Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The structurally unrelated immunosuppressants cyclosporine A (CsA) and tacrolimus (FK506) have both been proposed as positive modulators of the compensatory growth process after partial hepatectomy (PHx) in the rat, mouse, and dog (Garcia-Alonso et al., 1989; Francavilla et al., 1990; Kahn et al., 1990; Mazzaferro et al., 1990; Tanaka et al., 1993). However, CsA has been shown to lead to either cell proliferation, cell cycle arrest, or cell death in different cell types or organs, but its mechanisms of action have not yet been fully elucidated and the dual potential of CsA on cell cycle progression remains a paradox.

Present literature suggests that the apparent stimulatory effect of CsA on normal compensatory growth seems to be largely confined to the liver. Indeed, Francavilla et al. (1990) in a study on hepatocytes, kidney, and intestine were unable to identify any influence of the drug on the hyperplastic response of either the contralateral kidney after unilateral nephrectomy, or on the small intestine after partial resection. However, despite numerous reports indicating an increase in the hepatic regeneration phenomenon induced by CsA, several studies have noted that the liver weight restitution was not influenced by the drug at 3 or 7 days after surgery (Makowka et al., 1986; Kahn et al., 1990). It has also been suggested that CsA might have a "priming"-like effect on the liver because in the unhepatectomized rat, parameters indicative of putative regeneration processes such as increases in the hepatic content of ornithine decarboxylase, thymidine kinase, and the estrogen receptor have been observed (Kahn et al., 1990). Moreover, a study carried out in our laboratory has indicated that at the critical time of greatest loss in liver mass, CsA has only a selective influence on the biotransformation of cytochrome P450-dependent activities (Provencher et al., 1999). This observation indicates that the effect of CsA on the regeneration process does not translate into an overall accelerated recovery of the hepatic drug-metabolizing function.

It has now been well documented that hypocalcemia and vitamin D deficiency are accompanied by modifications in calcium signaling in hepatocytes such as impairment in inositol-1,4,5-trisphosphate (IP₃) synthesis, IP₃ receptor affinity, and in the responsiveness of several Ca²⁺-mobilizing hormones and growth factors, which seem related to low receptor affinity and/or to the size of the hormone/drug-sensitive intracellular Ca²⁺ pools (Gascon-Barré et al., 1997; Mailhot et al., 2000). Hypocalcemia has also been shown to retard the hepatic regenerative process (Éthier et al., 1990). These observations have raised the hypothesis that the defective regeneration process associated with hypocalcemia is due to defect(s) in the hepatocyte mitogenic signaling pathways (Sikorska et al., 1983; Rixon et al., 1989; Éthier et al., 1993; Goupil et al., 1997). Moreover, in several liver diseases where patients are candidates for liver transplantation, malabsorption of vitamin D and of calcium is suspected as being a major determinant of vitamin D depletion and of its accompanying poor systemic and cellular calcium metabolism. This contention is well illustrated by the increased incidence of secondary hyperparathyroidism and of bone diseases (osteomalacia as well as osteoporosis) in these patients (Krawitt et al., 1977; Kato et al., 1982; Mawer et al., 1985).

The goals of the present studies were to investigate the 1) role of calcium status on the positive effect of CsA on liver regeneration in vivo by investigating whether CsA could reverse the delay in the regeneration process induced by calcium deficiency, and 2) effect of CsA on intracellular Ca²⁺ homeostasis in isolated rat hepatocytes in vitro to further understand its mechanism of action.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

In Vivo Studies

Animal Model and Drug Regimen. Studies were conducted in normal rats (Charles River, St. Constant, QC, Canada) and in rats subjected to vitamin D depletion to induce a state of chronic hypocalcemia as described previously (Éthier et al., 1990). In a subset of hypocalcemic animals, oral calcium supplementation was achieved by providing a 3% calcium gluconate solution as drinking water to normalize circulating Ca²⁺ (Mailhot et al., 2000) and to investigate the interaction between CsA (Novartis, Montreal, QC, Canada) and the in vivo extracellular Ca²⁺ concentration. The rationale for the model used rests on previous reports showing that 1) hepatic regeneration is perturbed in vitamin D-depleted hypocalcemic animals but completely restored by either vitamin D₃ or 1,25-dihydroxyvitamin D₃ (calcitriol) administration (Éthier et al., 1990); 2) oral calcium repletion alone in vitamin D-depleted hypocalcemic rats fully restores DNA synthesis after 2/3 partial hepatectomy (Éthier et al., 1990); and 3) chronic hypocalcemia decreases the IP₃-sensitive Ca²⁺ pool content and attenuates cellular Ca²⁺ signaling by known Ca²⁺-mobilizing agents, a phenomenon also restored by vitamin D₃ or calcium repletion (Gascon-Barré et al., 1997; Mailhot et al., 2000).

