Remodelling of calcium signalling during liver regeneration in the rat

Journal of Hepatology 46 (2007) 247–256 www.elsevier.com/locate/jhep

Remodelling of calcium signalling during liver regeneration in the rat Alexandra Nicou1, Vale´rie Serrie`re1, Mauricette Hilly1, Sylvie Prigent1, Laurent Combettes1, Gilles Guillon3,4,5,6,7, Thierry Tordjmann1,2,* 1

INSERM U.757, Universite´ Paris Sud, baˆt. 443, 91405 Orsay, France 2 Universite´ Paris Sud, baˆt. 443, 91405 Orsay, France 3 INSERM U661, rue de la cardonille, 34094 Montpellier Cedex 5, France 4 CNRS UMR 5203, rue de la cardonille, 34094 Montpellier Cedex 5, France 5 IGF, rue de la cardonille, 34094 Montpellier Cedex 5, France 6 Universite´ Montpellier I, 34094 Montpellier Cedex 5, France 7 Universite´ Montpellier II, 34094 Montpellier Cedex 5, France

Background/Aims: During liver regeneration, a network of cytokines and growth factors interact with hepatocytes, helping to restore the liver mass and functions after partial tissue loss. Agonists that trigger Ca2+ signals in the liver contribute to this process, although little is known about calcium signalling during liver regeneration. Results: We observed two phases in which the hepatocyte response to calcium-mobilising agonists was greatly reduced versus control cells at 24 h and five days after partial hepatectomy. We found that both phases of hepatocyte desensitisation involved the down-regulation of cell surface receptors and the type II InsP3 receptor. Single cell studies with flash photolysis of caged InsP3 revealed that InsP3-mediated Ca2+ release was slower in regenerating hepatocytes at 24, 48 h and 5 days than in control cells. Also, the temporal pattern of vasopressin-elicited intracellular calcium oscillations studied on fura2-loaded cells was altered, with the duration of each Ca2+ peak being longer. Finally, we showed an association between hepatocyte desensitisation and progression through the cell cycle towards the S phase at 24 h after hepatectomy. Conclusions: Our study supports the remodelling of hepatocyte calcium signalling during liver regeneration, and that this change is partly linked with cell cycle progression.  2006 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. Keywords: Hepatocyte; Calcium oscillations; Receptors; Cell cycle

1. Introduction

Received 27 March 2006; received in revised form 6 July 2006; accepted 9 August 2006; available online 9 October 2006 * Corresponding author. Tel.: +33 1 69 15 70 72; fax: +33 1 69 15 58 93. E-mail addresses: [email protected] (A. Nicou), valerielanneau@ free.fr (V. Serrie`re), [email protected] (M. Hilly), [email protected] (S. Prigent), laurent.combettes@ibaic. u-psud.fr (L. Combettes), [email protected] (G. Guillon), [email protected] (T. Tordjmann). Abbreviations: AVP, arginine vasopressine; InsP3, D-myo-inositol 1,4,5-trisphosphate; [Ca2+]i, cytosolic free Ca2+ concentration; PP, periportal; PV, perivenous; HO-LVA, phenylacetyl1-D-Tyr (Mc)2Phe3-Gln4-Asn5-Arg6-Pro7-Arg8-NH2 (HO-LVA).

Liver regeneration is regulated by cytokines, growth factors, hormones and neurotransmitters that interact with hepatocytes to help restore liver mass and function within days after partial tissue loss [1]. A combination of paracrine, autocrine and endocrine interactions push quiescent liver cells towards the cell cycle, as well as maintaining normal differentiated hepatic functions in the regenerating tissue. Among the regulatory signalling pathways, a number of calcium-mobilising agonists including noradrenaline [2], arginine vasopressin (AVP) [3,4] and ATP [5], epidermal growth factor (EGF) and hepatocyte growth factor (HGF) [6] contribute to liver regeneration. Despite the large number of liver

0168-8278/$32.00  2006 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.jhep.2006.08.014

248

A. Nicou et al. / Journal of Hepatology 46 (2007) 247–256

regeneration studies, only a few have studied calcium signalling after partial hepatectomy. Calcium signalling regulates many cellular processes throughout the life of an organism, including secretion, metabolism and differentiation [7]. Intracellular calcium is also crucial for cell proliferation [8], including during liver regeneration, as suggested by early studies on hypocalcaemic rats [9]. Alterations in the calcium signalling machinery have been reported to occur during liver regeneration, although the data remain limited and conflicting [10–12]. Other reports have both directly and indirectly suggested that agonist-induced calcium signals are modified after partial hepatectomy in the rat [13,14]. In a recent study [4], we found that AVP increased hepatocyte entry into the cell cycle during liver regeneration. After a two-thirds hepatectomy, the expression and lobular distribution of the V1a AVP receptor were altered within the first 24 h, modifying the [Ca2+]i oscillation kinetics (delays and frequencies) of isolated hepatocytes stimulated with AVP. These alterations in AVP-elicited calcium signals were associated with an enhanced choleretic response to AVP. We observed two phases of reduced hepatocyte calcium response to AVP at 24 h and on the fifth day after partial hepatectomy, although the mechanism and functional implication of this remained unknown [4]. Alterations in the expression of agonist receptors in regenerating hepatocytes have been reported previously [2,15–19], although correlations between receptor expression, hepatocyte responsiveness and liver regeneration outcome have been poorly studied. In the present work, we investigated the different elements of the calcium signalling pathway involved in hepatocyte desensitisation after partial hepatectomy and found correlations between desensitisation and cell cycle progression. We found that the hepatocyte desensitisation occurring at 24 h and five days after partial hepatectomy involved both the downregulation of cell surface receptors and the modification of InsP3 receptor subtype expression. We also provide evidence linking hepatocyte desensitisation and progression through the cell cycle towards the S phase, during the first desensitisation phase. Taken together, our study suggests that hepatocyte calcium signalling is remodelled during liver regeneration, and that this is at least partly linked with cell cycle progression.

