Báo cáo khoa học: Cyclic ADP-ribose requires CD38 to regulate the release of ATP in visceral smooth muscle ppt

Using a small-volume superfusion assay and an HPLC technique with fluorescence detection, we measured the spontaneous and evoked release of ATP in bladder detrusor smooth muscles isolated

Cyclic ADP-ribose requires CD38 to regulate the release of ATP in visceral smooth muscle

Leonie Durnin and Violeta N Mutafova-Yambolieva

Department of Physiology and Cell Biology, University of Nevada School of Medicine, Reno, NV, USA

Keywords

ATP; bladder; cADP-ribose; CD38; NAD;

purinergic neurotransmission

Correspondence

V N Mutafova-Yambolieva, Department of

Physiology and Cell Biology, University

of Nevada School of Medicine, Center

for Molecular Medicine ⁄ MS 575, Reno,

NV 89557-0575, USA

Fax: +1 775 784 6903

Tel: +1 775 784 6274

E-mail: vmutafova@medicine.nevada.edu

(Received 30 April 2011, revised 24 June

2011, accepted 30 June 2011)

doi:10.1111/j.1742-4658.2011.08233.x

It is well established that the intracellular second messenger cADP-ribose (cADPR) activates Ca2+ release from the sarcoplasmic reticulum through ryanodine receptors CD38 is a multifunctional enzyme involved in the for-mation of cADPR in mammals CD38 has also been reported to transport cADPR in several cell lines Here, we demonstrate a role for extracellular cADPR and CD38 in modulating the spontaneous, but not the electrical field stimulation-evoked, release of ATP in visceral smooth muscle Using a small-volume superfusion assay and an HPLC technique with fluorescence detection, we measured the spontaneous and evoked release of ATP in bladder detrusor smooth muscles isolated from CD38+⁄ + and CD38) ⁄ ) mice cADPR (1 nM) enhanced the spontaneous overflow of ATP in blad-ders isolated from CD38+⁄ +mice This effect was abolished by the inhibi-tor of cADPR recepinhibi-tors on sarcoplasmic reticulum 8-bromo-cADPR (80 lM) and by ryanodine (50 lM), but not by the nonselective P2 puriner-gic receptor antagonist pyridoxal phosphate 6-azophenyl-2¢,4¢-disulfonate (30 lM) cADPR failed to facilitate the spontaneous ATP overflow in bladders isolated from CD38) ⁄ )mice, indicating that CD38 is crucial for the enhancing effects of extracellular cADPR on spontaneous ATP release Contractile responses to ATP were potentiated by cADPR, suggesting that the two adenine nucleotides may work in synergy to maintain the resting tone of the bladder In conclusion, extracellular cADPR enhances the spontaneous release of ATP in the bladder by influx via CD38 and subse-quent activation of intracellular cADPR receptors, probably causing an increase in intracellular Ca2+in neuronal cells

Introduction

Cyclic ADP-ribose (cADPR) is an intracellular second

messenger that can release Ca2+from

ryanodine-sensi-tive stores [1] in a wide variety of cells [2], including

cells in the nervous system [3] In mammals, cADPR is

generated from NAD by ADP-ribosyl cyclase

associ-ated with CD38, a multifunctional type II integral

membrane glycoprotein with ADP-ribosyl cyclase and

NAD-glycohydrolase activities [2,4,5] The catalytic

site of CD38 faces the ectocellular space [6,7], making

this enzyme suitable as a regulator of extracellular b-NAD+ and cADPR levels [8] Therefore, cADPR could be produced extracellularly in each system that releases b-NAD+ and expresses membrane-bound CD38 In 3T3 murine fibroblasts and HeLa cells, CD38 also mediates intracellular influx of cADPR [9,10] Furthermore, extracellular cADPR can stimu-late NG108-15 cells, a neurally derived clonal cell line, and elevate intracellular Ca2+levels [11] It is presently Abbreviations

ADPR, ADP-ribose; BoNTA, botulinum neurotoxin A; cADPR, cADP-ribose; CBX, carbenoxolone; cGDPR, cGDP-ribose; eADPR, 1,N 6 -etheno-ADPR; EFS, electrical field stimulation; FFA, flufenamic acid; NGD, nicotinamide guanine dinucleotide; PPADS, pyridoxal phosphate

6-azophenyl-2¢,4¢-disulfonate; PS, prestimuation; SE, standard error; TTX, tetrodotoxin.

