cell density-dependent changes in intracellular ca2+ mobilization via the p2y2 receptor in rat bone...

ORIGINAL ARTICLE 372J o u r n a l o fJ o u r n a l o f

CellularPhysiologyCellularPhysiology
Cell Density-Dependent Changes
in IntracellularCa2R Mobilization via theP2Y2 Receptor in Rat BoneMarrow Stromal Cells
JUN ICHIKAWA* AND HISAE GEMBA
Department of Physiology 2, Kansai Medical University, Osaka, Japan

Bone marrow stromal cells (BMSCs) are an interesting subject of research because they have characteristics of mesenchymal stem cells.We investigated intracellular Ca2þ signaling in rat BMSCs. Agonists for purinergic receptors increased intracellular Ca2þ levels ([Ca2þ]i).The order of potency followed ATP¼UTP>ADP¼UDP. ATP-induced rise in [Ca2þ]i was suppressed by U73122 and suramin, but notby pyridoxalphosphate-6-azophenyl-20,40-disulfonic acid (PPADS), suggesting the functional expression of G protein-coupled P2Y2

receptors. RT-PCR and immunohistochemical studies also showed the expression of P2Y2 receptors. [Ca2þ]i response to UTP changedwith cell density. The UTP-induced rise in [Ca2þ]i was greatest at high density. Vmax (maximum Ca2þ response) and EC50 (agonistconcentration that evokes 50% of Vmax) suggest that the amount and property of P2Y2 receptors were changed by cell density. Notethat UTP induced Ca2þ oscillation at only medium cell density. Pharmacological studies indicated that UTP-induced Ca2þ oscillationrequired Ca2þ influx by store-operated Ca2þ entry. Carbenoxolone, a gap junction blocker, enhanced Ca2þ oscillation.Immunohistochemical and quantitative real-time PCR studies revealed that proliferating cell nuclear antigen (PCNA)-positive cellsdeclined but the mRNA expression level of the P2Y2 receptor increased as cell density increased. Co-application of fetal calf serum withUTP induced Ca2þ oscillation at high cell density. These results suggest that the different patterns observed for [Ca2þ]i mobilization withrespect to cell density may be associated with cell cycle progression.

J. Cell. Physiol. 219: 372–381, 2009. � 2009 Wiley-Liss, Inc.

Abbreviations: BMSCs, bone marrow stromal cells; [Ca2þ]i,intracellular Ca2þ levels; PPADS, pyridoxalphosphate-6-azophenyl-20,40-disulfonic acid; SOC, store-operated Ca2þ entry; VDCC,voltage-dependent Ca2þ channel; CBX, carbenoxolone; PCNA,proliferating cell nuclear antigen; FCS, fetal calf serum; IP3, inositoltrisphosphate.

Contract grant sponsor: Kansai Medical University.Contract grant sponsor: Takeda Science Foundation.

*Correspondence to: Jun Ichikawa, Department of Physiology 2,Kansai Medical University, 10-15 Fumizono-cho, Moriguchi, Osaka570-8506, Japan. E-mail: [email protected]

Received 16 September 2008; Accepted 25 November 2008

Published online in Wiley InterScience(www.interscience.wiley.com.), 12 January 2009.DOI: 10.1002/jcp.21680

Bone marrow stromal cells (BMSCs) are non-hematopoieticcells residing in the marrow cavity (Krebsbach et al., 1999).They have a stem cell-like character that can differentiate intoosteoblasts, chondrocytes, adipocytes, myoblasts, and evenneurons (Prockop, 1997). Such pluripotent properties areretained even in the adult marrow-derived cells. Thus, they area promising source in the field of tissue regeneration andengineering, particularly because the use of human embryonicstem cells is limited by ethical considerations (Frankel, 2000).BMSCs are also important for construction of themicroenvironment of bone marrow in vivo. In the bonemarrow, BMSCs organize the three-dimensional reticulum andproduce the extracellular matrix scaffold (Krebsbach et al.,1999). BMSCs communicate with the hematopoietic stem cells(HSCs) via cell adhesion molecules and gap junction, as well assoluble mediators, and regulate the fate of HSCs, that is, self-renewal and differentiation into blood cells (Dorshkind, 1990;Krebsbach et al., 1999; Montecino-Rodriguez and Dorshkind,2001; Schwarz and Bhandoola, 2006).

Extracellular nucleotides have been recognized as autocrine/paracrine signaling molecules (Lazarowski and Boucher, 2001;Schwiebert and Fitz, 2008). Nucleotides are released from cellsin response to physiological and pathological stimulation, suchas mechanical stress, hypoxia, inflammation, and other agonists.Mechanisms of nucleotide release including exocytosis, ATP-binding cassette transporters, connexin hemichannels, andvoltage-dependent anion channels, have been considered(Burnstock, 2006). Released nucleotides diffuse into theextracellular area and activate purinergic receptors ofsurrounding cells. Nucleotides can be degraded quickly intoadenosine by ectonucleotidases that exist on the cell surface(Zimmermann, 1996). These mechanisms allow cell–cellcommunication in a limited area. Such signaling by nucleotideshas been reported in excitable and non-excitable cells, includinghuman BMSCs (Gallagher and Salter, 2003; Moskvina et al.,

� 2 0 0 9 W I L E Y - L I S S , I N C .

2003; Burnstock and Knight, 2004; Burnstock, 2006; Kawanoet al., 2006; Koizumi et al., 2007). In some cases, both thenucleotides and their metabolites (nucleosides and adenosine)are believed to act as messengers simultaneously (Burnstock,2006; Chen et al., 2006; Schwiebert and Fitz, 2008).

We found that rat BMSCs expressed G protein-coupledpurinergic receptor (P2Y2 subtype), and that Ca2þ signalingvia the P2Y2 receptor changed depending on cell density. Toinvestigate the true effects of cell density, we prepared differentcell densities of culture (low, medium, and high) at the sametime and used them for experiments on the same day of passage.The response of intracellular Ca2þ levels ([Ca2þ]i) to UTP,a potent agonist for the P2Y2 receptor, was stronger at highercell density. Only at the medium cell density, did UTPinduce Ca2þ oscillation. We also revealed that the growthstate of individual cells differed depending on cell density.Furthermore, co-application of fetal calf serum (FCS) and UTP

C E L L D E N S I T Y C H A N G E D C a 2þ S I G N A L I N G O F B M S C 373

induced Ca2þ oscillation. These results suggest that the celldensity-dependent differences in Ca2þ signaling of rat BMSCsmay be correlated with cell cycle progression.

Materials and MethodsCell culture

All animals were handled in accordance with the ‘‘Rules of AnimalExperimentation Committee, Kansai Medical University.’’ Thestandard medium consisted of minimum essential medium Eagle amodification (a-MEM; Sigma Chemical, St. Louis, MO)supplemented with 10% heat-inactivated FCS (Daiichi PureChemicals, Tokyo, Japan) and antibiotics (100 U/ml penicillin,100 mg/ml streptomycin, and 0.25 mg/ml amphotericin B; SigmaChemical).

