Review

Intracellular Ca2� release mechanisms: multiple pathways havingmultiple functions within the same cell type?

Cristina P. da Silva, Andreas H. Guse *University of Hamburg, University Clinic Hamburg-Eppendorf, Institute for Medical Biochemistry and Molecular Biology,

Division of Cellular Signal Transduction, Grindelallee 117, D-20146 Hamburg, Germany

Received 11 September 2000; accepted 12 September 2000

Abstract

The elevation of the cytosolic and nuclear Ca2� concentration is a fundamental signal transduction mechanism in almostall eukaryotic cells. Interestingly, three Ca2�-mobilising second messengers, D-myo-inositol 1,4,5-trisphosphate (InsP3), cyclicadenosine diphosphoribose (cADPR), and nicotinic acid adenine dinucleotide phosphate (NAADP�) were identified in aphylogenetically wide range of different organisms. Moreover, in an as yet very limited number of cell types, sea urchin eggs,mouse pancreatic acinar cells, and human Jurkat T-lymphocytes, all three Ca2�-mobilising ligands have been shown to beinvolved in the generation of Ca2� signals. This situation raises the question why during evolution all three messengers havebeen conserved in the same cell type. From a theoretical point of view the following points may be considered: (i) redundantmechanisms ensuring intact Ca2� signalling even if one system does not work, (ii) the need for subcellularly localised Ca2�

elevations to obtain a certain physiological response of the cell, and (iii) tight control of a physiological response of the cell bya temporal sequence of Ca2� signalling events. These theoretical considerations are compared to the current knowledgeregarding the three messengers in sea urchin eggs, mouse pancreatic acinar cells, and human Jurkat T lymphocytes. ß 2000Elsevier Science B.V. All rights reserved.

Keywords: Signal transduction; Calcium signalling; Inositol 1,4,5-trisphosphate; Cyclic ADP-ribose;Nicotinic acid adenine dinucleotide phosphate

1. Introduction

One of the fundamental intracellular signallingmechanisms that can be found in almost all eukary-

otic organisms is the elevation of the intracellularCa2� concentration ([Ca2�]i). In the normal, non-ac-tivated state of a cell ATP-driven Ca2� pumps de-crease the concentration of free Ca2� ions in the

0167-4889 / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved.PII: S 0 1 6 7 - 4 8 8 9 ( 0 0 ) 0 0 0 8 9 - 6

Abbreviations: [Ca2�]i, free intracellular Ca2� concentration; cADPR, cyclic adenosine diphosphoribose; CaM, calmodulin; cGMP,3P,5P-cyclic guanosine monophosphate; CICR, calcium-induced calcium release; EC50, half-maximal e¡ective concentration; ER, endo-plasmic reticulum; InsP3, D-myo-inositol 1,4,5-trisphosphate; InsP3R, D-myo-inositol 1,4,5-trisphosphate receptor(s); NAADP�, nicotinicacid adenine dinucleotide phosphate; NADase, NAD�-glycohydrolase; NAD(P)�, nicotinamide adenine dinucleotide (phosphate); Pi,inorganic phosphate; PKA, cAMP-dependent protein kinase; PKC, protein kinase C; PKG, cGMP-dependent protein kinase; PLC,phospholipase C; PtdInsP2, phospatidylinositol 4,5-bisphosphate; RyR, ryanodine receptor(s); TCR/CD3 complex, T-cell receptor/CD3complex

* Corresponding author. Fax: +49-40-42838-4275; E-mail : guse@uke.uni-hamburg.de

BBAMCR 14673 27-11-00

Biochimica et Biophysica Acta 1498 (2000) 122^133www.elsevier.com/locate/bba

cytosol and the nucleus to values of 100 nM or be-low. Thereby, these Ca2�-ATPases keep up highCa2�-gradients between the cytosol and both theextracellular space and the intracellular Ca2� stores.Activation of the cell by an external input signaloften results in an elevation of [Ca2�]i by openingof Ca2� channels located either in membranes ofintracellular Ca2� stores or in the plasma membrane.In this review we shall focus on the role of Ca2�

release from intracellular Ca2� stores.Currently, at least three Ca2�-mobilising intracel-

lular ligands are known. D-myo-Inositol 1,4,5-tris-phosphate has been discovered by Schulz andBerridge [1], while both cyclic adenosine diphosphor-ibose (cADPR) and nicotinic acid adenine dinucleo-tide phosphate (NAADP�) have been described forthe ¢rst time by Lee and co-workers [2^4]. BothInsP3 and cADPR have been accorded full statusas second messengers in di¡erent cell systems, mainlybased on the fact that, in addition to their intracel-lular Ca2�-release activity, an increase in their intra-cellular concentration upon stimulation of receptorsin the plasma membrane was observed. A similarresult has not yet been published for NAADP�,although enzymatic formation of NAADP� wasdemonstrated [5,6].

In addition to the three compounds mentionedabove, the lipids lysophosphatidic acid and phospha-tidic acid released Ca2� from intracellular stores inpermeabilised cells (Jurkat T cells [7,8]). The e¡ect ofthese lipids was independent of InsP3 or cADPR[7,8]. However, lysophosphatidic acid also exertedCa2�-releasing e¡ects on intact cells [8,9], raisingthe question whether it primarily acts on plasmamembrane-bound receptors or on intracellular tar-gets or on both. Because of these uncertainties, wewill not include the Ca2�-mobilising lipids in thisreview.

2. Why did evolution conserve three (or more)Ca2+-mobilising intracellular ligands within thesame cell type? Theoretical considerations

Since many aspects of the physiological role of thethree Ca2�-mobilising messengers InsP3, cADPR,and NAADP� in di¡erent cell systems are not yetclari¢ed, it is very di¤cult to say why at least the

InsP3- and the cADPR-system have been conservedbetween unicellular protozoa, e.g., Euglena gracilis[10] and many mammalian cells, including cells ofhuman origin (reviewed in [11,12]). For NAADP�

a role in lower, unicellular eukaryotes has not yetbeen described, but the fact that the NAADP� sys-tem appears to be highly conserved between sea ur-chin eggs [5], ascidian eggs [13], pancreatic acinarcells [6] and human Jurkat T cells [90] indicatesthat it may be similarly widespread as has been dem-onstrated for the InsP3- and cADPR-system.

The reasons for such long conservation of thesedi¡erent Ca2�-mobilising systems during evolutionare di¤cult to assess, but it may be worth consider-ing the following points.