unknown whether such mechanisms play a role in

complex systems such as visceral smooth muscle

Likewise, the role of extracellular cADPR in

modulat-ing neurotransmission at the nerve–smooth muscle

junction remains to be determined

In a number of smooth muscle tissues, the precursor

of cADPR b-NAD+is released at rest and upon firing

of action potentials, and serves as a neurotransmitter

and a neuromodulator [12–16] CD38 is expressed

exclusively on nerve terminals in some smooth muscle

preparations [14], and hence cADPR is present

extracellularly, probably because of degradation of

b-NAD+ by CD38 Exogenous cADPR modifies the

release of neurotransmitter in blood vessels [12], but it

is unclear whether neuromodulation by cADPR is

mediated by receptors on the presynaptic membrane or

by receptors on intracellular Ca2+ stores and

subse-quent changes in intracellular Ca2+ It is also

unknown whether cADPR can modulate equally the

spontaneous and evoked release of neurotransmitters

ATP is believed to be a cotransmitter with

acetyl-choline in the urinary bladder [17,18] To address some

of the aforementioned unresolved issues, we examined

how exogenous cADPR modulates the amounts of

ATP released in the bladder In particular, we studied

the effects of exogenous cADPR on spontaneous and

electrical field stimulation (EFS)-evoked overflow of

ATP in bladder detrusor smooth muscle isolated from

CD38-deficient (CD38) ⁄ )) mice and from control

C57⁄ BL6 mice, referred to as CD38+ ⁄ +mice

through-out this article We report here that exogenous cADPR

facilitates the spontaneous release of ATP, probably

because of influx of cADPR through CD38 and

subse-quent activation of intracellular ryanodine-sensitive

cADPR receptors The EFS-evoked release of ATP,

however, appears to be unaffected by extracellular

cADPR, suggesting that the spontaneous and

EFS-evoked release of ATP in the bladder are mediated

differentially by CD38

Results

Mechanisms of spontaneous and EFS-evoked

release of ATP in bladder detrusor muscles from

CD38+⁄ +and CD38) ⁄ )mice

We first determined the spontaneous and EFS-evoked

release of ATP in bladder detrusor smooth muscles

isolated from CD38+⁄ +and CD38) ⁄ )mice As shown

in Fig 1, superfusate samples collected before

stimula-tion [prestimulastimula-tion (PS)] or during EFS [16 Hz,

0.1 ms for 60 s; stimulation (ST)] of bladder detrusor

muscles from CD38+⁄ +and CD38) ⁄ )mice contained

ATP along with other adenine compounds, including ADP, AMP, b-NAD+, ADP-ribose (ADPR), cADPR and Ado, suggesting that there is spontaneous and evoked release of ATP in the murine bladder As demonstrated previously [12], b-NAD+, ADPR and cADPR eluted as one peak, owing to conversion to 1,N6-etheno-ADPR (eADPR) during etheno-derivatiza-tion of tissue superfusate samples (see Experimental procedures) There were no significant differences between the spontaneous and EFS-evoked overflow of ATP in CD38+⁄ + and CD38) ⁄ ) mice The EFS-evoked release of ATP, determined by the difference

ST) PS, was 3.18 ± 0.52 fmolÆmg)1tissue in bladders from CD38+⁄ + mice (n = 55) and 2.48 ± 0.41 fmo-lÆmg)1 tissue in bladders from CD38) ⁄ )mice (n = 40) (P > 0.05) Tetrodotoxin (TTX) (0.30.5 lm, for

30 min) had no effect on the spontaneous release of ATP in bladders isolated from CD38+⁄ + mice or CD38) ⁄ )mice (P > 0.05 versus controls; Fig 1) The EFS-evoked overflow of ATP was reduced by TTX

in bladders isolated from CD38+⁄ + mice (ST) PS was 0.18 ± 0.65 fmolÆmg)1 tissue, n = 12, P < 0.05 versus control), but not in bladders isolated from CD38) ⁄ ) mice (ST) PS was 2.05 ± 0.46 fmolÆmg)1

tissue, n = 22, P > 0.05 versus controls; Fig 1) Incubation of bladders isolated from CD38+⁄ +mice with botulinum neurotoxin A (BoNTA) (100–300 nm for 2.5 h) led to cleavage of SNAP25 (Fig 2, inset) The spontaneous overflow of ATP in BoNTA-treated tissues remained unchanged in bladders from CD38+⁄ + and CD38) ⁄ ) mice (Fig 2) (P > 0.05 versus PS values in nontreated tissues) As expected,

no additional overflow was observed upon EFS

As ATP release from cells can also occur via hemichannels [19–22], we next examined whether the spontaneous or evoked overflow of ATP is affected by two widely used hemichannel blockers, namely carbe-noxolone (CBX) and flufenamic acid (FFA) [19,22,23]

In bladders isolated from CD38+⁄ +mice, the sponta-neous overflow of ATP was as follows (fmolÆmg)1 tis-sue): 0.34 ± 0.08 (n = 4), 0.28 ± 0.04 (n = 4) and 0.58 ± 0.04 (n = 3) in the presence of vehicle, CBX (100 lm) and FFA (100 lm), respectively (P > 0.05 versus vehicle controls) The evoked overflow of ATP, determined from the ST) PS values, was as follows (fmolÆmg)1 tissue): 0.82 ± 0.21 (n = 4), 1.15 ± 0.27 (n = 4) and 0.36 ± 0.20 fmolÆmg)1 tissue in the pres-ence of vehicle, CBX and FFA, respectively (P > 0.05 versus controls) Therefore, neither the spontaneous nor the evoked release of ATP appeared to be affected

by CBX or FFA in bladders isolated from CD38+⁄ + mice Likewise, in bladders isolated from CD38) ⁄ ) mice, the spontaneous release of ATP was as follows

