Phosphatidylcholine/sphingomyelin metabolism crosstalk inside the ...

Biochem. J. (2008) 410, 381–389 (Printed in Great Britain)

381

doi:10.1042/BJ20070758

Phosphatidylcholine/sphingomyelin metabolism crosstalk inside the nucleus Elisabetta ALBI1 , Remo LAZZARINI and Mariapia VIOLA MAGNI Department of Clinical and Experimental Medicine, Physiopathology, Policlinico Monteluce, University of Perugia, Via Brunamonti, 06100 Perugia, Italy

It is known that phospholipids represent a minor component of chromatin. It has been highlighted recently that these lipids are metabolized directly inside the nucleus, thanks to the presence of enzymes related to their metabolism, such as neutral sphingomyelinase, sphingomyelin synthase, reverse sphingomyelin synthase and phosphatidylcholine-specific phospholipase C. The chromatin enzymatic activities change during cell proliferation, differentiation and/or apoptosis, independently from the enzyme activities present in nuclear membrane, microsomes or cell membranes. This present study aimed to investigate crosstalk in lipid metabolism in nuclear membrane and chromatin isolated from rat liver in vitro and in vivo. The effect of neutral sphingomyelinase activity on phosphatidylcholine-specific phospholipase C and sphingomyelin synthase, which enrich the intranuclear diacylglycerol pool, and the effect of phosphatidylcholine-specific

phospholipase C activity on neutral sphingomyelinase and reverse sphingomyelin synthase, which enrich the intranuclear ceramide pool, was investigated. The results show that in chromatin, there exists a phosphatidylcholine/sphingomyelin metabolism crosstalk which regulates the intranuclear ceramide/diacylglycerol pool. The enzyme activities were inhibited by D609, which demonstrated the specificity of this crosstalk. Chromatin lipid metabolism is activated in vivo during cell proliferation, indicating that it could play a role in cell function. The possible mechanism of crosstalk is discussed here, with consideration to recent advances in the field.

INTRODUCTION

lipids change as a result of N-SMase (neutral SMase) [21–26], SM synthase [27,28], PC-PLC [29,30] and RSM synthase [31] activity in the nucleus. The nuclear N-SMase activity is involved in cell proliferation [32] and/or apoptosis [33]. Purification of the subnuclear fractions has demonstrated that the enzymes localized in chromatin have different optimum pH and K m values from those present in nuclear membrane [25,27,29,31]. Currently, nuclear membrane and chromatin enzymes have not been purified and characterized, and therefore there are no specific antibodies that can be used to study their exact localization. Nevertheless, electron microscopy analysis on rat liver sections performed using N-SMase conjugated with colloidal gold particles showed that SM is present in nuclear domains active in DNA replication and transcription, and is probably also involved in different steps of mRNA processing (results not shown). On the other hand, SM, NSMase and SM synthase activity have been demonstrated recently in an intranuclear complex containing STAT3 (signal transducer and activator of transcription 3) and a newly synthesized doublestranded RNA, supporting their association with active chromatin [34]. Moreover, N-SMase activity plays various roles in different subnuclear fractions. During cell proliferation induced by partial hepatectomy, the activation of N-SMase present in nuclear membrane changes the level of SM and cholesterol [35], modifying the fluidity and the size of the pores [36] and consequently regulating the mRNA nucleus/cytoplasm efflux [37]. In the same experimental conditions, chromatin N-SMase is involved in DNA structure modification and in the molecular events that occur during the beginning of S-phase [25]. These results were supported by a previous study which showed that ciprofibrate treatment induced liver cell proliferation, and withdrawal of the

The involvement of SM (sphingomyelin) and PC (phosphatidylcholine) in signal transduction [1,2], and the role of DAG (diacylglycerol) and ceramides in cell proliferation, differentiation and apoptosis have been widely described [3–5]. SM is metabolized by SMase (sphingomyelinase), producing ceramide and PPC (phosphocholine) [6], and by RSM synthase (reverse SM synthase), which synthesizes PC using DAG and the PPC group of SM, releasing ceramide [7]. PC-PLC [PC-specific PLC (phospholipase C)] degrades PC, producing DAG and PPC [8], whereas SM synthase synthesizes SM using ceramide and the PPC group of PC, releasing DAG [9]. It is known that the breakdown of PC activates PKC (protein kinase C) [10], a family of serine/ threonine kinases which participates in signal transduction in many cells and mediates a wide number of intracellular functions [11]. On the other hand, it has been demonstrated that cellular DAG may activate SMase, resulting in the inactivation of PKC [12], whereas the activation of SM synthase causes an increase in DAG, with a positive effect on PKC [13]. The co-operation between different lipid pathways has been also reported in cells stimulated with TNFα (tumour necrosis factor α). TNFα acts on NF-κB (nuclear factor κB) through initial activation of PCPLC, resulting in production of DAG, which in turn stimulates SMase activity, with consequent ceramide formation and specific kinase isoform activation [14]. Moreover, it has been demonstrated that cellular SMase and SM synthase are involved in cell proliferation [15] and apoptosis [16] respectively. Previous evidence has highlighted the presence of a phospholipid fraction in the nucleus that changes in relation to cell function [17–20]. These

Key words: ceramide, chromatin, diacylglycerol (DAG), nuclear membrane, sphingomyelin (SM).

Abbreviations used: DAG, diacylglycerol; PC, phosphatidylcholine; PKC, protein kinase C; PLC, phospholipase C; PC-PLC, PC-specific PLC; PPC, phosphocholine; SM, sphingomyelin; SMase, sphingomyelinase; N-SMase, neutral SMase; RSM synthase, reverse SM synthase; TCA, trichloroacetic acid; TNFα, tumour necrosis factor α. 1 To whom correspondence should be addressed (email [email protected]). c The Authors Journal compilation c 2008 Biochemical Society

382

E. Albi, R. Lazzarini and M.Viola Magni

drug stimulated apoptosis [38]. It was also shown that the activity of chromatin N-SMase increased during cell proliferation when the SM synthase activity was depressed, whereas chromatin SM content increased during apoptosis as a consequence of SM synthase activation [38]. It has been suggested that a decrease in SM may favour the opening of the helix during DNA synthesis, whereas an increase in SM may hinder the DNA duplication. Since these results suggested strongly that modifications of intranuclear lipid metabolism occur in relation to cell life, it was possible to hypothesize the existence of lipid metabolic machinery that allows metabolism to take place inside the nucleus. In order to investigate this hypothesis, this present paper aimed to investigate the crosstalk between SM and PC metabolism inside the nucleus either in vitro or in vivo during liver regeneration.