Biochemical Analyses. Serum 25-hydroxyvitamin D₃ [25(OH)D₃] and calcitriol concentrations were measured using the 25(OH)D₃ and 1,25-dihydroxyvitamin D₃ assay kits (Incstar Corporation, Stillwater, MN) according to the manufacturer's instructions. Serum Ca²⁺ concentrations were measured with an ICA2 ionized calcium analyzer (Radiometer, Copenhagen, Denmark).

Parameters Indicative of Regeneration Process

[³H]Thymidine Incorporation into DNA and Liver Weight Restitution. At various time points after surgery (14-46 h), animals received a single i.v. injection of 50 µCi of [methyl-³H]thymidine (specific activity 70-90 Ci/mmol; ICN Canada Ltd, Mississauga, ON, Canada). Two hours later, animals were killed by exsanguination under anesthesia. The liver was then immediately perfused in situ with ice-cold saline, excised, and padded dry before being weighed. The liver was homogenized (10% w/v) and used for the determination of DNA and [³H]thymidine incorporation by measuring the ³H present in the perchloric acid (Laboratoire Mat, Beauport, QC, Canada) precipitate of 1 ml of liver homogenate as described previously (Éthier et al., 1990).

In Vitro Studies. To investigate the influence of CsA on the hepatic compensatory growth process, hepatocytes were obtained from livers of nonfasting normal rats, as described previously (Gascon-Barré et al., 1997). The freshly isolated hepatocytes were equilibrated in William E medium (Invitrogen, Burlington, ON, Canada), and cells were allowed to attach on collagen-coated coverslips (Fisher Scientific, Montreal, QC, Canada) during a period of 1 h at 37°C in 5% CO₂ atmosphere. Before calcium measurements, hepatocytes were exposed to 0, 0.1, 1.0, or 10 µg/ml CsA for a period of 30 min in medium containing Ca²⁺ concentrations similar to those observed in normocalcemia in vivo (1.25 mM Ca²⁺) or in Ca²⁺-free medium.

Intracellular Calcium Measurements

Cytoplasmic Ca²⁺ Measurements at Single Cell Level. Hepatocytes were loaded with 2 µM Fura-2/AM (Molecular Probes, Eugene, OR) for 30 min at 20°C. Dye-loaded cells were then transferred to a 10-µl plastic chamber on the stage of an inverted microscope (Nikon Diaphot; Nikon, Tokyo, Japan) equipped for epifluorescence measurement. All test compounds were superfused at a rate of 3 ml/min in calcium-free Krebs-Henseleit solution, pH 7.4, equilibrated with O₂/CO₂ (95:5, v/v). Intracellular calcium ([Ca²⁺]_i) measurements were made at the single cell level using an MCID dual excitation spectrofluorometer system (Imaging Research, St. Catherines, ON, Canada) and a refrigerated camera (C4880; Hamamatsu Photonics, Hamamatsu City, Japan), as described previously (Gascon-Barré et al., 1997; Mailhot et al., 2000).

IP₃-Sensitive Ca²⁺ Pools. IP₃-sensitive cellular pools were evaluated as described previously (Hofer et al., 1998). Briefly, Ca²⁺ content was monitored using 5 µg of mag-Fura-2/AM (Molecular Probes) as probe. After mag-Fura-2/AM loading (40 min at 20°C), hepatocytes were exposed to 100 µg of saponin/ml for a period of 1 min in a calcium-free buffer containing 125 mM KCl, 25 mM NaCl, 10 mM HEPES, 0.1 mM MgCl₂, and 1 mM ATP, pH 7.3, at 37°C to achieve plasma membrane permeabilization. Ca²⁺ mobilization from internal pools was achieved by applying 10 µg of IP₃ in the buffer described above but in the absence of saponin. Fluorescence signals were obtained at the single cell level using the imaging system described above to measure cytoplasmic Ca²⁺ concentrations. Data are presented as the signal ratios obtained at 340 and 380 nm.