In a simpli¢ed view, one may regard the di¡erentCa2�-mobilising second-messenger systems to end upin the very same ¢nal result, a global Ca2� responsethat can be seen throughout the entire cell, althoughwith cell type speci¢c di¡erences. Assuming that this¢nal result, a global Ca2� response, is required for aparticular physiological response, then it may be ofminor interest for the cell which Ca2� releasing sys-tem is used. On the other hand, if the particularphysiological response cannot take place without aglobal Ca2� response, evolution may have conservedtwo (or more) signalling systems (Fig. 1A, (a^c))which are activated in parallel, and therefore mayserve as redundant mechanisms to ensure that thecell can respond to a given extracellular signal.Blockade of one particular signalling system (Fig.1A, (c)) would neither inhibit the development ofthe Ca2� signal nor the physiological response.

Another reason for the use of more than oneCa2�-mobilising second-messenger system would beto elevate [Ca2�]i in di¡erent regions of the cell, e.g.,to activate a physiological response which can becarried out only in a localised area of the cell. Inthis case one would expect a discrete subcellular lo-calisation of second-messenger forming and/or me-tabolising enzymes, and/or of the intracellular Ca2�

stores sensitive to the di¡erent messengers, and eithera localised Ca2� signal or a regional or global Ca2�

wave (Fig. 1B). Such a scenario would also be help-ful for cells which respond to di¡erent input signalsby di¡erent, ¢ne-tuned Ca2� elevations. The latterthen may elicit di¡erent physiological responses.

The temporal aspect of Ca2� signalling may re-

BBAMCR 14673 27-11-00

C.P. da Silva, A.H. Guse / Biochimica et Biophysica Acta 1498 (2000) 122^133 123

quire as well the use of more than one second mes-senger. Two or more endogenous Ca2�-mobilisingcompounds formed (and degraded) in sequencemay o¡er more control points for the cell, e.g., incase that a physiological response is initiated, butneeds to be controlled tightly before a ¢nal decisionto launch a physiological response or not is made(Fig. 1C).

Finally, localised Ca2� responses of di¡erent am-plitude and length may be required for the ¢ne-tun-ing of physiological responses. Thus, from the evolu-tionary point of view di¡erent Ca2�-mobilisingsecond-messenger systems should vary in their signalamplitudes, as well as in temporal and spatial pat-tern.

3. Ca2+ signalling mediated by InsP3, cADPR andNAADP+

There is increasing evidence that internal Ca2�

stores can be mobilised by InsP3, cADPR orNAADP� via totally distinct mechanisms. Each ofthese three compounds and the Ca2� signalling path-ways in which they participate present di¡erent im-portant characteristics, which are brie£y described inthis section. A summary of some of the properties ofthese Ca2�-mobilising compounds and their recep-tors is shown in Table 1.

InsP3 is a well established second messenger in-volved in Ca2� signalling in many types of cells.Under resting conditions, intracellular concentra-tions of InsP3 are very low, but the levels of thiscompound increase rapidly, following stimulationof cell-surface receptors and subsequent activationof the enzyme phospholipase C (L or Q, dependingon the type of cell ; PLC) by G proteins or by recep-tor or non-receptor tyrosine kinases [14^18]. Acti-vated PLC promotes the hydrolysis of the membranephospholipid phospatidylinositol 4,5-bisphosphate(PtdInsP2) with production of the two second mes-sengers InsP3 and 1,2-diacylglycerol. Subsequent in-teraction of InsP3 with its speci¢c receptor ^ the ino-sitol 1,4,5-trisphosphate receptor (InsP3R)/channel,identi¢ed in three isoforms ^ induces the release ofCa2� from intracellular pools located in the endo-plasmic reticulum ([14,19], reviewed in [20]).

Studies performed in di¡erent cell types show thatInsP3Rs are activated by InsP3 in the nanomolar tolow micromolar concentration range [15,21^23] (Ta-ble 1). In addition to InsP3, other physiological mod-ulators of InsP3R-activity have been described,namely Ca2� and ATP, which can stimulate the re-ceptor, and Mg2� which has an inhibitory e¡ect onInsP3R/channel activity ([24,25], reviewed in [20]).The e¡ects of Ca2� on InsP3R-activity, however, de-pend on the subtype of InsP3R expressed, since theopen probability of type 1 InsP3R displays a biphasicdependence on cytoplasmic Ca2� and on calmodulin(CaM), whereas the open probability of type 3 InsP3Rincreases monotonically with increasing Ca2� con-centration [26^28].

Moreover, the activity of the InsP3R can also bemodulated by phosphorylation by cAMP-dependentprotein kinase (PKA) as well as by protein kinase C

Fig. 1. Potential roles for three Ca2�-mobilising messengerswithin the same cell type. The three Ca2�-mobilising messengersystems (a), (b), and (c) either act in a redundant fashion (A),or elevate Ca2� in one or more subcellular regions of the cell(B), or act in a temporal sequence giving the cell several check-points (CP) to decide whether the physiological response is ¢-nally launched or not (C).

BBAMCR 14673 27-11-00

C.P. da Silva, A.H. Guse / Biochimica et Biophysica Acta 1498 (2000) 122^133124

Table 1Characteristic properties of systems for Ca2� release from intracellular stores

Second messenger InsP3 Ref. cADPR Ref. NAADP� Ref.

Synthesisingenzyme

PLC-L,Q [14] ADP-ribosylcyclase; CD38 (?)

[3,32] ADP-ribosyl cyclase/CD38 (?)

[5]

[15] [46] NAD(P)-glyco-hydrolase (?)

[11]

[16] [47][17]

Mechanism ofactivation ofbiosynthesis

G-protein [16] PKG [41,42] ND

Tyr-kinases [17] G-protein [43,44][18] Tyr-kinases [45]

Substrate PtdInsP2 [14] L-NAD� [3] L-NADP� (?) [11]Enzymaticdegradation

InsP3-3-kinase [40] cADPR-hydrolase/NADase

[12,32] Alkalinephosphatase (?)