EXPERIMENTAL Materials

Radioactive [Me-14 C] SM (54.5 Ci/mol, 2.04 GBq/mmol) and [Me-3 H]PC (L-3-phosphatidyl[N-Me-3 H]choline 1,2-dipalmitoyl, 81.0 Ci/mmol, 3.03 TBq/mmol) and [3 H]UTP (41 Ci/mmol, 1.52 TBq/mmol) were from Amersham Biosciences, and GF/C filters were from Whatman. HionicFluor and Econofluor were from PerkinElmer, and Ecoscint A and Filtron-X were from National Diagnostics (Atlanta, GA, U.S.A.). PC, SM, ceramides, DAG, D609 (O-tricyclodec-9-yl dithiocarbonate, potassium salt), PMSF, EDTA and RNA polymerase were all purchased from Sigma. Animals

Sprague–Dawley rats (30 days old) of either sex were kept at a normal light/dark period and had free access to pelleted food and water. The animals were laparatomized after anaesthesia with 100 mg/kg farmotal (Farmitalia, Milano, Italy), and the liver was perfused via the portal vein with ice-cold 0.25 M sucrose, 3.3 mM CaCl2 and 1 mM PMSF (pH 7.2), and immediately removed. The animals were killed by cervical dislocation. Liver regeneration was induced by hepatectomy, which stimulates liver cells to proliferate. The hepatectomies were performed after anaesthesia as reported previously [39]. Sham-operated animals were used as controls. The animals were killed at different times after the operation from 0 to 24 h. All treatments were performed according to the international regulations of the National Institutes of Health. Nuclear membrane and chromatin purification

Nuclear membrane and chromatin were isolated and purified as described previously [31]. For chromatin extraction, purified hepatocyte nuclei were gently suspended for 5 min in 10 mM Tris/HCl containing 0.25 M sucrose, 1 mM PMSF and 0.3 % Triton X-100 (adjusted to pH 7.4 with 0.1 M NaOH) and centrifuged at 5000 g for 5 min at 4 ◦C in order to remove the outer nuclear membrane. The nuclear preparations were subsequently washed four times with a hypotonic solution containing 75 mM NaCl, 24 mM EDTA and 1 mM PMSF (pH 8.0), which resulted in swelling of the nuclei and destruction of the inner nuclear membrane. The chromatin extraction was completed by washing the denuded nuclei in a gradient of Tris concentrations [50 mM, 10 mM, 2 mM and 0.4 mM Tris/HCl with 1 mM PMSF (pH 8.0)]. After each wash step, the nuclear material was recovered by centrifugation at 8000 g for 10 min at 4 ◦C and resuspended each time with one to three strokes of a pestle at a low velocity. Each solution was used to wash the nuclei twice. Finally the c The Authors Journal compilation c 2008 Biochemical Society

nuclear material was resuspended in 1 mM PMSF (pH 8.0), and stirred overnight at 4 ◦C. The nuclei were then centrifuged at 90 000 g for 30 min at 4 ◦C. The chromatin-containing sediment was resuspended in 10 mM Tris/HCl and 1 mM PMSF (pH 8.6). Transcriptional activity of chromatin

Chromatin transcription was evaluated by using the method of Gould et al. [40] with the following modifications. The DNA content was evaluated by spectroscopy at 260 nm. Purified chromatin, corresponding to 20 µg of DNA, was incubated with 10 mM Tris/HCl (pH 8.0), 25 mM MnCl2 , 150 mM 2-mercaptoethanol, 6 mM ATP, 6 mM CTP, 6 mM GTP, 3 mM mercuriuridine, 20 µCi [3 H]UTP and 10 units of RNA polymerase, to a final volume of 250 µl. Incubations were performed at 37 ◦C for up to 4 h as indicated and the reaction was stopped by the addition of ice-cold 10 mM Tris/HCl and 10 % (v/v) TCA (trichloroacetic acid) (pH 8.0). The suspension was filtered with a GF/C filter and washed six times with 8 % (v/v) TCA until no radioactivity was recovered in the filtered solution. The radioactivity was measured in counting vials containing 10 ml of HionicFluor using a liquid scintillation analyser (1500 TRI-Carb; Packard Instrument Company, Meriden, CT, U.S.A.). The filter was dried by hot air and treated with 0.5 ml of Filtron-X at 60 ◦C for 30 min in the counting vials. When the filter had cooled, 20 µl of 96 % (v/v) acetic acid and 10 ml of Econofluor were added. The radioactivity was measured as stated above and was referred to RNA content (calculated as c.p.m./mg of RNA) [41]. Chromatin transcriptional activity was also evaluated after 60 and 90 min of treatment as required for the PC–SM metabolism study (see experiment 1 and experiment 2 below), with unlabelled molecules used at the same concentration of the radiolabelled molecules. Lipid analysis