Statistical Analysis. Results are expressed as means ± S.E.M. The qualitative evaluation of the time at which peak [³H]thymidine incorporation into DNA occurred was done using a cubic spline equation. Statistically significant differences between group means were evaluated by analysis of variance, or by the Student's t test as indicated in the table and figure legends.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of CsA on Parameters of Vitamin D and Calcium Metabolism

As illustrated in Fig. 1, CsA significantly increased circulating calcitriol concentrations in normal rats at all time points during CsA exposure, whereas it significantly decreased that of its precursor 25(OH)D₃, particularly after 3 days of CsA exposure.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Effect of CsA on the serum concentrations of calcitriol and 25(OH)D3 in normal rats. CsA was administered at a dose 3.33 mg/kg/day for a period of 12 days; n = 6 rats/time point in each group. Data are presented as mean ± S.E.M. The gray zone indicates the range of normal values for calcitriol. Normal values for the 25(OH)D3 range from 35 to 120 µM. 25(OH)D3 and calcitriol concentrations were not influenced by CsA administration (data not shown). Statistically significant differences between group means were analyzed by analysis of variance. The effect of CsA on calcitriol concentrations was p < 0.0004 and on 25(OH)D3 concentrations was p < 0.002. Individual contrasts relative to the zero time point were evaluated by the Student's t test using the Bonferroni/Dunn correction. Significantly different from control values: *, p < 0.04; **, p < 0.009; ***, p < 0.002.

Vitamin D depletion significantly decreased both 25(OH)D₃ and calcitriol concentrations. CsA administration, however, did not significantly influence the circulating concentrations of 25(OH)D₃ or calcitriol in vitamin D-depleted rats with serum calcitriol concentrations of 119 ± 3 pmol/l before CsA administration, and 134 ± 28 and 165 ± 35 pmol/l after 9 and 12 days of CsA exposure. Serum 25(OH)D₃ concentrations were similarly unaffected by CsA with circulating concentrations of 5.3 ± 0.4 and 4.6 ± 0.3 µM before and after 12 days of CsA administration. Vitamin D depletion, however, led to frank hypocalcemia (0.86 ± 0.08 mM), which was fully normalized by oral calcium feeding (1.27 ± 0.03 compared with 1.3 ± 0.03 mM in normal rats, N.S.).

At the time of euthanasia, no signs of toxicity or intolerance (such as peritonitis) were observed in the CsA-treated animals. CsA administration did not significantly affect the circulating concentrations of aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, and ionized calcium. Body weight and liver weight were also unaffected by CsA pretreatment.

Effect of CsA on Regeneration Process

Figure 2A presents the effect of CsA on [³H]thymidine incorporation into hepatic DNA in normal rats. As illustrated, CsA administration did not affect [³H]thymidine incorporation into DNA as illustrated in sham-operated animals. After 2/3 partial hepatectomy, CsA administration accelerated [³H]thymidine incorporation into DNA. This apparent acceleration in DNA synthesis was entirely due to the magnitude of the response at the 20-h time point as indicated by the highly significant difference between the CsA- and placebo-treated groups (p < 0.003) (Fig. 2A). The acceleration effect of CsA on the regeneration process was also qualitatively evaluated by cubic spline analysis. It was estimated that peak DNA synthesis occurred 20.3 h after liver resection in CsA-treated compared with 22.9 in control rats. Liver weight restitution was found to be similar in both groups during the time frame studied (Fig. 2B).