2. Materials and methods

2.2. Isolation of cells Hepatocytes were isolated as previously described [20,21]. Cell viability assessed by trypan blue exclusion was greater than 96%.

2.3. Determination of [Ca2+]i changes in hepatocytes 2.3.1. [Ca2+]i imaging with dye-loaded hepatocytes Freshly isolated hepatocytes were loaded with fura2 (Molecular Probes, US) and Ca2+ imaging experiments were done after one hour of plating, on an Axiovert 35 epifluorescence microscope (Zeiss), as previously described [20,21].

2.3.2. Spectrofluorimetry Calcium movements in hepatocyte populations were measured after loading with quin2 or fura2 as previously described 2+ [21]. Hepatocyte Ca responses remained similar during the 8 h after cell isolation.

2.3.3. Flash photolysis of microinjected caged InsP3 Inactive photolabile ‘‘caged’’ InsP3, the P-4 or P-5 1-(2-nitrophenyl) ethyl ester of InsP3 [20], and fluo3 (Molecular Probes, US) were microinjected together into cells using an Eppendorf microinjector (5242). The cells were then allowed to recover for at least 10 min. InsP3 was released from caged InsP3 by photolysis with a 1-ms pulse from a short arc xenon flashlamp producing a 2–3 mm diameter image across at the cell as described previously [22]. The subsequent changes in fluo3 fluorescence (kex = 480 nm, kem = 520 nm) due to intracellular calcium mobilisation were recorded [20].

2.4. [125I]HO-LVA binding on hepatocyte membranes Crude membranes were obtained from hepatocyte populations as previously described [20,21]. [125I]HO-LVA (2000 Ci/mmol) was prepared as previously described [23]. This iodinated AVP analogue labelled rat liver V1a AVP receptor with a high affinity (Kd = 8 pM). The binding protocol used was previously described [24].

2.5. Immunoblotting The V1a receptor, the type I and type II InsP3 receptors were resolved by electrophoresis and were semi-quantified by immunoblotting. We used both total microsomal and plasma membrane-enriched fractions for type I and type II InsP3 receptors, as previously described [25]. We compared subcellular fractions between sham and hepatectomised rats by: (1) measurement of fraction density on a percoll gradient; (2) 5 0 -nucleotidase assay for indicating plasma membrane enrichment; (3) measurement of SERCA2b expression as an endoplasmic reticulum marker. These parameters were found to be similar in sham and hepatectomised rats (data not shown). The anti-V1a receptor antibody (1/1200) was raised in rabbits against the third intracellular loop (Tran et al., unpublished data). Rabbit polyclonal antibodies directed against actin (1/800) were from Sigma. Type I InsP3 receptor antibody (1/1000) was from Affinity BioReagents and type II (1/1000) InsP3 receptor antibody was from Covalab (Lyon, France) and have been previously characterised [25]. Peroxidase-conjugated secondary antibodies were from Pierce Immunological Laboratories, and chemi-luminescence kit from Amersham Biosciences. Signal intensity was semi-quantified by densitometry.

2.1. Surgical procedure and treatments 2.6. Histology and immunohistochemistry Animal experiments were carried out according to the CEE directives for animal experimentation (decree 2001-131; ‘J.O.’ 06/02/01). Two-thirds hepatectomy was performed on adult female Wistar rats (200–250 g) (CERJ, France) [4]. At various times after surgery, liver fragments were removed, frozen in nitrogen-cooled isopentane, and stored at 80C until use.

Eight-micrometer-thick liver cryostat sections were air-dried and fixed in acetone for two minutes at room temperature. Sections were incubated at 4C for 12 h with the rabbit polyclonal antibody against the type II InsP3 receptor (Covalab, 1/300) and the rat monoclonal antibody against ZO1 (clone R40-76, gift from B. Stevensen, Alberta

A. Nicou et al. / Journal of Hepatology 46 (2007) 247–256

249

University, Edmonton, Canada). After rinsing in PBS, the sections were incubated at 37C for 30 min with the corresponding secondary antibodies conjugated to Alexa fluor 578–603 nm dye for ZO1, or the 495–519 nm dye for the type II InsP3 receptor (Molecular Probes, US). Fluorescence intensity was quantified (Image J software) in periportal (PP) and perivenous (PV) regions of interest, and PV/PP ratios were calculated for sham and hepatectomised rats. Bromodeoxyuridine (BrdU) immunohistochemistry was performed on frozen liver sections, as previously described [4].

tomy, could be due to over-stimulation by agonists, including AVP [4], endotoxin and cytokines [27], or nitric oxide [28], all of which increase in concentration in the liver after hepatectomy [29–31].

