www.elsevier.com/locate/npep

Neuropeptides 40 (2006) 283–290

Neuropeptides

Vasoactive intestinal polypeptide and pituitary adenylate cyclaseactivating polypeptide: Effects on insulin release in isolated mouse

islets in relation to metabolic status and age

Solveig Persson-Sjogren a,*, Sture Forsgren b, Per Lindstrom a

a Department of Integrative Medical Biology, Section for Histology and Cell Biology, Umea University, SE-901 87 Umea, Swedenb Department of Integrative Medical Biology, Section for Anatomy, Umea University, SE-901 87 Umea, Sweden

Received 6 January 2006; accepted 4 April 2006Available online 23 June 2006

Abstract

Obesity and development of the metabolic syndrome is related to an increased parasympathetic tone and hyperinsulinemia. Wehave now studied the effects of age and metabolic status on glucose-induced insulin release stimulated by the neuropeptides vaso-active intestinal polypeptide (VIP; 10 nM) and pituitary adenylate cyclase activating polypeptide (PACAP; 10 nM), that are constit-uents of the parasympathetic nerves in the islets, and the cholinergic agonists acetylcholine (ACh; 10 lM) and carbachol (10 lM), inisolated islets from female obese ob/ob mice and lean mice. Both VIP and PACAP enhanced insulin secretion in islets from 4-week-old hyperglycemic ob/ob mice. VIP did not increase 11.1 mM glucose-induced insulin release in islets from 4-week-old lean normo-glycemic mice and neither did PACAP in the absence of bicarbonate. The neuropeptides increased insulin release in islets from 9 to10-month-old mice but VIP and PACAP had no effect in islets from very old mice. ACh had no effect in islets from 9 to 10-monthsand older ob/ob mice in the absence of bicarbonate. The combination of VIP and cholinergic agonists had an additive effect in isletsfrom ob/ob mice, and PACAP combined with carbachol potentiated insulin release in islets from 4-week-old lean mice. VIPincreased early phase insulin release in perifused islets from young mice. A higher concentration of theophylline was needed topotentiate glucose-induced insulin release in islets from young lean mice than in islets from old lean mice and ob/ob mice. The pres-ent results demonstrate age-related dynamics in the effects of neuropeptides affecting cAMP in pancreatic islets. We suggest that VIPand PACAP contribute to the developing metabolic syndrome in ob/ob mice by aggravating hyperinsulinemia.� 2006 Elsevier Ltd. All rights reserved.

Keywords: Aged; ob/ob Mouse islets; HCO�3 ; Acetylcholine; Carbachol; VIP; PACAP; Theophylline

1. Introduction

The endocrine pancreas is innervated by parasympa-thetic neurons demonstrating acetylcholine (ACh) (Pers-son-Sjogren et al., 2001; Persson-Sjogren, 2001) and theneuropeptides vasoactive intestinal polypeptide (VIP)(Havel et al., 1997; Persson-Sjogren et al., 1996, 1998)

0143-4179/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.npep.2006.04.001

* Corresponding author. Tel.: +46 90 7865117; fax: +46 90 7866696.E-mail address: Solveig.Persson-Sjogren@histocel.umu.se (S. Pers-

son-Sjogren).

and pituitary adenylate cyclase-activating polypeptide(PACAP) (Tornøe et al., 1997). They are all releasedin the pancreas by activation of the vagus nerve (Ahren,2000). Colocalization of VIP and acetylcholinesterase(AChE) (Persson-Sjogren et al., 1996) and VIP andPACAP (Tornøe et al., 1997) in islet nerves suggestsan interplay between ACh and the neuropeptides. VIPand PACAP enhance glucose-induced insulin releasein vivo in mice (Ahren and Lundquist, 1981) andhumans (Filipsson et al., 1997) and in islets isolatedfrom mice (Filipsson et al., 1998) and rats (Jamenet al., 2002), and parasympathetic nerves are thought

284 S. Persson-Sjogren et al. / Neuropeptides 40 (2006) 283–290

to regulate insulin release during food consumption(Strubbe and Steffens, 1993).