(fmolÆmg)1 tissue): 0.25 ± 0.018 (n = 5), 0.30 ± 0.06

(n = 3) and 0.58 ± 0.19 (n = 4) in the presence of

vehicle, CBX and FFA, respectively (P > 0.05) The

EFS-evoked overflow of ATP (ST) PS values,

fmo-lÆmg)1 tissue) was as follows: 1.16 ± 0.14 (n = 5),

1.32 ± 0.18 (n = 3) and 0.69 ± 0.44 (n = 4) in the

presence of vehicle, CBX and FFA, respectively

(P > 0.05)

As shown in Fig 1, tissue superfusates contained not

only ATP, but also b-NAD+, as well as other adenine

compounds, including ADP, AMP, Ado, ADPR, and

cADPR These adenine compounds are metabolites of

either ATP, b-NAD+, or both: ADP is a direct

metab-olite of ATP, whereas AMP and Ado can be formed by

both ATP and b-NAD+ [2,4,24] Table 1 shows the

values of ADP, AMP, b-NAD++ ADPR + cADPR

(eluted as eADPR) and Ado accumulated in tissue

superfusates before (spontaneous overflow) and during

(evoked overflow) nerve stimulation in control

experi-ments in bladder detrusor muscles isolated from

CD38+⁄ + and CD38) ⁄ ) mice In control CD38+⁄ +

mice, the overflow of adenine purines was increased during nerve stimulation No significant differences were observed in the spontaneous overflow of all ade-nine purines in CD38+⁄ + and CD38) ⁄ )preparations The amounts of b-NAD++ ADPR + cADPR, adeno-sine and total purines were reduced in the samples col-lected during nerve stimulation of bladders isolated from CD38) ⁄ )mice

CD38 carries the ADP-ribosyl cyclase activity in the murine bladder detrusor muscle

Next, we tested whether ADP-ribosyl cyclase activity in the bladder is associated with CD38 We first examined whether there is a difference between the degradation

of nicotinamide guanine dinucleotide (NGD) to cGDP-ribose (cGDPR) in bladders isolated from CD38+⁄ + and CD38) ⁄ )mice as a measure of GDP-ribosyl (and possibly ADP-ribosyl) cyclase activity [4] As shown in Fig 3, production of cGDPR from NGD was increased during incubation of NGD with bladders

CD38 +/+

PS

ST

ATP ADP β-NAD + ADPR + cADPR AMP

Ado

ATP ADP β-NAD + ADPR + cADPR AMP

Ado

CD38 +/+

0 2

6

4

CD38 –/–

PS

ST

ATP ADP β-NAD + ADPR + cADPR AMP

Ado

ATP ADP β-NAD + ADPR + cADPR AMP

Ado

B A

C

ATP ADP β-NAD + ADPR + cADPR AMP

Ado

ST, TTX

Min

100 LU

Min

***

(55)

(55)

(12)

ATP ADP β-NAD + ADPR + cADPR AMP

Ado

ST, TTX

TTX

PS ST PS ST

Controls

(12)

CD38 –/–

0 2

6

4

D

TTX

PS ST PS ST

Controls

***

**

(22)

(22)

(40)

(40)

Fig 1 ATP is released at rest and during

EFS in murine bladder detrusor muscle.

(A, B) Original chromatograms of tissue

superfusate samples collected before EFS

(PS) and during EFS (16 Hz, 0.1 ms for 60 s;

ST) in CD38 + ⁄ +

mice and CD38) ⁄ )mice,

respectively Chromatograms from ST

samples collected during superfusion with

TTX (0.5 l M , 30 min) are also shown.

Spontaneous overflow of ATP and the

metabolites ADP, AMP and Ado, and

b-NAD + + ADPR + cADPR, occurred in PS

samples EFS (ST) resulted in increased

overflow of all nucleotides and nucleosides.

LU, luminescence units: scale applies to all

chromatograms (C, D) ATP overflow in

CD38 +⁄ + mice and CD38) ⁄ )mice,

respec-tively, before EFS (PS) and during EFS (ST)

in the absence and presence of TTX (0.3–

0.5 l M ) (averaged data in fmolÆmg)1tissue,

presented as means ± SE; ***P < 0.001,

**P < 0.05) Numbers of observations are in

parentheses Enhanced overflow of all

purines was observed during EFS TTX had

no effect on the spontaneous overflow of

ATP TTX significantly reduced the evoked

overflow of ATP during EFS of bladders

isolated from CD38 + ⁄ + mice, but not in

bladders isolated from CD38) ⁄ )mice.

isolated from CD38+⁄ +mice In contrast, bladders

iso-lated from CD38) ⁄ ) mice failed to degrade NGD

Thus, the entire GDP-ribosyl cyclase activity in the

murine bladder detrusor muscle appears to be

associ-ated with CD38

We next carried out an HPLC fraction analysis [12]

to determine whether cADPR and ADPR are present

in tissue superfusates from bladders isolated from

CD38) ⁄ ) mice along with their precursor b-NAD+

The amounts of ADPR and cADPR were negligible:

samples collected before EFS contained 94.71% ±

1.93% b-NAD+, 2.9% ± 0.69% ADPR, and

2.38% ± 1.24% cADPR, whereas samples collected

during EFS contained 98.42% ± 0.35% b-NAD+,

0.66% ± 0.31% ADPR, and 0.91% ± 0.42% cADPR

(n = 3, 12–16 chambers in each experiment)