2.7. Statistical analysis

Given that autocrine, paracrine and endocrine pathways are involved in orchestrating liver regeneration, alterations in surface receptor expression or coupling are likely to be crucial for compensatory growth [1]. We have previously shown by RT-PCR and autohistoradiography that V1a receptors are downregulated at 24 h after hepatectomy but not at 5 days [4]. In the present work, Western blotting and binding studies revealed a significant reduction in V1a receptor protein levels and in the number of binding sites both at 24 h and 5 days after hepatectomy (Figs. 2A and B). At 5 days after partial hepatectomy, the differences between our autohistoradiography data [4] and our present data suggest that the V1a receptor is located in a subcellular compartment that is not present in the membrane preparation used for the Western blotting and binding studies. Experiments carried out 48 h after partial hepatectomy (a time point between the two desensitisation phases, when cellular sensitivity had recovered, see Fig. 1A) revealed a similar V1a expression in hepatocytes as those from sham-operated rats (Figs. 2A and B). Other experiments showed that G protein (aq/a11) expression levels were similar between control hepatocytes and those after 24 h and 5 days regeneration (data not shown). Thus, it initially appears that at 24 h and 5 days after partial hepatectomy, the surface receptors of regenerating hepatocytes were desensitised to AVP at least. As already stated, alterations to the endotoxinaemia [27,29], paracrine

We used Student’s t-test to compare sample means with paired controls. Results are expressed as means ± SEM. P values of 60.05 were considered to be statistically significant.

3. Results and discussion 3.1. Biphasic desensitisation of calcium responses in regenerating hepatocytes As previously reported [4], hepatocyte sensitivity to AVP (in terms of calcium responses) was markedly altered after a two-thirds hepatectomy, particularly 24 h after surgery and, to a lesser extent, on the fifth day (Fig. 1A). We observed desensitisation as well for noradrenaline (Fig. 1A), angiotensin II (Fig. 1B) and ATP (Fig. 1C). The same results were observed at both high (1.8 mM) and low (100 nM) extracellular calcium concentrations (not shown), suggesting that calcium mobilization from internal stores (rather than calcium entry from the extracellular medium) was primarily altered after partial hepatectomy. Early studies had indirectly suggested that calcium-mediated hepatocyte responses during liver regeneration were reduced in the rat, although the mechanisms involved were unclear [13,17]. A loss in the sensitivity of hepatocytes to calcium mobilising hormones, at least at 24 h after hepatec-

A

B

C

1200 Sham (8)

60

**** 40

Hx Day 5 (4)

Sham (8) Hx H24 (4) Hx Day 5 (4)

1000

800

[Ca 2+ ]i (nM)

****

1400 1200

Hx H24 (4)

1000

80 [Ca 2+ ]i (nM)

[Ca 2+ ] i increase in Hx (% of sham)

100

3.2. V1a receptor is downregulated during hepatocyte desensitisation after hepatectomy

600

800 600

400

400

200

200

20 0 0

24 48 96 120 144 Time after hepatectomy (hours)

Hx (AVP 3 nM)

Hx (NOR 1 μM)

0

0 0

0.1 0.3 1 3 Angiotensin II (nM)

10

0

0.1

0.3 1 ATP (μM)

3

10

Fig. 1. Biphasic hepatocyte desensitisation during liver regeneration. (A) Time-course of hepatocyte sensitivity to AVP and noradrenaline. Increases in [Ca2+]i in response to AVP (3 nM) and noradrenaline (1 lM) in regenerating hepatocytes were compared (%) to those in sham rats and plotted against time after hepatectomy. ****p < 0.001. (B and C) Hepatocyte Ca2+ responses to angiotensin II (B) and ATP (C) in sham-operated and hepatectomised (HX) rats at 24 h and 5 days after surgery. [Ca2+]i (nM) was studied in quin2-loaded hepatocyte suspensions by spectrofluorimetry before and after agonist stimulation. Each data point in the dose–response curves is the mean ± SEM of at least 4 independent experiments.

250

A. Nicou et al. / Journal of Hepatology 46 (2007) 247–256 H48

H24

A

Hx Sham

Sham

15 μg

30 μg

15 μg

Day 5 30 μg

Sham Hx Sham

Hx

Hx

15 μg

30 μg

Sham Hx Sham

V1a receptor

Hx

Relative intensity (% of sham)

52 kD 140 120 100 80

****

****

60 40 20 0 H24

H48

Day 5

Bmax (fmol/mg prot)

B 180 150

***

120

***

90 60 30 0 SHAM (4)

H24 (4)

H48 (4)

Day 5 (4)

Fig. 2. Biphasic AVP V1a receptor downregulation during liver regeneration. (A) Western blot analysis of V1a receptor protein levels in hepatocyte membranes from sham-operated and hepatectomised rats, 24 and 48 h, and 5 days after surgery. The photograph is representative of four experiments. A semi-quantitative analysis of Western blots is shown. The relative levels of the V1a receptor were determined by scanning densitometry (n = 4 experiments in each group). ****p < 0.001. (B) Specific [125I]HO-LVA (V1a)-binding sites in hepatocyte membranes from sham-operated and hepatectomised rats, 24 and 48 h and 5 days after surgery. Crude membranes (2.4 lg per assay) from hepatocytes were incubated for 1 h at 37 C in the presence (non-specific binding) or absence (total binding) of 1 lM unlabelled AVP, with increasing amounts of [125I]HO-LVA (Free). Specific binding (Bound), calculated as the difference between total and non-specific binding, was determined as described by Grazzini et al. [24]. ***0.001 < p < 0.01.