Lipids were extracted from nuclear membrane and chromatin. SM and PC were separated by TLC, identified by comparison with SM and PC standards and scraped into test tubes for inorganic phosphorus determination [31]. Ceramide was separated according to the method of Previati et al. [42] on bidimensional TLC by using chloroform/methanol/acetic acid (190:9:1, by vol.) and diethyl ether/acetic acid (100:1, v/v) as the first and second mobile phases respectively. The spot, identified in relation to the standard, was scraped and eluted by washing twice in chloroform/methanol (2:1, v/v). The ceramide was dried and dissolved in pyridine/benzoyl chloride (1:2, v/v). The sample was heated for 1 h at 60 ◦C and the solution was evaporated under an efficient fume hood. Methanol (100 %, 4 ml) was added and the sample was heated again for 1 h at 70 ◦C. After methanol evaporation, 1 ml of 95 % (v/v) hexane and 1 ml of Na2 CO3 saturated 100 % methanol solution were added. The sample was centrifuged at 1500 g for 10 min at 4 ◦C. The lower phase was washed first with 1 ml of 100 % methanol containing 0.6 M HCl, and then with 100 % methanol. The lower phase was evaporated and the residue was dissolved in a known volume of hexane for analysis by HPLC using a Lichrosorb Si 60 column (125 mm × 4 mm, 5 µm internal particle size; Merck) with a flow rate of 0.5 ml/min, and hexane/propan-2-ol (49:1, v/v) was used as the mobile phase and measured at a wavelength of 240 nm. DAG was derivatized directly in the lipid extract by using Naproxen as a fluorescent label [43] and was purified by TLC using petroleum hexane/diethyl ether (3:1, v/v) as the mobile phase. The spot, identified in relation to the standard, was scraped and eluted by two washes with 2 ml of 99 % (v/v) acetonitrile. The analysis by HPLC was performed with a Lichrospher RP18 column

Nuclear phosphatidylcholine/sphingomyelin relationship

Figure 1

383

Experimental plan

In experiment 1, pre-incubation with [Me -14 C]SM induced activation of N-SMase. In the control sample, the pre-incubation step was performed without [Me -14 C]SM. The effect of N-SMase activity in the presence of specific substrates on PC-PLC and SM synthase activities was evaluated at different incubation times as reported in the Experimental section. In experiment 2, the pre-incubation with [Me -3 H]PC induced activation of PC-PLC. In the control sample, the pre-incubation step was performed without [Me -3 H]PC. The effect of PC-PLC activity in the presence of specific substrates on N-SMase and RSM synthase activities was evaluated at different incubation times as reported in the Experimental section.

(250 mm × 4.6 mm, 5 µm internal particle size; Merck) with a flow rate of 1.5 ml/min using acetonitrile/propan-2-ol (9:1, v/v) as the mobile phase and the fluorescence was detected (λex 230 nm, λem 352 nm). The total ceramide and DAG content was estimated by measurement of the area of all identified peaks. PC–SM metabolism (see Figure 1) Effect of endogenous N-SMase in nuclear membrane or chromatin on PC-PLC and SM synthase activities (Experiment 1)

[Me-14 C]SM was diluted 1:50 with non-radiolabelled SM to give a final radioactivity of 1.08 Ci/mol. The pre-incubation step was performed for 60 min at 37 ◦C with a reaction mixture (200 µl final volume) containing 0.1 M Tris/HCl (pH 7.6 or 8.4 for nuclear membrane or chromatin respectively), 6 mM MgCl2 , 0.1 % Triton X-100, 20 nCi [Me-14 C]SM, with 200 µg of nuclear membrane or chromatin suspended in 10 mM Tris/HCl (pH 7.6 or pH 8.4 for nuclear membrane or chromatin respectively) and 1 mM PMSF. In the control sample, the pre-incubation was performed without [Me-14 C]SM. At the end of the pre-incubation, considered to be the zero time point, 20 nCi [Me-14 C]SM, 114 nCi [Me-3 H]PC, 0.1 mM ceramide, 80 µg/ml sodium taurodeoxycholate and 2 mM CaCl2 were added to a final volume of 300 µl. The incubation was continued for 7.5, 15 and 30 min and was stopped by adding 3 ml of chloroform/methanol (1:1, v/v). The organic phase was washed with 0.2 volumes of 0.5 % NaCl. After 24 h, the upper phase was removed, measured and 500 µl of the upper phase was diluted in counting vials containing 10 ml of Ecoscint A and 1 ml of water to evaluate [3 H]PPC to determine PC-PLC activity and [14 C]PPC for N-SMase activity. The lower phase was used to evaluate SM synthase activity. The lower phase was dried under a nitrogen flow and the lipids were resuspended in 20 µl of 100 % chloroform and separated by TLC on silica gels

using chloroform/methanol/ammonia (65:25:4, by vol.) as the solvent. In the sample, non-radiolabelled SM was added to the tubes before chromatography in order to enhance identification of the spot. SM was identified on the basis of its migration in relation to a SM standard, localized with iodine vapour and scraped into counting vials containing 10 ml of Ecoscint A and 1 ml of water. The radioactivity was measured as reported above for the analysis of transcriptional activity of chromatin. For DAG and ceramide determination, samples containing 1 mg of protein were incubated as above for 0 and 30 min. The concentration of ceramide was calculated in both the control and experimental samples, taking into account the ceramide added to the incubation medium. Effect of endogenous PC-PLC activity in nuclear membrane or chromatin on N-SMase and RSM synthase (Experiment 2)

[Me-3 H]PC was diluted 1:100 with non-radiolabelled PC to give a final radioactivity of 1.27 Ci/mol. The pre-incubation step was performed for 60 min at 37 ◦C with a reaction mixture (200 µl final volume) containing 0.1 M Tris/HCl (pH 7.6 or 8.4 for nuclear membrane or chromatin respectively), 114 nCi [Me-3 H]PC, 2 mM CaCl2 and 0.1 % Triton X-100, with 200 µg of nuclear membrane or chromatin suspended in 10 mM Tris/HCl and 1 mM PMSF (pH 7.6 or 8.4 for nuclear membrane or chromatin respectively). In the control sample, the pre-incubation was performed without [Me-3 H]PC. At the end of the pre-incubation, considered to be the zero time point, 114 nCi [Me-3 H]PC, 60 nCi [Me-14 C]SM, 0.3 mM DAG and 6 mM MgCl2 were added to a final volume of 300 µl. The incubation was continued for 7.5, 15 and 30 min and was stopped by the addition of 3 ml of chloroform/methanol (1:1, v/v). The organic phase was washed with 0.2 volumes of 0.5 % NaCl. After 24 h, the upper phase was removed, measured and 500 µl of the upper phase was diluted in counting vials with c The Authors Journal compilation c 2008 Biochemical Society

384


Table 1 Lipid content and lipid metabolism enzyme activities in nuclear membrane and chromatin Values are expressed as nmol/mg of protein for SM, PC, ceramide and DAG, and pmol/mg of ∗∗ protein per min for enzyme activities. Results are means + − S.D. (n = 6), P < 0.01 compared with nuclear membrane. Lipid/enzyme