BMSCs were isolated from the femoral shaft of male Fischer344 rats (6 or 7 weeks old; CLEA Japan, Inc., Tokyo, Japan) using themethod described by Maniatopoulos et al. (1988). Both ends of therat femurs were cut at the epiphyses and the marrow was flushedout using 10 ml of culture medium expelled from a syringe througha 21-gauge needle. The bone marrow suspensions were cultured in60-mm-diameter culture dishes. The medium was changed firstafter 24 h to remove non-adherent cells, and subsequentlyrenewed every 2–3 days. Adherent cells were grown as monolayercultures in a humidified atmosphere at 5% CO2 and 378C for 1–2weeks until confluent. Cells were then dissociated using trypsin andplated on coverslips (0.13–0.17 mm thick, 12 mm diameter; FisherScientific, Pittsburgh, PA) at three different cell densities (lowdensity, 50 cells/cm2; medium density, 100 cells/cm2; high density,200 cells/cm2). Medium cell density was used in the experiments toidentify the subtype of purinergic receptors. Four days after plating,these cells were analyzed for [Ca2þ]i measurement orimmunohistochemistry.

Intracellular Ca2R measurement

The [Ca2þ]i was measured using a Ca2þ-sensitive fluorescentdye, fura-2/AM (Dojindo, Kumamoto, Japan). Cells cultured oncoverslips were incubated with 3mM fura-2/AM at 378C for 40 min.The coverslips were placed in a chamber, which was mounted onthe stage of an inverted microscope (IX71; Olympus, Tokyo, Japan)with an objective lens (Fluor 20�; N.A., 0.75). The chamber wasperfused continuously with the standard bath solution containing150 mM NaCl, 5 mM KCl, 0.8 mM MgCl2, 1.8 mM CaCl2, 5.6 mMglucose, and 10 mM HEPES–NaOH (pH 7.4). The change influorescence of the fura-2 dye was detected using anAQUACOSMOS imaging system (Hamamatsu Photonics,Shizuoka, Japan). Fura-2 was excited with light at 340 or 380 nmusing a xenon lamp. The emitted fluorescence was detected at510 nm using a cooled CCD camera (Hamamatsu Photonics). Thefluorescent images were collected at 5 sec intervals. The ratio ofthe two fura-2 fluorescence intensities excited at 340 and 380 nm(F340/F380) was used to indicate the relative changes in [Ca2þ]i.

Measurement of [Ca2þ]i response to FCS was performed usingfluo-4/AM (Dojindo) because bath application of FCS producedstrong artifactual fluorescence with fura-2/AM for reasonsunknown, which influenced the F340/F380 ratio. The cells wereincubated with 3 mM fluo-4/AM at 378C for 40 min. The sameimaging system was used as described for fura-2 detection, exceptfor the filter set (excitation at 480 nm, emission at 535 nm). Fluo-4fluorescence change (F/F0) was used to measure the changes in[Ca2þ]i. Calibration for converting the fluorescence change tointracellular Ca2þ concentration was achieved according to thecell-free calibration method, described previously (Ichikawaet al., 2000). Experiments were performed at room temperature(about 258C).

All agonists and antagonists were purchased from SigmaChemical. ADP, ATP, UDP, UTP, suramin, pyridoxalphosphate-6-azophenyl-20-40-disulfonic acid (PPADS), SKF96365, and

JOURNAL OF CELLULAR PHYSIOLOGY

carbenoxolone (CBX) were dissolved in distilled water as stocksolutions. U73122, U73343, thapsigargin, and nifedipine weredissolved in dimethyl sulfoxide (DMSO) as stock solutions. Alldrugs were diluted to working concentrations in a bath solutionprior to use. Caffeine was dissolved directly in the bath solutionbefore use. A Ca2þ-free solution was prepared by replacing CaCl2with NaCl and adding 0.5 mM ethylene glycol tetraacetic acid(EGTA). To maintain osmolarity in the bath solutions, equiosmolarNaCl was exchanged for CaCl2 when the extracellular Ca2þ

concentration was varied.

RT-PCR

The expression of mRNA encoding the P2Y2 receptor in ratBMSCs was determined using RT-PCR techniques. Total RNA wasisolated from rat BMSCs grown in 35-mm-diameter culture dishesusing TRIzol1 (Invitrogen Corp., Carlsbad, CA) according to themanufacturer’s instructions. The cDNA synthesis and PCRamplification were performed using the SuperscriptTM III One-StepRT-PCR system with Platinum1 Taq DNA polymerase (InvitrogenCorp.) according to the manufacturer’s instructions using 1 ng oftotal RNA. The P2Y2 primer sequences were as follows: P2Y2

sense, 50-ACCCGCACCCTCTATTACTCCTTC-30; P2Y2

antisense, 50-AGTAGAGCACAGGGTCAAGGCAAC-30. ThecDNA synthesis and PCR amplification were carried using aprogrammed thermal cycler (GeneAmp PCR System2400;PerkinElmer Inc., Waltham, MA) with an annealing temperature of558C to give a 129-bp product. Amplification products wereseparated on 2% agarose gels and then visualized by ethidiumbromide staining. RT-PCR was performed at seven time pointsfor separate mRNA preparations from seven rats.

Quantitative real-time PCR

Quantitative real-time PCR was carried out using a DNA EngineOpticon1 2 System (MJ Research Inc., Waltham, MA) according tothe instruction manual. The oligonucleotide primers for the P2Y2

receptor used in the amplification reaction were same as above.Rat 18S rRNA, an endogenous housekeeping gene, was used as aninternal control for assessing the overall cDNA content. The 18SrRNA primer sequences were as follows: 18S rRNA sense,50-TCCCCGAGAAGTTTCAGCACATCC-30; 18S rRNAantisense, 50-CTTCCCATCCTTCACGTCCTTCTG-30.Template of each sample was synthesized from total RNA usingHigh Capacity cDNA Reverse Transcription Kit (AppliedBiosystems, Foster City, CA) according to the manufacturer’sinstructions. Amplification was carried out in a 25 ml final volumecontaining 0.5 ml of the template, 200 nM of each primer, and12.5 ml of SYBR1 GreenERTM qPCR SuperMix Universal(Invitrogen Corp.). The amplification program were as follows:508C for 2 min, 958C for 10 min and, 60 cycles at 958C for 15 secand 608C for 1 min. Operating software of the DNA EngineOpticon1 2 System was used for automated data collection anddata analysis. The mRNA expression of the P2Y2 receptor at eachcell density was normalized to the overall cDNA content assessedby 18S rRNA expression. Eight independent experiments with fourdifferent rats were performed.

Immunohistochemistry and double labeling of P2Y2

receptor antibody with PCNA

Cells on coverslips were fixed in 4% paraformaldehyde for 30 min,and then washed with phosphate-buffered saline (PBS). Cells werepreincubated with 1% hydrogen peroxide in PBS containing 0.3%Triton X-100 (PBS-T) for 20 min and then rinsed with PBS. To blocknon-specific binding, cells were preincubated with PBS-Tcontaining 5% normal horse serum. Subsequently, they wereincubated with the anti-PCNA antibody (mouse monoclonal,dilution of 1:200; Novocastra Labs, Newcastle upon Tyne, UK) for

374 I C H I K A W A A N D G E M B A

1 h at room temperature. Proliferating cell nuclear antigen(PCNA), a type of intranuclear protein and an accessory protein ofDNA polymerase d (Byrnes et al., 1976; Prelich et al., 1987;Tsurimoto, 1999), is essential for eukaryotic chromosomal DNAreplication and expression in the S phase of the cell cycle (Celis andCelis, 1985); thus, PCNA has been widely used as a marker forproliferating cells. The avidin–biotin horseradish peroxidasemethod was used with a Vectastain Elite avidin–biotin–peroxidasecomplex (ABC) kit (Vector Laboratories, Burlingame, CA). Afterwashing with PBS, the biotinylated horse anti-mouse IgG wasapplied and then detected with avidin-coupled horseradish-peroxidase/nickel-intensified 0.02% diaminobenzidine (DAB). Forthe second layer of antibodies, cells were preincubated with PBS-Tcontaining 5% normal goat serum to block non-specific binding.They were then incubated with the anti-P2Y2 antibody (rabbitpolyclonal, dilution of 1:200; Alomone labs, Jerusalem, Israel) for1 h at room temperature. After washing with PBS, the biotinylatedgoat anti-rabbit IgG was applied, and avidin-coupledhorseradish-peroxidase was visualized with 0.02% DAB. Followingthe immunohistochemical steps, cells were dehydrated andcovered with Entellan resin. Control experiments were carried outby substituting primary antibodies with PBS, and the specificityof P2Y2 antibody was further determined by preabsorption withthe control peptide antigen.