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Effect of CsA on parameters indicative of the hepatic regeneration process in normal rats. A, hepatic [3H]thymidine incorporation into cellular DNA in sham-operated and 2/3 partially hepatectomized rats. PHx: CsA, - - - -; PHx: placebo, ; sham-operated: CsA,  · · · ; sham-operated: placebo,  · · · . CsA was administered for 10 days before PHx or sham operation. Placebo-treated rats received olive oil; n = 3 (sham) to 13 animals (PHx)/group. Statistically significant differences between control and CsA-treated rats were analyzed by two-way analysis of variance, p < 0.05. Pairwise comparisons between group means at each time point were evaluated by the Student's t test using the Bonferroni/Dunn correction; **, p < 0.003. B, [3H]thymidine incorporation patterns were also evaluated by cubic spline equations with peak DNA synthesis estimated to occur at 20.3 h after HPx in CsA-treated compared with 22.9 in control rats. Liver weight restitution after 2/3 partial hepatectomy: CsA-treated  and placebo-treated . Statistically significant differences between group means at each time point after PHx were analyzed by the Student's t test, N.S.

As expected, hypocalcemia (Fig. 3A) significantly impaired the regeneration process, an impairment not restored to the level of normal placebo-controls by CsA pretreatment. Moreover, CsA administration did not accelerate [³H]thymidine incorporation into DNA nor did it lead to a significant or persistent effect on the magnitude of the response as judged by analysis of variance (Fig. 3A) as well as by the evaluation of the cumulative [³H]thymidine incorporation into DNA over the entire 48 h studied.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Effect of CsA on parameters indicative of the hepatic regeneration process in vitamin D-depleted rats. A and B, hypocalcemic vitamin D-depleted rats; n = 4-7 rats/group except at the 28-h time point where n = 2/group. C and D, normocalcemic vitamin D-depleted rats; n = 4-7 rats/group. A and C, hepatic [3H]thymidine incorporation into cellular DNA. PHx: CsA, - - - -; PHx: placebo, ; sham-operated: CsA,  · · · ; sham-operated: placebo,  · · · . Statistically significant differences between control and CsA-treated rats were analyzed by two-way analysis of variance; hypocalcemic, p < 0.05, and normocalcemic, N.S. Pairwise comparisons between group means at each time point were evaluated by Student's t test using the Bonferroni/Dunn correction, hypocalcemic: *, p < 0.05. B and D, liver weight restitution after PHx in the hypocalcemic (B) and normocalcemic (D) animals. , CsA-treated; , placebo-treated. Statistically significant differences between group means at each time point after PHx were analyzed by the Student's t test, N.S.

Normalization of the circulating calcium concentration alone in vitamin D-depleted rats accelerated DNA synthesis compared with their hypocalcemic counterparts. Calcium supplementation, however, completely abrogated the stimulatory effect of CsA treatment on the pattern and magnitude of the DNA synthesis response (Fig. 3C). CsA had no significant effect on liver weight restitution compared with placebo-treated rats in either hypo- or normocalcemic vitamin D-depleted rats (Fig. 3, B and D, respectively).

Effect of CsA Exposure on Intracellular Ca²⁺

Cytoplasmic Ca²⁺ Concentrations. Figure 4 presents the resting basal cytoplasmic calcium concentrations ([Ca²⁺]_i) observed in normal rat hepatocytes exposed to CsA in normal extracellular Ca²⁺ concentrations ([Ca²⁺]_e) (1.25 mM) (Fig. 4A) as well as in Ca²⁺-free medium (Fig. 4B). As illustrated, in both cases, CsA led to a dose-dependent increase in basal resting cytoplasmic Ca²⁺ concentrations, which reached statistical significance at the 1- and 10-µg/ml doses of CsA. However, at the 10-µg/ml dose of CsA, the mean cytoplasmic Ca²⁺ concentration was found to be higher in calcium-free medium than in normal extracellular Ca²⁺ concentration, suggesting a translocation of Ca²⁺ from intracellular pools to the cytoplasmic compartment (p < 0.0001).

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Influence of CsA exposure on the resting cytoplasmic Ca2+ concentrations in single rat hepatocytes isolated from normal rat livers. Hepatocytes were exposed to CsA in either the presence of 1.25 mM (A) or in the absence (B) of extracellular Ca2+ for a period of 30 min before calcium measurements were made. Calcium measurements were made using the Ca2+-sensitive Fura-2/AM fluoroprobe. Data represents the mean ± S.E.M. Between 12 and 14 rats were used for each experimental condition with an average of 175 hepatocytes/condition. Significant differences between group means were evaluated by analysis of variance using the Bonferroni/Dunn test for all post hoc evaluations. Main effect: p < 0.0001; *, p < 0.005; **, p < 0.0004; ***, p < 0.0001.

