Báo cáo khoa học: Receptor- and calcium-dependent induced inositol 1,4,5-trisphosphate increases in PC12h cells as shown by fluorescence resonance energy transfer imaging pot

1,4,5-trisphosphate increases in PC12h cells as shown by fluorescence resonance energy transfer imaging Mitsuhiro Morita1,2, Fumito Yoshiki1, Akira Nakane1, Yoshiumi Okubo1 and Yoshihisa

1,4,5-trisphosphate increases in PC12h cells as shown by fluorescence resonance energy transfer imaging

Mitsuhiro Morita1,2, Fumito Yoshiki1, Akira Nakane1, Yoshiumi Okubo1 and Yoshihisa Kudo1

1 Laboratory of Cellular Neurobiology, School of Life Science, Tokyo University of Pharmacy and Life Science, Japan

2 Department of Neurosurgery, University of New Mexico, Albuquerque, NM, USA

Phosphatidylinositol hydrolysis and subsequent

increases in intracellular calcium, activated by

G-pro-tein-coupled receptors or receptor tyrosine kinases, are

important regulators of various cellular functions [1,2]

The initial step in receptor-mediated phosphatidylinosi-tol (PtdIns) metabolism involves the activation of phospholipase C (PLC), which in turn hydrolyzes PtdIns When the substrate is phosphatidylinositol

Keywords

calcium; fluorescent resonance energy

transfer; inositol 1,4,5-trisphosphate;

muscarinic acetylcholine receptor;

phospholipase C

Correspondence

M Morita, Laboratory of Cellular

Neurobiology, School of Life Science, Tokyo

University of Pharmacy and Life Science,

1432-1, Horinouchi, Hachioji, 192-0392,

Tokyo, Japan

Fax: +81 426 76 8841

Tel: +81 426 76 8963

E-mail: moritam@ls.toyaku.ac.jp

(Received 15 May 2007, revised 4 August

2007, accepted 8 August 2007)

doi:10.1111/j.1742-4658.2007.06035.x

The production and further metabolism of inositol 1,4,5-trisphosphate [Ins(1,4,5)P3] require several calcium-dependent enzymes, but little is known about subsequent calcium-dependent changes in cellular Ins(1,4,5)P3 To study the calcium dependence of muscarinic acetylcholine receptor-induced Ins(1,4,5)P3 increases in PC12h cells, we utilized an Ins(1,4,5)P3 imaging system based on fluorescence resonance energy trans-fer and using green fluorescent protein variants fused with the pleckstrin homology domain of phospholipase C-d1 The intracellular calcium con-centration, monitored by calcium imaging, was adjusted by thapsigargin pretreatment or alterations in extracellular calcium concentration, enabling rapid receptor-independent changes in calcium concentration via store-operated calcium influx We found that Ins(1,4,5)P3 production was increased by a combination of receptor- and calcium-dependent compo-nents, rather than by calcium alone The level of Ins(1,4,5)P3 induced by the receptor was found to be half that induced by the combined receptor and calcium components Increases in calcium levels prior to receptor acti-vation did not affect the subsequent receptor-induced Ins(1,4,5)P3 increase, indicating that calcium does not influence Ins(1,4,5)P3 production without receptor activation Removal of both the receptor agonists and calcium rapidly restored calcium and Ins(1,4,5)P3 levels, whereas removal of cal-cium alone restored calcal-cium to its basal concentration Similar calcal-cium- calcium-dependent increases in Ins(1,4,5)P3 were also observed in Chinese hamster ovary cells expressing m1 muscarinic acetylcholine receptor, indicating that the observed calcium dependence is common to Ins(1,4,5)P3 production

To our knowledge, our results are the first showing receptor- and calcium-dependent components within cellular Ins(1,4,5)P3

Abbreviations

BSS, basal salt saline; CCh, carbachol; CFP, cyan fluoresent protein; CHO, Chinese hamster ovary cells; FRET, fluorescent resonance energy transfer; GFP, green fluorescent protein; Ins(1,4,5)P3, inositol 1,4,5-trisphosphate; mAChR, muscarinic acetylcholine receptor; MDCK cells, Madin–Darby canine kidney cells; PHD, pleckstrin homology domain; PtdIns, phosphatidylinositol; PtdIns(4,5)P2,

phosphatidylinositol 4,5-bisphosphate; PLC, phospholipase C; SOC, store-operated calcium entry; Tg, thapsigargin; YFP, yellow fluoresent protein.

4,5-bisphosphate [PtdIns(4,5)P2], one of the hydrolysis

products, inositol 1,4,5-trisphosphate [Ins(1,4,5)P3],

can induce the release of calcium from intracellular

calcium stores via the Ins(1,4,5)P3 receptor, a

mecha-nism referred to as Ins(1,4,5)P3-induced calcium

release [3] The pattern of calcium increase via

Ins(1,4,5)P3-induced calcium release has been shown to

be diverse, namely transient, sustained and oscillatory

[4] In some cases, the intracellular Ins(1,4,5)P3

concen-tration oscillates simultaneously with the calcium

con-centration [2,5] Some PLCs contain calcium-binding

domains [6], and other Ins(1,4,5)P3-metabolizing

enzymes, including inositol polyphosphate 3-kinases

and inositol polyphosphate 5-phosphatases, are

regu-lated by calcium [7] Together, these findings suggest a

complicated mutual regulation between Ins(1,4,5)P3

metabolism and Ins(1,4,5)P3-induced calcium release,

which is thought to contribute to the overall

fine-tuning of cellular functions

Biochemical assays of PtdIns metabolism, using

radiolabeled compounds with restricted spatial and

temporal resolution, differ from cellular responses

under physiological conditions, because radiolabeling

usually requires lithium to enhance the accumulation

of PtdIns metabolism products [8] Alternatively,

imaging assays, using green fluorescent protein (GFP)

variants fused with either the C1 domain of protein

kinase C or the pleckstrin homology domain (PHD)

of PLCd can assess PtdIns metabolism at a temporal

resolution similar to that of calcium imaging [5]