Immunohistochemical images were viewed under an invertedmicroscope (IX70; Olympus, Tokyo, Japan) with an objective lens(10�, N.A., 0.30) and captured with a cooled color CCD camera(SPOT 2E, model 1.40; Diagnostic Instruments, Inc., SterlingHeights, MI). All images were captured under the same conditionsof transmitted light intensity, exposure time, and color balance.PCNA-positive nuclei were counted in three microscopic fields(1.27 mm� 0.87 mm per field) on a 12-mm-diameter coverslip.The threshold for counting was set at the optical density of themost intensely stained nuclei. More than four coverslips werecounted for each cell density (low, medium, and high) fromeach rat. Data were obtained from six rats. The percentage ofPCNA-positive cells was calculated by counting the PCNA-positivecells relative to the total cells.

Analysis of concentration–response curves

Concentration–response curves were fit using non-linear analysis(IGOR Pro, ver3.15; WaveMetrics, Inc., Lake Oswego, OR), whichautomatically calculated Vmax (maximum Ca2þ response) andEC50 (agonist concentration that evokes 50% of Vmax).

Statistical analysis

Data are presented as the mean� SEM. For comparison of morethan two groups, one-way analysis of variance (ANOVA) was used.Statistical significance was considered as P< 0.01.

ResultsRat BMSCs express functional P2Y2 receptors

Application of agonists for purinergic receptors increased[Ca2þ]i in the presence and absence of extracellular Ca2þ.When ATP or UTP was applied, a sustained increase in [Ca2þ]ifollowing the initial peak occurred under extracellular Ca2þ-existing conditions, while only a monophasic increase in [Ca2þ]iwas observed under Ca2þ-free conditions (Fig. 1A). Thebiphasic response to ATP/UTP is an established feature of Gprotein-coupled P2Y receptors (White et al., 2005). The initial[Ca2þ]i increase is due to the release of Ca2þ from intracellularstores by the G protein/PLCb/IP3 (inositol trisphosphate)/Ca2þ

pathway, and the second sustained [Ca2þ]i increase is due tostore-operated Ca2þ influx from extracellular space. The ATP-


induced [Ca2þ]i increase was suppressed by preincubation ofU73122, an inhibitor of phospholipase C (Fig. 1B). U73343,an inactive analog of U73122, did not inhibit the ATP-induced[Ca2þ]i increase (Fig. 1C). These observations are consistentwith G protein-coupled P2Y receptors (Maaser et al., 2002;White et al., 2005).

Pharmacological characterization of the P2Y receptorsubtypes was carried out by a concentration–responseexamination using P2Y receptor-specific agonists (ATP,UTP, ADP, UDP). Cells responded to all agonists in adose-dependent manner. The order of potency wasATP¼UTP>ADP¼UDP (Fig. 1D), consistent with thecharacter of P2Y2 (Burnstock and Knight, 2004; Burnstock,2007). All of the cells tested showed similar order of potency.Furthermore, we examined the effects of antagonists on P2Yreceptors. Preincubation with suramin completely inhibited theATP-induced [Ca2þ]i increase (Fig. 1E), but PPADS did not(Fig. 1F). The effects of using these blockers were consistentwith P2Y2, but not P2Y4, because the P2Y2 receptor is PPADS-insensitive (Lambrecht et al., 2002; Burnstock and Knight, 2004;Burnstock, 2007). These pharmacological observations suggestthe expression of the P2Y2 subtype in rat BMSCs.

RT-PCR analysis of the total RNA extracted from rat BMSCsshowed a positive band corresponding to the expected size forP2Y2 receptors (Fig. 1G). The immunohistochemical study alsoshowed the expression of P2Y2 receptors in rat BMSCs. BMSCswere stained positively with an antibody for the P2Y2 receptor(see Fig. 6A–C). Control immunohistochemistry resulted inabsence of any staining (data not shown). Thus, rat BMSCsappear to express functional P2Y2 receptors.

Ca2R mobilization via P2Y2 receptors changeddepending on cell density

The [Ca2þ]i response to UTP, a potent agonist of the P2Y2

receptor, was examined in rat BMSCs. We found that itspotency differed significantly depending on the cell density ofthe culture. To eliminate the possibility that the differences in[Ca2þ]i response were caused by the day of culture, weprepared the three densities of culture (low, medium, and high)on the same day and measured [Ca2þ]i response at the same dayof passage.

At low cell density, cells were sparsely located. Some cellswere contacted with neighboring cells, but many single cellswere observed (Fig. 2A). At medium cell density, single cellswere rarely seen. Most of cells were connected each other andthe cell edges were extended to the cell-free areas (Fig. 2B).At high cell density, all areas of culture dishes were occupiedwith cells (Fig. 2C).

Cells responded to UTP in a dose-dependent manner at allthree densities. However, the degree of [Ca2þ]i response toUTP was weak at the low cell density, and became strongeras cell density increased. At low density, cells did not respondto low concentrations of UTP (�1 mM), but cells at mediumdensity responded to �1 mM UTP. At high density, cells couldrespond to �0.1 mM UTP (Fig. 2D–F).

Dose–response curves at each cell density were plotted andfitted, and Vmax and EC50 were calculated (Fig. 3 and Table 1).[Ca2þ]i responses to UTP (peak amplitude) increased in a dose-dependent manner, reaching a plateau at 100 mM for alldensities (Fig. 3). Vmax for the low density was significantly lessthan those at the medium and high densities (P< 0.01)(Table 1). Vmax for the medium and high-density cultures werenot significantly different. The EC50 was lowered as the celldensity increased (Table 1). These results suggest that celldensity affects both Vmax and EC50, that is, affects the number ofP2Y2 receptors, the activity of the downstream pathway ofP2Y2 receptors, the affinity of P2Y2 receptors, and other factorsrelating to Ca2þ mobilization whereas the subtype of P2