and endocrine environment, as well as remodelling of the extracellular matrix [32], may repress surface receptor expression after partial hepatectomy. Thus, the consequences to signalling due to V1a receptor downregulation are likely to be part of a global hepatocyte adaptation to the changing microenvironment after partial hepatectomy. 3.3. InsP3 receptor is downregulated during hepatocyte desensitisation after hepatectomy We further investigated the calcium signalling pathway by studying InsP3 receptor expression in regenerating hepatocytes. It has been previously shown that rat hepatocytes contain (protein and mRNA expression) type II (81%) and type I (19%) InsP3 receptor isoforms, with type III isoforms being absent or very weakly expressed [33,34]. We observed a 50–60% reduction in the level of type II isoform protein at 24 h and 5 days regenerating versus control livers (Fig. 3A), whereas

we observed no significant change in type I isoform protein levels at these time points (Fig. 3B). The reduction in type II InsP3 receptor expression was observed in both plasma membrane-enriched and total microsomal liver fractions (Fig. 3A). Experiments carried out 48 h after hepatectomy (when the hepatocyte sensitivity had recovered) showed significantly reduced type I and type II InsP3 receptor expression in hepatectomised rats versus sham-operated rats (Fig. 3A), suggesting that after 48 h the recovered surface (V1a) receptor expression was enough to restore hepatocyte sensitivity (see Figs. 1 and 2). Conflicting results on the InsP3 receptor expression modification after partial hepatectomy have been reported [10,11], although Magnino et al. [12] reported a type II InsP3 receptor downregulation associated with an up-regulation of type I expression. However, this affected only mRNA levels, with protein expression being only slightly modified [12]. As already stated for cell desensitisation, hepatocyte over-stimulation by agonists and the resulting over-production of

A. Nicou et al. / Journal of Hepatology 46 (2007) 247–256

251

Fig. 3. InsP3 receptor downregulation during liver regeneration. (A and B) Western blot analysis of type II (40 lg protein) (A) and type I (B) InsP3 receptor protein levels in liver plasma membrane-enriched and total microsomal fractions [26] from sham-operated and hepatectomised rats, 24, 48 h and 5 days after surgery. Photographs are representative of four experiments on total microsomal fractions. Semi-quantitative analysis of Western blots are shown. The relative levels of the InsP3 receptor were determined by scanning densitometry and normalised on the actin expression level in the same samples (n = 4 experiments in each group). **p = 0.01; ns: non-significant.

InsP3 may be involved in the downregulation of InsP3 receptor, as reported in cell lines [35–37] and ovocytes [38]. Immunohistochemistry on liver sections from shamoperated and hepatectomised rats at 24 h (and 48 h, not shown) and 5–8 days after surgery revealed a similar InsP3 receptor intracellular location in the livers from all rat groups (n = 3) (Fig. 4), that is, at the canalicular pole, as previously reported [39]. This was supported by the co-labelling of the tight junction-associated protein ZO1, and suggested that the reduction in the level

of the type II InsP3 receptor was not due to its intracellular localisation being modified. Thus, desensitised regenerating hepatocytes (at 24 h and 5 days) appeared to be altered at two important checkpoints of calcium signalling. First, InsP3 production is reduced by hormone receptor downregulation [20,21] and, second, InsP3-mediated calcium release may be altered due to downregulation of the type II InsP3 receptor isoform. The relative proportions of the type I and type II InsP3 receptor isoforms in a cell are thought to determine the kinetics of the calcium signals [40]. At 24 h and five days

252

A. Nicou et al. / Journal of Hepatology 46 (2007) 247–256

Fig. 4. Type II InsP3 receptor intracellular distribution during liver regeneration. (A) After sham surgery, type II InsP3 receptor immunostaining was concentrated at the apical (canalicular) pole, as demonstrated by the co-labelling of the tight junction protein, ZO1. Twenty-four hours (B) and 5 days (C) after hepatectomy (Hx), type II InsP3 receptor immunostaining was still observed at the canalicular pole. Representative photomicrographs of four experiments at each time point. Scale bars = 10 lm.

after partial hepatectomy, when cells appear to be desensitised to calcium-mobilising agonists, the relative abundance (as compared with the total InsP3R population) of the type II isoform would be reduced (40% instead of 81%), whereas that of the type I isoform would be increased (60% instead of 19%). Interestingly, because both isoforms were similarly reduced (Fig. 3), these proportions would not be significantly altered at 48 hours, a time point when cell sensitivity to agonists appeared unaffected (Fig. 1). Alternatively, conserved hepatocyte sensitivity to calcium-mobilising agonists at 48 h may also indicate that surface receptor expression predominantly determines cellular responsiveness. 3.4. Hepatocyte sensitivity to InsP3 after partial hepatectomy We then checked whether the observed downregulation of InsP3 receptor expression was correlated with a reduction in InsP3-mediated calcium release. Saponintreated hepatocyte suspensions loaded with 45Ca2+, analyzed as described previously [21], revealed that the apparent sensitivity of the InsP3 receptor to InsP3,