These fusion proteins bind to the products of PtdIns

metabolism and change their intracellular localization

Recently, a fluorescence resonance energy transfer

(FRET)-based assay of the redistribution of PHD

fusion proteins from plasma membrane to cytosol

was found to be a more reliable and quantitative

method for measuring increases in cellular

Ins(1,4,5)P3[9] PHD binds to PtdIns(4,5)P2, and

more preferentially to Ins(1,4,5)P3, at a 20-fold higher

affinity [5] In addition, the PHD fusion proteins of

the cyan and yellow variants of GFP (cyan fluoresent

protein CFP–PHD and yellow fluoresent protein

YFP–PHD) bound to PtdIns(4,5)P2 before

stimula-tion, and were localized close to the plasma

mem-brane As the hydrolysis of PtdIns proceeds, the

subsequently formed PtdIns(4,5)P2 is converted to

Ins(1,4,5)P3, and the fusion proteins essentially lose

their ability to interact with each other, becoming

redistributed throughout the cytosol while bound to

Ins(1,4,5)P3 This loss of interaction between the

fusion proteins is reflected in the change in their

respective fluorescent intensities, as monitored by

FRET

In this study, we used FRET-based Ins(1,4,5)P3 imaging to monitor the response of the rat pheo-chromocytoma cell line (PC12h) towards muscarinic acetylcholine receptor (mAChR) activation Using this system, we quantitatively assessed the interaction between intracellular calcium and Ins(1,4,5)P3 metabo-lism Furthermore, we compared these results with those obtained from Chinese hamster ovary (CHO) cells expressing m1 mAChR to determine the variation between the two cell types

Results

FRET-based imaging of Ins(1,4,5)P3metabolism

in PC12h cells PHD proteins fused with the cyan and yellow variants

of GFP (CFP–PHD and YFP–PHD, respectively) were coexpressed in PC12h cells, and their interaction was analyzed by measuring changes in fluorescence inten-sity Increases in cellular calcium and Ins(1,4,5)P3were induced using carbachol (CCh; 100 lm), which specifi-cally activates mAChR and induces Ins(1,4,5)P3 pro-duction in PC12 cells, even at 500 lm[10] Following CCh stimulation, the fluorescence intensity of the FRET donor (CFP–PHD) increased, whereas the intensity of the corresponding acceptor (YFP–PHD) decreased, as expected from dissociation of the exchanging partners, causing a subsequent decrease in the overall YFP⁄ CFP ratio (Fig 1) After complete removal of CCh, fluorescence intensities returned to their original values over several minutes

Fig 1 Ins(1,4,5)P3FRET-based imaging applied to PC12h cells in response to mAChR activation (A) Representative response of PC12h cells expressing CFP–PHD and YFP–PHD to CCh (100 l M ,

1 min, bar) The fluorescence ratio (EYFP ⁄ ECFP) (red line) was cal-culated from the corresponding fluorescence changes in CFP (cyan line) and YFP (yellow line), which are shown in normalized form (F ⁄ F o ).

Calcium dependence of Ins(1,4,5)P3metabolism

in PC12h cells

To determine whether Ins(1,4,5)P3 metabolism in

PC12h cells responding to mAChR activation is

depen-dent on calcium, we stimulated cells in normal (2 mm

Ca2+) or Ca2+-free extracellular solution [Ca2+-free

basal salt saline (BSS)] In the presence of 2 mm

extra-cellular calcium, mAChR activation induced both a

transient and a sustained increase in intracellular

calcium, as measured by Fura 2⁄ AM-based calcium

imaging, together with a sustained increase in

Ins(1,4,5)P3, which peaked later than calcium

(Fig 2Aa) By contrast, when Ca2+-free BSS was

used, only a transient calcium increase was observed, and the level of Ins(1,4,5)P3 production was reduced (Fig 2Ab) This initial transient calcium increase was insensitive to extracellular calcium, a finding similar to that of other cells exposed to a variety of agonists [11] These findings indicate that transient calcium increases result from the release of Ca2+ from intracellular stores, whereas subsequent sustained calcium increases reflect calcium entry, including the so-called store-operated calcium (SOC), which is activated by deple-tion of the calcium store Because use of Ca2+-free BSS effectively prevented any sustained calcium increase, as well as reducing the production of Ins(1,4,5)P3 to 63.2 ± 15.8% (mean ± SEM, n¼ 7,

P < 0.01, t-test) of the peak amplitude in normal medium, a significant proportion of the Ins(1,4,5)P3 increase can be regarded as dependent on intracellular calcium

The calcium dependence of this increase in Ins(1,4,5)P3 was further examined using other stimula-tory conditions To prevent intracellular calcium increases triggered by CCh while still maintaining nor-mal extracellular calcium, cells were loaded with the calcium chelator BAPTA-AM (10 lm, 30 min) We found that CCh did not trigger any detectable increase

in calcium, as measured with Fura 2⁄ AM, a high-affin-ity indicator, but an increase in Ins(1,4,5)P3 was observed, albeit at a reduced level, 68.4 ± 13.2% (n¼ 4, P < 0.05, t-test) of the peak amplitude observed in the absence of treatment (Fig 2Ba) Although depolarization with 30 mm KCl induced a receptor-independent increase in calcium, it did not induce an increase in Ins(1,4,5)P3 (Fig 2Bb) Pretreat-ment with the sarco-endoplasmic calcium ATPase inhibitor, thapsigargin (Tg), which depletes calcium stores and induces SOC, and variations in extracellular calcium concentration induced a similar increase in cal-cium via SOC, but had no effect on Ins(1,4,5)P3 pro-duction By contrast, mAChR activation after the calcium increase induced an increase in Ins(1,4,5)P3 (Fig 2Bc), but less than that observed in the absence

of treatment (65.2 ± 18.8%; n¼ 5, P < 0.05, t-test), even in the presence of both receptor activation and calcium increase (Fig 2Bc) These findings suggest that calcium becomes less effective in enhancing Ins(1,4,5)P3 production if its level is increased prior to receptor stimulation Although the effect of calcium increases prior to receptor stimulation was not studied further, it is likely that the same effect may account for the finding that the total Ins(1,4,5)P3 increase induced by receptor activation and calcium entry was slightly larger in the absence than in the presence of preceding calcium treatment (Fig 3) Basal calcium