Fig. 1. Rat BMSCs express functional P2Y2 receptors. A: Purinergic receptor agonists increased [Ca2R]i in rat BMSCs. Data show a typicalresponse from one cell. Concentration of each agonist was 10 mM. Similar results were obtained from all of the cells tested (157 cells from sevenrats). B: Phospholipase C inhibitor U73122 (5 mM, pretreated for 10 min) reduced the [Ca2R]i response to ATP (10 mM). Data show a typicalresponse from one cell. Similar results were obtained from all of the cells tested (102 cells from five rats). C: U73343, the inactive analog of U73122(5mM,pretreatedfor10min)didnotsuppressthe[Ca2R]iresponsetoATP(10mM).Datashowatypicalresponsefromonecell.Similarresultswereobtained from all of the cells tested (134 cells from five rats). D: Concentration–response relationship of the various purinergic receptor agonists.Each point represents the mean W SEM of the [Ca2R]i peak amplitude of more than 210 cells in 7–10 different experiments from seven rats. Theorder of potency of the agonists to raise [Ca2R]i was ATP U UTP > ADP U UDP. No significant differences were found between the [Ca2R]i

increases due to ATP versus UTP (P > 0.01, ANOVA) as well as those of due to ADP versus UDP (P > 0.01, ANOVA). The increases in [Ca2R]i

induced by ATP and UTP were significantly higher at ‡5 mM than those observed with ADP and UDP (P < 0.01, ANOVA). E: Suramin (100 mM,pretreated for 10 min) suppressed the [Ca2R]i response to ATP (10mM). Data show a typical response from one cell. Similar results were obtainedfromall of thecells tested (93cells from fiverats).F: PPADS(100mM,pretreated for10min) didnot suppress the[Ca2R]i responsetoATP (10mM).Data show a typical response from one cell. Similar results were obtained from all of the cells tested (121 cells from five rats). G: mRNA forP2Y2 receptor was present in rat BMSCs. RT-PCR analysis of the total RNA extracted from rat BMSCs generated fragments of the expected size(rP2Y2 refers to the rat P2Y2 receptor; M refers to the marker).



Fig. 2. Celldensity-dependentchanges inUTP-induced increase in[Ca2R]i.Bright-field imagesofratBMSCsat(A) low, (B)medium,and(C)highcell density. Images were taken 4 days after seeding. Dose–response of UTP (0.1–1,000 mM) at (D) low, (E) medium, and (F) high cell density.Data show a typical response from one cell. Similar results were obtained from all of the cells tested (more than 142 cells from seven rats).


receptors remained unchanged (the order of potency wasATP¼UTP>ADP¼UDP at all cell densities; data not shown).

To investigate whether the content of Ca2þ stores changeswith cell density, we used thapsigargin, an inhibitor ofendoplasmic reticulum Ca2þ–ATPase, which releases Ca2þ

from IP3-sensitive Ca2þ stores (Thastrup et al., 1990), andcaffeine, an activator of most forms of the ryanodine receptor,which releases Ca2þ from ryanodine-sensitive Ca2þ stores(McPherson and Campbell, 1993). Both drugs were appliedseparately in Ca2þ-free solution and the amplitudes of theincrease in [Ca2þ]i were compared. The mean� SEM of peakamplitude of [Ca2þ]i in the presence of thapsigargin (1 mM)

Fig. 3. Concentration–response relationship for UTP at differentcell densities. Each point represents the mean W SEM peak amplitudeof the increase in [Ca2R]i from more than 142 cells in 10–17 differentexperiments from seven rats. The data at medium cell density are thesame as the UTP data in Figure 1D.


were 309� 10 nM at low cell density, 320� 23 nM at mediumcell density, and 311� 20 nM at high cell density (data wereobtained from more than 125 cells in 7–11 differentexperiments from seven rats). No significant difference wasfound among different cell densities (P> 0.01, ANOVA).Caffeine (10 mM) did not induce a rise in [Ca2þ]i at any celldensity (data not shown). Ca2þ stores in rat BMSCs werethapsigargin-sensitive at all three cell densities, andreleasable Ca2þ content from intracellular stores was constant,regardless of cell density.

Note that UTP induced Ca2þ oscillation only at medium celldensity (Fig. 2E). This Ca2þoscillation was displayed in 42% of allcells observed (160/380 cells in 15 different experiments fromseven rats by 10 mM UTP application). The cells at low and highdensities did not show Ca2þ oscillation at any concentration ofUTP. The Ca2þ oscillation at medium cell density was observedin a restricted range of UTP concentration from 10 to 100 mM(Fig. 2E). Ca2þ oscillation was not observed at lower or higherUTP concentrations.

UTP-induced Ca2R oscillation at medium cell density

Ca2þ oscillation at medium density was not synchronized withother oscillating cells, and the frequency varied. The oscillatingcells had no characteristic morphology and most were locatedsporadically. Ca2þ waves were observed in only about 10%

TABLE 1. Vmax and EC50 for UTP at different cell densities

Values were calculated from the concentration–response curves in Figure 3using IGOR Pro. Vmax is the maximum [Ca2þ]i response and EC50 is the agonistconcentration that evokes 50% of Vmax. Each value is the mean� SEM of morethan 142 cells in 10–17 different experiments from seven rats. �Significantlylower compared to other groups (P< 0.01, ANOVA). ��Significantly different(P< 0.01, ANOVA).

Fig. 4. ExtracellularCa2R influxviaSOCchannels is important forUTP-inducedCa2R oscillationatmediumcelldensity.UTPwastestedat10mMfor all experiments. Typical traces from one cell are shown. Similar results could be observed from more than 90 cells from five rats, in allexperiments.A:UTP inducedCa2R oscillation inthenormalbathsolutioncontainingCa2R,butnot in theCa2R-freesolution.B: Inthebathsolutioncontaining a low concentration of extracellular Ca2R, UTP failed to induce Ca2R oscillation. C: SKF96365 (50mM) suppressed UTP-induced Ca2R

oscillation. D: Nifedipine (10 mM) did not suppress UTP-induced Ca2R oscillation.

Fig. 5. UTP-induced Ca2R oscillation at medium cell density isenhanced by the gap junction blocker CBX. UTP was tested at 10mM.Data shown are the typical traces from one cell. Of the total cellsexamined, 52% (112/198 cells in 10 different experiments fromfive rats) showed such enhancements. A: CBX (10 mM) enhancedUTP-induced Ca2R oscillation. B: CBX (10 mM) induced Ca2R

oscillation in the cell that did not oscillate due to UTP.


of oscillating cells (17/160 cells in 15 different experimentsfrom seven rats). Spontaneous Ca2þ oscillation was notobserved at any cell density (data not shown).

This Ca2þ oscillation disappeared quickly when UTP waswashed out, and it was suppressed under Ca2þ-free conditions(Fig. 4A). When the extracellular Ca2þ concentration waslow, Ca2þ oscillation could not be observed (Fig. 4B). Thesedata indicate that Ca2þ influx from extracellular space is neededfor UTP-induced Ca2þ oscillation. SKF96365, a blocker forstore-operated Ca2þ influx (SOC) (Merritt et al., 1990),completely suppressed UTP-induced Ca2þ oscillation (Fig. 4C),but nifedipine, a blocker of the L-type voltage-dependent Ca2þ

channel (VDCC), did not affect UTP-induced Ca2þ oscillation(Fig. 4D). These data suggest that Ca2þ influx occurs throughthe SOC channel, but not L-type VDCC, and it plays acrucial role in activating and maintaining Ca2þ oscillation.

UTP-induced Ca2þ oscillation was also affected by CBX,a gap junction blocker. CBX enhanced Ca2þ oscillation(Fig. 5A,B) in about 52% of the total cells observed(112/198 cells in ten different experiments from five rats).

Expression of P2Y2 receptors and PCNA at differentcell densities

To investigate the changes that occurred at different celldensities, the correlation between growth conditions and theexpression of P2Y2 receptors was analyzed at three differentcell densities using double-immunostaining with a PCNAantibody and a P2Y2 receptor antibody. The expression of P2Y2

receptors was detected at all densities (Fig. 6A–C). In contrast,the expression of PCNA differed, depending on the cell density(Fig. 6A–C). The percentage of PCNA-positive cells wasgreatest for low-density cells, followed by the medium density,and was almost zero at high cell density (Fig. 6D). These cultureswere prepared using the same protocol as described for [Ca2þ]imeasurement (cells were seeded in different densities on thesame day, and analyzed on the same day of passage), and thus thechange in PCNA expression was not likely due to the growthday, but to cell confluence.