VIP binds to b-cell VIP2/PACAP (VPAC2) receptorsand PACAP binds to a PACAP-specific receptor(PAC1) and to VPAC2 receptors (Filipsson et al., 1998).The receptors are G-protein coupled and act by increasingcAMP production (Jamen et al., 2002) and elevating b-cellcytoplasmic Ca2+ (Wahl et al., 1993; af Klinteberg et al.,1996). Cholinergic agonists enhance glucose-inducedinsulin secretion by binding to muscarinic M3-receptors(Henquin and Nenquin, 1988; Boschero et al., 1995).M3-receptor activation increases formation of inositol1,4,5 trisphosphate (IP3) that triggers mobilization ofCa2+ from intracellular stores, and diacylglycerol(DAG) induced activation of protein kinase C (PKC)(Ahren et al., 1990). M3-receptor agonists also depolarizethe b-cell by increasing Na+ influx (Miura et al., 1996).

Obesity is linked to an increased parasympathetictone (Tassava et al., 1992) and both are suggested tobe involved in the development of insulin resistance(Ahren, 2000). Ob/ob mice are obese, hyperglycemicand hyperinsulinemic (Zhang et al., 1994). They do notshow signs of uncompensated type 2 diabetes and there-fore the ob/ob mouse is a good model for studying theeffects of obesity and hyperglycemia on insulin releasein a pre-diabetic state. Ob/ob mouse islets release moreinsulin in response to glucose (Zawalich and Zawalich,1996) and acetylcholine (Tassava et al., 1992; Persson-Sjogren and Lindstrom, 2004) than islets from lean mice.

Ageing is accompanied by changes in cholinergic neu-rons in both the central and the peripheral nervous sys-tems. Old rats have a lowered release of ACh in the CNS(Herzog et al., 2003) and there is a loss of cholinergicneurons in the myenteric plexus (Phillips et al., 2003).We have observed a reduced insulin response to M-receptor agonism in islets isolated from old ob/ob micesuggesting that chronic exposure to high glucose levelsleads to an impaired neural regulation of insulin release(Persson-Sjogren and Lindstrom, 2004). VIP andPACAP may control insulin release in situations of ahigh insulin demand (Filipsson et al., 1998). We havenow tested the effect of VIP and PACAP on insulinrelease in islets isolated from obese and lean mice of dif-ferent ages. VIP and ACh are released upon stimulationof parasympathetic nerves and VIP potentiates the effectof ACh in salivary glands (Lundberg et al., 1982). Wetherefore also studied the combined effects of M-recep-tor agonists and VIP/PACAP.

2. Materials and methods

2.1. Animals

Non-inbred, 4–16-week- and 9–17-month-old femaleUmea-ob/ob-mice (ob/ob) and their lean littermates

(lean mice) were used. Four-week- and 24-month-oldBALB/c female mice from Bomholtgard, Denmark wereused as controls. All animal experiments were carriedout in accordance with the principles set forth in the‘‘Guide for the Care and Use of Laboratory Animals’’(NIH publication No. 86-23, revised 1985) and wereapproved by the Local Ethical Committee on AnimalExperiments in Northern Sweden.

2.2. Drugs

Collagenase, dextran, carbachol, theophylline, VIP,PACAP, acetylcholine chloride and eserine (hemisul-phate salt) were from Sigma Chemical Co., St. Louis,MO, USA. BSA (fractionV) was from Miles, Slough,UK, Hepes was from Boehringer Mannheim GmbH,Germany. Redestilled water was used throughout. Allother chemicals were of analytical grade.

2.3. Methods

2.3.1. Islet isolationAll mice were starved overnight and the pancreases

were digested with collagenase to isolate individualislets. The medium used for the collagenase isolationwas a Krebs–Ringer medium (KRH) with the followingcomposition in (mM): NaCl, 130; KCl, 4.7; KH2PO4,1.2; MgSO4, 1.2; CaCl2, 2.56; 20 mM Hepes. The med-ium was supplemented with bovine serum albumin(BSA) at 1 mg/ml. NaOH was used to adjust the pHto a final pH of 7.4.