There-fore, the ADP-ribosyl cyclase activity in the murine

bladder detrusor appears to be attributable exclusively

to CD38

Effects of exogenous cADPR on spontaneous and evoked overflow of ATP

To determine whether extracellular cADPR is a neuro-modulator and can modify the release of ATP, we next examined the effects of exogenous cADPR (1 nm) on the spontaneous and EFS-evoked overflow of ATP cADPR caused a significant increase in the spontane-ous overflow of ATP in bladders isolated from CD38+⁄ + mice, but not in bladders isolated from CD38) ⁄ )mice (Fig 4), suggesting that CD38 is impor-tant for the enhancing effect of exogenous cADPR in the bladder However, cADPR (1 nm) did not enhance the EFS-evoked release of ATP in bladders isolated from either CD38+⁄ +mice or CD38) ⁄ )mice (Fig 5): The evoked release, determined by the difference in ATP amounts between ST and PS samples (ST ) PS), was 3.97 ± 1.88 fmolÆmg)1 tissue in bladders from CD38+⁄ + mice (n = 16) and 2.077 ± 0.87 fmolÆmg)1

PS

ST

ATP

ADP

β-NAD + ADPR + cADPR

AMP

Ado

ATP

ADP

β-NAD + ADPR + cADPR

AMP

Ado

CD38 –/–

PS

ST

ATP

β-NAD + ADPR + cADPR AMP

Ado

ATP ADP β-NAD + ADPR + cADPR AMP

Ado

B A

ATP

ADP

β-NAD + ADPR + cADPR

AMP

Ado

ST, BoNTA

Min

100 LU

Min

ATP ADP β-NAD + ADPR + cADPR AMP

Ado

ST, BoNTA

ADP

SNAP-25

25 kDa

Control BoNTA

–1 ti

–1 ti

0

2

4

D C

*

(4)

(4)

(4)

BoNTA

PS ST PS ST

Controls BoNTA

PS ST PS ST

Controls

(4)

*

(3) (3) (3)

(3)

0 2 4

Fig 2 Differential effects of BoNTA on the spontaneous and EFS-evoked release of ATP (A, B) Original chromatograms of tissue superfusate samples collected before EFS (PS) and during EFS (16 Hz, 0.1 ms for

60 s; ST) in CD38+⁄ +mice and CD38) ⁄ ) mice, respectively Chromatograms from ST samples collected during superfusion of BoNTA-treated (100 n M for 2.5 h) tissues are also shown EFS (ST) resulted in increased overflow of all nucleotides and nucleosides, and this was reduced by BoNTA LU, luminescence units: scale applies to all chromatograms (C, D) ATP overflow in CD38 + ⁄ +

mice and CD38) ⁄ ) mice, respectively, before EFS (PS) and dur-ing EFS (ST) in controls and BoNTA-treated tissues (averaged data in fmolÆmg)1, presented as means ± SE; *P < 0.05) Numbers of observations are in parenthe-ses Enhanced overflow of all purines was observed during EFS BoNTA significantly reduced the EFS-evoked, but not the spontaneous, overflow of ATP in bladders isolated from CD38 + ⁄ +

and CD38) ⁄ )mice (C) Inset: western immunoblot analysis of SNAP-25 shows a single band at 25 kDa in homogenates from control (vehicle-treated) tissues An additional 24-kDa band appears

in BoNTA-treated tissues, indicating cleav-age of SNAP-25 induced by BoNTA.

tissue in bladders from CD38) ⁄ ) mice (n = 11) (P > 0.05) These values were not significantly differ-ent from the ST) PS amounts of ATP in the absence

of cADPR Note that the peak of eADPR (standing for b-NAD++ ADPR + cADPR) was increased in the samples collected during superfusion with cADPR (Figs 4 and 5), because the exogenous cADPR was also derivatized to eADPR during the precolumn derivatization [12] Thus, the peaks of b-NAD++ ADPR + cADPR, AMP and Ado represented the amounts of endogenously formed nucleotides and nucleosides plus products of the degradation of the exogenous cADPR, and therefore were not analyzed in detail

The enhancing effect of cADPR on ATP overflow was not reduced by the nonselective P2 receptor antag-onist pyridoxal phosphate 6-azophenyl-2¢,4¢-disulfonate (PPADS) (30 lm) (Fig 6), suggesting that

prejunction-al P2 receptors were not involved in the facilitating effects of cADPR In contrast, the inhibitors of intra-cellular cADPR receptors 8-Br-cADPR (80 lm) and ryanodine (50 lm for 45 min) abolished the enhancing effect of cADPR (Fig 6) Therefore, the responses to exogenous cADPR are probably mediated by intracel-lular ryanodine-sensitive cADPR receptors

cADPR is hydrolyzed to ADPR [4], which is degraded to AMP by nucleotide pyrophosphatases [25] AMP, in turn, is degraded to Ado by ecto-5¢-nucleotidase [26], but AMP can also synthesize ADP and ATP via backward ecto-phosphotransfer reactions, provided that enzymes such as adenylate kinase, nucleoside diphosphate kinase and ATP syn-thase [27] are present on the cell surface Therefore, we next examined whether the increase in ATP during su-perfusion with cADPR is, rather, attributable to regen-eration of ATP from AMP or ADP, distant products

of cADPR The commercially available ADP sub-stance used in these experiments at a concentration of