Fig. 6. Immunohistochemistry of P2Y2 receptor and PCNA at (A) low, (B) medium, and (C) high cell densities of rat BMSCs. The expression ofP2Y2 receptorswasdetected in thecell bodies (brown) andPCNA expression wasdetected in thenuclei (black).P2Y2 receptors were expressed atallcelldensities.Almostallofthecellsat lowdensitywerestronglyPCNA-positive,andthepercentageofPCNA-positivecellsdecreasedatmediumdensity.Athighcelldensity,mostofcellswerePCNA-negative.D:QuantitativedataforPCNA-positivecells: low,0.5 W 0.1%;medium,54.0 W 7.3%;high, 93.3 W 2.9%. Data are presented as the mean W SEM from at least 400 cells for each cell density and were analyzed by ANOVA (MP < 0.01).E: Quantitative mRNA analysis for expression of the P2Y2 receptor. The value indicates the relative ratio compared to the P2Y2 receptorexpression at low density: low, 1.00 W 0.24; medium, 4.40 W 0.95; high, 6.73 W 1.73. Data are presented as the mean W SEM from eight independentexperiments from four different rats and were analyzed by ANOVA (MP < 0.01).

Fig. 7. Co-application of FCS (0.1%) and UTP (10 mM) inducesCa2R oscillation at high density. Typical responses from two cells areshown. The application of UTP or FCS alone did not induce Ca2R

oscillation, but co-application of FCS and UTP induced Ca2R

oscillation in 63% of the cells examined (120/192 cells in 12 differentexperiments from six rats, shown as a solid line). In the rest of cellsexamined, co-application of FCS and UTP did not induce Ca2R

oscillation (dotted line).


To investigate the change in the mRNA expression level ofthe P2Y2 receptor at different cell densities, we further carriedout quantitative real-time PCR analysis. The mean� SEM of therelative mRNA expression compared to the P2Y2 receptorexpression at low density were 1.00� 0.24 at low density,4.40� 0.95 at medium density, and 6.73� 1.73 at high density(Fig. 6E). Significant difference was found among low and highdensities (P< 0.01, ANOVA). The expression level among lowand medium/medium and high densities were not significantlydifferent (P> 0.01, ANOVA), but the relative ratio increased asthe cell density increased. These results indicate that the mRNAexpression of the P2Y2 receptor increased at higher celldensity.

Co-application of FCS and UTP induces Ca2R oscillation

The difference in cell density must be related closely to the cellgrowth condition of each cell. The Ca2þ oscillation observed inthe medium density cultures may also be related to the cellgrowth. The possibility that FCS, a potent growth inducer, mayaffect Ca2þ oscillation in rat BMSCs was investigated. FCS wastested at 0.1% because at higher concentrations, FCS affectedthe Ca2þ-fluorescence intensity and masked the changes in[Ca2þ]i for reasons unknown. We prepared the high densityculture, and ensured that UTP induced an increase in [Ca2þ]i

but not Ca2þ oscillation. Unexpectedly, co-application of FCSand UTP induced Ca2þ oscillation in 63% of the cells examined(120/192 cells in 12 different experiments from six rats)(Fig. 7). Ca2þ oscillation was not induced by the singleapplications of either FCS or UTP (Fig. 7). These results indicatethat FCS is the good inducer of Ca2þ oscillation with UTP.


Discussion

In this report, we provide evidence for functional P2Y2

receptors in rat BMSCs. Moreover, no other subtype of P2Xor P2Y receptors appear to be expressed, since all cellsshowed the P2Y2 receptor-specific [Ca2þ]i responses.


Ca2þ signaling via P2Y2 receptors in rat BMSCs changedaccording to cell density. To exclude the potential influence ofcell culture day and truly compare the effect of cell density,all cells were seeded on the same day and measured at thesame day of passage. The subtype of P2 receptors remainedunchanged across all densities. [Ca2þ]i response to UTP wasstronger at high cell density. The increase in [Ca2þ]i didnot come from changes in releasable Ca2þ stores, as thiswas constant among all cell densities. The analysis ofconcentration–response curves suggests that cell density affectboth the amount and the character (e.g., ligand affinity, thedownstream G protein/PLCb/IP3/Ca2þ pathway, and thethreshold number of receptors need to elicit the releaseof Ca2þ from stores, etc.) of P2Y2 receptors by a direct orindirect mechanism. Quantitative real-time PCR analysisdemonstrated that the mRNA expression of the P2Y2 receptorincreased as cell density increased. Cell density must affect atleast the amount of P2Y2 receptors. The amount of receptorsmay be in part responsible for the cell density-dependentchanges in Ca2þ responses.

Note that UTP only induced Ca2þ oscillation at the mediumcell density. These observations were restricted to UTPconcentrations within the range of 10–100 mM. Suchconcentration-dependent occurrence of Ca2þ oscillation wassimilar to an experiment involving ATP-induced Ca2þ

oscillation in HeLa cells (Missiaen et al., 2004) and pancreaticduct epithelial cells (Jung et al., 2004). As explained byShuttleworth and Mignen (2003), concentrations of agonist thatare too high must induce sustained elevation in [Ca2þ]i via SOCchannels, and reduced the amplitude of Ca2þ oscillation.Why did Ca2þ oscillation occur in only the medium density ratBMSCs? Ca2þ oscillation is regulated by a complex mechanism,including Ca2þ release from stores by IP3, Ca2þ influx, and Ca2þ

uptake by Ca2þ pumps (Keizer et al., 1995; Berridge, 2007). Weshowed that Ca2þ influx via SOC channels is important formaintaining Ca2þ oscillation as shown by Bird and Putney(2005). SOC was not the medium density-specific mechanismbecause thapsigargin-induced SOC was observed at all celldensities (J. Ichikawa, unpublished observations). The balanceof complex mechanisms consisting of Ca2þ release, SOC,and Ca2þ uptake at medium cell density may be appropriate forinducing Ca2þ oscillation.

Ca2þ oscillation was not synchronized and its frequency wasvaried. Most of the oscillating cells were locatedsporadically. Ca2þ wave to neighboring cells was observed, butthe rate was only 10% of all oscillating cells. CBX, a gap junctionblocker, enhanced UTP-induced Ca2þ oscillation. Theseobservations are not consistent with previous studies, whichreport that in many types of cells, Ca2þ oscillation wassynchronized and sometimes showed Ca2þ waves and that itwas suppressed by gap junction blockers (Stauffer et al., 1993;Deutsch et al., 1995; Fanchaouy et al., 2005). We suggest thatformation of connexin hemichannels is insufficient, and thatsmall molecules (Ca2þ and IP3) do not permeate to neighboringcells through the gap junction and failed to synchronize Ca2þ

oscillation in rat BMSCs. The enhancement of Ca2þ oscillationby CBX in rat BMSCs may also be explained by the samereasoning; that is, Ca2þ and IP3 cannot escape to neighboringcells through the gap junction, and IP3-induced Ca2þ oscillationof the individual cell is enhanced.