Fig 2 Calcium dependence of cellular calcium and Ins(1,4,5)P 3

lev-els in PC12h cells Cellular calcium (upper traces) and Ins(1,4,5)P 3

(lower traces) were measured in PC12h cells containing Fura 2⁄ AM

or expressing CFP–PHD and YFP–PHD, respectively (A) Effects of

extracellular calcium on responses induced by CCh (100 l M , bars).

Responses are shown as mean ± SEM (a) Normal BSS (2 m M

Ca 2+ ; n ¼ 7); (b) Ca 2+ -free BSS (Ca 2+ -free; n ¼ 7) (B) Effects of

cellular calcium on Ins(1,4,5)P 3 levels Representative changes of

(a) a cell pretreated with BAPTA-AM (10 l M , 30 min) and exposed

to CCh (100 l M , bar); (b) a cell depolarized with KCl (30 m M , bar);

and (c) a cell responding to SOC entries and mAChR activation In

(c), the cells were pretreated with Tg (1 l M , 5 min) in Ca2+-free

BSS prior to imaging, SOC entries were induced after substitution

of Ca 2+ -free BSS (white bar) with normal BSS (bold bar), and

stimu-lation with CCh (100 l M , bar) Scale bars: horizontal (30 s); vertical

(0.1) for DRatio (340 ⁄ 380) and (0.1) for DRatio (YFP ⁄ CFP).

and Ins(1,4,5)P3levels in all experiments in Fig 2 were

within 11.2 and 28.5%, respectively, of their mean

val-ues in containing 2 mm calcium (Fig 2Aa); these basal

levels were not altered significantly by pretreatment

with calcium-free medium, BAPTA-AM and Tg

Fluctuations in the basal Ins(1,4,5)P3 level were larger than those of calcium, presumably because of fluctua-tions in transfection efficiency of CFP–PHD and YFP–PHD expression vectors Taken together, these results suggest that increased calcium alone is not suffi-cient to increase Ins(1,4,5)P3 in PC12 cells Moreover, the requirement for receptor activation suggests that calcium has a modulatory effect on Ins(1,4,5)P3 production once it has been induced by receptor activation

Calcium-dependent enhancement of receptor-induced Ins(1,4,5)P3production in PC12h cells

To investigate the modulatory effect of calcium on receptor-induced Ins(1,4,5)P3 production, mAChRs were activated with CCh in Ca2+-free BSS (Fig 2Ab); the extracellular solution was replaced with normal BSS containing CCh, which induced a second calcium increase and a further increase in Ins(1,4,5)P3 (Fig 3Aa) Both of these calcium increases were accompanied by increased Ins(1,4,5)P3, indicating that calcium can enhance receptor-induced Ins(1,4,5)P3 increases The peak Ins(1,4,5)P3 increases were similar for both mAChR-activated and reintroduced extracel-lular calcium, indicating that calcium entry supple-mented a reduction in Ins(1,4,5)P3 production caused

by the restoration of intracellular calcium To separate the calcium-dependent component of Ins(1,4,5)P3 induction, PC12h cells were pretreated with Tg in

Ca2+-free BSS (1 lm, 5 min) to prevent the possible release of calcium, and were subjected to the same stimulation protocol We found that Tg pretreatment successfully prevented the release of calcium induced

by mAChR activation, but did not inhibit the calcium entry caused by re-addition of extracellular calcium (Fig 3Ab) The receptor-induced Ins(1,4,5)P3 increase was smaller in the absence of calcium release than in its presence [compare initial Ins(1,4,5)P3 peaks in Fig 3Aa,b] The reintroduction of extracellular cal-cium induced a larger Ins(1,4,5)P3 increase when the cells had been pretreated with Tg Figure 3B summa-rizes the Ins(1,4,5)P3 increases induced by receptor activation and the reintroduction of calcium, as well as the effects of Tg on the components of these responses The peak amplitudes were calculated as the difference between signals before and after mAChR activation (Fig 3B) For cells not pretreated with Tg, the recep-tor-induced Ins(1,4,5)P3 increase was 79.6 ± 7.4% of that induced by reintroduced calcium, whereas for the pretreated cells it was, 39.8 ± 3.0%, which was significantly lower (P < 0.01, t-test) than for untreated cells The total Ins(1,4,5)P3 increase induced by the

Fig 3 Effect of calcium release and calcium entry on cellular

Ins(1,4,5)P 3 levels in PC12h cells (A) Representative changes in

cellular calcium (upper trace) and Ins(1,4,5)P 3 (lower trace) levels.

Cells were stimulated with CCh (100 l M , bars) in Ca 2+ -free BSS

(white bars), which was replaced with normal BSS (bold bars).

