysosomal membranes (Thiele and Lipsky, 1992). In addition, it has been suggested that lysosomal disruption is involved in diseases such as prion encephalopathies (scrapie, Creutzfeld-Jakob disease and Gerstmann-Straussler-Scheinklersyndrome), cardiac myocyte ischemia and silicosis, as well as in keratinization of epidermal cells (Laszlo et al., 1992; Olsson et al., 1989). Lysosomal rupture has also been thought to be involved in the cytotoxic effects of PCT (Geze et al., 1993). However, lysosomal hydrolases cannot contribute to the cytotoxic effects of treatment with TPPS4 and light since they are highly sensitive to photochemical inactivation by TPPSI and no enzymatic activities are released to the cytosol (Fig. 5). The present results do indicate that PCT, using TPPS2, as a photosensitizer, can induce release of active lysosomal enzymes to cytosol even after exposure to non-toxic doses of light (Fig. 5). For cells treated with TPPSl and light the release of P-AGA increases with increasing inactivation of the cells, suggesting a possible correlation between cell inactivation and exposure of cytosol to lysosomal hydrolases. However, it should be emphasized that only equal (P-AGA) or much lower cathepsin (L + B) enzyme activities were found in cytosol after treatment with TPPSl and light as compared to treatment with TPPS,?,and light. Additionally, only a minor fraction of TPPS, is located in lysosomes (Berg et al., 1990a) and the cell inactivation is likely to be caused by the extralysosomal fraction of this dye. These results therefore indicate that NHIK 3025 cells are not killed by a PCT-induced release of lysosomal contents into the extralysosomal area. The possibility that release of the total (undamaged) lysosomal content can induce some cytotoxic effects can still not be excluded. However, the present results indicate that cells can survive a partial lysosomal disruption. Previous studies with the lysosomotropic agent C12-Im indicated substantial cell killing effects within less than 2 hr of treatment, suggesting a rapid destruction of cellular contents by the released enzymes (Miller et al., 1983). In contrast, only an insignificant number of cells showed signs of PCT-induced plasma membrane damage during the first 6-8 hr after treatment (Berg et aZ., 1990b). This was true even after the highest doses of light used in the present study. Thus, the cell

820

BERG AND MOAN 0

50

100

150 200

0

200

400

600

1

1 0.5

0.5

0.3

d

0

0.3

0.1

.A

0.1

-4 0.05

u

q.03 0.05

k

rcI

0.03

1 .?I

+ *$

0.5

0.5

0.3

0.3

0.1

0.1

k

9

0.05

0.05

0.03

0.03

0.01

0.01 0

40

80

120

0

20

40

60

L i g h t exposure ( s e c ) F~GURE10 - Fluence-response curves for inactivation of NHIK 3025 cells after treatment with TPPSp (u), TPPS2, (b), TPPS2, (c), TPPS, (d) or Photofrin (e) for 18 hr followed by 1 hr in drug-free medium before light exposure. The cells were treated with the photosensitizer only ( 0 )or co-incubated with 50 pgirnl E64 (0). TABLE 111-EFFECTS OF E64 O N RELEASE OF 8-AGA ACTIVITY FROM

of cathepsin L is much higher than that of cathepsin B and higher than that of cathepsin D, the second highly active lysosomal endoprotease, at cytosolic pH (Bohley and Seglen, 1992). Therefore, cathepsin L is likely to be the most important potentially cytotoxic lysosomal enzyme. The sensitivity of P-AGA and cathepsin (L B) to photochemical inactivation can only be directly compared in the case of TPPS4photosensitization, where probably no enzyme activities escape the lysosomal compartments (Fig. 5). It was found that cathepsin (L B) was approximately 2.5-fold more sensitive to PCT than P-AGA when these activities were measured on cytosolfree cells (Fig. 8). Assuming similar sensitivity differences between f3-AGA and cathepsin (L + B) as for treatment with TPPS4 and light, as much as 30% of cathepsin (L B) activity may be released to cytosol after treatment with TPPS2, and light. However, a much lower activity was found in the cytosol of treated cells (Fig. 6). Therefore, the present results may be consistent with a partial release of cathepsin L activity after TPPS2, and light, which is, however, inhibited by cytosolic cathepsin inhibitors. Altogether, we suggest that the lack of cytotoxic effect by PCT-induced lysosomal rupture is consistent with a high cathepsin sensitivityto photochemical inactivation and a sufficiently high cytosolic capacity of NHIK 3025 cells to inhibit the released lysosomal enzyme activities. The evidence for cysteine cathepsins as the major cytotoxic agents released from the lysosomes is partly based on the observation that pre-treatment with the cysteine cathepsin inhibitor E64 completely inhibited the cell killing effect of CI2-Im (Wilson ef a[., 1987). E64, which inhibited cathepsin activity by more than 90% in the present study, influenced neither the release of P-AGA after treatment with TPPS2, and light (Table 111) nor the survival of the cells (Fig. 10). These apparently contradictory results can be explained by the partial PCT-induced inactivation of cysteine cathepsins (Figs. 3, 8) and sufficient cytosolic cathepsin inhibitor capacity in