Kawano et al. (2006) reported the existence ofspontaneous Ca2þ oscillation in undifferentiated humanBMSCs. They showed that spontaneous Ca2þ oscillation wasinduced by the ATP autocrine/paracrine system and regulatedthe nuclear factor of activated T-cell (NFAT) translocationinto the nucleus (Kawano et al., 2006). We could not seespontaneous Ca2þ oscillation at any density of culture in ratBMSCs. We suggest that this difference in spontaneous Ca2þ

oscillation may come from the difference in species (human or


rat), or from culture/experimental conditions. Rat BMSCsthat have differentiated into adipocytes or osteoblasts havesometimes shown spontaneous Ca2þ oscillation (J. Ichikawa,unpublished observations). Spontaneous Ca2þ oscillation mayregulate gene expression associated with differentiation inrat BMSCs.

Ca2þ signaling is involved in various cellular functionsincluding fertilization, secretion, contraction, genetranscription, cell survival, cell proliferation, and cell death(Berridge et al., 2003). Different spatiotemporal patternsof Ca2þ signaling regulate different physiological events, such asdifferential gene transcription in the same cell (Berridge, 1997).For example, exocytosis was induced by sustained [Ca2þ]ielevation due to a high concentration of UTP, but not by Ca2þ

oscillation due to low UTP concentration in pancreatic ductepithelial cells (Jung et al., 2006). In rat BMSCs, thedifferent Ca2þ signaling must regulate various physiologicalevents at each cell density. Gregory et al. (2003) reported thatdickkopf-1 (Dkk-1), an inhibitor of the canonical Wnt signalingpathway, is expressed highly in the early log phase ofproliferation, but decreases in stationary phase in humanmesenchymal stem cells. They further reported that theaddition of Dkk-1 into the stationary phase cultures inducedproliferation, and that the addition of antibody to Dkk-1 atthe log phase suppressed proliferation (Gregory et al., 2003).Dkk-1 must closely regulate cell growth of BMSCs. It must bedeliberated in the discussion of our results because unlike ourexperiments, the results from Gregory et al. (2003) includedthe potential implication of the day of culture. However,investigating the cell density-dependent change in the secretionof Dkk-1 and the possible implication on P2Y2 receptorsignaling in rat BMSCs would be worthwhile because canonicalWnt pathway is regulated by increasing of [Ca2þ]i (Maye et al.,2004).

What is the physiological role of UTP-induced Ca2þ

oscillation observed at medium cell density of rat BMSCs? Thefact that UTP-induced Ca2þ oscillation occurred partly (42%) atmedium cell density may be related to the cell cycle. At mediumdensity, 54% of the total cells expressed PCNA, whereas theexpression at low and high density was nearly all or none. Thisresult suggests that the cells at medium density are in variousstages of the cell cycle. UTP-induced Ca2þoscillation may occurin a particular stage of the cell cycle, and may contribute tothe cell cycle progression in rat BMSCs. Kapur et al. (2007)reported that mouse embryonic stem cells showedspontaneous Ca2þ oscillation in 70% of G1/S phase cells, but inonly 15% of G2/M cells. They proposed that changes in IP3

receptor sensitivity or basal levels of IP3 induced theG1/S-confined Ca2þ oscillations and that IP3-mediatedCa2þ release drove the cell cycle progression through G1/S(Kapur et al., 2007). UTP-induced Ca2þ oscillation in rat BMSCsmay also be related to the cell cycle by changing theIP3-mediated signaling mechanism.

Several examples indicate that cell density must affect thephysiological conditions in some types of cells, but also thatthese aspects are multiple. IP3 receptor type 1 was localized inthe membrane region in the confluent state, and yet it waslocalized in the cytoplasm in the subconfluent state inMadin–Darby canine kidney (MDCK) cells (Zhang et al., 2003).In keratinocytes, the distribution of aryl hydrocarbon receptor(AhR), a transcription factor, is regulated by cell densityassociated with Ca2þ-dependent cell–cell contact (Ikuta et al.,2004). The level of reactive oxygen species (ROS) also changesdepending on the cell density in neural precursor cells of thehippocampus, which is associated with proliferation andmetabolic activity due to altered mitochondrial function (Limoliet al., 2004). In osteoblast cell lines, various Ca2þ responseswere reduced by cell confluence associated with the decreasein IP3 receptors and intracellular Ca2þ stores (Koizumi et al.,


2003). Sekiya et al. (2002) reported that seeding density andculture day of human BMSCs affected the content and potentialfor differentiation of the progenitor cells. Cell density mustbe important for BMSCs. In rat BMSCs, the percentage ofPCNA-positive cells varied with cell density. As described, thiswas not due to a difference in the day of culture. Cultures ateach cell density were stained at 4 days after seeding. Almost allcells at low density were PCNA-positive, but the percentagedecreased at medium cell density and all cells at the high celldensity were PCNA-negative. Our results indicate that almostall of the cells at low density were in the proliferation state,which went partially to the quiescent state of the cell cycle atmedium density, and they stopped proliferating at the high celldensity. This suggests that cell density closely regulates the cellcycle progression in rat BMSCs. Moreover, cell confluenceinhibits cell growth. In the developing cerebral cortex,proliferation of neural precursors was precisely regulated bycell density (Lien et al., 2006). Neural precursors expresscadherin at the cell surface, and cadherin is a transmembraneprotein that forms homophilic bonds between the surfaces ofneighboring cells. The cytoplasmic domain of cadherin binds tothe a- and b-catenin complex. When cells become confluentand are connected by cadherin, a-catenin is releasedand inhibits Hedgehog signaling and cell proliferation(DiCicco-Bloom, 2006). BMSCs may use a similar mechanismbecause BMSCs also express cadherin (Turel and Rao, 1998).In rat BMSCs, cell density affected cell cycle progression andphysiological changes such as UTP-induced [Ca2þ]i responses,including Ca2þ oscillation.

We showed that bath application of FCS with UTPinduces Ca2þ oscillation at high cell density, and Ca2þ

oscillation was not observed by applying either UTP or FCSalone. FCS may stimulate the Ca2þ oscillation-drivingmechanism that was lost at high cell density. FCS is a potentmitogen for all kinds of cells, so the effect of FCS on Ca2þ

oscillation must be strongly correlated with cell growth. Otherstudies have indicated the effect of FCS on Ca2þ oscillation.Fu et al. (1991) reported that FCS induced Ca2þ oscillation inras-transformed NIH3T3 cells that can grow in underlow-serum conditions. They suggested that Ca2þ oscillation isneeded to sustain efficient growth signals. Morita et al. (2003)reported the relationship between Ca2þ oscillation andgrowth factors in astrocytes. Astrocytes cultured in 10%FCS-containing medium showed various patterns of [Ca2þ]i

signals, including Ca2þ oscillation in response to glutamate orATP, and the Ca2þ oscillation occurred in almost all astrocytescultured in defined medium containing growth factors(Morita et al., 2003). They suggested that Ca2þ oscillationwas accompanied by cell proliferation and regulated via themitogen-activated protein kinase (MAPK) cascade (Moritaet al., 2003). In tumor (Bryant et al., 2004) and endothelial cells(Moccia et al., 2003), epidermal growth factor induced Ca2þ

oscillation. Similar regulation of Ca2þ oscillation by FCS andgrowth factors must exist in rat BMSCs. Many studies havereported that Ca2þ signaling is essential for cell proliferation(Berridge et al., 2003; Lipskaia and Lompre, 2004), and FCSmust drive the cell-cycle machinery again at the high cell densitywith Ca2þ oscillation in rat BMSCs.