Nuclear membrane

Chromatin

PC Ceramide DAG SMase SM synthase PC-PLC RSM synthase

61.07 + − 7.33 1.11 + − 0.72 2.20 + − 0.45 899.05 + − 17.58 768.43 + − 30.12 1640 + − 77.04 1.05 + − 0.20

4.69 + − 0.49 1.96 + − 0.31 3.07 + − 0.25∗∗ 25.07 + − 0.32 265.04 + − 23.11 215.99 + − 20.16∗∗ 34.85 + − 7.41

10 ml of Ecoscint A and 1 ml of water to evaluate [3 H]PPC for PC-PLC and [14 C]PPC for SMase. The lower phase was used for RSM synthase evaluation by the same method used for SM synthase evaluation, but using non-radiolabelled PC to identify the spot. The radioactivity was measured and DAG and ceramide content were determined as stated above. The content of DAG was calculated in both the control and experimental samples, taking into account the DAG added to the incubation medium. D609 treatment

Nuclear membrane or chromatin (200 µg protein) were incubated with 200 nmol D609 [29], and N-SMase [25], SM synthase [27], PC-PLC [29] and RSM synthase [31] activities were assayed after 0, 7.5, 15 and 30 min. Statistical analysis

Unless otherwise stated, results are means + − S.D. Statistical significance was determined using the Student’s t test. RESULTS Nuclear membrane and chromatin lipid fraction

The lipid content and enzyme activities obtained in purified nuclear membrane and chromatin are reported in Table 1. In nuclear membrane, the PC content was approx. 12-fold greater than the SM content, whereas the level of DAG was only 2-fold greater than the ceramide content. In fact, the DAG/ceramide ratio is 1:1.98. The enzymatic activities of N-SMase and SM synthase were similar, and their activities were approx. 50 % of that of PC-PLC, but the activity of RSM synthase was very low. In chromatin, the concentration of PC was 4.47-fold higher than that of SM. DAG concentration was 1.57-fold greater than the ceramide concentration. The activity of N-SMase, similar to that of RSM synthase, was 9 % of SM synthase activity. The PCPLC activity was similar to that of SM synthase. The results of the lipid composition and enzyme activities in nuclear membrane and chromatin support previous observations [17]. Comparison of the results indicates a strong difference in the PC/SM ratio in the two fractions and a very high SMase activity in the nuclear membrane, whereas RSM synthase activity is higher in chromatin. Characteristics of purified chromatin

The chromatin isolation procedure used in the present paper consists of many treatments that remove the outer nuclear membrane and then the inner nuclear membrane, with a progressive and c The Authors Journal compilation c 2008 Biochemical Society

Figure 2

Chromatin transcription

The incubation was performed as reported in the Experimental section, using 0.1 mM SM or 0.3 mM PC. The values are expressed as c.p.m./mg of RNA and the results are means + − S.D. of experiments performed in duplicate (n = 4). ∗∗ P < 0.01 compared with 0 min time point. The control sample is chromatin incubated without SM or PC. Ex 1, experiment 1; Ex 2, experiment 2.

gentle swelling of the nuclear material that permits good preservation of chromatin. The purified chromatin remains transcriptionally active for 2 h (Figure 2). Moreover, it is also active after the treatments used in the PC–SM metabolism crosstalk study, both after the pre-incubation time and after the incubation times as stated above in experiments 1 and 2 (Figure 2).

Effect of endogenous N-SMase on PC-PLC and SM synthase activities and of endogenous PC-PLC on N-SMase and RSM synthase activities in nuclear membrane

In nuclear membrane, the lipid metabolism enzyme activities in the control sample varied in a linear fashion with time. After the pre-incubation step (0 min time point) with [Me-14 C]SM (experiment 1) or [Me-3 H]PC (experiment 2), N-SMase activity was 52 + was − 11 nmol/mg of protein, whereas PC-PLC activity 14 97 + − 15 nmol/mg of protein. Pre-incubation with [Me- C]SM did not modify PC-PLC and SM synthase activities with respect to the control sample (Figure 3a). Moreover, pre-incubation with [Me-3 H]PC does not influence the N-SMase and RSM synthase activities (Figure 3b). No variations were observed in DAG and ceramide content at 30 min of incubation with respect to the control sample in experiment 1 (Figure 4a), whereas an increase in DAG was observed in experiment 2 (Figure 5a).

Effect of endogenous N-SMase on PC-PLC and SM synthase activities and of endogenous PC-PLC on SMase and RSM synthase activities in chromatin

In chromatin, the lipid metabolism enzyme activities in the control sample varied in a linear fashion with time. After pre-incubation (0 min time point) with [Me-14 C]SM (experiment 1) or [Me-3 H]PC (experiment 2), N-SMase activity was 1200 + − 321 pmol/mg of protein, whereas PC-PLC activity was 9560 + − 250 pmol/mg of protein. Pre-incubation with [Me-14 C]SM produced increases in PC-PLC activity of 1.63-, 6.64- and 5.56-fold compared with the control sample after 7.5, 15 and 30 min incubation respectively (Figure 6a). At the same incubation points, the SM synthase activity increased 2.06-, 1.99- and 10.94-fold compared with the control sample (Figure 6a). After 30 min incubation, the DAG content had increased 3.81-fold compared with the control sample, and the ceramide content had decreased 4.19-fold with respect to the control sample (Figure 4b). The DAG/ceramide ratio in the control sample was 3:2 and 8:3 after 0 and 30 min


385

Figure 3 Activity of SM and PC metabolism enzymes in the nuclear membrane (a) The effect of endogenous SMase activity (experiment 1) and (b) the effect of endogenous PC-PLC activity (experiment 2) in the presence of specific substrates. Enzyme activities are expressed as nmol/mg of protein, determined at different incubation times (min), and results are means + − S.D. of experiments performed in duplicate (n = 5). In (b), the right-hand ordinate refers to RSM synthase activity (determined as pmol/mg of protein). C, control sample; Ex, experimental sample.