Where indicated, cells were pretreated without (a) and with (b) Tg

(1 l M , 5 min) in Ca 2+ -free BSS Scale bars: horizontal (30 s); vertical

(0.1) for DRatio (340 ⁄ 380) and (0.1) for DRatio (YFP ⁄ CFP) (B) Effect

of Tg pretreatment on the total amplitude of Ins(1,4,5)P 3 increases

in response to mAChR activation, in the absence (white bars) and

presence (hatched bars) of extracellular calcium Results were

cal-culated from the traces as indicated in (A), with (Tg, n ¼ 16) and

without (–, n ¼ 20) Tg pretreatment Each bar represents the

mean ± SEM for the ratio of YFP ⁄ CFP, represented by differences

between the peak and basal values, prior to CCh stimulation (DR).

reintroduction of calcium in Tg-pretreated cells was

slightly, but significantly, larger than that in untreated

cells (P < 0.05, t-test), whereas the receptor-induced

portion of the Ins(1,4,5)P3 increase was significantly

smaller in Tg-pretreated than in untreated cells

(P < 0.01, t-test) These results strongly suggest that

calcium modulates the receptor-induced Ins(1,4,5)P3

production in PC12h cells

The calcium dependence of increased Ins(1,4,5)P3

production was further investigated by more accurately

controlling the amount of intracellular calcium,

because we had found that the total amounts of

cal-cium differed for released calcal-cium and calcal-cium entry

(Fig 4) When extracellular calcium was reintroduced

to Tg-pretreated PC12h cells, the calcium increase

was uniform and not affected by mAChR activation

(Fig 4, also see Fig 5) By contrast to our previous

results (Fig 3Ab), in which cells were also treated with

Tg, receptor activation accompanied by calcium entry

enhanced Ins(1,4,5)P3 production (Fig 4A) Although

removal of extracellular calcium restored the

intra-cellular calcium concentration, the increase in

Ins(1,4,5)P3 was sustained In addition, the subsequent

increase in calcium was less effective at enhancing the

initial receptor-activated Ins(1,4,5)P3 increase

Although similar to our previous results (Fig 3Aa),

these findings show more clearly that the presence

of calcium initially enhanced receptor-induced

Ins(1,4,5)P3 By contrast to the immediate restoration

of intracellular calcium, the removal of extracellular

calcium led to a gradual decrease in Ins(1,4,5)P3

Because the rates of Ins(1,4,5)P3 and calcium

restora-tion were similar following the terminarestora-tion of receptor

activation caused by the removal CCH, these findings

suggest that PC12h cells can clear Ins(1,4,5)P3 and⁄ or

re-synthesize PtdIns(4,5)P2 as rapidly as they clear

cal-cium This result could be observed with our imaging

system, which uses FRET to measure the rapid

trans-location of fluorescent proteins Because our results

suggested that mAChR induces Ins(1,4,5)P3

produc-tion even after the removal of calcium entry, we

com-pared the averaged time-dependent restoration plots

for Ins(1,4,5)P3 and calcium following removal of

extracellular calcium and removal of both calcium and

carbachol (Fig 4Ba,b) Although the removal of

cal-cium after 1 min had little effect on the level of

recep-tor- and calcium-induced Ins(1,4,5)P3, the removal of

both calcium and CCh reduced the receptor- and

calcium-induced Ins(1,4,5)P3 to 31% By contrast, the

time-dependent restoration of intracellular calcium was

similar under both conditions

The gradual loss of Ins(1,4,5)P3 production may

have been due to the retention of calcium in some

Ins(1,4,5)P3metabolism-related process, which includes calcium-dependent enzymes, even after intracellular calcium restoration To address this, we assayed the

Fig 4 Calcium and Ins(1,4,5)P 3 restoration rates following the removal of calcium entry or calcium entry and receptor activation (A) Representative changes in cellular calcium (upper trace) and Ins(1,4,5)P 3 (lower trace) levels Cells were pretreated with Tg (1 l M ,5 min) in Ca 2+ -free BSS and stimulated with CCh (100 l M , bar) in Ca 2+ -free BSS (white bars) or normal BSS (bold bars) mAChR activation (1 min) in the presence of extracellular calcium was followed by its removal (1 min) and reintroduction (1 min) Scale bars: horizontal (30 s); vertical (0.05) for DRatio (340 ⁄ 380) and (0.1) for DRatio(YFP ⁄ CFP) (B) Time-dependent restoration of cellular calcium and Ins(1,4,5)P 3 levels following removal of calcium

or calcium and CCh The average changes in Ins(1,4,5)P3(a, n ¼ 4) and intracellular calcium (b, n ¼ 14) are expressed as mean ± SEM All traces were calculated from the response to the stimulatory paradigm, as shown in (A) The black and red lines represent the changes occurring after flushing the calcium and calcium ⁄ CCh mixture, respectively The flushing process was initiated at the point indicated by the arrows Bar ¼ 10 s.

effects of increased calcium prior to receptor activation

by comparing the receptor- and calcium-induced

Ins(1,4,5)P3increases in Tg-pretreated cells in the

pres-ence (Fig 5A) and abspres-ence (Fig 5B) of preceding

cal-cium entry (1 min), which had been removed 1 min

prior to receptor activation Because a similar level of

Ins(1,4,5)P3 had been induced under both conditions,

the preceding calcium increase had no effect on either

receptor- or calcium-induced Ins(1,4,5)P3, suggesting

that the calcium-dependent process in Ins(1,4,5)P3

metabolism is unable to retain calcium without

recep-tor activation Therefore, in the absence of receprecep-tor

activation, Ins(1,4,5)P3 metabolism is likely either

insensitive to calcium or incapable of retaining it By

contrast, once the calcium-dependent process has been

activated by both receptor and calcium, the effects of

calcium are retained, even after the calcium is

removed

Calcium-dependent enhancement of

receptor-induced Ins(1,4,5)P3increases in CHO cells

expressing m1 mAChR

PC12h cells express m1 mAChR and utilize it in the

CCh-induced increases in calcium and Ins(1,4,5)P3 To

investigate the cell-type dependence of Ins(1,4,5)P3

production, we assayed receptor- and calcium-induced

Ins(1,4,5)P3 increases in CHO cells that express m1 mAChR (CHO-m1 cells), using the same stimula-tory conditions as described above (Fig 3) Because original CHO cells, which do not express m1 mAChR, lack both calcium and Ins(1,4,5)P3 responses to CCh [13], any responses observed were likely due to mAChR We found that mAChR activation, coupled with the reintroduction of extracellular calcium, induced both the release and entry of calcium, as well

as increasing Ins(1,4,5)P3 levels (Fig 6Aa) Further-more, although Tg pretreatment effectively inhibited the additional release of calcium, it had no effect on calcium entry Although the lack of calcium release essentially reduced the degree of receptor-induced Ins(1,4,5)P3, it enhanced Ins(1,4,5)P3 induced by the reintroduction of extracellular calcium (Fig 6Ab) The receptor-induced level of Ins(1,4,5)P3 in cells not pre-treated with Tg was 120 ± 12.8% of the Ins(1,4,5)P3 level induced by the reintroduction of extracellular cal-cium (Fig 6B) By contrast, in cells pretreated with