At present, some physiological studies on human BMSCshave shown the existence of various ion channels and receptorsthat control the intracellular ionic environment, for example,L-type Ca2þ channels (Heubach et al., 2004; Li et al., 2005;Zahanich et al., 2005), Kþ channels (Heubach et al., 2004;Li et al., 2005), Naþ channels (Li et al., 2005), TRP4 channels(Kawano et al., 2002), and P2X and P2Y1 receptors (Kawanoet al., 2006; Coppi et al., 2007). However, these studies havereported that not all of the cells observed expressed thesechannels. L-type Ca2þ current was observed in 15% and Naþ

current was observed in 29% of the recorded cells (Li et al.,


2005). Heubach et al. (2004) revealed the heterogeneity of ioncurrents in human BMSCs and discussed the possibility that thecells they prepared did not come from a homogeneouspopulation. Sorting the absolutely homogeneous population ofundifferentiated BMSCs is difficult because a perfect markerhas not yet been established. They also discussed anotherpossibility for heterogeneity, that is, various cell cycle stages(Heubach et al., 2004). We agree with their conclusion.

This is the first study that focused on cell density andrevealed the changes in Ca2þ mobilization in rat BMSCs.Future investigations are needed to examine thecell-density-dependent changes in other ion channels, asdescribed above, with the combined use of cell-cycle-specificmarkers.

What is the source of nucleotides that activates P2Y2

receptors of rat BMSCs in vivo? Kawano et al. (2006) reportedthat human mesenchymal stem cells secreted ATP via gapjunction. It is necessary to measure the secretion of ATP fromrat BMSCs, with considering the possibility that hematopoieticcells co-existing with BMSCs in bone marrow also secrete ATPand regulate BMSCs in vivo. To reveal the mutual interactionbetween BMSCs and hematopoietic cells by autocrine/paracrine signaling of ATP, the experiment for the visualizationof ATP secretion from a single cell and following activationof surrounding cells using co-culture system will be needed.

Acknowledgments

This study was supported by the research grant (D) from KansaiMedical University and Takeda Science Foundation.

Literature Cited

Berridge M. 1997. The AM and FM of calcium signalling. Nature 386:759–760.Berridge M. 2007. Inositol trisphosphate and calcium oscillations. Biochem Soc Symp 74:1–7.Berridge M, Bootman M, Roderick H. 2003. Calcium signalling: Dynamics, homeostasis and

remodelling. Nat Rev Mol Cell Biol 4:517–529.Bird G, Putney JJ. 2005. Capacitative calcium entry supports calcium oscillations in human

embryonic kidney cells. J Physiol 562:697–706.Bryant J, Finn R, Slamon D, Cloughesy T, Charles A. 2004. EGF activates intracellular

and intercellular calcium signaling by distinct pathways in tumor cells. Cancer Biol Ther 3:1243–1249.

Burnstock G. 2006. Historical review: ATP as a neurotransmitter. Trends Pharmacol Sci27:166–176.

Burnstock G. 2007. Physiology and pathophysiology of purinergic neurotransmission. PhysiolRev 87:659–797.

Burnstock G, Knight G. 2004. Cellular distribution and functions of P2 receptor subtypes indifferent systems. Int Rev Cytol 240:31–304.

Byrnes J, Downey K, Black V, So A. 1976. A new mammalian DNA polymerase with 30 to 50

exonuclease activity: DNA polymerase delta. Biochemistry 15:2817–2823.Celis J, Celis A. 1985. Cell cycle-dependent variations in the distribution of the nuclear

protein cyclin proliferating cell nuclear antigen in cultured cells: Subdivision of S phase. ProcNatl Acad Sci USA 82:3262–3266.

Chen Y, Corriden R, Inoue Y, Yip L, Hashiguchi N, Zinkernagel A, Nizet V, Insel P, Junger W.2006. ATP release guides neutrophil chemotaxis via P2Y2 and A3 receptors. Science314:1792–1795.

Coppi E, Pugliese A, Urbani S, Melani A, Cerbai E, Mazzanti B, Bosi A, Saccardi R, Pedata F.2007. ATP modulates cell proliferation and elicits two different electrophysiologicalresponses in human mesenchymal stem cells. Stem Cells 25:1840–1849.

Deutsch D, Williams J, Yule D. 1995. Halothane and octanol block Ca2þ oscillations inpancreatic acini by multiple mechanisms. Am J Physiol 269:G779–788.

DiCicco-Bloom E. 2006. Neuroscience Neuron, know thy neighbor. Science 311:1560–1562.Dorshkind K. 1990. Regulation of hemopoiesis by bone marrow stromal cells and their

products. Annu Rev Immunol 8:111–137.Fanchaouy M, Serir K, Meister J, Beny J, Bychkov R. 2005. Intercellular communication: Role

of gap junctions in establishing the pattern of ATP-elicited Ca2þ oscillations and Ca2þ-dependent currents in freshly isolated aortic smooth muscle cells. Cell Calcium 37:25–34.

Frankel M. 2000. In search of stem cell policy. Science 287:1397.Fu T, Sugimoto Y, Oki T, Murakami S, Okano Y, Nozawa Y. 1991. Calcium oscillation

associated with reduced protein kinase C activities in ras-transformed NIH3T3 cells. FEBSLett 281:263–2266.

Gallagher C, Salter M. 2003. Differential properties of astrocyte calcium waves mediated byP2Y1 and P2Y2 receptors. J Neurosci 23:6728–6739.

Gregory C, Singh H, Perry A, Prockop D. 2003. The Wnt signaling inhibitor dickkopf-1 isrequired for reentry into the cell cycle of human adult stem cells from bone marrow. J BiolChem 278:28067–28078.

Heubach J, Graf E, Leutheuser J, Bock M, Balana B, Zahanich I, Christ T, Boxberger S,Wettwer E, Ravens U. 2004. Electrophysiological properties of human mesenchymal stemcells. J Physiol 554:659–672.

Ichikawa J, Furuya K, Miyata S, Nakashima T, Kiyohara T. 2000. EGF enhances Ca2þ

mobilization and capacitative Ca2þ entry in mouse mammary epithelial cells. Cell BiochemFunct 18:215–225.

Ikuta T, Kobayashi Y, Kawajiri K. 2004. Cell density regulates intracellular localization of arylhydrocarbon receptor. J Biol Chem 279:19209–19216.


Jung S, Kim M, Hille B, Nguyen T, Koh D. 2004. Regulation of exocytosis by purinergicreceptors in pancreatic duct epithelial cells. Am J Physiol Cell Physiol 286:C573–C579.

Jung S, Kim K, Hille B, Nguyen T, Koh D. 2006. Pattern of Ca2þ increase determines the typeof secretory mechanism activated in dog pancreatic duct epithelial cells. J Physiol 576:163–178.

Kapur N, Mignery G, Banach K. 2007. Cell cycle-dependent calcium oscillations in mouseembryonic stem cells. Am J Physiol Cell Physiol 292:C1510–C1518.

Kawano S, Shoji S, Ichinose S, Yamagata K, Tagami M, Hiraoka M. 2002. Characterizationof Ca2þ signaling pathways in human mesenchymal stem cells. Cell Calcium 32:165–174.

Kawano S, Otsu K, Kuruma A, Shoji S, Yanagida E, Muto Y, Yoshikawa F, Hirayama Y,Mikoshiba K, Furuichi T. 2006. ATP autocrine/paracrine signaling induces calciumoscillations and NFAT activation in human mesenchymal stem cells. Cell Calcium 39:313–324.

