Introduction

Epidermal growth factor (EGF) and its receptor are key regulatory components of cell growth and differentiation in a variety of cell types1. The EGF receptor tyrosine kinase- mediated signal transduction pathways have been suggested to play the pivotal role under either physiological condition or pathological condition, for example, cancer. Among the multiple signal transduction pathways, EGF induces the intracellular free calcium ([Ca2+]i) increase in several cell types, but the underlying mechanism is so far unresolved. In some situation, the [Ca2+]i response has been considered to be linked to the hydrolysis of polyphosphoinositides by phospholipase Cγ, which is activated as a result of phosphorylation on tyrosine residues by the EGF receptor tyrosine kinase2. As a consequence, the second messengers 1,2-diacylglycerol and inositol-1, 4, 5-trisphosphate are produced. The former activates protein kinase C and the latter releases Ca2+ from internal stores in endoplasmic reticulum3, 4. Besides the induction of internal Ca2+ release, EGF induces the Ca2+ influx through the plasma membrane in some cell types. In these cells, the activation of the Ca2+ channels by EGF stimulation is independent of phospholipase C, but mediated by the activation of phospholipase A25. In the previous studies, we have demonstrated the EGF receptor gene expression and mitogenic effects of EGF in cells of a human hepatoma cell line BEL-74046. We have recently obtained a cell clone, which constitutively expresses the antisense EGF receptor sequence, from BEL-7404 cells7. These cell lines might be useful for studying the signalling of EGF.

In the present study, the effects of EGF on [Ca2+]i were examined in BEL-7404 cells and its derivatives by using the [Ca2+]i measurement in cell populations combined with single-cell [Ca2+]i analysis. The dependency of EGF-induced [Ca2+]i responses on extracellular Ca2+ concentration and intracellular Ca2+ storages was also studied.

Materials and Methods

Materials

Fluo-3/AM, fura-2/AM, thapsigargin were perchased from Sigma (St. Louis, MO, USA). EGF was from Gibco BRL (Gaithersburg, MD, USA). All other chemicals were from commercial sources.

Cell Culture

Human hepatoma cell line BEL-7404 cells and an antisense EGF receptor-expressed cell clone derived from BEL-7404 cells were grown in Dulbecco's modified Eagle's medium (DMEM, Gibco, Grand Island, NY, USA) supplemented with 13 % fetal calf serum, 100 μ/ml of penicillin and 100 μg/ml of streptomycin in 5 % CO2 incubator. After being cultured at 37 °C for 48 h, the cells were kept in the serum-free medium for 24 h before the determinations unless otherwise indicated.

Measurement of intracellular free Ca2+ concentration

For single-cell [Ca2+]i microspectrofluorimetry, cells were plated at a density of 1 × 104 cells/well on the 6-well Nunclon plate inserted with sterile glass coverslips. After the cells were cultured on the condition indicated above, the coverslips were incubated with assay buffer (NaCl 125 mM, KCl 5 mM, CaCl2 1 mM, MgCl2 1 mM, Glucose 10 mM, HEPES 25 mM, pH 7.05 and 0.1% bovine serum albumin) containing fura-2/AM (5 μM) at 37 °C for 45 min. At the end of the incubation, the coverslips were mounted on the recording chamber of a microscope equipped for microfluorimetry as described previovusly8. Fluorescence intensity was obtained with dual excitation wavelengths set at 350 and 380 nm and emission wavelength at 510 nm, respectively [Ca2+ ]i was valued from the 350 nm/380 nm fluorescence ratios.

For measurement of [Ca2+]i in cell populations, the cells were incubated with the assay buffer containing fluo-3/AM (5 μM) at 37°C for 30 min. At the end of the loading, the cells were detached by the treatment with 0.125 % trypsin in Ca2+-free and Mg2+-free Hank's solution for 3 min. After centrifugation at 250 × g for 3 min, the cell pellet was resuspended in 1 ml of the assay buffer. The fluorescence of individual cells was measured using a FACStar Plus flow cytometer (Becton and Dickinson, CA, USA). For each assay, 2,500 cells were examined and the mean value was taken. All the data were recorded and analyzed using specific FACStar research software. The change of [Ca2+]i was expressed as ΔF/F (%), the percentage of actual fluorescence to the control fluorescence level in unstimulated state.