+

Light exposure

E64

(wri \---,

0 60 0

60

-

-

+ +

P-AGA activity

Before sedimentation

*

100 6.0' 69 f 19 94 f 7.9 78 f 4.4

Non-sedimentable activity

Sedimentable activity

4.1 2 0.50 36 f 11 3.0 t 0.40 32 f 2.9

74 f 1.3 25 2 10 77 2 1.7 36 t 4.0

'Mean t range (from 2 experiments). killing following these 2 treatments is most likely caused by different mechanisms of action, or by differences in cell-type response (see below). The cytosol fraction of NHIK 3025 cells contains inhibitors of cathepsin (L B) activity (Figs. 6, 7) which are not destroyed by PCT. These inhibitors are most likely members of the cystatin superfamily (Turk and Bode, 1991). However, the cathepsins were only partly inhibited by the cytosol fraction (Fig. 7 ) . This may be due to differences in sensitivity of cathepsin B and L to cytosolic protease inhibitors. It has been shown that inhibition by stefins is weaker for cathepsin B than for cathepsin L and H (Turk and Bode, 1991), indicating that the presently observed cathepsin inhibition may be due to selective inhibition of cathepsin L. It should also be noted that, in lysosomes, a cathepsin-mediated endoproteolytic attack is a necessary initial step for the degradation of proteins (Seglen and Bohley, 1992). Furthermore, the endoproteolytic activity

+

+

+

821

LYSOSOMES AS PHOTOCHEMICAL TARGETS

NHIK 3025 cells to inhibit most of the residual proteolytic activity (Figs. 6, 7). It should be noted that large variations in the ratio between cysteine cathepsin activities and the amount of their inhibitors have been found in cells from different neoplastic diseases (Lah et al., 1992). The selected cell line may therefore influence the outcome of lysosomal rupture. In various treatments indicating cytotoxic effects of lysosoma1 rupture, the drugs or their vehicles may exert cytotoxic effects when released from the lysosomal compartment (Wilson et al., 1987). This is not valid for TPPS, since large quantities of TPPS, can be relocated to extralysosomal areas by light doses that do not inactivate the cells (Berg et al., 1991). Other secondary effects of different treatments, such as perturbation of cytosolic pH, should also be considered. Moreover, the present results cannot be explained by a rapid PCT-induced degradation of lysosomal enzymes after exposure to the cytosolic compartment, since both P-AGA and cathepsin activity remain constant for at least 2 hr after PCT (data not shown), the time needed for cell killing by CI2-Im. The present results indicate that the lysosomal membranes are targeted to a higher extent than P-AGA, which is located in the lysosomal matrix, by photoactivated TPPS2, and TPPSl than by TPPS4 (Fig. 5). TPPS4 is a relatively water-soluble dye with sulfonate groups located on each phenyl ring, while TPPS2, and TPPS, are amphiphilic dyes (Fig. 1). Altogether, these facts suggest that, within the lysosomes, TPPS4 is located in the matrix while TPPS2, and TPPSl are to a greater degree associated with the lysosomal membranes. It was also found that E64 to some extent attenuated the inactivation of P-AGA activity by TPPS2, and light (Table 111). This may be explained by an E64-induced increase in lysosomal size and thus a larger average distance between TPPS2, in the membranes and P-AGA. The inactivation of P-AGA (Fig. 2, 8) by TPPS4 and light indicates the existence of 2 compartments of this lysosomal enzyme, i.e. a photosensitive and a non-photosensitive one, the latter being most likely telolysosomal (Berg et al., 1993; Holtzman, 1991; Fig. 4). The inactivation of P-AGA and cathepsins in the photosensitive compartment seems to obey first-order kinetics (Berg et al., 1993; Fig. S), suggesting that TPPS4 and the lysosomal enzymes are evenly distributed. This