10 nm contained a small amount of ATP, which, nor-malized to tissue weight, is about 0.78 ± 0.09 fmo-lÆmg)1 tissue (n = 4) Perfusion with ADP did not result in additional formation of ATP: thus, the level

of ATP was 0.85 ± 0.06 fmolÆmg)1 tissue in the sam-ples collected during perfusion with ADP (n = 4,

P> 0.05 versus nontissue controls) Likewise, perfu-sion of tissue with AMP (10 nm) caused no additional formation of ATP: 0.514 ± 0.081 fmolÆmg)1in nontis-sue controls (n = 4), and 0.466 ± 0.023 fmolÆmg)1 tis-sue in bladders perfused with 10 nm AMP (n = 4,

P> 0.05) Therefore, superfusion of tissues with either ADP or AMP caused no additional formation of ATP

in tissue superfusates, suggesting that kinase activities mediating production of ATP from ADP or AMP

Spontaneous overflow

Evoked overflow

, 

, 

(and ultimately from cADPR) were undetectable under

our experimental conditions

To determine whether b-NAD+, a precursor of

cADPR, affects the spontaneous or EFS-evoked

over-flow in a manner similar to cADPR, we superfused

bladder detrusor muscles isolated from CD38+⁄ +mice

with b-NAD+ (1 nm) The resting overflow of ATP

was 1.81 ± 0.22 fmolÆmg)1 tissue (n = 12) and

3.72 ± 0.85 fmolÆmg)1 tissue (n = 12) in the absence

and presence of b-NAD+ (P > 0.05) The

EFS-evoked overflow of ATP was 5.91 ± 0.91 fmolÆmg)1

tissue (n = 12) in the presence of b-NAD+(P > 0.05

versus PS in b-NAD+-treated tissues; P > 0.05 versus

ST in controls)

To determine whether ADPR, a product of cADPR,

has an effect on the ATP release, we superfused

blad-ders isolated from CD38+⁄ + mice with 1 nm ADPR

The overflow of ATP was 3.56 ± 0.51 fmolÆmg)1tissue

(n = 6, P > 0.05 versus controls) in samples collected

before EFS and 10.07 ± 0.94 fmolÆmg)1tissue (n = 6,

P < 0.05 versus controls) in superfusate samples col-lected during EFS

It has been proposed that, in PC12, cells acetylcho-line induces the production of cADPR via CD38-medi-ated mechanisms [28] To determine whether acetylcholine that might have been released during EFS of murine bladder detrusor smooth muscles caused increased formation of ATP, we examined the effect of carbachol, a stable analog of acetylcholine, on the spontaneous overflow of ATP Carbachol (1 lm) caused no additional formation of ATP in bladder detrusor muscles isolated from CD38+⁄ + and CD38) ⁄ )mice: the amounts of ATP were 0.86 ± 0.14 and 0.70 ± 0.13 fmolÆmg)1 tissue in the absence and presence of carbachol, respectively (n = 4, P > 0.05) Therefore, stimulation of acetylcholine receptors or smooth muscle contraction per se did not induce addi-tional release of ATP

Min

CD38 +/+

cGDPR

NGD

(–) Tissue

(+) Tissue

CD38 –/–

cGDPR

NGD

200 LU

0

2 3

1

(–) Tissue (+) Tissue

200 LU

B A

D C

CD38 +/+

CD38 –/–

(+) Tissue

0

3

1 2

(–) Tissue

**

(–) Tissue

(+) Tissue

Min

(9)

(9)

(6) (6)

Fig 3 CD38 carries the GDP-ribosyl cyclase activity in bladder detrusor muscle (A) Original chromatograms showing the formation of cGDPR from NGD (0.2 m M ) in the absence of tissue [( )) tissue)] and in the presence of tissue for 2 min [(+) tissue)] in CD38 + ⁄ + mice A significant increase in cGDPR production occurred within 2 min of tissue contact LU, luminescence units (B) Averaged data (in nmolÆmg)1tissue) presented as means ± SE; **P < 0.01 (C) Original chromatograms showing the forma-tion of cGDPR from NGD (0.2 m M ) in the absence of tissue [( )) tissue)] and in the presence of tissue for 2 min [(+) tissue)] in

cGDPR from NGD did not occur within

2 min of tissue contact when CD38 was absent (P > 0.05) (D) Averaged data (nmolÆmg)1tissue) presented as means ± SE Numbers of observations are

in parentheses.