Tg, the level of receptor-induced Ins(1,4,5)P3 was 46.3 ± 4.1%, because Tg prevented any further release

of calcium The Ins(1,4,5)P3 increases induced by the reintroduction of extracellular calcium were about the same for cells pretreated with Tg (DR; 0.49 ± 0.06) and those without (DR; 0.49 ± 0.05) These results were similar to those obtained for PC12h cells, indicat-ing that the Ins(1,4,5)P3 metabolism of both cell types (PC12h and CHO) exhibits similar calcium depen-dence

Discussion

We have shown that a FRET-based Ins(1,4,5)P3 imag-ing system can be used to monitor the level of Ins(1,4,5)P3 production in mAChR-activated PC12h cells, with the main focus centered on the effects of intracellular calcium changes We found that removal of extracellular calcium had no effect on mAChR-induced calcium release, although it prevented subsequent calcium entry, thus reducing the overall production of Ins(1,4,5)P3 By contrast, calcium increases induced by depolarization or SOC without mAChR activation did not increase Ins(1,4,5)P3 production, indicating that Ins(1,4,5)P3 increases are likely to be modulated, but not activated, by calcium The effect of calcium on Ins(1,4,5)P3 production was further investigated by separately analyzing receptor- and calcium-induced Ins(1,4,5)P3 increases The level of receptor-induced Ins(1,4,5)P3was about half that induced by the combi-nation of receptor and calcium, and was further enhanced by the subsequent increase in intracellular calcium levels The increase in calcium levels just prior

Fig 5 Effect of prior calcium levels on cellular Ins(1,4,5)P 3

produc-tion in PC12h cells Representative changes in cellular calcium

(upper trace) and Ins(1,4,5)P3 (lower trace) levels In all

experi-ments, cells were pretreated with Tg (1 l M ,5 min) in Ca2+-free BSS

and stimulated with CCh (100 l M , bars) in Ca 2+ -free BSS (white

bars) and later in normal BSS (bold bar) (A) mAChR activation

(1 min) in the absence of extracellular calcium followed by its

re-introduction (1 min); (B) addition (1 min) and removal (1 min) of

extracellular calcium, followed by mAChR activation (1 min) in the

absence of extracellular calcium and its reintroduction (1 min).

Scale bars: horizontal (30 s); vertical (0.05) for DRatio(340 ⁄ 380) and

(0.1) for DRatio(YFP ⁄ CFP).

to receptor activation did not affect subsequent recep-tor- and calcium-induced Ins(1,4,5)P3, suggesting that calcium has no effect on Ins(1,4,5)P3production in the absence of mAChR preactivation Although removal of receptor agonists and calcium from the medium pro-moted the rapid recovery of both calcium and Ins(1,4,5)P3 levels, the removal of calcium alone led to the recovery of calcium, but had little or no effect on Ins(1,4,5)P3production These findings were confirmed

by assaying the receptor- and calcium-induced effects

on Ins(1,4,5)P3 production in CHO cells expressing m1 mAChR

Imaging of Ins(1,4,5)P3 with fluorescently labeled PHD fusion proteins relies on the cellular localization

of fluorescent fusion proteins by confocal microscopy [5,14] and measurement of the interaction between the fusion proteins using FRET, as shown here [9] Both methods depend on the redistribution of fusion proteins from the plasma membrane to the cytosol In our earlier study on PC12h cells, in which we used confocal microscopy, mAChR activation resulted in the production of a redistribution peak, which was sig-nificantly delayed relative to the increase in calcium levels [15] By contrast, when we used FRET, we observed that calcium increase and Ins(1,4,5)P3 pro-duction occurred within a similar time frame FRET is considered the more accurate measure of time depen-dence, because the interaction between fusion proteins changes almost instantaneously following the disrup-tion of PtdIns(4,5)P2 binding Therefore, the delay observed in our previous study indicates that the release of fusion proteins from the plasma membrane and their migration to the cytosol requires several tens

of seconds A similar, but shorter, delay has been reported in CHO cells, suggesting that the diffusion constant of the fusion proteins, or the spatial organiza-tion between the plasma membranes and confocal planes, varies among cell types

We could not determine whether the redistribution

of fluorescently labeled PHD fusion proteins from the plasma membrane to the cytosol results from an increase in Ins(1,4,5)P3 or a decrease in PtdIns(4,5)P2 Because PHD from PLCd1 has a 20-fold greater affinity for Ins(1,4,5)P3 than for PtdIns(4,5)P2[5], however, the fluorescently labeled PHD fusion proteins would likely favor Ins(1,4,5)P3 binding, thus moving from the plasma membrane to the cytosol immediately following

an increase in Ins(1,4,5)P3 This has been observed in Madin–Darby canine kidney (MDCK) cells, in which Ins(1,4,5)P3 injection caused the redistribution of fusion proteins, regardless of calcium production [5] Moreover, the introduction of inositol 5-phosphatase, which hydrolyzes Ins(1,4,5)P3, inhibited any further

Fig 6 Effect of calcium release and calcium entry on cellular

Ins(1,4,5)P3in CHO cells expressing an m1 receptor (A)

Represen-tative changes in cellular calcium (upper trace) and Ins(1,4,5)P 3

(lower trace) levels Cells were stimulated with CCh (100 l M , bars)

in Ca 2+ -free BSS (white bar), later replaced by normal BSS (bold

bar) Where indicated, cells were incubated in the absence (a) or

presence (b) of Tg pretreatment (1 l M , 5 min) in Ca2+-free BSS.