View larger version (35K): [in this window] [in a new window]   Fig. 5.   Representative traces illustrating the influence of CsA pretreatment on the response to calcium-mobilizing agonists in single rat hepatocytes isolated from normal rat livers. Hepatocytes were exposed to CsA in the presence of 1.25 mM (A-C) or in the absence (D-F) extracellular Ca2+ for a period of 30 min before intracellular Ca2+ measurements were made using the Fura-2/AM fluoroprobe. , indicate application; , withdrawal of each agonist.

View larger version (25K): [in this window] [in a new window]   Fig. 6.   Influence of CsA exposure on the response to calcium-mobilizing agonists in single rat hepatocytes isolated from normal rat livers and exposed in vitro to CsA in the presence of either 1.25 mM (A-C) or in the absence (D-F) extracellular Ca2+ for a period of 30 min before cellular Ca2+ measurements were made using the Fura-2/AM fluoroprobe. Cytoplasmic Ca2+ determinations were made in two to six rats per group with an average of 64 hepatocytes evaluated per group. Data represent the mean ± S.E.M. Significant differences between group means were evaluated by analysis of variance using the Bonferroni/Dunn test for all post hoc evaluations in relation to the 0 CsA dose. Phenylephrine was applied at a dose 2.5 mM, vasopressin at a dose of 10 nM, and EGF at a dose of 25 ng/ml. *, p < 0.05; ** p < 0.005; ***, p < 0.0001.

IP₃-Sensitive Ca²⁺ Pools. Evaluation of the IP₃-sensitive Ca²⁺ pools revealed that extracellular calcium did not influence the basal Ca²⁺ content of the IP₃-sensitive pools. CsA, however, increased the basal Ca²⁺ content of the IP₃-sensitive pools in the presence of extracellular calcium (Figs. 7A and 8A, a) but not in the absence of extracellular calcium (Figs. 7B and 8A, b). This observation is consistent with the observed lower agonist-stimulated Ca²⁺ mobilization presented in Figs. 5 and 6. The IP₃-mobilizable Ca²⁺ was shown to be increased at the 0.1-, 1.0-, and 10-µg/ml doses of CsA (Figs. 7A and 8B, a) in the presence of extracellular Ca²⁺, and at the 1.0- and 10-µg/ml doses of CsA (Figs. 7B and 8B, b) in the absence of extracellular Ca²⁺. Upon withdrawal of IP₃ and application of 1 mM Ca²⁺, an expected increase in the cellular Ca²⁺ pools was observed in both groups (Fig. 7, A and B), illustrating the adequacy of the experimental preparations.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Representative traces illustrating the influence of CsA exposure on the IP3-sensitive Ca2+ pools in single rat hepatocytes isolated from normal rat livers. Hepatocytes were exposed to CsA in the presence of 1.25 mM (A) or in the absence (B) of extracellular Ca2+ for a period of 30 min before cellular Ca2+ measurements were made using the mag-Fura-2/AM fluoroprobe. , indicates application; , withdrawal of IP3; , indicates application; , withdrawal of calcium.