InsP3-sensitive calcium pools and ionomycin-sensitive calcium pools were not significantly different in regenerating (either 24 or 5 days) and control hepatocytes (data not shown). Previous studies on the relationship between InsP3 receptor density and calcium release suggested that only large differences in receptor density cause any observable leftward shift in the dose-response relationship [41]. Thus, any differences in InsP3-mediated calcium release would be seen only at very low InsP3 concentrations and on short timescale. Differences in the kinetics of InsP3-elicited calcium release between single regenerating and control cells were investigated using flash photolytic release of encaged InsP3 in fluo3-loaded hepatocytes, to analyse the rise-time of the calcium responses, which has been previously described to correlate with InsP3 receptor density [42]. We found that regenerating hepatocytes at 24 h and 5 days had a significantly slower InsP3-elicited [Ca2+]i rise-time than cells from sham-operated rats (Figs. 5A and B). We observed similar results 48 h after hepatectomy, suggesting that surface receptor recovery might compensate for InsP3 receptor loss, allowing hepatocytes to be fully sensitive to agonists (Fig. 5). Thus,

A. Nicou et al. / Journal of Hepatology 46 (2007) 247–256

253

B 400

A 100

80

300

sham

40

H24

t1/2 (ms)

250

60

(%)

fluorescence increase

350

H48

20

200 150 100

D5

50

0 0

250

500

750

0

1000 1250 1500

Sham (44)

time (ms) C

Hx-H24 Hx-H48 (12) (34)

Hx-D5 (37)

D

control Hx-H24

40

2+

ΔF/Fo=20%

Ca peak duration (s)

AVP 0.1 nM

30

20

10

0 0

50

100

150 time,s

200

250

Sham (4) Hx-H24 (4)

Fig. 5. Intracellular calcium release and [Ca2+]i oscillations in hepatocytes are altered during liver regeneration. (A) Photolytic release of InsP3 from caged InsP3 was carried out in fluo3-loaded hepatocytes from sham-operated (d) and hepatectomised rats 24 h (s), 48 h (m) and 5 days (h) after surgery, as explained in Section 2. Ca2+-associated changes in fluorescence were recorded at 1 image/ 6 120 ms, and expressed as a percentage of normalised fluorescence intensity. Flashlamp intensity was 100 mV. The plots of InsP3-mediated [Ca2+]i increase are representative of those obtained in three independent experiments in each rat group. (B) The rate of change of [Ca2+]i ([Ca2+]i rise time), calculated (t1/2, half of the time to reach maximum amplitude) and taken as a measure of InsP3-mediated Ca2+ flux, as previously reported [42], was significantly longer in regenerating hepatocytes at 24 h (0.295 ± 90 s, n = 12), 48 h (0.275 ± 0.023 s, n = 34) and 5 days (0.203 ± 0.022 s, n = 37) than in control cells (0.137 ± 0.015 s, n = 44) in three independent experiments (p < 0.05). (C) Videomicroscopic analysis of [Ca2+]i oscillations in response to AVP (0.1 nM) in fura2-loaded hepatocytes from sham-operated and HX rats 24 h after surgery. We analysed 42 cells in five sham-operated and 37 cells in four hepatectomised rats. For each cell, the oscillation pattern was recorded after stimulation with AVP (0.1 nM). 25 control and 19 regenerating hepatocytes responded to 0.1 nM AVP. Among the HX responding cells, 42% (8 hepatocytes) exhibited [Ca2+]i peaks considerably larger than other cells. Plots representative of the [Ca2+]i time course under AVP (0.1 nM) in sham-operated (full lane) and HX (dotted lane) rats are shown. Frame-rate: 1 image/3 s. For technical reasons, traces began 30 s after addition of the agonist. (D) Semiquantitative analysis of [Ca2+]i peak duration, measured at half amplitude in hepatocytes from sham-operated (25 cells) and HX (19 cells) rats, 24 h after surgery.

the downregulation of the type II InsP3 receptor and the related changes in the proportion of the different receptor isoforms may cause subtle alterations in the kinetics of Ca2+ release during liver regeneration. This may have an impact on the initiation and propagation of intracellular Ca2+ oscillations and waves [43]. 3.5. AVP-mediated [Ca2+]i oscillations are altered during liver regeneration Previous studies on cells expressing only one InsP3 receptor isoform reported that each receptor type can

generate a particular pattern of calcium signals, with the type II isoform generating the more robust [Ca2+]i oscillations [40,44]. Thus, in the light of the alteration of InsP3 receptor isoform proportions observed during liver regeneration (see above), we investigated the temporal pattern of calcium oscillations in fura2-loaded regenerating hepatocytes. We found that AVP-elicited [Ca2+]i oscillations were significantly different in regenerating hepatocytes at 24 h than in control cells (Fig. 5C). In 30–40% of the hepatocytes from livers 24 h after partial hepatectomy, the oscillation frequency was reduced (Fig. 5C),

254

A. Nicou et al. / Journal of Hepatology 46 (2007) 247–256

as previously shown [4], the duration of the [Ca2+]i peak was significantly larger (Fig. 5D), and the oscillations were less regular (not shown). We obtained similar but less marked results at day 5 after partial hepatectomy (data not shown). The downregulation of the surface receptors and the resulting fall in InsP3 production [20,21] would be involved in the slowing down of [Ca2+]i oscillations, although observed changes in Ca2+ peak regularity and duration may be due to changes in InsP3 receptor expression [40,44]. Although calcium-dependent processes in hepatocytes are not completely defined, these alterations in [Ca2+]i oscillations may affect bile secretion regulation [4,21,45], hepatocyte metabolism [46,47] and, potentially, cell cycle progression through transcription factor activation [48] and gene expression [7]. Thus, the changes to hepatocyte calcium signalling may be physiologically relevant at a critical time point, when the hepatocytes have to simultaneously enter the cell cycle and maintain their differentiated functions.

