- Email: [email protected]
Differential Osmosensing Signalling Pathways and G-protein Involvement in Human Cervical Cells with Different Tumour Potential
Cell. Signal. Vol. 10, No. 2, pp. 113–120, 1998 Copyright 1998 Elsevier Science Inc.
ISSN 0898-6568/98 $19.00 PII S0898-6568(97)00115-0
Differential Osmosensing Signalling Pathways and G-protein Involvement in Human Cervical Cells with Different Tumour Potential Meng-Ru Shen, Cheng-Yang Chou,* Min-Lee Wu, Ko-En Huang Department of Obstetrics and Gynecology, National Cheng Kung University Medical College, 138 Sheng Li Road, Tainan 704, Taiwan
ABSTRACT. Previous studies show that the regulatory volume decrease (RVD) in human cervical cells with different tumour potential may be mediated by different ion channels. The signalling events involved in regulating these channel activities are not clear. To screen the possible mechanisms involved in cell volume regulation in these cells, we examine intracellular mechanisms and second messengers listed as follows: phospholipase C (PLC), phospholipase A2 (PLA2), tyrosine kinase (TK), protein kinase C (PKC), protein kinase A (PKA), and cAMP. The involvement of G-protein was also studied. Our results showed that PLC signalling with downstream activation of PKC was involved in the cell volume regulation of cervical cancer cells. On the other hand, different PKC isoforms that were not related to upstream PLC regulation were involved in the RVD of human papillomavirus (HPV)-immortalised and normal cervical epithelia. Furthermore, GTP-gamma S facilitated the process of RVD in cervical cancer cells, while pertussis toxin retarded this process. In contrast, neither GTP-gamma S nor pertussis toxin showed effect on the RVD responses of HPV-immortalised and normal cervical cells. cell signal 10;2:113–120, 1998. 1998 Elsevier Science Inc. KEY WORDS. Human cervical epithelia, Cell volume regulation, Regulatory volume decrease, Protein kinase C, G-protein
INTRODUCTION The maintenance of homeostasis is a fundamental property of the cells that form the tissues of the body. Two fundamental cellular homeostatic mechanisms are the regulation of cell volume and cellular pH. All cells possess mechanisms to regulate their volume precisely during mitosis and osmotic challenge. Cell volume homeostasis does not simply mean volume constancy, but rather cellular hydration state is an important determinant of cell function. Hormones, oxidation and nutrients exert their effects on metabolism and gene expression in part by a modification of cell volume [1, 2]. The mechanisms responsible for RVD may differ among different cell types but in general involve the activation of ionic transport systems in the plasma membrane. These ionic mechanisms have been reviewed extensively in the past [3, 4]. Swelling causes cells to activate transport systems for K1, Cl2, and certain organic molecules, resulting in water loss and restoration of the original cell volume [5]. In one common type of RVD, KCl leaves the cell via distinct K1 and Cl2 channels [6, 7]. However, the signal *Author to whom all correspondence should be addressed. E-mail: chou [email protected] Received 17 February 1997; and accepted 16 June 1997.
transduction pathways that control the RVD response to osmotic exposure are much less well understood. Several second messengers have been suggested as potential mediators of RVD, including calcium, calmodulin-dependent protein kinase, protein kinase C, cAMP and protein kinase A, and arachidonic acid and its metabolites [4, 8]. In our previous studies [9–11], the volume-sensitive Cl2 channels, leading to RVD, were distinctly activated in cervical cancer cell lines as well as in the primary culture cells of carcinoma in situ of the uterine cervix and invasive cancer, but not in HPV-immortalised cells and primary cultures of normal cervical epithelia. In addition, we have recently found that various potassium channel blockers inhibited the RVD processes in HPV-immortalised and normal cervical cells, suggesting the involvement of K1 channels in cell volume regulation of these cells (Chou et al., unpublished observations). Although the signalling pathways mediating RVD in these cells remain to be determined, it is, however, tempting to speculate that the mechanisms of signalling involved in this process may be different in different cervical cell types. Accordingly, the present study was conducted to screen and compare the intracellular mechanisms and second messengers involved in RVD regulation in different cell types of cervical epithelia. Six possible signalling path-
114
ways, namely PLC, PLA2, TK, PKC, PKA, and cAMP were studied. Furthermore, because we previously demonstrated that the GTP-gamma S-activated Cl2 current was similar to the volume-sensitive Cl2 current in the primary culture cells of cervical cancer [11], the role of G protein on cell volume regulation of cervical epithelial cells was thus further investigated. The results showed a PLC signalling with downstream activation of PKC and a pertussis-toxin sensitive G protein were involved during the process of RVD in cervical cancer cells. On the other hand, PKC isoforms unrelated to PLC activation were involved, while G protein was not implicated, in the RVD of HPV-immortalised and normal cervical epithelia. MATERIALS AND METHODS Cell Culture Two cervical cancer cell lines, HT-3 and CasKi, were obtained from American Type Culture Collection (Rockville, MD, USA). These cancer cells have been characterised and described in detail [12]. A HPV-immortalised cervical epithelial cell line, Z183A, was kindly provided by Dr. TzerMing Chen (National Taiwan University, Taipei, Taiwan). Z183A cells were established by electroporation of human cervical cells, and reportedly display alterations in growth and differentiation closely analogous to those seen in lowgrade cervical intraepithelial neoplasia [13]. The growth medium for cervical cancer cells was DMEM (Gibco Lab., Grand Island, NY, USA) supplemented with 10% foetal calf serum (Gibco Lab.), 80 IU/mL penicillin and 80 mg/mL streptomycin (Sigma, St. Louis, MO, USA). HPV-immortalised cells were maintained in DMEM containing 10% NuSerum IV (Collaborative Research, Bedford, MA, USA) in place of foetal calf serum. Primary cultures of normal cervical epithelial cells were derived from the portio surface and transformation zone of cervix uteri as described [10]. Normal cervical cells were a mixture of cells obtained from uterine cervix of four surgical specimens. The serum-free keratinocyte medium (Keratinocyte-SFM, Gibco Lab.) was employed for cell culturing. Cervical cells were maintained at 378C in a CO2/O2: 5%/95% atmosphere and used 2–3 d after subculturing. In order to eliminate the possibility that maintaining cells in different media may have consequences on the signalling pathways, the cervical cells were all adapted to grow in Keratinocyte-SFM 2–3 d before use.
M.-R. Shen et al.