is not the case when TPPS2, or TPPSl are used as photosensitizers (Fig. 5, 6). The lysosomal enzymes were not inactivated by small doses of light, while additional exposure to light gave approximately first-order kinetics (Fig. 5 and additional results with higher doses of light). This may be due to a relocation of these sensitizers during PCT from the lysosomal membranes to the lysosomal matrix closer to the location of the lysosomal enzymes. It is unlikely that the diffusion length of '02 increases during light exposure. This might occur if all the targets of '02 attack in the close vicinity of its formation are damaged and made unreactive by small doses of light. However, we consider this unlikely because of the enormous number of molecules capable of reacting with lo2in cells. The present results indicate that NHIK 3025 cells are not killed by lysosomal disruption and raise the question of the cause of PCT-induced cytotoxicity. It has previously been shown that TPPS,-treated cells accumulate in mitosis after light exposure (Berg et al., 1990b). This is due to damage to the unpolymerized form of the microtubule (MT) component (Berg et al., 1992). Cells which accumulate in mitosis after PCT have a reduced ability to form colonies and PCT-induced cell inactivation is to a large extent due to an irreversible block in the metaphase-anaphase transition (Berg et al., 1992). Nocodazole is an MT inhibitor which alone exerts a completely reversible depolymerization of MTs. Nocodazole and PCT were found to act synergisticallyin cell inactivation, in accumulation of cells in mitosis and in prevention of MT formation (Berg et al., 1992). The PCT-induced inactivation of MT assembly is not a result of generalized damage to cytosolic components since LDH was found to be photochemically resistant. On the basis of the present results we suggest that photochemical inactivation of NHIK 3025 cells using TPPS, as sensitizers is mainly due to inhibition of MT assembly induced by TPPS, located in the cytosol or escaping lysosomes during light exposure. ACKNOWLEDGEMENTS

This work was supported by the Research Council of Norway.

REFERENCES

J.C. and WINKELMAN, J.W., Cellular BARRETT,A.J., KEMBHARI,A.A., BROWN, M.A., KIRSCHKE, H., KNIGHT, BERG,K., MOAN,J., BOMMER, C.G., TAMAI,M. and HANADA,K., L-trans-cpoxysuccinyI-leucyl- inhibition of microtubule assembly by photoactivated sulfonated amido(4-guanidin0)butane (E-64) and its analogues as inhibitors of tetraphenylporphins. Int. J. Radial. BioL, 58,475-487 (1990b). cysteine proteinases including cathepsin B, H and L. Biochem. J., 201, BOHLEY,P. and SEGLEN,P.O., Proteases and proteolysis in the 189-198 (1982). lysosomes. Experienfia,48,151-157 (1992). BARRETT, A.J. and KIRSCHKE,H., Cathepsin B, cathepsin H, cathepsin BROWN,W.J., GOODHOUSE, J. and FARQUHAR, M.G., Mannose-6L. Meth. Enzymol., 80,535-561 (1981). phosphate receptors for lysosomal enzymes cycle between the Golgi BEAUFAY, H., AMAR-COSTESEC, A,, FEYTMANS, E., THINES-SEMPOUX,complex and endosomes. J. CellBioL, 103,1235-1247 (1986). D., WIBO, M., ROBBI,M. and BERTHET,J., Analytical study of DE DUVE,C., DE BARSY,T., POOLE,B., TROUET, A,, TULKENS, P. and microsomes and isolated subcellular membranes from rat liver. J. Cell VAN HOFF, F., Lysosomotropic agents. Biochem. Pharmacol., 23, Biol., 61,188-200 (1974). 2495-2531 (1974). BERG,K., MADSLIEN, K., BOMMER, J.C., OFTEBRO,R., WINKELMAN, GEZE,M., MORLIERE, P., MAZIERE, J.C., SMITH,K.M. and SANTUS, R., J.W. and MOAN,J., Light induced relocalization of sulfonated meso- Lysosomes, a key target of hydrophobic photosensitizers roposed for tetraphenylporphines in NHIK 3025 cells and effects of dose fraction- photochemotherapeutic applications. J. Phorochem. Plotobiol., 20, ation. Phorochem. Photobiol., 53,203-210 (1991). 23-35 (1993). BERG,K., PRYDZ,K. and MOAN,J., Photochemical treatment with the HENDERSON, B. and DOUGHERTY, T.J., How does photodynamic lysosomally localized dye tetra(4-sulfonatophenyl)porphine results in therapy work? Photochem. Photobiol., 55,145-157 (1992). lysosomal release of the dye but not of P-N-acetyl-D-glucosaminidase HOLTZMAN, E., acidification; Membrane properties; Permeability and activity. Biochim. biophys.Acta, 1158,300-306 (1993). Transport In: E. Holtzman (ed.), Lysosomes. Plenum, New York BERG,K., STEEN,H.B., WINKELMAN, J.W. and MOAN,J., Synergistic (1991). p. 93-160. effects of photoactivated tetra(4-sulfonatophenyl)porphineand nocodazole on microtubule assembly, accumulation of cells in mitosis and KHALFAN,H.A., Study of thiol proteases of normal human skin fibroblasts. Cell biochem. Funct., 9,55-62 (1991). cell survival.J. Photochem. Photobiol., 13,59-70 (1992). H. and BARRETT, A.L., Chemistry of lysosomal proteases. BERG,K., WESTERN,A., BOMMER,J.C. and MOAN,J., Intracellular KIRSCHKE, localization of sulfonated meso-tetraphenylporphine in a human In: H. Glaumann and F.J. Ballard (eds.), Lysosomes: their role inprotein breakdown, pp. 193-238, Academic Press, London (1987). carcinoma cell line. Photochem. Photobiol., 52,481-487 (1990~).