incubation respectively, whereas the ratios were 1:1 and 40:1 in the experimental sample at the same time points. Therefore endogenous N-SMase activity induces an enrichment in ceramide which is responsible for PC-PLC activation, leading to SM synthase activation, which enriches the DAG pool and reduces the pool of ceramide in chromatin. The endogenous PC-PLC activity increased N-SMase activity 1.75-, 1.5- and 1.62-fold compared with the control sample after 7.5, 15 and 30 min incubation respectively (Figure 6b). The RSM synthase activity was the same as determined in the control sample at 7.5 min, and had increased 3.2-fold at 15 min and 2.90fold at 30 min with respect to the control sample (Figure 6b). Therefore PC-PLC stimulates N-SMase activity first, followed by stimulation of RSM synthase. After 30 min incubation, the DAG content had decreased 1.76-fold and the ceramide content had increased 2.90-fold with respect to the control sample. The DAG/ceramide ratio is 30:19 and 3:1 in the control sample after 0 and 30 min, whereas in the experimental sample the ratios were 9:2 and 4:7 respectively (Figure 5b). Therefore in chromatin, PC-PLC activity results in an enrichment of the DAG pool that is responsible for N-SMase activation, which is followed by RSM synthase activation, resulting in enrichment of the ceramide pool and a reduction in the DAG pool.

Effect of D609 on enzyme activities

To verify the interaction between PC and SM metabolism, incubation of nuclear membrane and chromatin with D609 was performed. The results show that in nuclear membrane, the PC-PLC

Figure 4 DAG and ceramide content after endogenous SMase activity in the presence of specific substrates DAG and ceramide content was determined in nuclear membrane (a) and in chromatin (b). The results were obtained at 0 min (0 ) and after 30 min (30 ) incubation and were calculated taking into account the added ceramide. Results are expressed as nmol/mg of protein and ∗∗ are means + − S.D. of experiments performed in duplicate (n = 5). P < 0.01 compared with control. C, control sample; Ex, experimental sample.

(Figure 7a) and SM synthase (Figure 7b) activities were inhibited 99 % and 35 % respectively after 30 min incubation with D609 compared with the control sample. No inhibition was observed for N-SMase (Figure 7c) and RSM synthase (Figure 7d). In chromatin, D609 treatment induced a reduction in PC-PLC activity of 22 %, 48 % and 40 % compared with the control sample after 7.5, 15 and 30 min respectively (Figure 8a). At the same time points, treatment with D609 resulted in inhibition of SM synthase (59 %, 75 % and 86 %, Figure 8b), N-SMase (1.7 %, 10 % and 48 %, Figure 8c) and RSM synthase (14 %, 13 % and 45 %, Figure 8d) compared with the control samples. Therefore D609 acts on SM synthase initially, and then on PC-PLC, and the reduction of the DAG pool could be responsible for the inhibition of N-SMase and RSM synthase activities.

PC–SM metabolism during rat liver regeneration

In order to determine if the crosstalk of chromatin PC and SM metabolism demonstrated in vitro could occur in vivo, the activity of PC-PLC, N-SMase, SM synthase and RSM synthase were evaluated in the regenerating rat liver after partial hepatectomy. The results showed that in nuclear membrane, PC-PLC and NSMase activity increased progressively over time up to 24 h after the hepatectomy. SM synthase activity reached a peak level after 18 h, whereas RSM synthase was unaffected by the procedure (Figure 9a). In chromatin, the activity of PC-PLC peaked at 12 h post-hepatectomy and had reduced by 18 h, when N-SMase and RSM synthase activities peaked. However, the c The Authors Journal compilation c 2008 Biochemical Society

386


Figure 5 DAG and ceramide content after endogenous PC-specific PLC activity in the presence of specific substrates DAG and ceramide content was determined in nuclear membrane (a) and in chromatin (b). The results were obtained at 0 min (0 ) and after 30 min (30 ) incubation and were calculated taking into account the added DAG. Results are expressed as nmol/mg of protein and are means + − S.D. of experiments performed in duplicate (n = 5). ∗∗ P < 0.01 compared with control. C, control sample; Ex, experimental sample.

SM synthase activity increased strongly at 24 h, together with PC-PLC activity, which peaked again at 24 h (Figure 9b). No significant modification in enzyme activities were reported for sham-operated animals. DISCUSSION

These results highlight clearly that in nuclear membrane, the metabolism of PC is independent from that of SM and viceversa, whereas in chromatin, a crosstalk between the two lipid metabolism pathways exists. To study these metabolic interactions, each enzyme has been activated by the addition of its substrate, since physiologic activators or inducers are not currently available. Watanabe et al. [21] have shown that Fas activates N-SMase and inhibits SM synthase in nuclei isolated from human leukaemia Jurkat T-lymphoid cells. These authors have studied the whole nuclei, whereas our present results refer to nuclear membrane and chromatin purified from isolated nuclei, which have different physico-chemical properties [25,27,29,31]. In chromatin, the mechanism of enzyme regulation was very finely regulated. The N-SMase activity reduced the level of SM detected, with a consequent enrichment of the ceramide pool that is responsible for the stimulation of PC utilization, first by PCPLC, and then by SM synthase activation. The PC-PLC activity degrades PC, releasing PPC and DAG, and the SM synthase activity permits resynthesizing of SM using free ceramide and PC as sources of PPC, with consequent release of DAG. In this way, two stages of DAG production exist (early and late) which result in a reduction of the ceramide pool, which occurs with a c The Authors Journal compilation c 2008 Biochemical Society

Figure 6

Activity of SM and PC metabolism enzymes in chromatin

(a) The effect of endogenous SMase activity (experiment 1) and (b) the effect of endogenous PC-PLC activity (experiment 2) in the presence of specific substrate. Enzyme activities are expressed as pmol/mg of protein, determined at different incubation times (min), and results ∗∗ are means + − S.D. of experiments performed in duplicate (n = 5). P < 0.01 compared with control. C, control sample; Ex, experimental sample.