In order to monitor the change of cell size, the microscope was coupled to a video camera system with magnification up to 4003 and continuously stored on a video cassette recorder (National Inc., Tokyo, Japan). The majority of cells observed were spheroid. Therefore, we defined the cell diameter as the mean of the longest axis plus the shortest axis. The data were presented as the percentage of starting volume (V/Vo), as a function of time. Signal Transduction Pathways To determine possible signal transduction pathways involved during the process of RVD, the following reagents were used: neomycin, a PLC blocker; 4-bromophenacyl bromide (pBPB), a PLA2 inhibitor; erbstatin analogue, a tyrosine kinase blocker; staurosporine and H7 dihydrochloride (H7), PKC blockers; phorbol 12-myristate 13-acetate (PMA), a PKC activator; protein kinase A inhibitor (PKI), a PKA inhibitor; 3-isobutyl-1-methylxanthine (IBMX), a PKA activator; and dibutyryl cAMP (db-cAMP), a cell membrane-permeable c-AMP analogue. Micromanipulation of Cells In the screening of signalling pathways involved in RVD, some blockers that were poorly permeable to cell membrane were delivered into the cytoplasm by microinjection. These blockers included: neomycin, H7, and PKI. This technique is efficient for the transfer of macromolecule into culture cells. In addition, the procedure of micromanipulation is reported to have no effect on cell function of RVD [16, 17]. For preparing the micropipettes for microinjection, the GD-1 glass capillaries (Narishige Scientific Instrument Lab, Tokyo, Japan) were heated and pulled by gravity using a two-step, vertical micropipette puller (PC-10; Narishige Scientific Instrument Lab., Tokyo, Japan). Coarse positioning manipulator (ONM-1) and ergonomic joystick micromanipulator (ONO-125; Narishige Scientific Instrument Lab., Tokyo, Japan) were used to position the micropipette near the cell. Cells were injected with 10 ml of various blockers by microinjector (IM-6, Narishige Scientific Instrument Lab., Tokyo, Japan). The composition of pipette solution is as follows (in mM): KCl 130.0, MgCl2 2.0, Hepes 5.0 (buffered by NaOH to pH 7.2). Role of G Protein on RVD
Measurement of Cell Volume Cell volume was measured as described previously [14, 15]. Briefly, cells were grown on tissue culture flask and were harvested by trypsinisation. The 3 3 106 cells were suspended in 5 mL of an isotonic (300 mOsm/L) solution for 10 min. Then, 500 mL aliquots of the cells in the isotonic solution were suspended in a 2 mL bath which was continuously superfused at a rate of 2 mL/min with the isotonic solution, the hypotonic solution, or the hypotonic solution containing different channel blockers at room temperature.
GTP-gamma S, a G protein activator, and pertussis toxin, which inhibits a certain subgroup of G proteins were used to characterise the G protein involved in the processes of RVD in different cervical cells. Chemicals and Solutions Staurosporine, IBMX, H7, erbstatin analogue, PMA were acquired from Research Biochemicals International (Natick, MA, USA). The other chemicals were purchased from Sigma Chemical (St. Louis, MO, USA). The osmolarity of
G-protein Involvement in Cell Volume Regulation
115
112, KCl 5.4, CaCl2 1.8, MgCl2 0.53, glucose 5.5, HepesNaOH buffer 5 (pH 7.4). Statistics All data are expressed as mean 6 SE Student’s paired or unpaired t-test and analysis of variance (ANOVA), as appropriate, are used for the statistical analyses. Differences between values are considered significant when P , 0.05. RESULTS Volume Response to Hypotonic Stress In cervical cancer cell lines (HT-3 and CasKi, Fig. 1, control group), Z183A cells, and normal cervical epithelia, the typical volume response can be divided into three phases on exposure to 250 mOsm/L hypotonic solution: first, an initial and rapid, osmotic swelling to a peak volume at 20–30 min; then a rapid shrinkage in the following 20 min; and finally a more gradual decrease in cell volume to the original cell size at 50–60 min. Signal Transduction Pathways in HT-3 and CasKi Cells
FIGURE 1. Time course of volume changes of cervical cancer
cell lines (CasKi and HT-3) following superfusion with 250 mOsm/L hypotonic bath solution. Neomycin (1 mM), a PLC blocker, was transferred into cancer cells by micromanipulation. Ten minutes after micromanipulation, hypotonic solution was perfused. The effect of 0.5 mM neomycin was almost identical to that induced by 1 mM neomycin, and therefore was not shown. The y axis (V/Vo) depicts the ratio of cell volume at time 0 min divided by the cell volume at the indicated times. Each point represents mean 6 SE (n 5 20 cells).
the solution was measured using a freezing-point osmometer (Roebling, Germany). According to our previous experiments, cancer cells, HPV-immortalised cells and normal cervical cells can well function the RVD process under 250 mOsm/L hypotonic stress. However, only cancer cells can well regulate their cell sizes below 220 mOsm/L hypotonic stress (Chou et al., submitted for publication). Therefore, 250 mOsm/L was chosen as the hypotonic condition in this study. The composition of isotonic solution (300 6 10 mOsm/L) is as follows (in mM): NaCl 136.5, KCl 5.4, CaCl2 1.8, MgCl2 0.53, glucose 5.5, Hepes-NaOH buffer 5 (pH 7.4). The composition of 250 mOsm/kg hypotonic solution (250 6 10 mOsm/L) is as follows (in mM): NaCl
To investigate the role of PLC in the regulation of RVD, neomycin (0.5 or 1 mM), a PLC blocker, was transferred into cervical cells by micromanipulation. Ten minutes after micromanipulation, hypotonic solution was perfused. As shown in Figure 1, neomycin increased the initial, rapid osmotic swelling, attenuating the shrinkage phase, and inhibited the gradual, slower decrease in cell volume, compared to cancer cells not treated with neomycin. The inhibitory effect induced by 0.5 mM neomycin was almost identical to that induced by 1 mM neomycin (data not shown). When HT-3 cells were exposed to hypotonic stress in the presence of either 1 mM staurosporine or 100 mM H7, the regulation of RVD was obviously inhibited (Fig. 2). In the presence of H7, cell volume at 30 min was 1.43 6 0.06 (n 5 20) of initial volume, compared to a 30-min volume of 1.18 6 0.03 (n 5 20) for cells without H7 treatment. In addition, cell volume was maintained as 1.30 6 0.02 of initial measurement after 60-min perfusion with hypotonic solution in the presence of H7. In contrast, cancer cells, not treated with H7, can return to the original cell size after 50–60 min. Another PKC blocker, staurosporine, also showed similar inhibitory effect on RVD. On the other hand, when 0.1 mM PMA, a PKC activator, was added in the hypotonic medium, the time course of RVD was shortened significantly (Fig. 3). Under the control circumstance, cell volume reached the maximal cell size at 30 6 5 min and returned to the original cell size at 50–60 min after exposure to hypotonic solution (Fig. 3). Subsequent return to isotonic solution caused a regulatory volume increase (RVI). Hereafter, cells were perfused with hypotonic solution added with 0.1 mM PMA. As shown in Figure 3, the RVD process accelerated: HT-3 cells reached the maximal cell volume at about 18 6 3 min and the cell volume returned to the steady state
116
M.-R. Shen et al.
FIGURE 2. Effect of PKC blockers on cell volume regulation of HT-3 cells during exposure to 250 mOsm/L hypotonic solution. The y axis (V/Vo) depicts the ratio of cell volume at time 0 min divided by the cell volume at the indicated times. Each point represents mean 6 SE (n 5 20 cells).
FIGURE 3. Time course of volume changes of HT-3 cells following superfusion with 250 mOsm/L hypotonic bath solution, in the pres-
ence or absence of 0.1 mM PMA. Neomycin (1 mM), a PLC blocker, was transferred into cancer cells by micromanipulation. The y axis (V/Vo) depicts the ratio of cell volume at time 0 min divided by the cell volume at the indicated times. The sequences of changes in the superfusion solution are illustrated above the panel. Each point represents mean 6 SE (n 5 20 cells). PMA: phorbol 12-myristate 13-acetate.