View larger version (32K): [in this window] [in a new window]   Fig. 8.   Influence of CsA exposure on the IP3-sensitive Ca2+ pools in single rat hepatocytes isolated from normal rat livers and exposed in vitro to CsA in the presence of 1.25 mM Ca2+ (A) or in the absence of extracellular Ca2+ (B) for a period of 30 min before cellular Ca2+ measurements were made using the mag-Fura-2/AM fluoroprobe. Intracellular Ca2+ measurements were made in three to five rats per group with an average of 82 hepatocytes per group in the presence of 1.25 mM Ca2+ and in two to three rats/group with an average of 40 hepatocytes per group in the absence of extracellular calcium. Data represents the mean ± S.E.M. Significant differences between group means were evaluated by analysis of variance using the Bonferroni/Dunn test for all post hoc evaluations in relation to the 0 CsA dose. A, basal [Ca2+]i in the presence of 1.25 mM [Ca2+]e, p < 0.0001, and in calcium-free medium, N.S. B, IP3-induced increases in [Ca2+]i: main effect in the presence of 1.25 mM [Ca2+]e, p < 0.0001, and in calcium-free medium, p < 0.0001. *, p < 0.05; **, p < 0.008; ***, p < 0.0001.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Our data indicate that, in normal rats, CsA increased hepatic DNA synthesis after 2/3 partial hepatectomy. The most striking difference between CsA- and placebo-treated animals was, not only, the intensity of the response but also an acceleration and a sharpening in peak DNA synthesis. This observation clearly suggests that the drug mediated an increase in the progression through the G₁/S phase(s) of the cell cycle and reinforce an earlier observation where CsA was shown to intensify the regeneration gradient (Garcia-Alonso et al., 1990). This phenomenon was not, however, observed in hypocalcemic animals in which a defective regeneration process and hyporesponsiveness to EGF have been shown to occur. Surprisingly, vitamin D-depleted rats fed a high-calcium diet were not responsive to CsA, despite a complete reversal of the EGF hyporesponsiveness by calcium feeding (Éthier et al., 1990; Bilodeau et al., 1995), and the reestablishment of normal DNA synthesis as shown in the present studies. These observations illustrate not only that calcium is an important mediator of DNA synthesis in the regeneration model associated with partial liver resection but also that CsA cannot superimpose its stimulatory effect on the calcium effect.

Although CsA has been shown to influence calcium metabolism in several ways, the mechanisms by which the ion mediates the action of CsA on compensatory hepatic hyperplasia is not known. In the present studies, CsA has been shown to influence calcium homeostasis in vivo by mediating an early and sustained increase in the concentrations of the calcium-regulating hormone calcitriol as well as by influencing the hepatocyte resting and stimulated cytoplasmic Ca²⁺ responses and the size of the mobilizable IP₃-sensitive Ca²⁺ pools in vitro. Other investigators have also reported that CsA plays a role in calcium signaling as illustrated by an increase in the overall Ca²⁺ content in hepatocytes (Nicchitta et al., 1985), a reduction in the level of the calcitriol-dependent 28-kDa calcium binding protein despite increases in calcitriol synthesis (Stein et al., 1991; Steiner et al., 1996), a potentiating effect of known Ca²⁺ agonists such as vasopressin and phenylephrine on the cellular Ca²⁺ response in kidney mesengial cells (Skorecki et al., 1992) and C6 glioma cells (Ikesue et al., 2000), an enhancement at the IP₃ receptor (without increases in IP₃ production) of agonist-induced cellular Ca²⁺ mobilization in kidney cells (Gordjani et al., 2000), and an induction of a hepatic hypermetabolic state associated with increases in [Ca²⁺]_i mobilization and prostaglandin G₂ synthesis in Kupffer cells (Zhong et al., 2001). In addition, CsA has been shown to increase the synthesis and release of potent calcium agonists such as endothelin and norepinephrine and to augment their reactivity in vivo (Moss et al., 1985; Scherrer et al., 1990; Bunchman and Brookshire, 1991; Takeda et al., 1992).

Whether CsA acts as a hepatomodulator of compensatory growth via a calcium-mediated mechanism is still uncertain but Ca²⁺ has been demonstrated to be essential for the normal cell cycle progression (Lu and Means, 1993), whereas several Ca²⁺-mobilizing agents have been reported to positively influence hepatic compensatory growth. For example, norepinephrine has not only been shown to increase [Ca²⁺]_i mobilization but also to stimulate the hepatic regeneration process via an ₁-adrenergic receptor-mediated signaling pathway and as such to act as a comitogen in hepatocytes (Cruise et al., 1985; Michalopoulos and DeFrances, 1997). In addition, several growth factors (EGF/transforming growth factor , hepatocyte growth factor, acidic fibroblast growth factor, which are considered complete liver mitogens and known cellular Ca²⁺ mobilizers) are essential for the hepatic compensatory growth process, but their role in the CsA-mediated effect on DNA synthesis in the regenerating liver has not been investigated (Michalopoulos and DeFrances, 1997). Mazzaferro et al. (1990) have suggested that CsA seems to have an effect similar to that of insulin, a permissive but incomplete hepatic mitogen, an observation analogous to that proposed for ₁-adrenergic agents. The data obtained during the present studies are entirely in agreement with the latter suggestion because they unequivocally show that CsA does not have any effect on hepatic DNA synthesis in animals not subjected to partial hepatectomy, hence clearly showing that CsA is not a complete hepatic mitogen.