[11]

InsP3-5-phosphatase [39] Pyrophosphatase (?) [32]Di¡usioncoe¤cient(Wm2/s)

283 [33] ND ND

Intracellularreceptor/channel

InsP3R (1,2,3) [20] RyR2 [50,51] ND

[24] RyR3E¡ectiveconcentration

EC50 = 5 nM^10 WM [14] EC50 = 3.8 nM^10 WM

[11] Low nanomolarrange

[11]

[21] [12][22] EC50 = 10^50 nM [68][23]

Physiologicalmodulators:

activators InsP3, Ca2� [20] cADPR; Ca2� [52^54] NAADP� (nanomolarconcentration)

[5]

[27] CaM [57] [32]ATP [25] Pi [58] [68]PKC [20] Palmitoyl-CoA [60]PKA [20] PKA [55,61]CaM-kinase II [20] PKG [55,61]Tyr-kinase [29] CaM-kinase [55,61]

inhibitors Mg2� [24] Mg2� [56] NAADP� (highconcentration, WM)

[32]

Ca2� (high concentration,type 1 InsP3R)

[26] Spermine and otherpolyamines

[59] [72]

[28] PKC [62]Non-physiologicalmodulators:

activators Adenophostins [32] Ryanodine [11] NDinhibitors Ca¡eine [11]

Heparin [30] Ruthenium red,procaine

[11] L-type Ca2�-channelantagonists

[73]

Xestospongin C [31] [11]8-NH2-cADPR [12] and

[63,64]

BBAMCR 14673 27-11-00

C.P. da Silva, A.H. Guse / Biochimica et Biophysica Acta 1498 (2000) 122^133 125

(PKC) and Ca2�-calmodulin-dependent protein ki-nase II (CaM kinase-II) (reviewed in [20]). Otherstudies also demonstrated regulation of the InsP3Rby tyrosine phosphorylation [29].

Apart from these physiological modulatory e¡ects,other non-physiological compounds have beenshown to a¡ect the activity of InsP3R/Ca2� channelsand are very useful as pharmacological tools forstudies on InsP3R-mediated Ca2� signalling. Amongsuch compounds, heparin and xestospongin C areused as potent speci¢c antagonists of the InsP3Rand InsP3-mediated Ca2� release [30,31], whereas ad-enophostins are used as activators of this system (re-viewed in [32]).

Studies of the conduction properties of the InsP3-gated Ca2� channel showed that, under physiologicalconditions, the Ca2� current through the InsP3R is0.5 pA, which is about four times less than the Ca2�

current through the ryanodine receptor (RyR) [24].The di¡usion coe¤cient of InsP3 as measured in

cytosolic extracts from Xenopus laevis oocytes [33] is283 Wm2/s, which corresponds to an e¡ective range ofmessenger action of about 24 Wm. When comparedwith corresponding values obtained for Ca2� (13^65Wm2/s and 0.1^5 Wm, depending on the free Ca2�

concentration), these parameters make InsP3 agood candidate for the mobile messenger in Ca2�

signalling in most cells.Release of Ca2� from InsP3-sensitive intracellular

stores can give rise to elementary, highly localisedand short-duration events of Ca2� signalling calledpu¡s, which individually contribute a quantalamount of Ca2�, as described in Xenopus oocytes

[34], HeLa cells [35], smooth muscle [36] or pancre-atic acinar cells [37]. The spatio-temporal propertiesof Ca2� pu¡s arising from InsP3R have been studiedand are characterised by a duration of about 500 ms,reaching ranges of 4^6 Wm (diameter of Ca2� signal)and maximal amplitudes of elevation of [Ca2�]i of140 nM [38]. Global Ca2� signals result from thecoordination and summation of such elementaryCa2� release units to produce regenerative waves ofCa2� that spread throughout the cells, whereas local-isation of these Ca2� release units to a particularregion of the cell can give rise to spatially-restrictedCa2� signals.

Degradation of InsP3 is carried out by the enzymeIns(1,4,5,)P3 3-kinase, which is activated by Ca2�/cal-modulin, and/or by Ins(1,4,5,)P3 5-phosphatase[39,40].

cADPR is a cyclic derivative of L-nicotinamideadenine dinucleotide (NAD�) which is synthesisedby ADP-ribosyl cyclase ([3], reviewed in [11,12,32]).ADP-ribosyl cyclases are found both in membrane-bound and soluble forms and the latter can be acti-vated by a cGMP-dependent protein kinase, as de-scribed in sea urchin eggs and PC12 cells [41,42] orby G-protein coupled [43,44] or tyrosine-kinasecoupled receptors [45]. The membrane-bound ADP-ribosyl cyclases usually have a prominent NAD� gly-cohydrolase (NADase) activity and, in a side reac-tion, catalyse the synthesis of cADPR [46]. The sur-face antigen CD38, mainly expressed on activated Tand B lymphocytes, natural killer cells and erythro-cytes and in minor levels in other major organs, alsoshows NADase activity and a minor ADP-ribosyl

Table 1 (continued)

Second messenger InsP3 Ref. cADPR Ref. NAADP� Ref.

8-OCH3-cADPR8-Bromo-cADPR7-Deaza-8-bromo-cADPR

Channel conduction(pA)

0.5 [24] 2.0 [24] ND

Elementary Ca2�

signalsPu¡s [38] Sparks [38] ND

duration (ms) V500 6 200 NDrange (Wm) 4^6 3 NDamplitude (nM) 140 170 ND

BBAMCR 14673 27-11-00

C.P. da Silva, A.H. Guse / Biochimica et Biophysica Acta 1498 (2000) 122^133126

cyclase activity [47]. However, the exact role of CD38in the synthesis of intracellular cADPR is still con-troversial and may di¡er between cellular systems.Receptor-mediated increases in ADP-ribosyl cyclaseactivity and concomitant formation of cADPR havebeen reported in a number of cellular systems[43,45,48].

cADPR is enzymatically degraded by NADases(reviewed in [11,12,32]). Speci¢c cADPR hydrolaseshave not yet been isolated.

The Ca2�-releasing e¡ects of cADPR are thoughtto be mediated through its interaction, either director via accessory proteins, with RyR. RyR are knownfor their role in the mechanism of Ca2�-inducedCa2� release (CICR) [49]. Three isoforms of RyRhave been identi¢ed and speci¢c stimulatory e¡ectsof cADPR on RyR subtypes 2 and 3 have been de-scribed [50,51]. Several pharmacological studies indi-cate that cADPR is a physiological modulator ofCICR [52^54]. In addition to cADPR, the activityof the RyR/Ca2� channel is also modulated by otherphysiological ligands, namely Ca2�, CaM, inorganicphosphate (Pi), palmitoyl-CoA, which have stimula-tory e¡ects, and Mg2� (at high concentration) andvarious polyamines, which are negative modulators[11,32,55^60]. It should be mentioned that, in con-trast to the InsP3-gated channel (type 1), whichshows both positive and negative feedback regulationby Ca2�, the RyR behaves as a Ca2�-activated chan-nel in the physiological range of Ca2� concentra-tions.