G-protein Involvement in Cell Volume Regulation
at about 35 min. Similar effects of staurosporine, H7, and PMA on RVD were also observed in CasKi cells. Because both PLC and PKC were implicated in the regulation of RVD in cervical cancer cells, it is tempting to speculate that isoform of conventional PKC (types I, II, and III) which is activated through an upstream PLC signalling [18] may be involved in that regulation. Two more series of experiments were thus performed to investigate the interaction of PLC and PKC: (1) HT-3 cancer cells were injected with 1 mM neomycin by micromanipulation. Ten minutes after that procedure, cells were perfused with hypotonic solution containing 0.1 mM PMA. (2) Cells were perfused with hypotonic solution containing 0.1 mM PMA at first. Then, neomycin was transferred into cells at different time intervals: 10, 20, 30 min after hypotonic shock. As depicted in Figure 3, there was no significant difference in the RVD processes between cells exposed to hypotonic solution containing PMA and cells injected with neomycin, then exposed to hypotonic solution containing PMA. In addition, Fig. 4 revealed that the acceleration of RVD process by PMA was not affected by injection of neomycin at different time intervals. Similar experiments were done on CasKi cells, and identical results were obtained. Altogether, these data confirmed that PKC activation through an upstream PLC signalling is involved during RVD of cervical cancer cells. When HT-3 cells were exposed to hypotonic solution, the regulation of RVD was not significantly affected in the presence of IBMX, db-cAMP, pBPB, PKI, or erbstatin analogue. These data suggest that none of the PLA2, tyrosine
FIGURE 4. Time course of vol-
ume changes of HT-3 cells following superfusion with 250 mOsm/L hypotonic bath solution containing 0.1 mM PMA (dash line, control group). The arrowhead and close symbols indicated 1 mM neomycin was microinjected at different time intervals: 10, 20, 30 min after cell exposure to hypotonic solution. The y axis (V/Vo) depicts the ratio of cell volume at time 0 min divided by the cell volume at the indicated times. Each point represents mean 6 SE (n 5 6 cells).
117
kinase, PKC, PKA, or cAMP pathways are involved in the regulation of RVD in HT-3 cells.
Signal Transduction Pathway in HPV Immortalised Cells and Normal Cervical Epithelia The RVD process in Z183A cells was nearly completely inhibited by PKC blockers, H7 and staurosporine (Fig. 5A). In the presence of either staurosporine or H7, Z183A cells increased the initial, rapid osmotic swelling, attenuating the shrinkage phase, and inhibited the gradual, slower decrease in cell volume, compared to Z183A cells exposed to hypotonic solution without PKC blockers. In addition, the time course of RVD was significantly shortened by 0.1 mM PMA (Fig. 5B). Under the control circumstance, cell volume reached the maximal cell size at 23 6 5 min and returned to the original cell size at 50 6 5 min after exposure to hypotonic stress. In the presence of PMA, Z183A cells reached the maximal cell volume at about 12 6 5 min and the cell volume returned to the steady state at 38 6 5 min. Figures 5C and 5D showed the time course of cell volume changes of normal cervical epithelium, following hypotonic stress in the presence or absence of PKC activator or PKC inhibitors. The RVD processes of normal cervical cells were significantly inhibited by 1 mM staurosporine and 100 mM H7 (Fig. 5C). On the other hand, 0.1 mM PMA accelerated the RVD process (Fig. 5D). The processes of RVD in Z183A cells or normal cervical epithelia were not significantly affected by the addition of
118
M.-R. Shen et al.
FIGURE 5. Time course of volume changes of HPV immortalised cells (Z183A) and normal cervical epithelia following superfusion
with 250 mOsm/L hypotonic solution, in the presence of PKC blockers (H7 and staurosporine) or PKC activator (PMA). The y axis (V/Vo) depicts the ratio of cell volume at time 0 min divided by the cell volume at the indicated times. Each point represents mean 6 SE (n 5 20 cells). PMA: phorbol 12-myristate 13-acetate.
neomycin (0.5 or 1 mM), IBMX, db-cAMP, pBPB, PKI, or erbstatin analogue during hyposmotic challenge. Role of G Protein on RVD Figure 6 showed that when 100 mM GTP-gamma S was added in the hypotonic medium, the RVD process of HT-3 cells significantly accelerated. In addition, the function of RVD was blocked when HT-3 cells were pre-incubated with 100 ng/mL pertussis toxin for 12 h. In contrast, the process of RVD or RVD reversal in Z183A and normal cervical cells was not affected by either GTP-gamma S or pertussis toxin (Fig. 6). These results suggested that in cervical cancer cells the regulation of RVD is associated with pertussis toxin sensitive G protein, while RVD in Z183A and normal cervical cells was not regulated by G protein. DISCUSSION Our data suggest the involvement of different PKC isoforms and the differential activation of a pertussis-toxin sensitive G protein in signalling events of RVD in human cervical epithelial cells with different tumour potential. The active phorbol ester, PMA, which stimulates PKC, accelerates
RVD; while the blockers of PKC, staurosporine and H7, impair the RVD function. PMA can substitute for diacylglycerol (DAG), binding to PKC, and subsequently increasing the affinity of PKC for Ca21 [19]. Therefore, in the presence of PMA, entry of Ca21 to the cell during cell swelling will lead to greater activation of PKC and, subsequently, solute exit pathways. The PKC family is a heterogeneous family of phospholipid-dependent kinase that can be divided into three categories on the basis of cofactor requirements and structure [18, 20]. Conventional PKCs require calcium and DAG or phorbol ester as cofactors. Activated receptors of tyrosine kinase and G-protein coupled classes are able to recruit conventional PKC for intracellular signalling. This is primarily mediated through receptor-induced hydrolysis of phosphatidylinositol biphosphate (PIP2) by PLC, which generates two second messengers, inositol 1,4,5-trisphosphate (IP3) and DAG. DAG activates PKC directly and IP3 indirectly stimulates PKC via its promotion of the release of Ca21 from intracellular storage in the endoplasmic reticulum. Novel PKCs require only DAG or phorbol ester, whereas atypical PKCs do not require calcium or DAG for maximal activity. This report presents the findings that the signalling pathways mediating RVD in different cervical cell types may involve differential activation of different
G-protein Involvement in Cell Volume Regulation
FIGURE 6. Role of G protein on cell volume regulation of cervi-
cal epithelial cells with different tumour potential. The y axis (V/Vo) depicts the ratio of cell volume at time 0 min divided by the cell volume at the indicated times. Each point represents mean 6 SE (n 5 20 cells). Hypo: hypotonic medium; GTPrS: hypotonic medium plus 100 mM GTP-gamma S; Pertussis toxin: cells pre-incubated with 100 ng/mL pertussis toxin before hypotonic exposure. *P , 0.05, compared with the volume ratio in the hypotonic condition at 60 min, paired t-test.