822

BERG AND MOAN

KOPITZ,J., KEEN, G., GORDON,P., BOHLEY,P. and SEGLEN,P.O., Nonselective autophagy of cytosolic enzymes by isolated rat hepatocytes. J. Cell Biol., 111,941-953 (1990). LAH, T.T., KOKAU-KUNOVAR, M., STRUKEU,B., PUNGERCAR, J., BARLIC-MAGANJA, D., DROBNIC-KOSOROK, M., KASTELIC, L., BABNIK, J., GOLOUH,R. and TURK,V., Stefins and lysosomal cathepsins B, L and D in human breast carcinoma. Int. J. Cancer, 50,36-44 (1992). LASZLO, L., LOWE,J., SELF,T., KENWARD, N., LANDON, M., MCBRIDE, T., FARQUHAR, C., MCCONNEL, I., BROWN,J., HOPE,J. and MAYER, R.J., Lysosomes as key organelles in the pathogenesis of prion encephalopathies. J. Pathol., 166,333-341 (1992). LENNEY, J.F., Inhibitors associated with the proteinases of mammalian cells and tissues. Cur. Top. Cell Reg., 17,25-55 (1980). LIN,C.-W., SHULOK, J.R., KIRLEY,S.D., BACHELDER, C.M., FLOTTE, T.J., SHERWOOD, M.E., CINCOTTA, L. and FOLEY,J.W., Photodynamic destruction of lysosomes mediated by Nile blue photosensitizers. Photochem. Photobiol., 58,81-91 (1993). MILLER,D.K., GRIFFITHS, E., LENARDS, J. and FIRESTONE, R.A., Cell killing by lysosomotropic detergents. J. Cell Biol., 97,1841-1851 (1983). MOAN,J. and BERG,K., The photodegradation of porphyrins in cells can be used to estimate the lifetime of singlet oxygen. Photochem. Photobiol., 53,549-553 (1991). MOAN,J . and BERG,K., Photochemotherapy of cancer. Experimental research. Photochem Photobiol., 55,931-948 (1992). MOAN,J., BERG,K., KVAM,E., EVENSEN, J.F., MALIK,Z., RUCK,A. and SCHNECKENBURGER, H., Intracellular localisation of photosensitizers. In; G. Bock and S. Harnett (eds.), Photosensitizing compounds: their chemistry, biology and clinical use, (Ciba Foundation Symposium 146) pp. 95-111, Wiley, Chichester (1989).

MOAN,J. and SOMMER, S., Fluorescence and absorption properties of components of hematoporphyrin derivative. Phorobiochem. Phorobio&s., 3,93-103 (1981). OLSSON, G.M., SVENSSON, I., ZDOLSEK,J.M. and BRUNK,U.T., Lysosomal enzyme leakage during the hypoxanthineixanthine oxidase reaction. VirchowsArch.B. Cell Pathol., 56,385-391 (1989). REEVES,W.J. and FIMOGNARI, G.M., Meth. Enzymol., 9, 288-294 (1966). SANTUS,R., KOHEN,C., KOHEN,E., REYFTMANN, J.P., MORLIERE, P., DUBERTRET, L. and TOCCI,P.M., Permeation of lysosomal membranes in the course of photosensitization with methylene blue and hematoporphyrin: study by cellular microspectrofluorometry. Photochem. Photobiol., 38,11-17 (1983). SEGLEN, P.O. and BOHLEY, P., Autophagy and other vacuolar protein degradation mechanisms. Experientiu, 48,158-172 (1992). THIELE.D.L. and LIPSKY.P.E.. Auoutosis is induced in cells with cytolytic potential by L-leucyl-L-leu& methyl ester.L Imrnunol., 148, 3950-3957 (1992). TURK,V. and BODE,W., The cystatins: protein inhibitors of cysteine proteinases. FEBSLett., 285,213-219 (1991). WESSELS, J.M., STAUSS, J.M., SEIDLITZ, H.K., RUCK,A. and SCHNECKENBURGER, H., Intracellular localization of meso-tetraphenylporphine tetrasulfonate probed by time-resolved and microscopic fluorescence spectroscopy. J. Photochem. Photobiol., 12,275-284 (1992). WILSON,P.D., FIRESTONE, R.A. and LENARD, J., The role of lysosomal enzymes in killing of mammalian cells by the lysosomotropic detergent N-dodecylimidacole. J. CeN Biol., 104,1223-1229 (1987).