Scale bars: horizontal (30 s); vertical (0.1) for DRatio (340⁄ 380) and

(0.1) for DRatio (YFP ⁄ CFP) (B) Effect of Tg pretreatment on the

total amplitude of Ins(1,4,5)P 3 response to mAChR activation, in

the absence (white bar) or presence (hatched bar) of extracellular

calcium Values were calculated from the traces in (A) with (Tg,

n ¼ 14), and without (–, n ¼ 19) Tg pretreatment Each bar

repre-sents the mean ± SEM for the ratio of YFP ⁄ CFP, using the

differ-ences between peak and basal values prior to CCh stimulation

(DR).

redistribution of the fusion proteins [5] By contrast,

others have failed to observe the Ins(1,4,5)P3-dependent

redistribution of fusion proteins [9] For example, in

N1E-115 cells, weak photoactivation of caged

Ins(1,4,5)P3has been found to induce a total release of

calcium, although uncaged Ins(1,4,5)P3 was not

suffi-cient to promote the redistribution of fusion

pro-teins [9] In addition, using radiolabeled inositol in

adrenal glomerular cells, enhanced accumulation of

Ins(1,4,5)P3, but not the redistribution of fusion

pro-teins, was observed in the presence of an

inosi-tol 3-kinase inhibitor (Sr3+) [9] A recent study, testing

the response of N1E-115 cells to bradykinin, suggested

that the redistribution of fusion protein is reflected by

the combination of increases and decreases in

Ins(1,4,5)P3and PtdIns(4,5)P2[16]

To address the CCh-induced FRET signal changes

in PC12h cells to increased Ins(1,4,5)P3 or to

decreased PtdIns(4,5)P2, we tested the effects of

inosi-tol 5-phosphatase expression, which had been utilized

to attribute PHD fusion protein translocation to

Ins(1,4,5)P3 increase in previous studies [5,17], on

CCh-induced calcium release and Ins(1,4,5)P3

metabo-lism We found that this enzyme significantly, but

partially suppressed these responses (Fig S1) The

partial inhibition of calcium release suggests that this

enzyme does not possess sufficient activity for

com-plete hydrolysis of the Ins(1,4,5)P3 resulting from

CCh stimulation, and we therefore could not use this

enzyme to determine the process underlying the

fluo-rescence changes These fluofluo-rescence changes,

how-ever, were completely abolished in 4 of 36 cells

expressing inositol 5-phosphatase, but in 0 of 15 cells

not expressing this enzyme, indicating that these

fluo-rescence changes were likely due to an increase in

Ins(1,4,5)P3 production

The effect of calcium on Ins(1,4,5)P3 production has

been described in cell types other than PC12h, by

measuring the translocation of GFP and PHD fusion

protein In cerebellar Purkinje cells, increased

intracell-ular calcium induced a more efficient increase in

Ins(1,4,5)P3 than did the activation of the group I

metabotropic glutamate receptor [17] In bovine

adre-nal glomerular cells, both intracellular calcium and

receptor induced an increase in Ins(1,4,5)P3[18] In

both MDCK and CHO cells, the apparently

synchro-nized oscillation of intracellular calcium and

Ins(1,4,5)P3 following receptor activation may be due

to the calcium dependence of Ins(1,4,5)P3 production

[5,19] That is, positive and negative calcium feedback

in Ins(1,4,5)P3 metabolism may induce Ins(1,4,5)P3

oscillation, which in turn promotes calcium oscillation

By contrast, the PC12 cell line is believed to be

incapa-ble of generating an oscillatory calcium response Several of our results, however, are in apparent disagreement with this oscillation hypothesis For example, we found similar calcium-dependent Ins(1,4,5)P3 increases in PC12h and CHO cells, and calcium was retained in Ins(1,4,5)P3 metabolism-related enzymes even after the restoration of calcium levels In HEK293 cells, mAChR-induced calcium oscillation was not accompanied by Ins(1,4,5)P3 oscillation, suggesting that calcium oscillation may not always require Ins(1,4,5)P3 oscillation [20] To determine the physiological role of the calcium depen-dence of Ins(1,4,5)P3 metabolism, a more detailed analysis of the effects of calcium and the cross-talk between other signaling systems (e.g protein kinase C)

is required

Although calcium has been found to affect receptor desensitization processes via G-protein receptor kinase

or PKC [21,22], it is less likely that calcium increases Ins(1,4,5)P3 metabolism by inhibiting receptor desen-sitization, inasmuch as the reduction in Ins(1,4,5)P3 metabolism by receptor desensitization during CCh stimulation was almost negligible It is therefore likely that the effect of calcium on Ins(1,4,5)P3 production may be a reflection of the calcium dependence of PLC Other enzymes involved in Ins(1,4,5)P3 meta-bolism, such as inositol polyphosphate 3-kinase and inositol polyphosphate 5-phosphatase, are calcium dependent, but their activation by calcium is expected

to promote Ins(1,4,5)P3 degradation, thus reducing the amount of cellular Ins(1,4,5)P3 The calcium-dependent activation of these enzymes may explain the reduced cellular Ins(1,4,5)P3 content in MDCK cells observed at higher extracellular calcium concen-trations [5] Four PLC enzyme subtypes, designated

b, c, d and e, have been identified to date, with all, except for e, possessing a calcium-binding domain and requiring calcium for proper activation [6] Although PC12 cells express PLCb, -c and -d[23], there is little evidence of PLCe expression in this cell line PLCe is usually expressed in neurons [24], and PC12 cells exhibit considerable neuronal character [25], suggesting that PLCe can be expressed in PC12h cells Calcium-dependent PLC isozymes are inactive