PKC isoforms. In cervical cancer cells, PKC activation through an upstream PLC signalling was involved during the process of RVD. On the other hand, PKCs that are not related to PLC activation were involved in the RVD of HPV-immortalised and normal cervical epithelia. The present study suggests that the PKC isoform involved in the signalling of volume control of cancer cells may be downstream components of phosphatidylinositol (PI)- specific PLC signalling pathways. Thus, it is likely to be conventional PKCs. On the other hand, PKCs involved in the volume regulation of
119
HPV-immortalised and normal cervical cells are favoured as the category of novel PKCs due to the evidence of activation by phorbol ester and the lack of correlation with PIPLC signalling pathways. Further studies should be done to characterise the individual PKCs responsible for the signalling events involved in the process of RVD in each cervical cell type. PKC is reportedly involved in basic cellular functions, including regulation of growth, differentiation, and gene expression [20]. Potent inhibitors of PKC are shown to have cytostatic and chemosensitising properties [21]. Furthermore, evidence showing that overexpression of different PKC isoforms may induce opposite effects on growth, anchorage dependence, and tumorogenicity [22], and that differential localisation of PKCs or different amount of PKCbinding protein and substrate in normal and transformed cells has been noted [23, 24]. These data, together with our present findings that the signalling pathways of RVD in cervical cells with different tumour potential involve differential activation of PKC isoforms, suggest the possible correlation of PKC differential activation with cervical carcinogenesis. However, the mechanism underlying this differential activation remains unknown, and thus the association between the differential signalling events and carcinogenesis needs to be critically examined. The changes in volume-sensitive transport pathways in response to cell volume changes require the following mechanisms [4]: a signal of volume change, a sensor of the signal of the change, transduction of the signal to the effector, i.e., the transporter or channels, and changes in the function of the transporter or channels. These mechanisms are necessary for all transport pathways involved in the regulation of cell volume. The signal transduction pathways during hypotonic stress in cervical cancer cells may be hypothesised as follows. Cell membranes sense the signal of hypotonic shock and degradation of membrane phospholipid happens. IP2 is degraded by PLC into two signalling components: IP3 and DAG. IP3 causes the release of Ca21 from internal stores, while DAG activates PKC. Both may lead to the activation of volume-sensitive Cl2 channels and regulation of cell volume. Furthermore, both PLC and pertussis-toxin sensitive G protein are implicated in the signalling events of RVD, whereas tyrosine kinase inhibitor did not affect that signalling. Although the direct interaction of PLC and pertussis-toxin sensitive G protein cannot be demonstrated in the present study, the above-mentioned data suggest that a pertussis-toxin sensitive G protein may regulate the activation of PLC. Further characterisation of PKC, PLC and G protein should better define the mechanisms of signalling involved in this process. PKC and PLC have variable effects on RVD in other cell types. In rat hepatocytes, swelling causes a transient increase in IP3 [25]. Hypotonicity is shown to stimulate phosphatidylcholine hydrolysis and generate DAG in skate erythrocytes [26]. In proximal tubule cells of Rana temporaria, chloride conductance is activated by PKC-mediated phosphorylation and plays a role in volume regulation [27].
120
However, neither PKC nor PLC pathways are involved in the regulation of volume-sensitive Cl2 channel in human endothelial cells [28]. Altogether, it seems likely that the role of PKCs as a whole and individual PKC in the signalling events involved in the process of RVD may be cell type-specific. In conclusion, PKC activation through an upstream PLC signalling and a pertussis-toxin sensitive G protein were involved during the process of RVD in cervical cancer cells. On the other hand, PKCs that are not regulated by PLC activation were involved, whereas G protein was not implicated in the RVD of HPV-immortalised and normal cervical cells. Further studies are required to elucidate the role of PKCs in cervical carcinogenesis. This study was supported by a grant NSC87-2314-B006-132 from National Science Council, Taiwan. We thank Dr. Tzer-Ming Chen (National Taiwan University, Taipei, Taiwan) for providing the HPV-immortalized cells. We also thank Shu-Hua Cheng for technical assistance.
References 1. Ha¨ussinger D. and Lang F. (1992) Trends Pharmacol. Sci. 13, 371–373. 2. Ha¨ussinger D. (1996) Biochem. J. 313, 697–710. 3. Ha¨ussinger D., Lang F. and Gerok W. (1994) Am. J. Physiol. 267, E343–E355. 4. Hoffmann E. K. and Dunham P. B. (1995) Int. Rev. Cytol. 161, 173–262. 5. Hoffmann E. K. and Simonsen L. O. (1989) Physiol. Rev. 69, 315–373. 6. Rothstein A. and Bear C. (1989) Ann. New York Acad. Sci. 574, 294–308. 7. Hazama A. and Okada Y. (1988) J. Physiol. 402, 687–702. 8. McCarty N. and O’Neil R. G. (1992) Physiol. Rev. 72, 1037– 1061.
M.-R. Shen et al. 9. Shen M. R., Chou C. Y. and Wu S. N. (1995) FASEB J. 6, A356. 10. Chou C. Y., Shen M. R. and Wu S. N. (1995) Cancer Res. 55, 6077–6083. 11. Shen M. R., Wu S. N. and Chou C. Y. (1996) Biochim. Biophys. Acta 1315, 138–144. 12. Scheffner M., Munger K., Byrne J. C. and Howley P. M. (1991) Proc. Natl. Acad. Sci. USA 88, 5523–5527. 13. Pecoraro G., Lee M., Morgan D. and Defendi V. (1991) Am. J. Pathol. 138, 1–8. 14. Suzuki M., Kawahara K., Ogawa A., Morita T., Kawaguchi Y., Kurihara S. and Sakai O. (1990) Am. J. Physiol. 258, F690– F696. 15. Park K. P., Beck J. S., Douglas I. J. and Brown P. D. (1994) J. Membr. Biol. 141, 193–201. 16. Kleuss C., Schultz G. and Wittig B. (1994) Methods Enzymol. 237, 345–355. 17. Ridley A. J. (1995) Methods Enzymol. 256, 313–320. 18. Jaken S. (1996) Curr. Opin. Cell Biol. 8, 168–173. 19. Rasmussen H., Isales C. M., Calle R., Throckmorton D., Anderson M., Gasalla-Herraiz J. and McCarthy R. (1995) Endocr. Rev. 16, 649–681. 20. Nishizuka Y. (1995) FASEB J. 9, 484–496. 21. Powis G. (1992) Trends Pharmacol. Sci. 12, 188–294. 22. Mischak H., Goodnight J., Kolch W., Martiny-Baron G., Schaechtles C., Kazanietz M. G., Blumberg P. M., Pierce J. H. and Mushinski J. F. (1993) J. Biol. Chem. 268, 6090–6096. 23. Hyatt S. L., Liao L., Chapline C. and Jaken S. (1994) Cell Growth Differ. 5, 495–502. 24. Chapline C., Mousseau B., Ramsay K., Duddy S., Li Y., Kiley S. C. and Jaken S. (1996) J. Biol. Chem. 271, 6417–6422. 25. Dahl S. V., Hallbrucker C., Lang F. and Ha¨ussinger D. (1991) Eur. J. Biochem. 198, 73–83. 26. Musch M. W. and Goldstein L. (1990) J. Biol. Chem. 365, 13055–13059. 27. Robson L. and Hunter M. (1994) J. Physiol. 480, 1–7. 28. Nilius B., Oike M., Zahradnik I. and Droogmans G. (1994) J. Gen. Physiol. 103, 787–805.