There is also signi¢cant evidence that the activityof the RyR can be modulated by phosphorylation bydi¡erent protein kinases, including PKA, PKG, PKCand CaM-kinase. However, the e¡ects of receptorphosphorylation on Ca2� release may be di¡erent,depending on the type of cell or tissue (reviewed in[55,61,62]).

Several compounds are presently used as pharma-cological tools for studying cADPR-mediated Ca2�-release e¡ects. Among these, ryanodine and ca¡einehave been shown to act as selective activators ofRyR, whereas ruthenium red, procaine and varioussynthetic derivatives of cADPR, namely 8-NH2-,8-methoxy, 8-bromo- and 7-deaza-8-bromo-cADPR,are used as selective inhibitors of this receptor([32,45,63]; for complete listing see [12,64]).

Self-desensitisation is another characteristic of the

cADPR-mediated Ca2�-mobilisation system. In mostcellular systems studied so far cADPR exerts Ca2�

release from intracellular stores in the nanomolarand low micromolar range (EC50 ranging between3.8 nM and 10 WM, using permeabilised cells andcell homogenates) (reviewed in [11]) (Table 1).

The ionic conduction properties of the RyR are, ingeneral, similar to those reported for the InsP3R,except for the di¡erence that ionic currents throughthe RyR are approximately four times higher (2 pA),as compared to the InsP3R [24].

In analogy to the InsP3R, RyR are also involvedin the generation of elementary Ca2� release eventscalled sparks, as observed, for example, in cardiacmuscle cells [38,65,66]. The spatio-temporal proper-ties of Ca2� sparks arising from RyR are, however,slightly di¡erent from the ones described for Ca2�

pu¡s arising from activation of InsP3R. Ca2� sparksshow a duration of less than 200 ms and spatialranges of about 3 Wm and have maximal amplitudesof elevation of [Ca2�]i of 170 nM [38].

In contrast to the InsP3- and cADPR-signallingpathways, many aspects of NAADP�-mediatedCa2� signalling are still unknown. In spite of theexperimental evidence for an NAADP�-synthesisingactivity in some types of cells and tissue homogenates[6,67], and the observations that ADP-ribosyl cyclaseor CD38 can catalyse the synthesis of NAADP�

from L-NADP� and nicotinic acid at acidic pH [5],the identity of the enzyme involved in intracellularsynthesis of NAADP� is still not known. Also theintracellular receptor for NAADP� has so far notbeen identi¢ed. However, results from di¡erent typesof cells show that, in general, NAADP� inducesCa2� release at much lower concentrations (lownanomolar range), as compared to InsP3 andcADPR ([68], reviewed in [11,12]). The NAADP�-induced Ca2�-release mechanism is clearly distinctfrom those controlled by InsP3 and cADPR, since(i) it is not blocked by known antagonists of InsP3

and cADPR [4] and (ii) there is no cross-desensitisa-tion between NAADP�- and InsP3/cADPR-inducedCa2�-release mechanisms [4,69].

Moreover, NAADP�-induced Ca2� release is ki-netically faster than the corresponding mechanismsinduced by either cADPR or InsP3 [4] and, in con-trast to the latter, NAADP�-triggered Ca2� release isindependent of the cytoplasmic Ca2� concentration

BBAMCR 14673 27-11-00

C.P. da Silva, A.H. Guse / Biochimica et Biophysica Acta 1498 (2000) 122^133 127

and, therefore, does not behave as a CICR system[70]. In addition, this mechanism is not a¡ected bydivalent cations such as Mg2� [70]. Another observa-tion is that the NAADP�-sensitive Ca2�-pools seemto be localised in an intracellular compartmentclearly separated from those Ca2�-pools sensitive toInsP3 and cADPR, i.e., not contained within the ER[71]. Further important characteristics of theNAADP�-mediated Ca2� release are its self-desensi-tisation and self-inactivation. Sub-threshold concen-trations of NAADP� can completely and irreversiblyblock any further Ca2� release by this compound insea urchin eggs, probably due to a complete inacti-vation of receptors [71,72]. Due to this unique prop-erty, it is unlikely that NAADP� itself is responsiblefor long term Ca2� oscillations. In addition, higher(WM) concentrations of NAADP� desensitise theCa2�-release system without mediating any Ca2� re-lease.

A pharmacological study in sea urchin eggsshowed that Ca2� release mediated by NAADP�

can be speci¢cally inhibited by some L-type Ca2�

channel blockers [73].Although the precise mechanism of degradation of

NAADP� in cells is also not known, potential path-ways include the enzymes alkaline phosphatase andnucleotide pyrophosphatase [11,32].

4. InsP3-, cADPR- and NAADP+-mediated Ca2+

signalling: multiple pathways with distinct functionswithin one cell?

Evidence for an active role of all three Ca2�-re-lease pathways has only been obtained in a verylimited number of cell systems. Examples of theseare sea urchin eggs, pancreatic acinar cells and Tlymphocytes. In this section, some considerationsabout the possible functions of three di¡erentCa2�-signalling pathways in these cell types will bepresented.