consequent increase in the DAG/ceramide ratio. Alternatively, PC-PLC activity causes an enrichment of the DAG pool that is responsible for N-SMase and RSM synthase activation. NSMase degrades SM, releasing PPC and ceramide, whereas RSM synthase resynthesizes PC using free DAG and PPC derived from SM, resulting in the release of ceramide. Consequently, as reported above for DAG, early and late ceramide production steps also exist. Therefore in chromatin, SM and PC metabolism pathways are highly correlated. These results are supported by treatment with D609, a drug known to be an inhibitor of PC-PLC [44] and SM synthase [45] regardless of whether the substrates are localized in cell membranes or in the nucleus [27]. This effect is evident in nuclear membrane, in which the drug acts at the same time on the two enzymes, even though they are inhibited by D609 to different extents. No inhibitory effect of D609 on N-SMase or RSM synthase activity is shown. In chromatin, the varying effect of D609 on the various enzymes tested is evident at different times. It has been demonstrated that inhibition of SM synthase occurs first, with inhibition of PC-PLC activity occurring later; the latter is responsible for the subsequent N-SMase and RSM synthase inhibition. Therefore a mechanism of internal regulation of lipid metabolism exists in chromatin, probably to maintain an equilibrium either between SM and PC or between ceramide and DAG. To study the possible role of chromatin SM– PC crosstalk in vivo, liver cell proliferation was stimulated by partial hepatectomy. Liver regeneration represents a good model of synchronous proliferation in vivo. The hepatocytes enter G1 phase and begin to synthesize DNA 18 h after hepatectomy, and divide after 24 h. Our results show that during liver regeneration, the metabolism of PC and SM localized at nuclear membrane were not related, whereas the metabolism of PC and SM in


Figure 7

387

Effect of D609 treatment on lipid enzymatic activities present in nuclear membrane

The effect of D609 on (a) PC-PLC, (b) SM synthase, (c) N-SMase and (d) RSM synthase activities was examined. The enzyme activities are expressed as nmol/mg of protein and results are ∗∗ means + − S.D. of experiments performed in duplicate (n = 4). P < 0.01 compared with control. C, control sample not treated with D609; Ex, experimental sample treated with D609.

Figure 8

Effect of D609 treatment on lipid enzymatic activities present in chromatin

The effect of D609 on (a) PC-PLC, (b) SM synthase, (c) N-SMase and (d) RSM synthase activities was examined. The enzyme activities are expressed as nmol/mg of protein and results are ∗∗ means + − S.D. of experiments performed in duplicate (n = 4). P < 0.01 compared with control. C, control sample not treated with D609; Ex, experimental sample treated with D609.

chromatin represents a finely-regulated mechanism. The activity of chromatin PC-PLC increased at the beginning of S-phase and was reduced during S-phase when N-SMase and RSM synthase activities increased, whereas SM synthase activity increased strongly only at the end of S-phase. Currently there is not sufficient evidence to allow us to describe the exact mechanism of action of ceramide derived from N-SMase on PC-PLC and DAG derived from PC-PLC on N-SMase. It is possible to suppose that different DAG- [44] or ceramide-dependent [45] PKC isoforms may be involved. It has been demonstrated that PKCα is stimulated by

DAG [46] and inhibited by ceramide [45]. In HL-60 (human promyelocytic leukaemia) cells, the depletion of PKCα from the cytosol abolished vitamin D3 -dependent stimulation of NSMase, an enzyme which is inhibited by PKCδ [47]. Depletion of PKCδ with specific antibodies significantly increased SMase activity [47]. On the other hand, PKCδ is a ceramide-activated kinase [48]. It is possible that DAG stimulates SM hydrolysis [49] by activating PKCα or inhibiting PKCδ. In nuclei isolated from rat liver, PLC activity increases PKCα [50], whereas in nuclei isolated from BL6 murine melanoma cells that do not c The Authors Journal compilation c 2008 Biochemical Society

388

Figure 9


Behaviour of lipid enzyme activities during rat liver regeneration

Cell proliferation is induced by partial hepatectomy as reported in the Experimental section, with nuclear membrane (a) and chromatin (b) prepared at different times after the operation (0, 6, 12, 18 or 24 h). The enzyme activities are expressed as pmol/mg of protein per min and results ∗∗ are means + − S.D. of experiments performed in duplicate (n = 4). P < 0.01 compared with 0 h time point. In (a), the right-hand ordinate refers to PC-PLC activity.

express PKCδ [51], overexpression of PKCδ inhibits N-SMase activity and stimulates SM synthase [41]. Therefore it is possible to hypothesize that at the nuclear level, ceramide could have an inhibitory effect on DAG-activated kinases, such as PKCα, which could act as a regulatory mechanism by stimulating PCPLC, resulting in an increase in the DAG pool. Vice-versa, DAG could have an inhibitory effect on ceramide-activated kinases, such as PKCδ, with consequent stimulation of N-SMase leading to an increase in the ceramide pool. Treatment with D609, which causes inhibition of SM synthase and PC-PLC activity, leading to a reduction in the level of DAG, does not result in inhibition of PKCδ and therefore results in inhibition of N-SMase activity. In conclusion, our results show that SM and PC can be synthesized and hydrolysed directly in chromatin, and a regulatory mechanism exists, suggesting that SM and PC could have a specific role to play in relation to cell function. We thank PRIN (Ministero dell’ Universit`a e Ricerca), ASI (Agenzia Spaziale Italiana) and the Fondazione Cassa di Risparmio di Perugia for financial support. We also thank Dr Elisa Bartoccini, Dr IIaria Bernardini, Dr Giacomo Cascianelli and Dr Francesca Marini (Department of Clinical and Experimental Medicine, University of Perugia, Perugia, Italy) for technical assistance, and Silvano Pagnotta (Department of Clinical and Experimental Medicine, University of Perugia, Perugia, Italy) for preparation of the Figures.