ISSN 0898-6568/98 $19.00 PII S0898-6568(97)00115-0
Differential Osmosensing Signalling Pathways and G-protein Involvement in Human Cervical Cells with Different Tumour Potential Meng-Ru Shen, Cheng-Yang Chou,* Min-Lee Wu, Ko-En Huang Department of Obstetrics and Gynecology, National Cheng Kung University Medical College, 138 Sheng Li Road, Tainan 704, Taiwan
ABSTRACT. Previous studies show that the regulatory volume decrease (RVD) in human cervical cells with different tumour potential may be mediated by different ion channels. The signalling events involved in regulating these channel activities are not clear. To screen the possible mechanisms involved in cell volume regulation in these cells, we examine intracellular mechanisms and second messengers listed as follows: phospholipase C (PLC), phospholipase A2 (PLA2), tyrosine kinase (TK), protein kinase C (PKC), protein kinase A (PKA), and cAMP. The involvement of G-protein was also studied. Our results showed that PLC signalling with downstream activation of PKC was involved in the cell volume regulation of cervical cancer cells. On the other hand, different PKC isoforms that were not related to upstream PLC regulation were involved in the RVD of human papillomavirus (HPV)-immortalised and normal cervical epithelia. Furthermore, GTP-gamma S facilitated the process of RVD in cervical cancer cells, while pertussis toxin retarded this process. In contrast, neither GTP-gamma S nor pertussis toxin showed effect on the RVD responses of HPV-immortalised and normal cervical cells. cell signal 10;2:113–120, 1998. 1998 Elsevier Science Inc. KEY WORDS. Human cervical epithelia, Cell volume regulation, Regulatory volume decrease, Protein kinase C, G-protein
INTRODUCTION The maintenance of homeostasis is a fundamental property of the cells that form the tissues of the body. Two fundamental cellular homeostatic mechanisms are the regulation of cell volume and cellular pH. All cells possess mechanisms to regulate their volume precisely during mitosis and osmotic challenge. Cell volume homeostasis does not simply mean volume constancy, but rather cellular hydration state is an important determinant of cell function. Hormones, oxidation and nutrients exert their effects on metabolism and gene expression in part by a modification of cell volume [1, 2]. The mechanisms responsible for RVD may differ among different cell types but in general involve the activation of ionic transport systems in the plasma membrane. These ionic mechanisms have been reviewed extensively in the past [3, 4]. Swelling causes cells to activate transport systems for K1, Cl2, and certain organic molecules, resulting in water loss and restoration of the original cell volume [5]. In one common type of RVD, KCl leaves the cell via distinct K1 and Cl2 channels [6, 7]. However, the signal *Author to whom all correspondence should be addressed. E-mail: chou [email protected] Received 17 February 1997; and accepted 16 June 1997.
transduction pathways that control the RVD response to osmotic exposure are much less well understood. Several second messengers have been suggested as potential mediators of RVD, including calcium, calmodulin-dependent protein kinase, protein kinase C, cAMP and protein kinase A, and arachidonic acid and its metabolites [4, 8]. In our previous studies [9–11], the volume-sensitive Cl2 channels, leading to RVD, were distinctly activated in cervical cancer cell lines as well as in the primary culture cells of carcinoma in situ of the uterine cervix and invasive cancer, but not in HPV-immortalised cells and primary cultures of normal cervical epithelia. In addition, we have recently found that various potassium channel blockers inhibited the RVD processes in HPV-immortalised and normal cervical cells, suggesting the involvement of K1 channels in cell volume regulation of these cells (Chou et al., unpublished observations). Although the signalling pathways mediating RVD in these cells remain to be determined, it is, however, tempting to speculate that the mechanisms of signalling involved in this process may be different in different cervical cell types. Accordingly, the present study was conducted to screen and compare the intracellular mechanisms and second messengers involved in RVD regulation in different cell types of cervical epithelia. Six possible signalling path-
114
ways, namely PLC, PLA2, TK, PKC, PKA, and cAMP were studied. Furthermore, because we previously demonstrated that the GTP-gamma S-activated Cl2 current was similar to the volume-sensitive Cl2 current in the primary culture cells of cervical cancer [11], the role of G protein on cell volume regulation of cervical epithelial cells was thus further investigated. The results showed a PLC signalling with downstream activation of PKC and a pertussis-toxin sensitive G protein were involved during the process of RVD in cervical cancer cells. On the other hand, PKC isoforms unrelated to PLC activation were involved, while G protein was not implicated, in the RVD of HPV-immortalised and normal cervical epithelia. MATERIALS AND METHODS Cell Culture Two cervical cancer cell lines, HT-3 and CasKi, were obtained from American Type Culture Collection (Rockville, MD, USA). These cancer cells have been characterised and described in detail [12]. A HPV-immortalised cervical epithelial cell line, Z183A, was kindly provided by Dr. TzerMing Chen (National Taiwan University, Taipei, Taiwan). Z183A cells were established by electroporation of human cervical cells, and reportedly display alterations in growth and differentiation closely analogous to those seen in lowgrade cervical intraepithelial neoplasia [13]. The growth medium for cervical cancer cells was DMEM (Gibco Lab., Grand Island, NY, USA) supplemented with 10% foetal calf serum (Gibco Lab.), 80 IU/mL penicillin and 80 mg/mL streptomycin (Sigma, St. Louis, MO, USA). HPV-immortalised cells were maintained in DMEM containing 10% NuSerum IV (Collaborative Research, Bedford, MA, USA) in place of foetal calf serum. Primary cultures of normal cervical epithelial cells were derived from the portio surface and transformation zone of cervix uteri as described [10]. Normal cervical cells were a mixture of cells obtained from uterine cervix of four surgical specimens. The serum-free keratinocyte medium (Keratinocyte-SFM, Gibco Lab.) was employed for cell culturing. Cervical cells were maintained at 378C in a CO2/O2: 5%/95% atmosphere and used 2–3 d after subculturing. In order to eliminate the possibility that maintaining cells in different media may have consequences on the signalling pathways, the cervical cells were all adapted to grow in Keratinocyte-SFM 2–3 d before use.
M.-R. Shen et al.