in these cells at calcium concentrations below 100 nm, but become active following physiological activation [26] Therefore, the calcium-independent component

of the mAChR-induced Ins(1,4,5)P3 increase, which was approximately half the total Ins(1,4,5)P3 increase, may reflect the activity of calcium-independent PLCe This PLC subtype is activated by mAChR via a small G-protein Rho [27], with a similar mechanism hypothesized in PC12h cells The calcium-dependent

activation of receptor-induced Ins(1,4,5)P3 increases

we observed in PC12h cells is reflected in the

corre-sponding activation of calcium-dependent PLCs,

par-ticularly PLCb and -d, with PLCb activated by Gq in

a calcium-dependent manner [28] and purified PLCd

activated by calcium alone [26] Thus, although PC12

cells express significant amounts of PLCd, an increase

in intracellular calcium in the absence of receptor

activation would induce Ins(1,4,5)P3 production only

when PLCd is overexpressed [29] The lack of activity

of endogenous PLCd in PC12 cells and other cell

types has yet to be established, but changes in

cellu-lar PtdIns(4,5)P2 content may provide one possible

explanation For example, the ratio of PtdIns to

PtdIns(4,5)P2 (100 : 2.8) in NIE-115 cells has been

reported to decrease almost immediately (within 10 s)

after bradykinin stimulation [16] Because the binding

of PHD to PtdIns(4,5)P2 is essential for PLCd

acti-vation [26], the low PtdIns(4,5)P2 content prior to

receptor activation may lead to inhibition of this

enzyme Recently, a Gq-coupled calcium-sensing

receptor was shown to induce PtdIns(4,5)P2

produc-tion through the utilizaproduc-tion of a small G-protein [30],

suggesting that this receptor activates Gq-dependent

PLCb and calcium-dependent PLCd via PtdIns(4,5)P2

production If mAChR-activation in PC12h cells

fol-lows a similar pathway, receptor-induced PtdIns(4,5)

P2 may provide the PtdIns(4,5)P2 necessary for the

calcium-dependent activation of PLCd

The Ins(1,4,5)P3 imaging method utilized here

reveals some interesting quantitative and

time-depen-dent properties of Ins(1,4,5)P3 metabolism Because

PtdIns metabolism is a fundamental mechanism that

controls several major cellular processes, with varying

dynamics, this method, coupled with the

overexpres-sion of enzymes involved in Ins(1,4,5)P3metabolism or

their suppression by RNA interference, may lead to a

greater understanding of the molecular mechanisms

involved in many cellular functions

Experimental procedures

Recombinant DNA

The expression vectors for CFP–PHD and YFP–PHD were

constructed as described for the GFP fusion protein [15],

using an expression vector containing the SRa promoter

[31]

Cell culture and transfection

PC12h cells were seeded on 12-mm diameter polyethylene–

imine precoated cover slips (1 lgÆmL)1) in Dulbecco’s

modified Eagle’s medium-high (Asahi Technoglass, Funab-ashi, Japan) containing 5% horse serum (Gibco BRL, Gaithersburg, MD) and 5% semifetal bovine serum (Mitsubishi Kagaku, Tokyo, Japan) The CFP–PHD and YFP–PHD expression vectors were transfected using Trans-Fast (Promega, Madison, WI) CHO cells expressing m1 mAChR, the gift of T Haga (Gakushuin University, Tokyo, Japan), were seeded on 12-mm cover slips in Nutri-ent Mixture (Ham) F-12 containing 10% fetal bovine serum (Equitech-Bio, Ingram, TX), whereas the CFP–PHD and YFP–PHD expression vectors were transfected using Lipo-fectAMINE2000 (Gibco BRL) All cells were imaged for 48–72 h

Imaging

Extracellular BSSs were used in all physiological experi-ments (normal BSS; 130 mm NaCl, 5.4 mm KCl, 5.5 mm glucose, 2 mm CaCl2, 1 mm MgCl2, 20 mm Hepes,

pH 7.4) Calcium-free extracellular solution (Ca2+-free BSS) lacked CaCl2and contained 0.5 mm EGTA To load the calcium indicator, the cells were incubated for 45 min

at 30C in normal BSS containing Fura 2 ⁄ AM (7.5 lm; Dojin-kagaku, Kumamoto, Japan), washed three times, and incubated at room temperature for 20 min For cal-cium imaging, sulfinpyrazone (100 lm) was added to nor-mal BSS at each step after repeated washing Fluorescence images in Figs 1–4 and 6 were obtained using an E600FN upright microscope (Nikon, Tokyo, Japan) equipped with

a Polychrome IV, high-speed tunable scanning monochro-matic light source (T.I.L.L Photonics GmbH, Gra¨felfing, Germany) and a C6790 CCD camera (Hamamatsu Pho-tonics, Hamamatsu, Japan), while the images in Figs 5 and 6 were obtained using an IX70 inverted microscope, fitted with an OSP-EXA filter exchanger (Olympus, Tokyo, Japan) and a C6790 CCD camera The image data were analyzed using aquacosmos software (Hamamatsu Photonics) For FRET imaging, the fluorescence was split

by a W-View dichroic mirror system (Hamamatsu Photon-ics) equipped with a dichroic mirror (510LP) and the barrier filters 480DF30 and 535DF25 for YFP–PHD and CFP–PHD, respectively Calcium and Ins(1,4,5)P3 increases were expressed as changes in the ratio of cence intensities (DR) For calcium imaging, the fluores-cence ratio was calculated by dividing the fluoresfluores-cence intensity obtained at 510 nm after excitation at 340 nm by the intensity at 510 nm after excitation at 380 nm For Ins(1,4,5)P3 imaging, the fluorescence ratio was calculated

by dividing the fluorescence intensity of YFP–PHD (535 nm) by the intensity of CFP–PHD (489 nm), both of which were obtained by excitation at 430 nm The fluores-cence ratio of the FRET response is larger in Figs 3–6 than Figs 1–2, because of the different neutral density fil-ter configuration for excitation, and each of YFP–PHD and CFP–PHD