To investigate the potential role of changes in cellular Ca²⁺ homeostasis in the CsA-induced effect on the early hepatic regeneration process, an investigation of the IP₃-sensitive Ca²⁺ pools and of the response to Ca²⁺-mobilizing agonists known to be either hepatocytic mitogen (EGF) or comitogens (vasopressin and phenylephrine) during the compensatory growth process was undertaken. Data show that CsA induced an extracellular Ca²⁺-dependent increase in the IP₃-sensitive Ca²⁺ pools. In addition, IP₃-mediated Ca²⁺ mobilization was significantly increased at all CsA doses in the presence of extracellular Ca²⁺, whereas IP₃ induced significant Ca²⁺ responses at the 1- and 10-µg/ml does of CsA in the absence of extracellular Ca²⁺. Interestingly, we have previously shown that repletion of hypocalcemic rats with oral calcium also increased the hepatocyte content of the hormone-sensitive Ca²⁺ pool as well as that of calreticulin (Mailhot et al., 2000), a protein known to regulate the functional size of IP₃-sensitive Ca²⁺ pools (Bastianutto et al., 1995; Mery et al., 1996). It is, therefore, possible that the acceleration in DNA synthesis observed in calcium-supplemented animals may be due to the cellular calcium effect induced by the nutritional repletion, an effect that cannot be further increased by CsA exposure.

Our data also indicate that CsA significantly modifies the response to Ca²⁺-mobilizing agonists and that the presence of extracellular Ca²⁺ is essential for the response to vasopressin and EGF, whereas a blunted response to phenylephrine (54% less than in the presence of [Ca²⁺]_e) was still observed at the highest CsA dose in the absence of extracellular Ca²⁺. This observation is in accordance with that of others where extracellular Ca²⁺ has been shown to be essential to the CsA-mediated rise in intracellular Ca²⁺. Because it is known that hepatic mitogens such as EGF/TGF and HGF and comitogens such as vasopressin and ₁-adrenergic agonists act at the G₀-G₁ and/or at the early G₁ phase of the cell cycle, a positive CsA-mediated effect on these agents would be expected to translate into an acceleration of the G₁ to S phase of the cell cycle and hence by an earlier peak in DNA synthesis. Our observations indicating that DNA synthesis was accelerated by CsA exposure in normal rats in vivo would concur with the in vitro data, indicating a significant CsA-mediated increased resting as well as in stimulated intracellular Ca²⁺ mobilization, most particularly in the presence of normal extracellular Ca²⁺. Of interest also is the observation that CsA led to a significantly sustained increase in the circulating calcitriol concentrations in normal animals. This increase in serum calcitriol may also have played a role in the intensity of the response to CsA because the hormone has been shown to increase intracellular Ca²⁺ concentration as well as to accelerate DNA synthesis and liver weight restitution after partial hepatectomy in the rat (Éthier et al., 1990; Mailhot et al., 2000).

Despite the apparent acceleration effect of CsA on DNA synthesis, liver weight restitution was shown not to be influenced by the drug in any of the groups studied at the early time points studied. These observations indicate that the early recovery of liver mass is not significantly influenced by CsA, an observation also made by others where liver weight restitution was studied for longer periods of time (Grant et al., 1988; Kahn et al., 1990; Kapan et al., 1996). The absence of a sustained effect of CsA on liver mass restitution indicates, contrary to earlier reports, little effect of the drug on the organ recovery, an observation also supported by our previous observations showing that CsA has only a selective effect on the drug metabolism recovery after 2/3 partial hepatectomy (Provencher et al., 1999). Our data, thus, indicate that the extracellular calcium status either directly or indirectly via modulation of intracellular homeostasis plays an essential role as mediator of the cellular calcium response after partial hepatectomy. The data are consistent with CsA accelerating the hepatic compensatory growth process by, in part, increasing the hepatocyte Ca²⁺ pools and the response to growth factors and Ca²⁺-mobilizing hormones known to be comitogens for hepatocytes.