In order to monitor the change of cell size, the microscope was coupled to a video camera system with magnification up to 4003 and continuously stored on a video cassette recorder (National Inc., Tokyo, Japan). The majority of cells observed were spheroid. Therefore, we defined the cell diameter as the mean of the longest axis plus the shortest axis. The data were presented as the percentage of starting volume (V/Vo), as a function of time. Signal Transduction Pathways To determine possible signal transduction pathways involved during the process of RVD, the following reagents were used: neomycin, a PLC blocker; 4-bromophenacyl bromide (pBPB), a PLA2 inhibitor; erbstatin analogue, a tyrosine kinase blocker; staurosporine and H7 dihydrochloride (H7), PKC blockers; phorbol 12-myristate 13-acetate (PMA), a PKC activator; protein kinase A inhibitor (PKI), a PKA inhibitor; 3-isobutyl-1-methylxanthine (IBMX), a PKA activator; and dibutyryl cAMP (db-cAMP), a cell membrane-permeable c-AMP analogue. Micromanipulation of Cells In the screening of signalling pathways involved in RVD, some blockers that were poorly permeable to cell membrane were delivered into the cytoplasm by microinjection. These blockers included: neomycin, H7, and PKI. This technique is efficient for the transfer of macromolecule into culture cells. In addition, the procedure of micromanipulation is reported to have no effect on cell function of RVD [16, 17]. For preparing the micropipettes for microinjection, the GD-1 glass capillaries (Narishige Scientific Instrument Lab, Tokyo, Japan) were heated and pulled by gravity using a two-step, vertical micropipette puller (PC-10; Narishige Scientific Instrument Lab., Tokyo, Japan). Coarse positioning manipulator (ONM-1) and ergonomic joystick micromanipulator (ONO-125; Narishige Scientific Instrument Lab., Tokyo, Japan) were used to position the micropipette near the cell. Cells were injected with 10 ml of various blockers by microinjector (IM-6, Narishige Scientific Instrument Lab., Tokyo, Japan). The composition of pipette solution is as follows (in mM): KCl 130.0, MgCl2 2.0, Hepes 5.0 (buffered by NaOH to pH 7.2). Role of G Protein on RVD
Measurement of Cell Volume Cell volume was measured as described previously [14, 15]. Briefly, cells were grown on tissue culture flask and were harvested by trypsinisation. The 3 3 106 cells were suspended in 5 mL of an isotonic (300 mOsm/L) solution for 10 min. Then, 500 mL aliquots of the cells in the isotonic solution were suspended in a 2 mL bath which was continuously superfused at a rate of 2 mL/min with the isotonic solution, the hypotonic solution, or the hypotonic solution containing different channel blockers at room temperature.
GTP-gamma S, a G protein activator, and pertussis toxin, which inhibits a certain subgroup of G proteins were used to characterise the G protein involved in the processes of RVD in different cervical cells. Chemicals and Solutions Staurosporine, IBMX, H7, erbstatin analogue, PMA were acquired from Research Biochemicals International (Natick, MA, USA). The other chemicals were purchased from Sigma Chemical (St. Louis, MO, USA). The osmolarity of
G-protein Involvement in Cell Volume Regulation
115
112, KCl 5.4, CaCl2 1.8, MgCl2 0.53, glucose 5.5, HepesNaOH buffer 5 (pH 7.4). Statistics All data are expressed as mean 6 SE Student’s paired or unpaired t-test and analysis of variance (ANOVA), as appropriate, are used for the statistical analyses. Differences between values are considered significant when P , 0.05. RESULTS Volume Response to Hypotonic Stress In cervical cancer cell lines (HT-3 and CasKi, Fig. 1, control group), Z183A cells, and normal cervical epithelia, the typical volume response can be divided into three phases on exposure to 250 mOsm/L hypotonic solution: first, an initial and rapid, osmotic swelling to a peak volume at 20–30 min; then a rapid shrinkage in the following 20 min; and finally a more gradual decrease in cell volume to the original cell size at 50–60 min. Signal Transduction Pathways in HT-3 and CasKi Cells
FIGURE 1. Time course of volume changes of cervical cancer
cell lines (CasKi and HT-3) following superfusion with 250 mOsm/L hypotonic bath solution. Neomycin (1 mM), a PLC blocker, was transferred into cancer cells by micromanipulation. Ten minutes after micromanipulation, hypotonic solution was perfused. The effect of 0.5 mM neomycin was almost identical to that induced by 1 mM neomycin, and therefore was not shown. The y axis (V/Vo) depicts the ratio of cell volume at time 0 min divided by the cell volume at the indicated times. Each point represents mean 6 SE (n 5 20 cells).
the solution was measured using a freezing-point osmometer (Roebling, Germany). According to our previous experiments, cancer cells, HPV-immortalised cells and normal cervical cells can well function the RVD process under 250 mOsm/L hypotonic stress. However, only cancer cells can well regulate their cell sizes below 220 mOsm/L hypotonic stress (Chou et al., submitted for publication). Therefore, 250 mOsm/L was chosen as the hypotonic condition in this study. The composition of isotonic solution (300 6 10 mOsm/L) is as follows (in mM): NaCl 136.5, KCl 5.4, CaCl2 1.8, MgCl2 0.53, glucose 5.5, Hepes-NaOH buffer 5 (pH 7.4). The composition of 250 mOsm/kg hypotonic solution (250 6 10 mOsm/L) is as follows (in mM): NaCl
To investigate the role of PLC in the regulation of RVD, neomycin (0.5 or 1 mM), a PLC blocker, was transferred into cervical cells by micromanipulation. Ten minutes after micromanipulation, hypotonic solution was perfused. As shown in Figure 1, neomycin increased the initial, rapid osmotic swelling, attenuating the shrinkage phase, and inhibited the gradual, slower decrease in cell volume, compared to cancer cells not treated with neomycin. The inhibitory effect induced by 0.5 mM neomycin was almost identical to that induced by 1 mM neomycin (data not shown). When HT-3 cells were exposed to hypotonic stress in the presence of either 1 mM staurosporine or 100 mM H7, the regulation of RVD was obviously inhibited (Fig. 2). In the presence of H7, cell volume at 30 min was 1.43 6 0.06 (n 5 20) of initial volume, compared to a 30-min volume of 1.18 6 0.03 (n 5 20) for cells without H7 treatment. In addition, cell volume was maintained as 1.30 6 0.02 of initial measurement after 60-min perfusion with hypotonic solution in the presence of H7. In contrast, cancer cells, not treated with H7, can return to the original cell size after 50–60 min. Another PKC blocker, staurosporine, also showed similar inhibitory effect on RVD. On the other hand, when 0.1 mM PMA, a PKC activator, was added in the hypotonic medium, the time course of RVD was shortened significantly (Fig. 3). Under the control circumstance, cell volume reached the maximal cell size at 30 6 5 min and returned to the original cell size at 50–60 min after exposure to hypotonic solution (Fig. 3). Subsequent return to isotonic solution caused a regulatory volume increase (RVI). Hereafter, cells were perfused with hypotonic solution added with 0.1 mM PMA. As shown in Figure 3, the RVD process accelerated: HT-3 cells reached the maximal cell volume at about 18 6 3 min and the cell volume returned to the steady state
116
M.-R. Shen et al.
FIGURE 2. Effect of PKC blockers on cell volume regulation of HT-3 cells during exposure to 250 mOsm/L hypotonic solution. The y axis (V/Vo) depicts the ratio of cell volume at time 0 min divided by the cell volume at the indicated times. Each point represents mean 6 SE (n 5 20 cells).
FIGURE 3. Time course of volume changes of HT-3 cells following superfusion with 250 mOsm/L hypotonic bath solution, in the pres-
ence or absence of 0.1 mM PMA. Neomycin (1 mM), a PLC blocker, was transferred into cancer cells by micromanipulation. The y axis (V/Vo) depicts the ratio of cell volume at time 0 min divided by the cell volume at the indicated times. The sequences of changes in the superfusion solution are illustrated above the panel. Each point represents mean 6 SE (n 5 20 cells). PMA: phorbol 12-myristate 13-acetate.