Two different incubation media were used. The med-ium without HCO�3 =CO2 will be referred to as ‘‘KRH’’and was the same as that used in the collagenase isola-tion of islets. The other buffer was gassed with O2/CO2

(19:1) and will be referred to as ‘‘KRBH’’. The‘‘KRBH’’ buffer had the following composition: (mM):NaCl, 115; KCl, 4.7; KH2PO4, 1.2; MgSO4, 1.2; CaCl2,2.56, NaHCO3, 25; Hepes 20. BSA (10 mg/ml) and glu-cose (3 mM) were added. The pH was adjusted withNaOH to a final pH of 7.4. Glucose was added as indi-cated in the tables and figures.

All experiments with ACh included eserine, 10 lM.

2.3.2. Insulin release

For batch incubation studies, 3–10 islets were prein-cubated in 1 ml medium (KRH or KRBH) supple-mented with 3 mM glucose and then incubated for60 min in 300 ll medium with 11.1 mM glucose, testsubstances being added as indicated in the tables.

For dynamic studies, batches of 40 islets were placedin a perifusion chamber housed in an infant incubator at37 �C (Lindstrom and Sehlin, 1982). The islets were peri-fused at a rate of 1 ml/min with KRBH supplementedwith glucose and neurotransmitters as indicated in thefigures. After a 40 min preperifusion with 3 mM glucose

Fig. 1. Insulin release in batch incubated islets isolated from 4-week-old ob/ob (white columns) and lean mice (black columns). The isletswere preincubated with KRH buffer 3 mM glucose for 30 min and thenincubated with 11.1 mM glucose for 60 min in the presence of VIP(10 nM), ACh (10 lM) and PACAP (10 nM) as indicated. Data areexpressed as mean ± SEM using Kruskal–Wallis non-parametric testcombined with Fisher’s PLSD test. ***, P < 0.001; *, P < 0.05 vs.control (11.1 mM glucose alone). Number of experiments is indicatedin the columns.

S. Persson-Sjogren et al. / Neuropeptides 40 (2006) 283–290 285

the medium was changed to KRBH supplemented with11.1 mM glucose or 11.1 mM glucose + 10 nM VIP. Theislets were perifused for 20 min with test media andfinally for 20 min with 3 mM glucose. During perifusionin test medium samples of effluent were collected over 1-min intervals during the first 5 min and then over inter-vals of 2, 3 and 2 · 5 min. During the perifusion in3 mM glucose, samples were collected over intervals of15, 3 and 2 min.

After the batch-incubations or perifusions, the isletswere transferred to small pieces of aluminium foil andfreeze-dried overnight (�40 �C, 0.1 Pa). Islets wereweighed on a quartz-fibre balance. Samples from theincubation buffers were taken for radioimmunologicalanalysis of insulin by using 125I-labelled porcine insu-lin (EuroDiagnostica) and crystalline mouse insulinas reference. Free and antibody-bound insulin wereseparated by precipitation with 81% (v/v) ethanol.The quantity of insulin was related to the islet dryweight.

2.4. Statistical analysis

Statistical analyses were performed by using thetwo-tailed Student’s t-test for paired samples or if sostated, Kruskal–Wallis test or ANOVA supplementedwith Fisher’s protected least significant difference(ANOVA).

3. Results

3.1. Effect of VIP, PACAP and M-receptor agonists on

insulin release in isolated islets from 4-week-old mice

VIP stimulated insulin release in islets from youngob/ob mice, but not in islets from young lean mice.Ten nM VIP stimulated insulin release in islets from 4-week-old ob/ob mice both in the presence and absenceof bicarbonate (Fig. 1, Table 1). VIP did not increaseglucose-induced insulin release in islets from 4-week-old lean (Table 1) or BALB/c mice [11.1 mM glucose:0.46 ± 0.17 ng/lg dry islet weight · h�1 (ng/lg) vs.VIP: 0.53 ± 0.10 ng/lg; n = 8].

ACh, 10 lM, increased insulin secretion in islets fromboth ob/ob and lean mice (Fig. 1, Table 1). There was noadditional effect when VIP was also added (Fig. 1, Table1).