REFERENCES 1 Levade, T. and Jaffrezou, J. P. (1999) Signalling sphingomyelinases: which, where, how and why? Biochim. Biophys. Acta 1438, 1–17 c The Authors Journal compilation c 2008 Biochemical Society

2 Wright, M. M., Howe, A. G. and Zaremberg, V. (2004) Cell membranes and apoptosis: role of cardiolipin, phosphatidylcholine, and anticancer lipid analogues. Biochem. Cell Biol. 82, 18–26 3 Lang, D., Leray, C., Mestre, R., Massarelli, R., Dreyfus, H. and Freysz, L. (1995) Molecular species analysis of 1,2-diglycerides on phorbol ester stimulation of LA-N-1 neuroblastoma cells during proliferation and differentiation. J. Neurochem. 65, 810–817 4 Hannun, Y. A. and Obeid, L. M. (2002) The ceramide-centric universe of lipid-mediated cell regulation: stress encounters of the lipid kind. J. Biol. Chem. 277, 25847–25850 5 Luberto, C., Kraveka, J. M. and Hannun, Y. A. (2002) Ceramide regulation of apoptosis versus differentiation: a walk on a fine line. Lessons from neurobiology. Neurochem. Res. 27, 609–617 6 Spence, M. W. (1993) Sphingomyelinases. Adv. Lipid Res. 26, 3–23 7 Marggraf, W. D. and Kanfer, J. N. (1984) The phosphorylcholine acceptor in the phosphatidylcholine:ceramide cholinephosphotransferase reaction. Is the enzyme a transferase or a hydrolase? Biochim. Biophys. Acta 793, 346–353 8 Exton, J. H. (1994) Phosphatidylcholine breakdown and signal transduction. Biochim. Biophys. Acta 1212, 26–42 9 Allan, D. and Quinn, P. (1988) Resynthesis of sphingomyelin from plasma-membrane phosphatidylcholine in BHK cells treated with Staphylococcus aureus sphingomyelinase. Biochem. J. 254, 765–771 10 van Dijk, M., Muriana, F. J., van Der Hoeven, P. C., de Widt, J., Schaap, D., Moolenaar, W. H. and van Blitterswijk, W. J. (1997) Diacylglycerol generated by exogenous phospholipase C activates the mitogen-activated protein kinase pathway independent of Ras- and phorbol ester-sensitive protein kinase C: dependence on protein kinase Cζ . Biochem. J. 323, 693–699 11 Cross, T. G., Scheel-Toellner, D., Henriquez, N. V., Deacon, E., Salmon, M. and Lord, J. M. (2000) Serine/threonine protein kinases and apoptosis. Exp. Cell Res. 256, 34–41 12 Kolesnick, R. N. and Clegg, S. (1988) 1,2-Diacylglycerols, but not phorbol esters, activate a potential inhibitory pathway for protein kinase C in GH3 pituitary cells. Evidence for involvement of a sphingomyelinase. J. Biol. Chem. 263, 6534–6537 13 Hampton, R. Y. and Morand, O. H. (1989) Sphingomyelin synthase and PKC activation. Science 246, 1050 14 Schutze, S., Potthoff, K., Machleidt, T., Berivic, D., Wiegmann, K. and Kronke, M. (1992) TNF activates NF-κB by phosphatidylcholine-specific phospholipase C-induced “acidic” sphingomyelin breakdown. Cell 71, 765–776 15 Maupas-Schwalm, F., Auge, N., Robinet, C., Cambus, J. P., Parsons, S. J., Salvayre, R. and Negre-Salvayre, A. (2004) The sphingomyelin/ceramide pathway is involved in ERK1/2 phosphorylation, cell proliferation, and uPAR overexpression induced by tissue-type plasminogen activator. FASEB J. 18, 1398–1400 16 Van der Luit, A. H., Budde, M., Zerp, S., Caan, W., Klarenbeek, J. B., Verheij, M. and Van Blitterswijk, W. J. (2007) Resistance to alkyl-lysophospholipid-induced apoptosis due to downregulated sphingomyelin synthase 1 expression with consequent sphingomyelin- and cholesterol-deficiency in lipid rafts. Biochem. J. 401, 541–549 17 Albi, E. and Viola Magni, M. P. (2004) The role of intranuclear lipids. Biol. Cell 96, 657–667 18 Ledeen, R. W. and Wu, G. (2004) Nuclear lipids: key signalling effectors in the nervous system and other tissues. J. Lipid Res. 45, 1–8 19 Ledeen, R. W. and Wu, G. (2006) Sphingolipids of the nucleus and their role in nuclear signaling. Biochim. Biophys. Acta 1761, 588–598 20 Martelli, A. M., Fala, F., Faenza, I., Billi, A. M., Cappellini, A., Manzoli, L. and Cocco, L. (2004) Metabolism and signaling activities of nuclear lipids. Cell. Mol. Life Sci. 61, 1143–1156 21 Watanabe, M., Kitano, T., Kondo, T., Yabu, T., Taguchi, Y., Tashima, M., Umehara, H., Domae, N., Uchiyama, T. and Okazaki, T. (2004) Increase of nuclear ceramide through caspase-3-dependent regulation of the “sphingomyelin cycle” in Fas-induced apoptosis. Cancer Res. 64, 1000–1007 22 Tamiya-Koizumi, K., Umekawa, H., Yoshida, S. and Kojima, K. (1989) Existence of Mg2+ -dependent, neutral sphingomyelinase in nuclei of rat ascites hepatoma cells. J. Biochem. 106, 593–598 23 Neitcheva, T. and Peeva, D. (1995) Phospholipid composition, phospholipase A2 and sphingomyelinase activities in rat liver nuclear membrane and matrix. Int. J. Biochem. Cell Biol. 27, 995–1001 24 Romanenko, E. B., Demidenko, Z. N. and Vanyushin, B. F. (1998) RNA-polymerase, DNA-polymerase, DNA-methyltransferase and sphingomyelinase activities in liver nuclei of rats of different age. Biochemistry 63, 159–163 25 Albi, E. and Viola Magni, M. P. (1997) Chromatin neutral spingomyelinase and its role in hepatic regeneration. Biochem. Biophys. Res. Commun. 236, 29–33 26 Mizutani, Y., Tamiya-Koizumi, K., Nakamura, N., Kobayashi, M., Hirabayashi, Y. and Yoshida, S. (2001) Nuclear localization of neutral sphingomyelinase 1: biochemical and immunocytochemical analyses. J. Cell Sci. 114, 3727–3736 27 Albi, E. and Viola Magni, M. P. (1999) Sphingomyelin-synthase in rat liver nuclear membrane and chromatin. FEBS Lett. 460, 369–372