G-protein Involvement in Cell Volume Regulation
at about 35 min. Similar effects of staurosporine, H7, and PMA on RVD were also observed in CasKi cells. Because both PLC and PKC were implicated in the regulation of RVD in cervical cancer cells, it is tempting to speculate that isoform of conventional PKC (types I, II, and III) which is activated through an upstream PLC signalling [18] may be involved in that regulation. Two more series of experiments were thus performed to investigate the interaction of PLC and PKC: (1) HT-3 cancer cells were injected with 1 mM neomycin by micromanipulation. Ten minutes after that procedure, cells were perfused with hypotonic solution containing 0.1 mM PMA. (2) Cells were perfused with hypotonic solution containing 0.1 mM PMA at first. Then, neomycin was transferred into cells at different time intervals: 10, 20, 30 min after hypotonic shock. As depicted in Figure 3, there was no significant difference in the RVD processes between cells exposed to hypotonic solution containing PMA and cells injected with neomycin, then exposed to hypotonic solution containing PMA. In addition, Fig. 4 revealed that the acceleration of RVD process by PMA was not affected by injection of neomycin at different time intervals. Similar experiments were done on CasKi cells, and identical results were obtained. Altogether, these data confirmed that PKC activation through an upstream PLC signalling is involved during RVD of cervical cancer cells. When HT-3 cells were exposed to hypotonic solution, the regulation of RVD was not significantly affected in the presence of IBMX, db-cAMP, pBPB, PKI, or erbstatin analogue. These data suggest that none of the PLA2, tyrosine
FIGURE 4. Time course of vol-
ume changes of HT-3 cells following superfusion with 250 mOsm/L hypotonic bath solution containing 0.1 mM PMA (dash line, control group). The arrowhead and close symbols indicated 1 mM neomycin was microinjected at different time intervals: 10, 20, 30 min after cell exposure to hypotonic solution. The y axis (V/Vo) depicts the ratio of cell volume at time 0 min divided by the cell volume at the indicated times. Each point represents mean 6 SE (n 5 6 cells).
117
kinase, PKC, PKA, or cAMP pathways are involved in the regulation of RVD in HT-3 cells.
Signal Transduction Pathway in HPV Immortalised Cells and Normal Cervical Epithelia The RVD process in Z183A cells was nearly completely inhibited by PKC blockers, H7 and staurosporine (Fig. 5A). In the presence of either staurosporine or H7, Z183A cells increased the initial, rapid osmotic swelling, attenuating the shrinkage phase, and inhibited the gradual, slower decrease in cell volume, compared to Z183A cells exposed to hypotonic solution without PKC blockers. In addition, the time course of RVD was significantly shortened by 0.1 mM PMA (Fig. 5B). Under the control circumstance, cell volume reached the maximal cell size at 23 6 5 min and returned to the original cell size at 50 6 5 min after exposure to hypotonic stress. In the presence of PMA, Z183A cells reached the maximal cell volume at about 12 6 5 min and the cell volume returned to the steady state at 38 6 5 min. Figures 5C and 5D showed the time course of cell volume changes of normal cervical epithelium, following hypotonic stress in the presence or absence of PKC activator or PKC inhibitors. The RVD processes of normal cervical cells were significantly inhibited by 1 mM staurosporine and 100 mM H7 (Fig. 5C). On the other hand, 0.1 mM PMA accelerated the RVD process (Fig. 5D). The processes of RVD in Z183A cells or normal cervical epithelia were not significantly affected by the addition of
118
M.-R. Shen et al.
FIGURE 5. Time course of volume changes of HPV immortalised cells (Z183A) and normal cervical epithelia following superfusion
with 250 mOsm/L hypotonic solution, in the presence of PKC blockers (H7 and staurosporine) or PKC activator (PMA). The y axis (V/Vo) depicts the ratio of cell volume at time 0 min divided by the cell volume at the indicated times. Each point represents mean 6 SE (n 5 20 cells). PMA: phorbol 12-myristate 13-acetate.
neomycin (0.5 or 1 mM), IBMX, db-cAMP, pBPB, PKI, or erbstatin analogue during hyposmotic challenge. Role of G Protein on RVD Figure 6 showed that when 100 mM GTP-gamma S was added in the hypotonic medium, the RVD process of HT-3 cells significantly accelerated. In addition, the function of RVD was blocked when HT-3 cells were pre-incubated with 100 ng/mL pertussis toxin for 12 h. In contrast, the process of RVD or RVD reversal in Z183A and normal cervical cells was not affected by either GTP-gamma S or pertussis toxin (Fig. 6). These results suggested that in cervical cancer cells the regulation of RVD is associated with pertussis toxin sensitive G protein, while RVD in Z183A and normal cervical cells was not regulated by G protein. DISCUSSION Our data suggest the involvement of different PKC isoforms and the differential activation of a pertussis-toxin sensitive G protein in signalling events of RVD in human cervical epithelial cells with different tumour potential. The active phorbol ester, PMA, which stimulates PKC, accelerates
RVD; while the blockers of PKC, staurosporine and H7, impair the RVD function. PMA can substitute for diacylglycerol (DAG), binding to PKC, and subsequently increasing the affinity of PKC for Ca21 [19]. Therefore, in the presence of PMA, entry of Ca21 to the cell during cell swelling will lead to greater activation of PKC and, subsequently, solute exit pathways. The PKC family is a heterogeneous family of phospholipid-dependent kinase that can be divided into three categories on the basis of cofactor requirements and structure [18, 20]. Conventional PKCs require calcium and DAG or phorbol ester as cofactors. Activated receptors of tyrosine kinase and G-protein coupled classes are able to recruit conventional PKC for intracellular signalling. This is primarily mediated through receptor-induced hydrolysis of phosphatidylinositol biphosphate (PIP2) by PLC, which generates two second messengers, inositol 1,4,5-trisphosphate (IP3) and DAG. DAG activates PKC directly and IP3 indirectly stimulates PKC via its promotion of the release of Ca21 from intracellular storage in the endoplasmic reticulum. Novel PKCs require only DAG or phorbol ester, whereas atypical PKCs do not require calcium or DAG for maximal activity. This report presents the findings that the signalling pathways mediating RVD in different cervical cell types may involve differential activation of different
G-protein Involvement in Cell Volume Regulation
FIGURE 6. Role of G protein on cell volume regulation of cervi-
cal epithelial cells with different tumour potential. The y axis (V/Vo) depicts the ratio of cell volume at time 0 min divided by the cell volume at the indicated times. Each point represents mean 6 SE (n 5 20 cells). Hypo: hypotonic medium; GTPrS: hypotonic medium plus 100 mM GTP-gamma S; Pertussis toxin: cells pre-incubated with 100 ng/mL pertussis toxin before hypotonic exposure. *P , 0.05, compared with the volume ratio in the hypotonic condition at 60 min, paired t-test.