cADPR facilitates the contractile responses to

ATP

ATP at 1–10 lm for 1 min caused transient contractile

responses in bladder detrusor strips cADPR (1 nm)

did not cause measurable changes in the resting

smooth muscle tone, but the responses to ATP were

enhanced in the presence of cADPR (Fig 7)

Discussion

This study demonstrates several new features of

presynaptic neuromodulation in a visceral smooth

muscle Stimulation of intrinsic neurons in murine

bladder detrusor muscle caused release of ATP and

b-NAD+ b-NAD+ was degraded by CD38 to

cADPR and ADPR cADPR enhanced the

spontane-ous release of ATP but not the release of ATP evoked

by action potential firings The enhancing effect of

cADPR on spontaneous release of ATP was: (a)

unaf-fected by inhibition of P2 purinoreceptors; (b)

abol-ished by inhibition of intracellular cADPR receptors;

(c) eliminated by prolonged treatment with ryanodine; and (d) absent in bladders isolated from mice lacking the CD38 gene These data suggest that, in the bladder detrusor muscle, extracellular cADPR can be trans-ported by CD38 to the cytosol, activate cADPR recep-tors on ryanodine-sensitive Ca2+ stores, and facilitate spontaneous ATP release

ATP is a proposed neurotransmitter at the nerve– smooth muscle junction in the urinary bladder [17,29], enteric nervous system [30–32], and blood vessels [33] b-NAD+ is another adenine-based nucleotide that is released upon stimulation of neurosecretory cells [34] and nerves in the bladder [12,13], mesenteric blood vessels [12,14], and large intestine [15,16] In all

of these tissues, ATP and b-NAD+ coexist in tissue superfusates, and, in some cases, b-NAD+mimics the effects of the endogenous neurotransmitter better than ATP [15,16] b-NAD+ is degraded to ADPR and cADPR by NAD-glycohydrolase and ADP-ribosyl cyclase, respectively [2,4] In mammals, both enzymatic activities are associated with CD38 [2,10] The cyclase activity of CD38 is relatively weak [2], but even small

CD38 +/+

ATP ADP eADPR for β-NAD + ADPR + cADPR

AMP

Ado

ATP ADP eADPR for cADPR (1 n M )

ATP ADP eADPR for β-NAD + ADPR + cADPR

AMP

Ado

ATP ADP AMP

Ado

CD38 –/–

Control, no EFS Control, no EFS

cADPR, no EFS cADPR, no EFS

B A

Min

25 LU

Min eADPR for cADPR (1 n M )

0

6

2 4

Control cADPR (1 n M )

C

0

6

2

4

Control cADPR (1 n M )

***

(40)

(11)

(55)

(12)

CD38 –/–

Fig 4 cADPR enhances the spontaneous

overflow of ATP (A, B) Original

chromato-grams showing spontaneous overflow of

ATP in the absence (upper panels) and

presence of cADPR (1 n M ) (lower panels) in

CD38+⁄ +mice and CD38) ⁄ )mice,

respec-tively cADPR caused a significant increase

in the spontaneous overflow of ATP in

CD38+⁄ +mice In CD38) ⁄ )mice,

spontane-ous overflow of ATP was not increased in

the presence of cADPR (P > 0.05) LU,

luminescence units: scale applies to all

chromatograms (C, D) Averaged data

(fmolÆmg)1tissue) presented as

means ± SE; ***P < 0.001 Numbers of

observations are in parentheses.

amounts of the second messenger cADPR [1,2] might

have an effect on the release of cotransmitters in the

smooth muscle CD38, in addition to producing

cAD-PR from extracellular b-NAD+, can also transport

cADPR in the intracellular compartment [9–11] This

might not be a universal mechanism, however, as some

cells, such as T-lymphocytes [35], do not express

CD38-mediated transport of cADPR If this

mecha-nism were present in ATP-releasing nerve terminals,

then cADPR, formed extracellularly, would affect the

release of neurotransmitters, a process that depends

heavily on elevated Ca2+in the cytosol [36,37] To test

this hypothesis, we used murine bladder detrusor

mus-cle as a smooth musmus-cle organ with established

puriner-gic cotransmission in the parasympathetic nervous

system [17,18,29] In agreement with previous studies

in the bladder [12,13], we found that both ATP and

b-NAD+ are released spontaneously and upon action

potential firing As expected, the evoked release of

ATP in bladders isolated from CD38+⁄ + mice was

inhibited by TTX, and ATP during EFS therefore

appeared to originate from excitable cells containing fast Na+ channels, such as neurons Interestingly, the evoked release of ATP in bladders isolated from CD38) ⁄ ) mice demonstrated lack of sensitivity to TTX, despite the large number of observations Fur-ther studies are warranted to examine the mechanisms underlying the switch to TTX-resistant release of ATP during EFS in bladders from CD38) ⁄ ) mice As expected, the EFS-evoked release in bladders isolated from both CD38+⁄ +and CD38) ⁄ )mice was abolished

by BoNTA, suggesting that this release was mediated

by SNAP-25-dependent vesicle exocytosis

Multiple mechanisms may be involved in the basal release of ATP from cells [38], including numerous types of membrane channel, such as connexin and pannexin hemichannels [39,40], maxi-ion channels [41], volume-regulated anion channels [42], the P2X7 receptor [43], ATP-binding cassette transporters [44],

or vesicle exocytosis [45] The mechanisms responsible for this release may differ among different types of cell In the present study, the spontaneous release of

cADPR, 16 Hz

ATP

ADP

eADPR for cADPR (1 n M )