4.1. Redundant Ca2+-signalling pathways ensurefertilisation of sea urchin eggs

Analysis of fertilisation of sea urchin eggs showsthat a transient rise in [Ca2�]i occurs immediatelyafter sperm^egg fusion and propagates across the

cytoplasm [11]. This Ca2� wave, which is requiredfor metabolic activation of the quiescent egg, is dueto Ca2� release from intracellular stores. When mi-croinjected in sea urchin eggs, InsP3 and cADPR are

Fig. 2. Roles for the three Ca2�-mobilising messengers InsP3,cADPR, and NAADP� in sea urchin eggs, pancreatic acinarcells, and Jurkat T lymphocytes. (A) In sea urchin eggs eitherthe InsP3 or the cADPR pathway is su¤cient for fertilisation.Blockade of one of these pathways ^ for this ¢gure the cADPRpathway was chosen randomly ^ does not inhibit fertilisation.The NAADP�/Ca2� release system is active in sea urchin eggs;however, its role in the process of fertilisation is not yet clari-¢ed. (B) In pancreatic acinar cells InsP3 is generated uponcholecystokinin and acetylcholine stimulation. InsP3 acts at theluminal pole of the cell. The local rise in Ca2� at this site thenallows secretory granules to fuse to the plasma membrane.NAADP� and cADPR may be involved in the response tocholecystokinin, but not to acetylcholine, according to the mod-el of Petersen and Cancela [83,84]. A Ca2� wave then swapsfrom the luminal to the basolateral pole of the cell which iscarried by cADPR-mediated Ca2� release [77]. The site ofNAADP�-mediated Ca2� release has not yet been localised.(C) In T cells NAADP�, InsP3, and cADPR may be formed ina temporal sequence resulting in long-lasting Ca2� signals thatare controlled stepwise by the di¡erent messengers. This mayallow the cell to use the generation and/or metabolism of thesemessengers as checkpoints to tightly control the physiologicalresponse, e.g., proliferation of T cells.

BBAMCR 14673 27-11-00

C.P. da Silva, A.H. Guse / Biochimica et Biophysica Acta 1498 (2000) 122^133128

both able to trigger independently Ca2�-mobilisationduring fertilisation [53,74]. Inhibition of either InsP3-mediated or cADPR-mediated Ca2� release was in-su¤cient to prevent the cortical reaction and prop-agation of Ca2� waves associated with fertilisation,whereas inhibition of both signalling pathways pre-vented these phenomena [53,74]. More recently, anadditional Ca2�-release system triggered byNAADP� has been described in sea urchin egghomogenates as well as in intact cells [4,69,75].NAADP� induces Ca2� release by a mechanism dis-tinct from those triggered by InsP3 and cADPR, asdemonstrated by (i) a lack of inhibition by the InsP3

and cADPR antagonists heparin and 8-amino-cADPR, and (ii) no cross-desensitisation betweenNAADP�- and InsP3/cADPR-induced Ca2�-releasemechanisms. In addition, once fertilisation occurred,no further e¡ects of microinjected NAADP� onCa2� release were observed, suggesting that fertilisedeggs are desensitised to NAADP� [75].

Taken together, these data demonstrate that InsP3-,cADPR- as well as NAADP�-dependent Ca2�-re-lease mechanisms may all be involved in fertilisationin sea urchin eggs. Although the signi¢cance of thecomplex control of Ca2� release during fertilisationremains to be elucidated, the available experimentaldata indicate that the cADPR- and the InsP3 path-ways act in parallel and independently of each other.Thereby, these independent pathways for Ca2�-mo-bilisation constitute redundant control mechanismsto ensure the development of the Ca2� waves re-quired for metabolic activation of fertilised eggs(Fig. 2A).

4.2. Localised Ca2+ stores allow Ca2+ signalling in allregions of pancreatic acinar cells

Pancreatic acinar cell are structurally and func-tionally polarised having the secretory granules con-centrated at the apical (luminal) pole. Fluid and en-zyme secretion by acinar cells is mediated by agonist-induced increases in [Ca2�]i which start in a triggerzone at the luminal cell pole and spread as a Ca2�

wave to the basal cell side [76,77]. Electrophysiolog-ical studies showed that agonist-induced Ca2� signalsin pancreatic acinar cells can be mimicked, at least inpart, by infusion of InsP3 [37] and selective injectionsof InsP3 into di¡erent parts of the cell showed that

the Ca2� release mechanism has a much higher sen-sitivity for this messenger in the secretory pole thanin the basal part of the cell [78]. In addition, type 3InsP3R have been identi¢ed and localised in the lu-minal pole of pancreatic acinar cells [79]. On theother hand, propagation of agonist-induced Ca2�

waves from the luminal to the basal pole could bereduced by ryanodine and ca¡eine [80], known mod-ulators of RyR, while microinjection of cADPR orof a sub-threshold concentration of Ca2� into pan-creatic acinar cells induced increases in Ca2� andrepetitive Ca2� spikes which originated in a regiondistinct from the trigger zone [78,81]. cADPR-in-duced Ca2� spikes as well as Ca2� signals elicitedby the peptide hormone cholecystokinin were inhib-ited by ryanodine [81]. Furthermore, evidence hasalso been obtained for expression of the type 2RyR in pancreatic cells and for its polarised distri-bution throughout the cell, mainly localised in thebasolateral region and, to a much lesser extent, inthe apical region [82].

Taken together, these ¢ndings suggest that ago-nist-induced Ca2� signals in pancreatic acinar cellsstart with formation of InsP3 in the basal pole andits rapid di¡usion (di¡usion coe¤cient; see Table 1)to its receptor sites followed by Ca2� release in theapical region. Subsequent propagation of Ca2�

waves then proceeds via a CICR mechanism basedon activation of RyR and mediated by cADPR.The di¡erential, polarised spatial distribution of in-tracellular receptors/Ca2� channels in acinar cells,i.e., with InsP3R localised exclusively in the apicalpole and RyR essentially in the basal pole, there-fore accounts for and correlates with the patternand mechanism of Ca2� wave propagation in thiscellular system (Fig. 2B). The observation thatcADPR-induced Ca2� signals can be blocked bythe InsP3 antagonist heparin suggests, in addition,a close link between the InsP3- and the cADPR-mediated signalling mechanisms in pancreatic acinarcells [81].

In spite of this, NAADP� was recently shown toinduce also repetitive Ca2� spikes in pancreatic aci-nar cells, at a concentration much lower than thatrequired for InsP3 and cADPR to induce similar ef-fects (50 nM versus 10 WM) [6]. Interestingly, in con-trast to observations in other cell systems, theNAADP�-mediated Ca2� signals in pancreatic aci-

BBAMCR 14673 27-11-00

C.P. da Silva, A.H. Guse / Biochimica et Biophysica Acta 1498 (2000) 122^133 129

nar cells could be inhibited by the cADPR antagonist8-NH2-cADPR and by the InsP3R antagonist hepa-rin [6]. Self-desensitisation was also observed at highconcentrations of NAADP�. Such desensitising con-centrations of NAADP� markedly inhibited chole-cystokinin-induced Ca2� signalling [6]. These exper-imental evidences, therefore, indicate also theinvolvement of NAADP�, in addition to InsP3 andcADPR, in agonist-induced Ca2� signalling in pan-creatic acinar cells. Based on these experimentaldata, Petersen and Cancela recently proposed a mod-el in which NAADP�-induced Ca2� release functionsas a trigger, while the cADPR- and InsP3-mediatedCa2� release systems would act as Ca2� ampli¢er andCa2� oscillator [83]. However, due to the lack ofinformation regarding the identity and intracellularlocalisation of the NAADP� receptor and theNAADP�-sensitive Ca2� pools, it is di¤cult atpresent to establish further correlations between thespatial intracellular organisation of Ca2� channelsand the above model in which three di¡erent Ca2�

release systems cooperate to regulate the complexspatio-temporal patterns of Ca2� signals in pancre-atic acinar cells.