PKC isoforms. In cervical cancer cells, PKC activation through an upstream PLC signalling was involved during the process of RVD. On the other hand, PKCs that are not related to PLC activation were involved in the RVD of HPV-immortalised and normal cervical epithelia. The present study suggests that the PKC isoform involved in the signalling of volume control of cancer cells may be downstream components of phosphatidylinositol (PI)- specific PLC signalling pathways. Thus, it is likely to be conventional PKCs. On the other hand, PKCs involved in the volume regulation of
119
HPV-immortalised and normal cervical cells are favoured as the category of novel PKCs due to the evidence of activation by phorbol ester and the lack of correlation with PIPLC signalling pathways. Further studies should be done to characterise the individual PKCs responsible for the signalling events involved in the process of RVD in each cervical cell type. PKC is reportedly involved in basic cellular functions, including regulation of growth, differentiation, and gene expression [20]. Potent inhibitors of PKC are shown to have cytostatic and chemosensitising properties [21]. Furthermore, evidence showing that overexpression of different PKC isoforms may induce opposite effects on growth, anchorage dependence, and tumorogenicity [22], and that differential localisation of PKCs or different amount of PKCbinding protein and substrate in normal and transformed cells has been noted [23, 24]. These data, together with our present findings that the signalling pathways of RVD in cervical cells with different tumour potential involve differential activation of PKC isoforms, suggest the possible correlation of PKC differential activation with cervical carcinogenesis. However, the mechanism underlying this differential activation remains unknown, and thus the association between the differential signalling events and carcinogenesis needs to be critically examined. The changes in volume-sensitive transport pathways in response to cell volume changes require the following mechanisms [4]: a signal of volume change, a sensor of the signal of the change, transduction of the signal to the effector, i.e., the transporter or channels, and changes in the function of the transporter or channels. These mechanisms are necessary for all transport pathways involved in the regulation of cell volume. The signal transduction pathways during hypotonic stress in cervical cancer cells may be hypothesised as follows. Cell membranes sense the signal of hypotonic shock and degradation of membrane phospholipid happens. IP2 is degraded by PLC into two signalling components: IP3 and DAG. IP3 causes the release of Ca21 from internal stores, while DAG activates PKC. Both may lead to the activation of volume-sensitive Cl2 channels and regulation of cell volume. Furthermore, both PLC and pertussis-toxin sensitive G protein are implicated in the signalling events of RVD, whereas tyrosine kinase inhibitor did not affect that signalling. Although the direct interaction of PLC and pertussis-toxin sensitive G protein cannot be demonstrated in the present study, the above-mentioned data suggest that a pertussis-toxin sensitive G protein may regulate the activation of PLC. Further characterisation of PKC, PLC and G protein should better define the mechanisms of signalling involved in this process. PKC and PLC have variable effects on RVD in other cell types. In rat hepatocytes, swelling causes a transient increase in IP3 [25]. Hypotonicity is shown to stimulate phosphatidylcholine hydrolysis and generate DAG in skate erythrocytes [26]. In proximal tubule cells of Rana temporaria, chloride conductance is activated by PKC-mediated phosphorylation and plays a role in volume regulation [27].
120
However, neither PKC nor PLC pathways are involved in the regulation of volume-sensitive Cl2 channel in human endothelial cells [28]. Altogether, it seems likely that the role of PKCs as a whole and individual PKC in the signalling events involved in the process of RVD may be cell type-specific. In conclusion, PKC activation through an upstream PLC signalling and a pertussis-toxin sensitive G protein were involved during the process of RVD in cervical cancer cells. On the other hand, PKCs that are not regulated by PLC activation were involved, whereas G protein was not implicated in the RVD of HPV-immortalised and normal cervical cells. Further studies are required to elucidate the role of PKCs in cervical carcinogenesis. This study was supported by a grant NSC87-2314-B006-132 from National Science Council, Taiwan. We thank Dr. Tzer-Ming Chen (National Taiwan University, Taipei, Taiwan) for providing the HPV-immortalized cells. We also thank Shu-Hua Cheng for technical assistance.
References 1. Ha¨ussinger D. and Lang F. (1992) Trends Pharmacol. Sci. 13, 371–373. 2. Ha¨ussinger D. (1996) Biochem. J. 313, 697–710. 3. Ha¨ussinger D., Lang F. and Gerok W. (1994) Am. J. Physiol. 267, E343–E355. 4. Hoffmann E. K. and Dunham P. B. (1995) Int. Rev. Cytol. 161, 173–262. 5. Hoffmann E. K. and Simonsen L. O. (1989) Physiol. Rev. 69, 315–373. 6. Rothstein A. and Bear C. (1989) Ann. New York Acad. Sci. 574, 294–308. 7. Hazama A. and Okada Y. (1988) J. Physiol. 402, 687–702. 8. McCarty N. and O’Neil R. G. (1992) Physiol. Rev. 72, 1037– 1061.
M.-R. Shen et al. 9. Shen M. R., Chou C. Y. and Wu S. N. (1995) FASEB J. 6, A356. 10. Chou C. Y., Shen M. R. and Wu S. N. (1995) Cancer Res. 55, 6077–6083. 11. Shen M. R., Wu S. N. and Chou C. Y. (1996) Biochim. Biophys. Acta 1315, 138–144. 12. Scheffner M., Munger K., Byrne J. C. and Howley P. M. (1991) Proc. Natl. Acad. Sci. USA 88, 5523–5527. 13. Pecoraro G., Lee M., Morgan D. and Defendi V. (1991) Am. J. Pathol. 138, 1–8. 14. Suzuki M., Kawahara K., Ogawa A., Morita T., Kawaguchi Y., Kurihara S. and Sakai O. (1990) Am. J. Physiol. 258, F690– F696. 15. Park K. P., Beck J. S., Douglas I. J. and Brown P. D. (1994) J. Membr. Biol. 141, 193–201. 16. Kleuss C., Schultz G. and Wittig B. (1994) Methods Enzymol. 237, 345–355. 17. Ridley A. J. (1995) Methods Enzymol. 256, 313–320. 18. Jaken S. (1996) Curr. Opin. Cell Biol. 8, 168–173. 19. Rasmussen H., Isales C. M., Calle R., Throckmorton D., Anderson M., Gasalla-Herraiz J. and McCarthy R. (1995) Endocr. Rev. 16, 649–681. 20. Nishizuka Y. (1995) FASEB J. 9, 484–496. 21. Powis G. (1992) Trends Pharmacol. Sci. 12, 188–294. 22. Mischak H., Goodnight J., Kolch W., Martiny-Baron G., Schaechtles C., Kazanietz M. G., Blumberg P. M., Pierce J. H. and Mushinski J. F. (1993) J. Biol. Chem. 268, 6090–6096. 23. Hyatt S. L., Liao L., Chapline C. and Jaken S. (1994) Cell Growth Differ. 5, 495–502. 24. Chapline C., Mousseau B., Ramsay K., Duddy S., Li Y., Kiley S. C. and Jaken S. (1996) J. Biol. Chem. 271, 6417–6422. 25. Dahl S. V., Hallbrucker C., Lang F. and Ha¨ussinger D. (1991) Eur. J. Biochem. 198, 73–83. 26. Musch M. W. and Goldstein L. (1990) J. Biol. Chem. 365, 13055–13059. 27. Robson L. and Hunter M. (1994) J. Physiol. 480, 1–7. 28. Nilius B., Oike M., Zahradnik I. and Droogmans G. (1994) J. Gen. Physiol. 103, 787–805.