AMP

Ado

Control, 16 Hz

ATP ADP eADPR for β-NAD + ADPR + cADPR

AMP

Ado

cADPR, 16 Hz

ATP ADP eADPR for cADPR (1 n M )

AMP

Ado

ATP

ADP

eADPR for β-NAD + ADPR + cADPR

Control, 16 Hz

B A

D C

0

6 8

2 4

Control cADPR (1 n M ) 0

6

8

2

–1 tissue)

–1 tissue) 4

Control cADPR (1 n M )

25 LU

Min

Min

(40) (55)

(12)

(11)

Fig 5 cADPR does not change the EFS-evoked overflow of ATP (A, B) Original chromatograms showing EFS-evoked (16 Hz, 0.1 ms for 60 s) overflow of ATP in the absence (upper panels) and presence of cADPR (1 n M ) (lower panels) in CD38+⁄ + mice and CD38) ⁄ )mice, respectively cADPR did not affect the EFS-evoked over-flow of ATP in CD38 +⁄ + mice or CD38) ⁄ ) mice (P > 0.05) LU, luminescence units: scale applies to all chromatograms (C, D) Averaged data (fmolÆmg)1tissue) presented

as means ± SE Numbers of observations are in parentheses.

ATP in bladders from both CD38+⁄ + and CD38) ⁄ )

mice was insensitive to inhibition of fast Na+channels

with TTX, inhibition of connexin and pannexin

hemi-channels with CBX and FFA, and cleavage of

SNAP-25 with BoNTA Importantly, the spontaneous release

of ATP in the bladder was activated by stimulation of

intracellular cADPR receptors with cADPR (discussed

below) The spontaneous release of ATP also tended

to be reduced by inhibition of ryanodine

recep-tor⁄ channels, although statistical significance was not

reached The precise mechanisms of spontaneous

release of ATP in the bladder remain to be determined,

but the present study suggests that this release is not

induced by action potential firing in peripheral nerves,

by opening of hemichannels, or by vesicle exocytosis,

and requires intact ryanodine-sensitive and

cADPR-sensitive intracellular Ca2+stores

cADPR is formed in the murine bladder, as it does

express ADP-ribosyl cyclase activity measured as

GDP-ribosyl cyclase activity Although the

ADP-ribo-syl cyclase and GDP-riboADP-ribo-syl cyclase activities are not

always equivalent [46], in the mouse bladder the

cyclase activities appear to be carried entirely by

CD38: bladders isolated from CD38) ⁄ ) mice failed to form cGDPR from NGD, which is in contrast to the findings in bladders isolated from CD38+⁄ + mice Furthermore, tissue superfusates from bladders iso-lated from CD38) ⁄ ) mice contained b-NAD+, but almost no cADPR and ADPR (the present study), whereas bladders isolated from CD38+⁄ + mice also contained the b-NAD+ metabolites cADPR and ADPR [12] cADPR, in particular, constituted  12%

of the b-NAD++ ADPR + cADPR cocktail in the

PS samples in bladders isolated from CD38+⁄ + mice [12], whereas the PS samples from CD38) ⁄ ) bladders contained < 2% cADPR in the b-NAD++ ADPR + cADPR mixture Furthermore, the overflow

of Ado and total purines was reduced in the bladders isolated from CD38) ⁄ ) mice, suggesting that, in control tissues, a significant proportion of Ado is formed by the degradation of b-NAD+ via CD38 The data from the overflow experiments and HPLC fraction analysis demonstrate that ATP and cADPR can simultaneously exist in the vicinity of the neuro-muscular junction at rest and during action potential firing

0

4

8

***

(55)

(12)

(9)

(6)

(4) (4)

(40) (11) (7) (3)

CD38 –/–

CD38 +/+

Fig 6 Effects of cADPR on spontaneous overflow of ATP in

blad-der detrusor smooth muscle isolated from CD38 +⁄ + mice or

means ± SE Numbers of observations are in parenthesis cADPR

(1 n M ) significantly increased the spontaneous overflow of ATP in

CD38 +⁄ + mice (***P < 0.001) The enhancing effect was also

observed in the presence of PPADS (30 l M ), a nonselective P2

pur-ine receptor antagonist (***P < 0.001) The inhibitor of intracellular

cADPR receptors, 8-Br-cADPR (80 l M ), and ryanodine (50 l M )

abol-ished the enhancing effect on spontaneous ATP overflow

(P > 0.05) cADPR did not affect spontaneous ATP overflow when

CD38 was absent (CD38) ⁄ ), P > 0.05).