Very recently, it has been reported that acetyl-choline stimulation of Ca2� signalling in acinar cells,in contrast to cholecystokinin, is neither inhibitedby desensitisation of the NAADP� system nor byinhibition of the cADPR system [84]. Cancela et al.concluded that activation by cholecystokinin involvesthe NAADP� as the trigger and the cADPR andInsP3 system as ampli¢er/oscillator, whereas in thecase of acetylcholine the InsP3 system serves both astrigger and ampli¢er/oscillator [84]. Remarkably, us-ing confocal Ca2� imaging there was no signi¢cantdi¡erence between acetylcholine or cholecystokininstimulation indicating that the e¡ects of bothNAADP� and cADPR must be very low in am-plitude or short in duration [84]. Interestingly,the physiological responses of acinar cells to chole-cystokinin and acetylcholine are not identical :both agonists stimulate £uid and enzyme secre-tion, but cholecystokinin in addition promotes pan-creatic growth [84]. This may indicate the involve-ment of both the NAADP� and the cADPRsystem re£ect the need for additional ¢ne tuning ofthe Ca2� signal to achieve this growth promotinge¡ect.

4.3. T-Cell activation can be controlled bysequentially recruited Ca2+-signalling pathways

Stimulation of T lymphocytes via the T-cell recep-tor (TCR)/CD3 complex results in activation of sev-eral signal transduction pathways, including an ele-vation of [Ca2�]i which is essential for cellproliferation and clonal expansion (reviewed in[85]). The increase of [Ca2�]i is due to release ofCa2� from intracellular stores and subsequent Ca2�

entry from the extracellular medium [85]. InsP3 andcADPR are both involved in the process of Ca2�

release in T lymphocytes [86^89] and, in addition, asecond-messenger role for these compounds has beendemonstrated by (i) increases in their intracellularconcentration upon stimulation of the TCR/CD3complex, and (ii) the inhibitory e¡ect of the InsP3

antagonist xestospongin C and the cADPR antago-nist 7-deaza-8-Br-cADPR on TCR/CD3-mediatedCa2� signalling [45,89]. InsP3 acts primarily in theinitial phase of Ca2� signalling in T cells, whereascADPR is involved in both the initial and the sub-sequent sustained phase. More recently, an addi-tional, important role of NAADP� in Ca2� sig-nalling in Jurkat T cells has been described [90].Microinjection of NAADP� into intact T cells stimu-lated Ca2� signalling in a speci¢c and dose-depen-dent manner. In addition, both InsP3- and cADPR-mediated Ca2� release depended upon a functionalNAADP�/Ca2�-release system. Moreover, self-inac-tivation of the NAADP�/Ca2� release completely in-hibited Ca2� signalling mediated by activation of theTCR/CD3 complex [90]. Based on these experimentalobservations, a model is proposed in which each ofthe three Ca2�-mobilising messengers acts in a tem-poral sequence to regulate Ca2� signalling in T cells.In spite of the lack of information about changesin NAADP� concentration upon T-cell stimulation,it is proposed that NAADP� (either in basal nano-molar concentrations or eventually in higher stimu-lated levels) acts as the ¢rst messenger to provide atrigger-Ca2� which subsequently activates the othertwo Ca2�-mobilising systems. The NAADP�-in-duced £ow of trigger Ca2� sequentially activatesthe InsP3/Ca2�-release system and, together with in-creased levels of InsP3, which are formed rapidly andtransiently after cell stimulation, promotes furtherincreases in Ca2�. This would subsequently lead to

BBAMCR 14673 27-11-00

C.P. da Silva, A.H. Guse / Biochimica et Biophysica Acta 1498 (2000) 122^133130

activation of RyR and the cADPR/Ca2�-release sys-tem so that, in concert with cADPR which is ele-vated in the sustained phase of Ca2� signalling [45],further major increases in Ca2� occur.

In summary, the model presented above consti-tutes an example of how three di¡erent Ca2�-mobi-lising messengers, with their speci¢c and de¢nedfunctions and characteristic signalling pathways,may provide a temporal regulation of Ca2� signallingin T cells. This sequential organisation may allow theT cell to use each system as a sort of checkpoint;commitment of T cells to activation may be tightlycontrolled by each of these di¡erent checkpoints(Fig. 2C). Although no information is presentlyavailable on the kinetics of synthesis of NAADP�

in T cells, this model is in accordance with the ki-netics of InsP3 and cADPR formation, i.e., rapid andmarked elevation of InsP3 in the ¢rst minutes aftercell stimulation and slower increase in intracellularlevels of cADPR, which then remain elevated for alonger period of time [45].

5. Conclusion

Three cellular models have been reviewed to ob-tain an insight of understanding into why duringevolution the three Ca2�-mobilising messengers In-sP3, cADPR, and NAADP� have been conservedin the same cell types. The experimental data ob-tained so far indicate cell type-speci¢c di¡erences re-garding the individual role of each of the three Ca2�

mobilisers. Although many experimental details havebeen collected in the past, large gaps in our currentknowledge of the detailed intracellular processes donot allow the development of more conclusive mod-els. In spite of this, the ideas described in this articlehopefully will stimulate research in this fascinating¢eld and thus may be used as working hypothesisto be veri¢ed or falsi¢ed in the future.

Acknowledgements

We are grateful to G.W. Mayr (Hamburg) forcritically reading the manuscript. Research in the au-thors' laboratory was supported by grants from theDeutsche Forschungsgemeinschaft (grant no. Gu

360/1 to 5, jointly with G.W. Mayr, Hamburg), theWellcome Trust (research collaboration grant no.051326 jointly with B.V.L. Potter, Bath, UK) andthe Deutsche Akademische Austauschdienst (PRO-COPE program grant no. D/9822786, jointly withP. Deterre, Paris, France).