Results

1. EGF induced [Ca2+]i increase in human hepatoma BEL-7404 cells

Measurement of [Ca2+]i in cell populations loaded with fluorescent indicator fluo-3 demonstrated that EGF elicited the increase of fluorescence intensity, which reflected the increase of [Ca2+]i, in a dose-dependent manner, with the maximal dose around 300 ng/ml (Fig 1). Different doses (from 3 ng/ml to 300 ng/ml) of EGF produced different magnitudes but similar kinetics of Ca2+ signals (see Fig 1, inset). Therefore, in the following experiments, the maximal dose (300 ng/ml) of EGF was used unless otherwise indicated. Single-cell [Ca2+]i measurement in fura-2-1oaded cells revealed that EGF (300 ng/ml) induced the transient [Ca2+]i increase in BEL-7404 cells (Fig 2). The peak of [Ca2+]i response was observed around 1 min after EGF application, and the ratio (350 nm/380 nm) increased about 3-fold. Then, the [Ca2+]i level rapidly decreased within 2 min of the stimulation. As shown in Fig 3 a, EGF (300 ng /ml) induced the rapid increase of [Ca2+]i in cell populations, which was initially identical to that obtained in single-cell [Ca2+]i measurement except a slow falling phase following the [Ca2+]i peak.

Fig 1
figure 1

Dose-dependence increase of EGF-induced [Ca2+]i in fluo-3-loaded BEL7404 cells with flow cytometry. [Ca2+]i levels were measured in cell populations after the application of indicated doses of EGF for 1 min. Inset, [Ca2+]i levels were recorded in cell populations after application of different doses of EGF (3, 30 and 300 ng/ml, respectively) for indicated periods.

Full size image
Fig 2
figure 2

EGF-induced [Ca2+]i increase in fura-2-1oaded BEL-7404 cells measured by microspectrofluorimetry with single-cell [Ca2+]i measurement. EGF (300 ng/ml) was added at the point indicated by arrow.

Full size image
Fig 3
figure 3

Effect of extracellular Ca2+ on EGF-induced [Ca2+]i increase in fluo-3-1oaded BEL-7404 cells. (a), [Ca2+]i levels were measured in cell populations after the application of EGF (300 ng/ml) for 1, 2, 5, 10 min respectively, in the presence of 1 mM Ca2+. (b), the cells were preincubated in Ca2+-free assay buffer containing 1 mM EGTA for 3 min, and then [Ca2+]i levels were measured in cell populations after the addition of EGF (300 ng/ml) to the above buffer for indicated periods.

Full size image

2. Dependency of EGF-induced [Ca2+]i increase on extracellular Ca2+ concentration

As shown in Fig 3 b, in Ca2+-free assay buffer, pretreatment of cells with 1 mM EGTA for 3 min partially inhibited the EGF-induced [Ca2+]i increase, especially in the early stage, for example 1 min after the stimulation, the inhibition was about 50% as compared with the [Ca2+]i response in 1 mM Ca2+-containing assay buffer. The results suggested the contribution of Ca2+ influx to the total [Ca2+]i increase was triggered by EGF.

3. The effects of thapsigargin on [Ca2+]i with or without the EGF stimulation

As expected, thapsigargin (TG) at 100 nM concentration did produce the pronounced [Ca2+]i increase in human hepatoma BEL-7404 cells (Fig 4A). In the Ca2+-free assay buffer, TG also induced some [Ca2+]i increase but to a much less extent(Fig 4B). Interestingly, when 1 mM Ca2+ was re-added to the above sample, the [Ca2+]i was rapidly elevated again to the level comparable to the peak of early [Ca2+]i response in the cells incubated with 100 nM TG in 1 mM Ca2+-containing assay buffer, which suggested that Ca2+ influx could occur as a consequence of the TG- induced Ca2+ store depletion.