1 mN

ATP cADPR, 1 nM

0

2

1

**

ATP

30 s

(11)

(11)

A

B

Fig 7 Exogenous cADPR facilitates the contractile responses to ATP in bladder smooth muscle strips (A) ATP (1 l M ) caused tran-sient contractile responses, which were enhanced in the presence

of cADPR (1 n M ) (B) Averaged data (mN force) presented as means ± SE Numbers of observations are in parentheses.

The amounts of cADPR produced by released

b-NAD+may be relatively low, given that the

mamma-lian ADP-ribosyl cyclase associated with CD38 converts

only 2% of b-NAD+ to cADPR [2,10] We therefore

sought to determine whether low concentrations of

cADPR can affect the amounts of released ATP in the

bladder We found that a low nanomolar concentration

of cADPR enhances the spontaneous overflow of ATP,

but does not change the release of ATP evoked by

action potential firing These differential effects of

cAD-PR can be explained by differences in the dependence

of ‘spontaneous’ and ‘evoked’ release of

neurotransmit-ters on extracellular and intracellular Ca2+ For

example, it is well accepted that physiological

neurotransmitter release is largely triggered by action

potential-evoked Ca2+ influx through voltage-gated

Ca2+channels localized on presynaptic nerve terminals

[36] Unlike this ‘evoked’ release, the ‘spontaneous’

release of neurotransmitters is not triggered by action

potential firing Spontaneous vesicle fusion is thought

to be a Ca2+-independent process, because it occurs

both in the absence of action potentials and without

any apparent stimulus However, increasing evidence

shows that this form of neurotransmitter release can be

modulated by changes in intracellular Ca2+

concentra-tion [37,47] Modulaconcentra-tion of spontaneous discharge at

the level of the release machinery is not always

accompanied by corresponding modulation of action

potential-evoked release, suggesting that two

indepen-dent processes underlie spontaneous and action

potential-evoked exocytosis [47] In agreement with this

notion, the present study demonstrates that exogenous

cADPR modulates the spontaneous but not the action

potential-evoked release of ATP Therefore, the

neuro-modulator effects of cADPR are not mediated by influx

of extracellular Ca2+, but are probably caused by Ca2+

release from intracellular stores Similar to cADPR, its

precursor b-NAD+did not affect the evoked release of

ATP, but tended to increase the spontaneous release of

ATP, suggesting that the effects of b-NAD+might be

mediated by its metabolite cADPR ADPR, a product

of both b-NAD+and cADPR [2,10], did not enhance

the spontaneous overflow of ATP, suggesting that the

effect of cADPR was not caused by its breakdown

product ADPR Unlike cADPR and b-NAD+,

however, ADPR facilitated the EFS-evoked release of

ATP Further studies are needed to determine the

mechanisms of purine-mediated presynaptic

neuromod-ulation in the bladder

The enhancing effect of cADPR on the spontaneous

release of ATP is not caused by activation of

membrane-bound P2 purinoceptors, backward

ecto-phosphotransfer reactions and formation of ATP from

either ADP or AMP [27] potentially produced by the exogenous cADPR, or acetylcholine-induced produc-tion of cADPR [28] Instead, the enhancing effect of cADPR on the spontaneous release of ATP is inhibited by 8-Br-cADPR, a specific antagonist of cADPR receptors in intracellular Ca2+stores [48], and

by ryanodine, which, at higher concentrations and with prolonged application, also inhibits Ca2+release chan-nels (receptors) in intracellular Ca2+stores [49] These findings suggest that the effect of exogenous cADPR

on the spontaneous release of ATP is mediated by receptors localized in the intracellular compartment Mechanisms for cADPR influx must, then, be present

in this preparation Of particular importance is the finding that exogenous cADPR failed to increase the spontaneous release of ATP in the absence of CD38

In other words, the presence of CD38 is mandatory for the occurrence of intracellular actions of extracellu-lar cADPR Low concentrations of cADPR, which do not produce measurable changes in mechanical force

in bladder preparations, potentiated the contractile responses to ATP, suggesting that our observations that cADPR enhances the spontaneous release of ATP may imply novel mechanisms of cotransmission that might be important for the fine tuning of bladder functions

In conclusion, the present study suggests that the enhancing effects of extracellular cADPR on ATP release are mediated by the triggering of intracellular signal transduction pathways in response to cADPR transported into the cytosol via membrane-bound CD38 Thus, similar to studies in some cell lines [9,10], the present study suggests that extracellular cADPR can be transported into the cytosol by CD38 on nerve cell membranes in a smooth muscle organ The extracellular b-NAD+–cADPR system, together with CD38, may thus participate in the complex mech-anisms of synaptic regulation of smooth muscle functions

Experimental procedures Animals used

C57BL⁄ 6 mice (45–60 days of age; Charles River Laborato-ries, Wilmington, MA, USA) and CD38 knockout mice (CD38) ⁄ ); The Jackson Laboratory, Bar Harbor, ME, USA) were anesthetized with isoflurane and decapitated after cervical dislocation This method is approved by the Institutional Animal Care and Use Committee at the University of Nevada Urinary bladders were dissected out and placed in oxygenated cold (10C) Krebs solution with the following composition: 118.5 mm NaCl, 4.2 mm KCl,