A

3.6. Is there any correlation between cell cycle progression and hepatocyte desensitisation? During liver regeneration after two-thirds hepatectomy, periportal (PP) hepatocytes enter the cell cycle about 24 h before perivenous (PV) cells [49]. Accordingly, we found predominant BrdU incorporation in the PP area at 24 h after hepatectomy (Fig. 6A). InsP3 receptor immunohistochemistry revealed a predominant labelling of the type II InsP3 receptor in the PV area in control rats (PV/PP = 1.82 ± 0.2, n = 4) (Figs. 6B and D). This lobular gradient was strongly accentuated in regenerating livers at 24 h due to a strong loss of expression in PP areas (PV/PP = 6.99 ± 1.1, n = 3) (Figs. 6C and D), and returned to values close to those observed in sham rats 5 and 8 days after hepatectomy (Fig. 6D). We have previously shown a decrease in V1a receptor expression predominantly in PP hepatocytes [4] 24 h after partial hepatectomy, suggesting these hepatocytes were primarily exposed to desensitising signals that may have included endotoxin, cytokines and NO [28–32]. Thus, both the

B portal

CV

D

C

9 8

CV

PV/PP InsP3RII labelling

portal

7 6 5 4 3 2 1 0 SHAM (4)

HX-H24 (3)

HX-H48 (3)

HX-D5 (3)

HX-D8 (3)

Fig. 6. Intralobular distribution of type II InsP3 receptor is altered during liver regeneration. (A) Lobular distribution of BrdU incorporation at 24 h after hepatectomy. Liver samples were harvested 2 h after intraperitoneal BrdU injection, and processed for BrdU immunodetection as described in Section 2. S phase hepatocytes were predominantly located in periportal (PP) areas. The photomicrograph is representative of four experiments. Original magnification: 100·. (B) After sham surgery, type II InsP3 receptor immunostaining (see Fig. 4 for intracellular localisation) was distributed all along hepatocyte plates, with a slight perivenous (PV) concentration. Twenty-four hours (C) after hepatectomy (Hx), type II InsP3 receptor immunostaining was strongly concentrated in the PV region, around the central vein (CV), and was almost undetectable in the PP region around the portal vein branches (‘‘portal’’). (D) Semi-quantitative analysis of the type II InsP3 receptor distribution (PV/PP ratios) in sham-operated and hepatectomised rats. Photomicrographs were analysed with Image J software by determining the fluorescence intensity in PP and PV regions of interest and by calculating PV/PP ratios for sham-operated and hepatectomised rats. Representative photomicrographs of four experiments at each time point. Original magnification: 160·. Scale bars = 100 lm.

A. Nicou et al. / Journal of Hepatology 46 (2007) 247–256

255

Fig. 7. Schematic representation of the remodelling of calcium signalling during the course of liver regeneration. After partial hepatectomy, surface receptors (V1a vasopressin receptors) are downregulated at 24 h and 5 days, whereas they recover at 48 hours. The type I InsP3 receptor (InsP3RI) is downregulated at 48 h, although its level of expression does not change at 24 h and 5 days. The type II InsP3 (InsP3RII) receptor is downregulated at 24, 48 h and at 5 days. Resulting alterations in agonist-induced calcium oscillations are depicted schematically at each time point. S: S phase; M: M (mitosis) phase.

type II InsP3 receptor and the V1a receptor were lost primarily in the PP region at 24 h, an area in which most hepatocytes are in the S phase at this time. This chronological coincidence was reinforced by preliminary data on S phase-synchronised cell lines that showed desensitised calcium responses to AVP or ATP (not shown). Thus, hepatocyte desensitisation (and the related modification of calcium signalling) and the progression in S phase appeared to be closely associated during the early stages of liver regeneration. Although calcium signals have been reported to be necessary during G1 to S phase progression [8], it is not clear if changes in hepatocyte sensitivity to calcium-mobilising agonists help or are secondary to cell proliferation. A similar chronological association between calcium signalling patterns, particularly the occurrence of [Ca2+]i oscillations, and the progression through the different phases of the cell cycle has been reported in the ovocyte experimental model [8,38]. Taken together, our data suggest that during liver regeneration, calcium signalling undergoes a complex remodelling, the two major targets of which are the surface agonist receptor and the intracellular type II InsP3 receptor (Fig. 7). The resulting alterations in [Ca2+]i oscillation kinetics may be important for hepatocyte physiology, and may, in particular, be related to hepatocyte entry into the S phase. Acknowledgements This work was supported by the Association pour la Recherche contre le Cancer (ARC, No. 3435). Vale´rie Serrie`re was supported by the Fondation pour la Recherche Me´dicale (FRM). We thank Alex Edelman and Associates for their help editing the manuscript, and Jean-Pierre Mauger for fruitful discussion of the data. We also thank Sylviane Boucherie, Josiane Simon, Nicole Stelly and Laura Lagoudakis for excellent technical assistance.