Fig 4
figure 4

Effects of TG (100 nM) on [Ca2+]i in fluo-3-loaded BEL-7404 cells. [Ca2+]i levels were measured in cell populations. A, TG was added into the assay buffer containing 1 mM Ca2+, and [Ca2+]i levels were recorded after the application of TG for 1, 5, 10, 20 min, respectively. B, Before stimulation, the cells were washed twice with 1 mM EGTA-containing Ca2+-free assay buffer. TG was added into Ca2+-free assay buffer for 10 min, then 1 mM Ca2+ was re-added into the sample at the point indicated by arrows. Measurements of [Ca2+]i were performed at the points as shown in the figure.

Full size image

In order to investigate the contribution of intracellular Ca2+ storage to the EGF-induced [Ca2+]i increase, we did some [Ca2+]i measurements either through the addition of EGF (300 ng/ml) to suspended human hepatoma BEL-7404 cells pretreated with TG (100 nM) for 20 min or through the addition of TG (100 nM) to the cells pretreated with EGF (300 ng/ml) for 10 min. The data showed that the pretreatment with maximal dose of EGF (300 ng/ml) for 10 min exhibited no inhibitory effect on TG-induced [Ca2+]i increase, whereas preincubation with TG (100 nM for 20 min) completely inhibited the EGF-induced [Ca2+]i increase in these cells (Tab 1).

Tab 1 Effects of TG and EGF on [Ca2+]i in human hepatoma BEL-7404 cells*
Full size table

4. Expression of antisense EGF receptor suppressed the EGF-induced [Ca2+]i increase in human hepatoma BEL-7404 cells

Recently, we have obtained a cell clone, JX-1, constitutively expressing the antisense EGF receptor sequence, and a control cell clone, JX-0, transferred with the vector plasmid in BEL-7404 cells7. We found that EGF-induced [Ca2+]i increase in JX-1 cells were obviously inhibited in comparison with those in control JX-0 cells, although JX-0 cells showed a lower amplitude of [Ca2+]i increase than that in BEL-7404 cells for an unknown reason. When JX-1 cells were serum-starved for 24 h, the [Ca2+]i response to EGF stimulation were nearly abolished (Fig 5). Without serum-starvation, the EGF-induced [Ca2+]i increase in JX-1 cells were inhibited by 29.5% to 50.3% in the presence of different doses of EGF (data not shown).

Fig 5
figure 5

Increase of [Ca2+]i induced by EGF in fluo-3-1oaded JX-0 and JX-1 cells. The cells were serum-starved for 24 h before the determinations and [Ca2+]i was recorded after the application of indicated doses of EGF for 1 min.

Full size image

Discussion

The use of fluorescent indicators sensitive to Ca2+ and readily introduced into cells represented a major advance in the measurement of [Ca2+]i. The first of these, quin-2, has been largely superseded by fura-29. Fluo-3, a more recent derivative, has the major advantage of having an excitation spectrum that is visible and thus does not require expensive quartz optics10. In the present study, a combination of single-cell [Ca2+]i analysis with the [Ca2+]i measurement in cell populations was made to resolve the regulation of EGF-induced increase of [Ca2+]i in human hepatoma ceils in vitro. Single-cell [Ca2+]i was recorded continuously in fura-2-1oaded cells by using microflurometry, which made it possible to gauge with a reasonable degree of accuracy of the magnitude and the rapidity of intracellular [Ca2+]i increase during EGF stimulation. The [Ca2+]i was represented as the ratio of emission at 510 nm with excitations at 350 and 380 nm (350 nm/380 nm), which is a monotonic function of [Ca2+]i. On the other hand, the [Ca2+]i determination in cell populations was performed in fluo-3-1oaded cells by using flow cytometry, which had the advantage that measurements were made on large number of single cells, and [Ca2+]i was represented as the changes of the fluo-3 fluorescence intensity. As shown in Figs 1, 2, 3, EGF-induced [Ca2+]i responses in BEL-7404 cells were not only dosedependent, but also time-dependent. The data obtained by the two methods were comparable, although the falling phase of the [Ca2+]i response was more sustained in suspended cells. The explanation for such discrepancy might be that the onset of [Ca2+]i response to EGF stimulation is different in individual cells. Since the [Ca2+]i spikes following the initial transient [Ca2+]i increase were often observed in singlecell [Ca2+]i analysis, an alternative explanation is that the [Ca2+]i spikes in single cells may account for the sustained [Ca2+]i elevation observed in cell suspension.