Nuclear phosphatidylcholine/sphingomyelin relationship 28 Kavok, N. S., Krasilnikova, O. A. and Babenko, N. A. (2003) Increase in diacylglycerol production by liver and liver cell nuclei at old age. Exp. Gerontol. 38, 441–447 29 Albi, E. and Viola Magni, M. P. (1999) Phosphatidylcholine-dependent phospholipase C in rat liver chromatin. Biochem. Biophys. Res. Commun. 265, 640–643 30 Ramoni, C., Spadaro, F., Barletta, B., Dupuis, M. L. and Podo, F. (2004) Phosphatidylcholine-specific phospholipase C in mitogen-stimulated fibroblasts. Exp. Cell Res. 299, 370–382 31 Albi, E., Lazzarini, R. and Viola Magni, M. (2003) Reverse sphingomyelin-synthase in rat liver chromatin. FEBS Lett. 549, 152–156 32 Alessenko, A. and Chatterjee, S. (1995) Neutral sphingomyelinase: localization in rat liver nuclei and involvement in regeneration/proliferation. Mol. Cell. Biochem. 143, 169–174 33 Tsugane, K., Tamiya-Koizumi, K., Nagino, M., Nimura, Y. and Yoshida, S. (1999) A possible role of nuclear ceramide and sphingosine in hepatocyte apoptosis in rat liver. J. Hepatol. 31, 8–17 34 Rossi, G., Viola Magni, M. and Albi, E. (2007) Signal transducer and activator of transcription 3 and sphingomyelin metabolism in intranuclear complex during cell proliferation. Arch. Biochem. Biophys. 464, 138–143 35 Albi, E., Peloso, I. and Viola Magni, M. P. (1999) Nuclear membrane sphingomyelincholesterol changes in rat liver after hepatectomy. Biochem. Biophys. Res. Commun. 262, 692–695 36 Tomassoni, M. L., Albi, E. and Viola Magni, M. P. (1999) Changes of nuclear membrane fluidity during rat liver regeneration. Biochem. Mol. Biol. Intern. 47, 1049–1059 37 Tomassoni, M. L., Amori, D. and Magni, M. V. (1999) Changes of nuclear membrane lipid composition affect RNA nucleocytoplasmic transport. Biochem. Biophys. Res. Commun. 258, 476–481 38 Albi, E., Pieroni, S., Viola Magni, M. P. and Sartori, C (2003) Chromatin sphingomyelin changes in cell proliferation and/or apoptosis induced by ciprofibrate. J. Cell Physiol. 196, 354–361 39 Albi, E., Rossi, G., Maraldi, N. M., Viola Magni, M. P., Castaldi, S., Solimando, L. and Zini, N. (2003) Involvement of nuclear phosphatidylinositol-dependent phospholipases C in cell cycle progression during rat liver. J. Cell Physiol. 197, 181–188 40 Gould, H. J., Maryanka, D., Fey, S. J., Cowling, G. J. and Allan, J. (1978) The assay of globin gene transcription in reconstituted chromatin. Methods Cell Biol. 19, 387–422

389

41 Schneider, W. C. (1957) Determination of nucleic acids by pentose analysis. Methods Enzymol. 3, 680–684 42 Previati, M., Bertolaso, L., Tramarin, M., Bertagnolo, V. and Capitani, S. (1996) Low nanogram range quantitation of diglycerides and ceramide by high-performance liquid chromatography. Anal. Biochem. 233, 108–114 43 Rastegar, A., Pelletier, A., Duportail, G., Freysz, L. and Leray, C. (1990) Sensitive analysis of phospholipid molecular species by high-performance liquid chromatography using fluorescent naproxen derivatives of diacylglycerols. J. Chromatogr. 518, 157–165 44 Li, F., Wu, N., Su, R. B., Zheng, J. Q., Xu, B., Lu, X. Q., Cong, B. and Li, J. (2006) Involvement of phosphatidylcholine-selective phospholipase C in activation of mitogen-activated protein kinase pathways in imidazoline receptor antisera-selected protein. J. Cell Biochem. 98, 1615–1628 45 Meng, A., Luberto, C., Meier, P., Bai, A., Yang, X., Hannun, Y. A. and Zhou, D. (2004) Sphingomyelin synthase as a potential target for D609-induced apoptosis in U937 human monocytic leukemia cells. Exp. Cell Res. 292, 385–392 46 Albi, E., La Porta, C. A. M., Cataldi, S. and Viola Magni, M. P. (2005) Nuclear sphingomyelin-synthase and protein kinase Cδ in melanoma cells. Arch. Biochem. Biophys. 438, 156–161 47 Stahelin, R. V., Digman, M. A., Medkova, M., Ananthanarayanan, B., Rafter, J. D., Melowic, H. R. and Cho, W. (2004) Mechanism of diacylglycerol-induced membrane targeting and activation of protein kinase Cδ. J. Biol. Chem. 279, 29501–29512 48 Abboushi, N., El-Hed, A., El-Assaad, W., Kozhaya, L., El-Sabban, M. E., Bazarbachi, A., Badreddine, R., Bielawska, A., Usta, J. and Dbaibo, G. S. (2004) Ceramide inhibits IL-2 production by preventing protein kinase C-dependent NF-κB activation: possible role in protein kinase Cθ regulation. J. Immunol. 173, 3193–3200 49 Shen, Y. M., Chertihin, O. I., Biltonen, R. L. and Sando, J. J. (1999) Lipid-dependent activation of protein kinase Cα by normal alcohols. J. Biol. Chem. 274, 34036–34044 50 Visnjic, D., Batinic, D. and Banfic, H. (1999) Different roles of protein kinase Cα and δ isoforms in the regulation of neutral sphingomyelinase activity in HL-60 cells. Biochem. J. 344, 921–928 51 Gopee, N. V. and Sharma, R. P. (2003) Sphingolipid bases and their phosphatases: transient activation and delayed repression of protein kinase C isoforms and their possible involvement in fumonisin B1 cytotoxicity. Toxicology 187, 239–250

Received 5 June 2007/14 November 2007; accepted 14 November 2007 Published as BJ Immediate Publication 14 November 2007, doi:10.1042/BJ20070758

c The Authors Journal compilation c 2008 Biochemical Society