References

[1] H. Streb, R.F. Irvine, M.J. Berridge, I. Schulz, Nature 306(1983) 67^69.

[2] D.L. Clapper, T.F. Walseth, P.J. Dargie, H.C. Lee, J. Biol.Chem. 262 (1987) 9561^9568.

[3] H.C. Lee, T.F. Walseth, G.T. Bratt, R.N. Hayes, D.L. Clap-per, J. Biol. Chem. 264 (1989) 1608^1615.

[4] H.C. Lee, R. Aarhus, J. Biol. Chem. 270 (1995) 2152^2157.[5] R. Aarhus, R.M. Grae¡, D.M. Dickey, T.F. Walseth, H.C.

Lee, J. Biol. Chem. 270 (1995) 30327^30333.[6] J.M. Cancela, G.C. Churchill, A. Galione, Nature 398 (1999)

74^76.[7] S. Sakano, H. Takemura, K. Yamada, K. Imoto, M. Kane-

ko, H. Ohshika, J. Biol. Chem. 271 (1996) 11148^11155.[8] H. Takemura, K. Imoto, S. Sakano, M. Kanko, H. Ohshika,

Biochem. J. 319 (1996) 393^397.[9] Y. Xu, G. Casey, G.B. Mills, J. Cell Physiol. 163 (1995) 441^

450.[10] W. Masuda, S. Takenaka, S. Tsuyama, M. Tokunaga, R.

Yamaji, H. Inui, K. Miyatake, Y. Nakano, Comp. Biochem.Physiol. 118 (1997) 279^283.

[11] H.C. Lee, Physiol. Rev. 77 (1997) 1133^1164.[12] A.H. Guse, J. Mol. Med. 78 (2000) 26^35.[13] M. Albrieux, H.C. Lee, M. Villaz, J. Biol. Chem. 273 (1998)

14566^14572.[14] M.J. Berridge, Nature 361 (1993) 315^325.[15] M.J. Berridge, R.F. Irvine, Nature 341 (1989) 197^205.[16] T.A. Brock, P.A. Dennis, K.K. Griendling, T.S. Diehl, P.F.

Davies, Am J. Physiol. 255 (1988) C667^673.[17] A. Ullrich, J. Schlessinger, Cell 61 (1990) 203^212.[18] J.D. Fraser, D. Straus, A. Weiss, Immunol. Today 14 (1993)

357^362.[19] K. Mikoshiba, Trends Pharmacol. Sci. 14 (1993) 86^89.[20] C.D. Ferris, S.H. Snyder, Annu. Rev. Physiol. 54 (1992)

469^488.[21] S. Kurasawa, I. Schultz, G.S. Kopf, Dev. Biol. 133 (1989)

295^304.[22] S.K. Ambler, P. Taylor, Mol. Pharmacol. 35 (1989) 369^374.[23] Y. Nakamura, S. Ohtaki, Endocrinology 120 (1987) 2302^

2307.[24] I. Bezprozvanny, B. Erlich, J. Gen. Physol. 104 (1994) 821^

856.[25] I. Bezprozvanny, B. Erlich, Neuron 10 (1993) 1175^1184.[26] I. Bezprozvanny, J. Watras, B. Erlich, Nature 351 (1991)

751^754.

BBAMCR 14673 27-11-00

C.P. da Silva, A.H. Guse / Biochimica et Biophysica Acta 1498 (2000) 122^133 131

[27] R.E. Hagar, A.D. Burgstahler, M.H. Nathanson, B.E. Ehr-lich, Nature 396 (1998) 81^84.

[28] T. Michikawa, J. Hirota, S. Kawano, M. Hiraoka, M. Ya-mada, T. Furuichi, K. Mikoshiba, Neuron 23 (1999) 799^808.

[29] T. Jayaraman, K. Ondrias, E. Ondriasova, A.R. Marks, Sci-ence 272 (1996) 1492^1494.

[30] T.D. Hill, P.O. Berggren, A.L. Boynton, Biochem. Biophys.Res. Commun. 149 (1987) 897^901.

[31] J. Gafni, J.A. Munsch, T.H. Lam, M.C. Catlin, L.G. Costa,T.F. Molinski, I.N. Pessah, Neuron 19 (1997) 723^733.

[32] T.D. Dousa, E.N. Chini, K.W. Beers, Am. J. Physiol. 271(1996) C1007^C1024.

[33] N. Allbittron, T. Meyer, L. Stryer, Science 258 (1992) 1812^1815.

[34] Y. Yao, J. Choi, I. Parker, J. Physiol. 482 (1995) 533^553.[35] M.D. Bootman, M.J. Berridge, Curr. Biol. 6 (1996) 855^

861.[36] M.T. Nelson, H. Cheng, M. Rubart, L.F. Santana, A.

Bonev, H. Knot, W.J. Lederer, Science 270 (1995) 633^637.[37] P. Thorn, A.M. Lawrie, P.M. Smith, D.V. Gallacher, O.H.

Petersen, Cell 74 (1993) 661^668.[38] M.D. Bootman, M.J. Berridge, Cell 83 (1995) 675^678.[39] B. Verjans, C. Moreau, C. Erneux, Cell. Signal. 6 (1994)

335^344.[40] K. Takazawa, M. Lemos, A. Delvaux, C. Lejeune, J.E. Du-

mont, C. Erneux, Biochem. J. 268 (1990) 213^217.[41] R.M. Grae¡, L. Franco, A. De Flora, H.C. Lee, J. Biol.

Chem. 273 (1998) 118^125.[42] E. Clementi, M. Piccio, C. Sciorati, G. Nistico, J. Meldolesi,

J. Biol. Chem. 271 (1996) 117739^117745.[43] H. Higashida, FEBS Lett. 418 (1997) 355^356.[44] H. Higashida, S. Yokoyama, M. Hashii, M. Taketo, M.

Higashida, T. Takayasu, T. Ohshima, S. Takasawa, H. Oka-moto, M. Noda, J. Biol. Chem. 272 (1997) 31272^31277.

[45] A.H. Guse, C.P. da Silva, I. Berg, A. Skapenko, K. Weber,P. Heyer, M. Hohenegger, G. Ashamu, H. Schulze-Koops,B.V.L. Potter, G.W. Mayr, Nature 398 (1999) 70^73.