Although the intracellular [Ca2+]i increase in hepatocytes may be brought about either as a result of mobilizing intracellular Ca2+ or as a result of influx of the ion from extracellular medium11. The data presented in this paper demonstrated that EGF- induced [Ca2+]i increase is dependent on both extracellular Ca2+ concentration and internal Ca2+ storage in endoplasmic reticulum of human hepatoma BEL-7404 cells. Chelating extracellular Ca 2+ by excess EGTA partially inhibited the EGF-induced [Ca2+]i increase, indicating that the Ca2+ influx occurs during EGF stimulation. There are diverse Ca 2+ entry pathways in a variety of vertebrate cells according to the electrophysiological and biochemical criteria12. The nature of the stimulated Ca2+ entry in cultured human hepatoma BEL-7404 cells was also investigated by using Mn2+, an indicator of divalent cation entry13. In Ca2+-free assay buffer containing 50 μM Mn2+, EGF induced the fluorescence quenching in fura-2-1oaded BEL-7404 cells, indicating the occurrence of bivalent cation influx during the stimulation (data not shown). Mn2+ possess three desirable properties. Firstly, it quenches fura-2 fluorescence so that its entry into the cytoplasma is readily detected. Secondly, Mn2+ entry appears to be activated under the same situation as that of Ca2+, suggesting a common pathway. Thirdly, since there is no endogenous agonist-releasable Mn2+ store, a fluorescence quench unambiguously indicates that Mn2+ enters the cytoplasm from outside of the cells and not from an internal store. On the other hand, thapsigargin (TG), a tumor- promoting sesquiterpene lactone, has been proved to be able to discharge intracellular Ca2+ stores in rat hepatocytes by specific inhibition of endoplasmic reticulum Ca2+-ATPase, as it does in many vertebrate cell types14, 15. Preincubation of BEL-7404 cells with 100 nM TG for 20 min, completely abolished EGF-induced [Ca2+]i responses (Tab 1), which together with the observed facts of a rapid formation of inositol-1, 4, 5-trisphosphate in EGF-stimulated BEL-7404 cells (data not shown) and of the expression of antisense EGF receptor sequence in the cells suppressing the [Ca2+]i response to EGF suggested that EGF stimulation linked intracellular Ca2+ pool was covered by TG-sensitive Ca2+ pool and that EGF induced Ca2+ release was the initial response during the stimulation. The heterologous desensitization occurred in EGF-stimulated cells pretreated with TG, but not in TG-treated cells preincubated with EGF, might reflect different mechanisms of Ca2+ release from ER in EGF-treatment and TG-treatment. The former was due to inositol-1, 4, 5-trisphosphate-induced Ca2+ release, and therefore, the depleted Ca2+ pool could be recovered rapidly with the rapid metabolism of inositol-1, 4, 5-trisphosphate and reuptake of Ca2+ by Ca2+-ATPase in surface of ER. The latter was due to the net loss of Ca2+ from ER by the inhibition of ATP driven uptake of Ca2+ from cytosol and thus the exhausted Ca2+ pool could not be refilled rapidly. Besides the induction of Ca2+ discharge from endoplasmic reticulum, TG itself like EGF, triggered the Ca2+ influx in BEL- 7404 cells (Fig 4B), implicating that in either case, the decrease in Ca2+ content of the pool leads to the activation of a plasma membrane Ca2+ channel. Our finding, therefore, support the capacitative model proposed by Putney16 that the depletion of inositol-1, 4, 5-trisphosphate-sensitive Ca2+ pool activates Ca2+ influx from outside of the cells. An alternative possibility is that the increase of [Ca2+]i, due to the discharge from ER in either case, activates the Ca2+-dependent phospholipase A2 and then activates the Ca2+ channel in plasma membrane.

Taken together, EGF receptor-mediated [Ca2+]i increase in the human hepatoma BEL-7404 cells are essentially dependent on the Ca2+ storage in endoplasmic reticulum and the internal Ca2+ release might be followed by the Ca2+ influx. The effects of the cellular Ca2+ movements on either cell growth and apoptosis in these cells are worth studying further.