Keizer J, Li Y, Stojilkovic S, Rinzel J. 1995. InsP3-induced Ca2þ excitability of the endoplasmicreticulum. Mol Biol Cell 6:945–951.

Koizumi T, Hikiji H, Shin W, Takato T, Fukuda S, Abe T, Koshikiya N, Iwasawa K, Toyo-oka T.2003. Cell density and growth-dependent down-regulation of both intracellular calciumresponses to agonist stimuli and expression of smooth-surfaced endoplasmic reticulum inMC3T3-E1 osteoblast-like cells. J Biol Chem 278:6433–6439.

Koizumi S, Shigemoto-Mogami Y, Nasu-Tada K, Shinozaki Y, Ohsawa K, Tsuda M, Joshi B,Jacobson K, Kohsaka S, Inoue K. 2007. UDP acting at P2Y6 receptors is a mediator ofmicroglial phagocytosis. Nature 446:1091–1095.

Krebsbach P, Kuznetsov S, Bianco P, Robey P. 1999. Bone marrow stromal cells:Characterization and clinical application. Crit Rev Oral Biol Med 10:165–181.

Lambrecht G, Braun K, Damer M, Ganso M, Hildebrandt C, Ullmann H, Kassack M, Nickel P.2002. Structure-activity relationships of suramin and pyridoxal-50-phosphate derivatives asP2 receptor antagonists. Curr Pharm Des 8:2371–2399.

Lazarowski E, Boucher R. 2001. UTP as an extracellular signaling molecule. News Physiol Sci16:1–5.

Li G, Sun H, Deng X, Lau C. 2005. Characterization of ionic currents in human mesenchymalstem cells from bone marrow. Stem Cells 23:371–382.

Lien W, Klezovitch O, Fernandez T, Delrow J, Vasioukhin V. 2006. AlphaE-catenin controlscerebral cortical size by regulating the hedgehog signaling pathway. Science 311:1609–1612.

Limoli C, Rola R, Giedzinski E, Mantha S, Huang T, Fike J. 2004. Cell-density-dependentregulation of neural precursor cell function. Proc Natl Acad Sci USA 101:16052–16057.

Lipskaia L, Lompre A. 2004. Alteration in temporal kinetics of Ca2þ signaling and control ofgrowth and proliferation. Biol Cell 96:55–68.

Maaser K, Hopfner M, Kap H, Sutter A, Barthel B, von Lampe B, Zeitz M, Scherubl H. 2002.Extracellular nucleotides inhibit growth of human oesophageal cancer cells via P2Y2-receptors. Br J Cancer 86:636–644.

Maniatopoulos C, Sodek J, Melcher A. 1988. Bone formation in vitro by stromal cells obtainedfrom bone marrow of young adult rats. Cell Tissue Res 254:317–330.

Maye P, Zheng J, Li L, Wu D. 2004. Multiple mechanisms for Wnt11-mediated repression ofthe canonical Wnt signaling pathway. J Biol Chem 279:24659–24665.

McPherson P, Campbell K. 1993. The ryanodine receptor/Ca2þ release channel. J Biol Chem268:13765–13768.

Merritt J, Armstrong W, Benham C, Hallam T, Jacob R, Jaxa-Chamiec A, Leigh B, McCarthy S,Moores K, Rink T. 1990. SK&F 96365, a novel inhibitor of receptor-mediated calciumentry. Biochem J 271:515–522.


Missiaen L, Van Acker K, Van Baelen K, Raeymaekers L, Wuytack F, ParysJ, De Smedt H, Vanoevelen J, Dode L, Rizzuto R, Callewaert G. 2004. Calciumrelease from the Golgi apparatus and the endoplasmic reticulum in HeLa cells stablyexpressing targeted aequorin to these compartments. Cell Calcium36:479–487.

Moccia F, Berra-Romani R, Tritto S, Signorelli S, Taglietti V, Tanzi F. 2003. Epidermal growthfactor induces intracellular Ca2þ oscillations in microvascular endothelial cells. J CellPhysiol 194:139–150.

Montecino-Rodriguez E, Dorshkind K. 2001. Regulation of hematopoiesis by gap junction-mediated intercellular communication. J Leukoc Biol 70:341–347.

Morita M, Higuchi C, Moto T, Kozuka N, Susuki J, Itofusa R, Yamashita J, Kudo Y. 2003. Dualregulation of calcium oscillation in astrocytes by growth factors and pro-inflammatorycytokines via the mitogen-activated protein kinase cascade. J Neurosci23:10944–10952.

Moskvina E, Unterberger U, Boehm S. 2003. Activity-dependent autocrine-paracrineactivation of neuronal P2Y receptors. J Neurosci 23:7479–7488.

Prelich G, Kostura M, Marshak D, Mathews M, Stillman B. 1987. The cell-cycle regulatedproliferating cell nuclear antigen is required for SV40 DNA replication in vitro. Nature326:471–475.

Prockop D. 1997. Marrow stromal cells as stem cells for nonhematopoietic tissues. Science276:71–74.

Schwarz B, Bhandoola A. 2006. Trafficking from the bone marrow to the thymus: Aprerequisite for thymopoiesis. Immunol Rev 209:47–57.

Schwiebert E, Fitz J. 2008. Purinergic signaling microenvironments: An introduction.Purinergic Signal 4:89–92.

Sekiya I, Larson B, Smith J, Pochampally R, Cui J, Prockop D. 2002. Expansion of humanadult stem cells from bone marrow stroma: Conditions that maximize the yields ofearly progenitors and evaluate their quality. Stem Cells 20:530–541.

Shuttleworth T, Mignen O. 2003. Calcium entry and the control of calcium oscillations.Biochem Soc Trans 31:916–919.

Stauffer P, Zhao H, Luby-Phelps K, Moss R, Star R, Muallem S. 1993. Gap junctioncommunication modulates [Ca2þ]i oscillations and enzyme secretion in pancreatic acini.J Biol Chem 268:19769–19775.

Thastrup O, Cullen P, Drøbak B, Hanley M, Dawson A. 1990. Thapsigargin, a tumorpromoter, discharges intracellular Ca2þ stores by specific inhibition of the endoplasmicreticulum Ca2þ-ATPase. Proc Natl Acad Sci USA 87:2466–2470.

Tsurimoto T. 1999. PCNA binding proteins. Front Biosci 4:D849–D858.Turel K, Rao S. 1998. Expression of the cell adhesion molecule E-cadherin by the human bone

marrow stromal cells and its probable role in CD34þ stem cell adhesion. Cell Biol Int22:641–648.

White N, Ryten M, Clayton E, Butler P, Burnstock G. 2005. P2Y purinergic receptors regulatethe growth of human melanomas. Cancer Lett 224:81–91.

Zahanich I, Graf E, Heubach J, Hempel U, Boxberger S, Ravens U. 2005. Molecular andfunctional expression of voltage-operated calcium channels during osteogenicdifferentiation of human mesenchymal stem cells. J Bone MinerRes 20:1637–1646.

Zhang S, Mizutani A, Hisatsune C, Higo T, Bannai H, Nakayama T, Hattori M, Mikoshiba K.2003. Protein 4.1N is required for translocation of inositol 1,4,5-trisphosphate receptortype 1 to the basolateral membrane domain in polarized Madin-Darby canine kidney cells.J Biol Chem 278:4048–4056.

Zimmermann H. 1996. Extracellular purine metabolism. Drug Dev Res 39:337–352.

cell density-dependent changes in intracellular ca2+ mobilization via the p2y2 receptor in rat bone...

Documents