References [1] Taub R. Liver regeneration: from myth to mechanism. Nat Rev Mol Cell Biol 2004;5:836–847. [2] Cruise JL, Muga SJ, Lee YS, Michalopoulos GK. Regulation of hepatocyte growth: alpha-1 adrenergic receptor and ras p21 changes in liver regeneration. J Cell Physiol 1989;140: 195–201. [3] Russell WE, Bucher NLR. Vasopressin modulates liver regeneration in the Brattleboro rat. Am J Physiol 1983;245:G321–G324. [4] Nicou A, Serriere V, Prigent S, Boucherie S, Combettes L, Guillon G, Alonso G, et al. Hypothalamic vasopressin release and hepatocyte Ca2+ signaling during liver regeneration: an interplay stimulating liver growth and bile flow. FASEB J 2003;17: 1901–1903. [5] Thevananther S, Sun H, Li D, Arjunan V, Awad SS, Wyllie S, Zimmerman TL, et al. Extracellular ATP activates c-jun Nterminal kinase signaling and cell cycle progression in hepatocytes. Hepatology 2004;39:393–402. [6] Mine T, Kojima I, Ogata E, Nakamura T. Comparison of effects of HGF and EGF on cellular calcium in rat hepatocytes. Biochem Biophys Res Commun 1991;181:1173–1180. [7] Berridge MJ, Bootman MD, Roderick HL. Calcium signalling: dynamics, homeostasis and remodelling. Nat Rev Mol Cell Biol 2003;4:517–529. [8] Kahl C, Means A. Regulation of cell cycle progression by calcium/calmodulin-dependent pathways. Endocr Rev 2003;24: 719–736. [9] Whitfield JF, Boynton AL, MacManus JP, Rixon RH, Sikorska M, Tsang B, Walker PR, Swierenga SH. The roles of calcium and cyclic AMP in cell proliferation. Ann NY Acad Sci 1980;339: 216–240. [10] Feng L, Kraus-Friedmann N. Changes in 1,4,5-inositol trisphosphate binding following partial hepatectomy. Biochem Biophys Res Commun 1994;205:291–297. [11] Diaz-Munoz M, Canedo-Merino R, Gutierrez-Salinas J, Hernandez-Munoz R. Modifications of intracellular calcium release channels and calcium mobilization following 70% hepatectomy. Arch Biochem Biophys 1998;349:105–112. [12] Magnino F, St-Pierre M, Luthi M, Hilly M, Mauger JP, Dufour JF. Expression of intracellular calcium channels and pumps after partial hepatectomy in rat. Mol Cell Biol Res Commun 2000;3:374–379. [13] Huerta-Bahena J, Villalobos-Molina R, Corvera S, Garcia-Sainz JA. Sensitivity of liver cells formed after partial hepatectomy to

256

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

A. Nicou et al. / Journal of Hepatology 46 (2007) 247–256 glucagon, vasopressin and angiotensin II. Biochim Biophys Acta 1983;763:120–124. Kitamura T, Watanabe S, Ikejima K, et al. Different features of Ca2+ oscillations in differentiated and undifferentiated hepatocyte doublets. Hepatology 1995;21:1395–1404. Leffert H, Alexander NM, Faloona G, Rubalcava B, Unger R. Specific endocrine and hormonal receptor changes associated with liver regeneration in adult rats. Proc Natl Acad Sci USA 1975;72:4033–4036. Earp HS, O’Keefe EJ. Epidermal growth factor receptor number decreases during rat liver regeneration. J Clin Invest 1981;67: 1580–1583. Huerta-Bahena J, Villalobos-Molina R, Garcia-Sainz JA. Roles of alpha 1- and beta-adrenergic receptors in adrenergic responsiveness of liver cells formed after partial hepatectomy. Biochim Biophys Acta 1983;763:112–119. Marti U, Hug M. Acinar and cellular distribution and mRNA expression of the epidermal growth factor receptor are changed during liver regeneration. J Hepatol 1995;23:318–327. Chari RS, Price DT, Sue SR, Meyers WC, Jirtle RL. Down-regulation of transforming growth factor beta receptor type I, II, and III during liver regeneration. Am J Surg 1995;169: 126–131. Tordjmann T, Berthon B, Jacquemin E, Clair C, Stelly N, Guillon G, Claret M, et al. Receptor-oriented intercellular calcium waves evoked by vasopressin in rat hepatocytes. EMBO J 1998;17: 4695–4703. Serrie`re V, Berthon B, Boucherie S, Jacquemin E, Guillon G, Claret M, Tordjmann T. Vasopressin receptor distribution in the liver controls calcium wave propagation and bile flow. FASEB J 2001;15:1484–1486. Walker JW, Feeney J, Trentham DR. Photolabile precursors of inositol phosphates. Preparation and properties of 1(2-nitrophenyl)ethyl esters of myo-inositol 1,4,5-trisphosphate. Biochemistry 1989;28:3272–3280. Barberis C, Balestre MN, Jard S, Tribollet E, Arsenijevic Y, Dreifuss JJ, et al. Characterization of a novel, linear radioiodinated vasopressin antagonist: an excellent radioligand for vasopressin V1a receptors. Neuroendocrinology 1995;62:135–146. Grazzini E, Loedboerer AM, Perez-Martin A, Joubert D, Guillon G. Molecular and functional characterization of V1b vasopressin receptor in rat adrenal medulla. Endocrinology 1996;137: 3906–3914. Picard L, Ibarrondo J, Coquil JF, Hilly M, Mauger JP. Ligandbinding affinity of the type 1 and 2 inositol 1,4,5-trisphosphate receptors: effect of the membrane environment. Biochem Pharmacol 2000;59:131–139. Lie`vremont JP, Hill AM, Hilly M, Mauger JP. The inositol 1,4,5trisphosphate receptor is localized on specialized sub-regions of the endoplasmic reticulum in rat liver. Biochem J 1994;300: 419–427. Bucher M, Hobbhahn J, Taeger K, Kurtz A. Cytokine-mediated downregulation of vasopressin V(1A) receptors during acute endotoxemia in rats. Am J Physiol Regul Integr Comp Physiol 2002;282:R979–R984. Patel S, Gaspers LD, Boucherie S, Memin E, Stellato KA, Guillon G, Combettes L, et al. Inducible nitric-oxide synthase attenuates vasopressin-dependent Ca2+ signaling in rat hepatocytes. J Biol Chem 2002;277:33776–33782. Cornell RP, Liljequist BL, Bartizal KF. Depressed liver regeneration after partial hepatectomy of germ-free, athymic and lipopolysaccharide-resistant mice. Hepatology 1990;11:916–922. Hortelano S, Dewez B, Genaro AM, Diaz-Guerra MJ, Bosca L. Nitric oxide is released in regenerating liver after partial hepatectomy. Hepatology 1995;21:776–786.