CCh was purchased from Wako Chemicals (Tokyo, Japan)

All other chemicals were purchased from Sigma (St Louis,

MO)

Acknowledgements

This work was supported by Grant-in-Aid 10214204 for

Scientific Research on Priority Areas (B) on ‘Regulation

of Neural Transduction by Glial Cells’, Grant-in-Aid

15082101 for Scientific Research on Priority Areas on

‘Elucidation of Glia–Neuron Network-Mediated

Infor-mation Processing Systems’ and Grant-in-Aid 14780582

for Young Scientists on ‘Application of an Insect

Receptor to the Investigation of Neuronal Networks’

all from the Ministry of Education, Science and Culture

in Japan We thank Ms Hiromi Yanaka and Ms Keiko

Suzuki for their excellent secretarial assistance We also

thank Dr John A Conner of the University of New

Mexico for scientific criticism of the manuscript

References

1 Berridge MJ, Lipp P & Bootman MD (2000) The

versa-tility and universality of calcium signalling Nat Rev

Mol Cell Biol 1, 11–21

2 Carafoli E (2002) Calcium signaling: a tale for all

sea-sons Proc Natl Acad Sci USA 99, 1115–1122

3 Irvine RF & Schell MJ (2001) Back in the water: the

return of the inositol phosphates Nat Rev Mol Cell Biol

2, 327–338

4 Fewtrell C (1993) Ca2+oscillations in non-excitable

cells Annu Rev Physiol 55, 427–454

5 Hirose K, Kadowaki S, Tanabe M, Takeshima H &

Iino M (1999) Spatiotemporal dynamics of inositol

1,4,5-trisphosphate that underlies complex Ca2+

mobili-zation patterns Science 284, 1527–1530

6 Rhee SG (2001) Regulation of phosphoinositide-specific

phospholipase C Annu Rev Biochem 70, 281–312

7 Woodring PJ & Garrison JC (1997) Expression,

purifi-cation, and regulation of two isoforms of the inositol

1,4,5-trisphosphate 3-kinase J Biol Chem 272, 30447–

30454

8 Berridge MJ, Downes CP & Hanley MR (1982) Lithium

amplifies agonist-dependent phosphatidylinositol

responses in brain and salivary glands Biochem J 206,

587–595

9 van der Wal J, Habets R, Varnai P, Balla T & Jalink K

(2001) Monitoring agonist-induced phospholipase C

activation in live cells by fluorescence resonance energy

transfer J Biol Chem 276, 15337–15344

10 Vicentini LM, Ambrosini A, Di Virgilio F, Pozzan T &

Meldolesi J (1985) Muscarinic receptor-induced

phos-phoinositide hydrolysis at resting cytosolic Ca2+ con-centration in PC12 cells J Cell Biol 100, 1330–1333

11 Bennett DL, Bootman MD, Berridge MJ & Cheek TR (1998) Ca2+entry into PC12 cells initiated by ryanodine receptors or inositol 1,4,5-trisphosphate receptors Biochem J 329, 349–357

12 Ebihara T & Saffen D (1997) Muscarinic acetylcholine receptor-mediated induction of zif268 mRNA in PC12D cells requires protein kinase C and the influx of extracel-lular calcium J Neurochem 68, 1001–1010

13 Burford NT, Tobin AB & Nahorski SR (1995) Differen-tial coupling of m1, m2 and m3 muscarinic receptor subtypes to inositol 1,4,5-trisphosphate and adenosine 3¢,5¢-cyclic monophosphate accumulation in Chinese hamster ovary cells J Pharmacol Exp Ther 274, 134–142

14 Varnai P & Balla T (1998) Visualization of phospho-inositides that bind pleckstrin homology domains: calcium- and agonist-induced dynamic changes and rela-tionship to myo-[3H]inositol-labeled phosphoinositide pools J Cell Biol 143, 501–510

15 Morita M, Yoshiki F & Kudo Y (2003) Simultaneous imaging of phosphatidyl inositol metabolism and Ca(2+) levels in PC12h cells Biochem Biophys Res Commun 308, 673–678

16 Xu C, Watras J & Loew LM (2003) Kinetic analysis of receptor-activated phosphoinositide turnover J Cell Biol

161, 779–791

17 Okubo Y, Kakizawa S, Hirose K & Iino M (2001) Visualization of IP(3) dynamics reveals a novel AMPA receptor-triggered IP(3) production pathway mediated

by voltage-dependent Ca(2+) influx in Purkinje cells Neuron 32, 113–122

18 Balla T, Nakanishi S & Catt KJ (1994) Cation sensitiv-ity of inositol 1,4,5-trisphosphate production and metabolism in agonist-stimulated adrenal glomerulosa cells J Biol Chem 269, 16101–16107

19 Young KW, Nash MS, Challiss RA & Nahorski SR (2003) Role of Ca2+feedback on single cell inositol 1,4,5-trisphosphate oscillations mediated by G-protein-coupled receptors J Biol Chem 278, 20753–20760

20 Sinnecker D & Schaefer M (2004) Real-time analysis of phospholipase C activity during different patterns of receptor-induced Ca2+responses in HEK293 cells Cell Calcium 35, 29–38

21 Haga K, Kameyama K, Haga T, Kikkawa U, Shiozaki

K & Uchiyama H (1996) Phosphorylation of human m1 muscarinic acetylcholine receptors by G protein-coupled receptor kinase 2 and protein kinase C J Biol Chem

271, 2776–2782

22 Pitcher JA, Freedman NJ & Lefkowitz RJ (1998) G protein-coupled receptor kinases Annu Rev Biochem 67, 653–692

23 Mazzoni M, Bertagnolo V, Neri LM, Carini C, Marchisio M, Milani D, Manzoli FA & Capitani S