[46] H.E. Kim, E.L. Jacobson, M.K. Jacobson, Science 261(1993) 1330^1333.

[47] M. Howard, J.C. Grimaldi, J.F. Bazan, F.E. Lund, L. San-tos-Argumedo, R.M.E. Parkhouse, T.F. Walseth, H.C. Lee,Science 262 (1993) 1056^1059.

[48] H. Kuemmerle, G.M. Makhlouf, J. Biol. Chem. 270 (1995)25488^25494.

[49] A. Fabiato, Am. J. Physiol. 245 (1993) C1^C14.[50] L.G. Meszaros, J. Bak, A. Chu, Nature 364 (1993) 76^79.[51] A. Sonnleitner, A. Conti, F. Bertocchini, H. Schindler, V.

Sorrentino, EMBO J. 17 (1998) 2790^2798.[52] A. Galione, Trends Pharmacol. Sci. 13 (1992) 304^306.[53] A. Galione, A. MacDougall, W.B. Busa, N. Willmott, I.

Gillot, M. Whitaker, Science 253 (1993) 1143^1146.[54] Y. Cui, A. Galione, D.A. Terrar, Biochem. J. 342 (1999)

269^273.[55] G. Meissner, Annu. Rev. Physiol. 56 (1994) 485^508.

[56] R.M. Grae¡, R.J. Podein, R. Aarhus, H.C. Lee, Biochem.Biophys. Res. Commun. 206 (1995) 786^791.

[57] H.C. Lee, R. Aarhus, R. Grae¡, M.E. Gurnack, T.F. Wal-seth, Nature 370 (1994) 307^309.

[58] A.H. Guse, C.P. Silva, K. Weber, G.A. Ashamu, B.V.L.Potter, G.W. Mayr, J. Biol. Chem. 271 (1996) 23946^23954.

[59] E.N. Chini, K.W. Beers, C.C.S. Chini, T.P. Dousa, Am. J.Physiol. 269 (1995) C1042^C1047.

[60] E.N. Chini, T.P. Dousa, Am. J. Physiol. 270 (1996) C530^C537.

[61] J. Hain, S. Nath, M. Mayrleitner, S. Fleischer, H. Schindler,Biophys. J. 67 (1994) 1823^1833.

[62] A.D. Bonev, J.H. Jaggar, M. Rubart, M.T. Nelson, Am. J.Physiol. 273 (1997) C2090^C2095.

[63] T.F. Walseth, H.C. Lee, Biochim. Biophys. Acta 1178 (1993)235^242.

[64] F.J. Zhang, Q.M. Gu, C.J. Sih, Bioorg. Med. Chem. 7 (1999)653^664.

[65] H. Cheng, W.J. Lederer, M.B. Cannel, Science 262 (1993)740^744.

[66] P. Lipp, E. Niggli, J. Physiol. 508 (1998) 801^809.[67] E.N. Chini, T.P. Dousa, Biochem. Biophys. Res. Commun.

270 (1995) 167^174.[68] H.C. Lee, R. Aarhus, K.R. Gee, T. Kestner, J. Biol. Chem.

272 (1997) 4172^4178.[69] E.N. Chini, K.W. Beers, T.P. Dousa, J. Biol. Chem. 270

(1995) 3216^3223.[70] E.N. Chini, T.P. Dousa, Biochem. J. 316 (1996) 709^711.[71] A.A. Genazzani, A. Galione, Biochem. J. 315 (1996) 721^

725.[72] R. Aarhus, D.M. Dickey, R.M. Grae¡, K.R. Gee, T.F. Wal-

seth, H.C. Lee, J. Biol. Chem. 271 (1996) 8513^8516.[73] A.A. Genazzani, A. Galione, Trends Pharmacol. Sci. 18

(1997) 108^110.[74] H.C. Lee, R. Aahrus, T.F. Walseth, Science 261 (1993) 352^

355.[75] C.M. Perez, E.N. Chini, S.S. Shen, T.P. Dousa, D.E. Clap-

ham, Biochem. J. 312 (1995) 955^959.[76] H. Kasai, G.J. Augustine, Nature 348 (1990) 735^738.[77] I. Schulz, E. Krause, A. Gonzalez, A. Go«bel, L. Sternfeld, A.

Schmid, Biol. Chem. 380 (1999) 903^908.[78] H. Kasai, Y.X. Li, Y. Miyashita, Cell 74 (1993) 669^677.[79] M.H. Nathanson, M.B. Fallon, P.J. Pad¢eld, A.R. Maranto,

J. Biol. Chem. 269 (1994) 4693^4696.[80] F. Pfei¡er, L. Sternfeld, A. Schmid, I. Schulz, Am. J. Phys-

iol. 274 (1998) C663^C672.[81] P. Thorn, O. Gerasimenko, O.H. Petersen, EMBO J. 13

(1994) 2038^2043.[82] M.F. Leite, J.A. Drano¡, L. Gao, M.H. Nathanson, Bio-

chem. J. 337 (1999) 305^309.[83] O.H. Petersen, J.M. Cancela, Trends Neurosci. 22 (1999)

488^494.[84] J.M. Cancela, O.V. Gerasimenko, J.V. Gerasimenko, A.V.

Tepikin, O.H. Petersen, EMBO J. 19 (2000) 2549^2557.[85] A.H. Guse, Crit. Rev. Immunol. 18 (1998) 419^448.

BBAMCR 14673 27-11-00

C.P. da Silva, A.H. Guse / Biochimica et Biophysica Acta 1498 (2000) 122^133132

[86] A.H. Guse, E. Roth, F. Emmrich, Biochem. J. 291 (1993)447^451.

[87] A.H. Guse, C.P. da Silva, F. Emmrich, G.A. Ashamu,B.V.L. Potter, G.W. Mayr, J. Immunol. 155 (1995) 3353^3359.

[88] A.H. Guse, I. Berg, C.P. da Silva, B.V.L. Potter, G.W.Mayr, J. Biol. Chem. 272 (1997) 8546^8550.

[89] J. Ng, J. Gustavsson, M. Jondal, T. Andersson, Biochim.Biophys. Acta 1053 (1990) 97^105.

[90] I. Berg, B.V.L. Potter, G.W. Mayr, A.H. Guse, J. Cell Biol.150 (2000) 581^588.

BBAMCR 14673 27-11-00

C.P. da Silva, A.H. Guse / Biochimica et Biophysica Acta 1498 (2000) 122^133 133