[31] Trautwein C, Rakerman T, Niehof M, Rose-John S et Manns P. Acute-phase response factor, increased binding, and target gene transcription during liver regeneration. Gastroenterology 1996;110:1854–1862. [32] Serandour AL, Loyer P, Garnier D, Courselaud B, Theret N, Glaise D, Guguen-Guillouzo C, Corlu A. TNFalpha-mediated extracellular matrix remodeling is required for multiple division cycles in rat hepatocytes. Hepatology 2005;41:478–486. [33] De Smedt H, Missiaen L, Parys JB, Bootman MD, Mertens L, Van Den Bosch L, Casteels R. Determination of relative amounts of inositol trisphosphate receptor mRNA isoforms by ratio polymerase chain reaction. J Biol Chem 1994;269:21691–21698. [34] Wojcikiewicz RJ. Type I, II, and III inositol 1,4,5-trisphosphate receptors are unequally susceptible to down-regulation and are expressed in markedly different proportions in different cell types. J Biol Chem 1995;270:11678–11683. [35] Sipma H, Deelman L, Smedt HD, Missiaen L, Parys JB, Vanlingen S, Henning RH, et al. Agonist-induced down-regulation of type 1 and type 3 inositol 1,4,5-trisphosphate receptors in A7r5 and DDT1 MF-2 smooth muscle cells. Cell Calcium 1998;23:11–21. [36] Bokkala S, Joseph SK. Angiotensin II-induced down-regulation of inositol trisphosphate receptors in WB rat liver epithelial cells. Evidence for involvement of the proteasome pathway. J Biol Chem 1997;272:12454–12461. [37] Zhu CC, Furuichi T, Mikoshiba K, Wojcikiewicz RJ. Inositol 1,4,5-trisphosphate receptor down-regulation is activated directly by inositol 1,4,5-trisphosphate binding. Studies with bindingdefective mutant receptors. J Biol Chem 1999;274:3476–3484. [38] Jellerette T, Kurokawa M, Lee B, Malcuit C, Yoon SY, Smyth J, Vermassen E, et al. Cell cycle-coupled [Ca2+]i oscillations in mouse zygotes and function of the inositol 1,4,5-trisphosphate receptor-1. Dev Biol 2004;274:94–109. [39] Hirata K, Pusl T, O’Neill AF, Dranoff JA, Nathanson MH. The type II inositol 1,4,5-trisphosphate receptor can trigger Ca2+ waves in rat hepatocytes. Gastroenterology 2002;122:1088–1100. [40] Dupont G, Combettes L. Modelling the effect of specific inositol 1,4,5-triphosphate receptor isoforms on cellular Ca2+ signals. Biol Cell 2006;98:171–182. [41] Joseph SK, Rice HL, Nicchitta CV. Characteristics of GTPmediated microsomal Ca2+ release. Biochim Biophys Acta 1988;945:185–194. [42] Ogden D, Capiod T. Regulation of Ca2+ release by InsP3 in single guinea pig hepatocytes and rat Purkinje neurons. J Gen Physiol 1997;109:741–756. [43] Tovey SC, de Smet P, Lipp P, Thomas D, Young KW, Missiaen L, De Smedt H, et al. Calcium puffs are generic InsP(3)-activated elementary calcium signals and are downregulated by prolonged hormonal stimulation to inhibit cellular calcium responses. J Cell Sci 2001;114:3979–3989. [44] Miyakawa T, Maeda A, Yamazawa T, Hirose K, Kurosaki T, Iino M. Encoding of Ca2+ signals by differential expression of IP3 receptor subtypes. EMBO J 1999;18:1303–1308. [45] Pusl T, Nathanson MH. The role of inositol 1,4,5-trisphosphate receptors in the regulation of bile secretion in health and disease. Biochem Biophys Res Commun 2004;322:1318–1325. [46] Kraus-Friedmann N, Feng L. The role of intracellular Ca2+ in the regulation of gluconeogenesis. Metabolism 1996;45:389–403. [47] De Koninck P, Schulman H. Sensitivity of CaM kinase II to the frequency of Ca2+ oscillations. Science 1998;279:227–230. [48] See V, Rajala NK, Spiller DG, White MR. Calcium-dependent regulation of the cell cycle via a novel MAPK-NF-kappaB pathway in Swiss 3T3 cells. J Cell Biol 2004;166:661–672. [49] Rabes HM. Kinetics of hepatocellular proliferation after partial resection of the liver. Prog Liver Dis 1976;5:83–99.