PACAP had no effect in islets from 4-week-old leanmice when bicarbonate was omitted (Fig. 1). BothPACAP and carbachol increased insulin secretion inislets from young ob/ob and lean mice in the presenceof bicarbonate (Table 2). PACAP combined with carba-chol potentiated insulin release stimulated by 11.1 mMglucose in islets from 4-week-old lean mice (Table 2)but not in islets from ob/ob mice (Table 2).

3.2. Effect of VIP, PACAP and M-receptor agonists on

insulin release in isolated islets from 9 to 24-month-oldmice

VIP stimulated insulin release in islets from old ob/ob

mice and lean mice, but the effect of ACh was reduced inislets from old mice.

VIP, 10 nM, stimulated insulin release in islets from 9to 10-month-old ob/ob mice and lean mice (Fig. 2, Table3). The effect of VIP was also observed in islets fromlean mice incubated in KRBH (VIP + 11.1 mM glucose:3.48 ± 0.58 ng/lg vs. 11.1 mM glucose alone: 1.82 ±0.36 ng/lg; P < 0.01, n = 8). VIP stimulated glucose-induced insulin release in islets from 17-month-old obesemice when bicarbonate was omitted (Table 4), but VIPhad no effect in islets from 24-month-old BALB/c mice(VIP: 1.34 ± 0.36 ng/lg vs. 11.1 mM glucose: 1.18 ±0.11 ng/lg, n = 7). The stimulatory effect was the samein islets from young (134 ± 32%) and old (108 ± 14%)obese mice.

ACh and carbachol stimulated insulin secretion inislets from 9 to 10-month-old ob/ob mice in the presenceof bicarbonate (Table 3) but the effect of ACh was lowerin islets from 9 to 10-month-old ob/ob mice than in isletsfrom 4-week-old ob/ob mice (207 ± 68% compared with11.1 mM glucose alone: Table 3 vs. 397 ± 40%: Table 2;P < 0.01, ANOVA). ACh had no effect when bicarbon-ate was omitted (Fig. 2). Neither was there any effect ofACh alone (1.28 ± 0.42 ng/lg), or the combination of

Table 2Effect of PACAP and carbachol on insulin release in KRBH buffer

Test substance Insulin release (ng/lg dry islet weight · h�1)

ob/ob mouse Compared with number Lean mouse Compared with number

1 1.29 ± 0.21 (10) 0.28 ± 0.05 (9)2 PACAP 2.30 ± 0.42 (10) 1 P < 0.05 0.42 ± 0.07 (10) 1 P < 0.053 Carbachol 8.84 ± 0.77 (10) 1 P < 0.001 3.05 ± 0.29 (10) 1 P < 0.0014 PACAP, carbachol 9.65 ± 0.77 (10) 2 P < 0.01, 3 n.s. 4.73 ± 0.37 (9) 2 P < 0.001, 3 P < 0.01

Islets from 5-week-old mice were prepared and preincubated for 30 min in the presence of 3 mM glucose as described in Section 2.3 and thenincubated in 11.1 mM glucose for 60 min at 37 �C in the presence of PACAP (10 nM) and carbachol (10 lM) as indicated in the table. Data areexpressed as mean values ± SEM for the number of experiments indicated in parentheses. n.s. denotes P > 0.05 for difference from control usingStudent’s t-test for paired data.

Table 1Effect of VIP and ACh on insulin release in KRBH buffer

Test substance Insulin release (ng/lg dry islet weight · h�1)

ob/ob mouse Compared with number Lean mouse Compared with number

1 2.09 ± 0.26 (8) 0.49 ± 0.24 (5)2 VIP 4.51 ± 0.52 (8) 1 P < 0.01 0.68 ± 0.08 (7) 1 n.s.3 ACh 9.88 ± 0.73 (8) 1 P < 0.001 5.87 ± 0.71 (7) 1 P < 0.014 VIP, ACh 12.48 ± 1.62 (8) 2 P < 0.001, 3 n.s. 10.01 ± 1.92 (7) 2 P < 0.001, 3 n.s.

Islets from 4-week-old ob/ob and lean mice, were prepared and preincubated for 30 min in the presence of 3 mM glucose as described in Section 2.3and then incubated in 11.1 mM glucose for 60 min at 37 �C in the presence of VIP (10 nM) and ACh (10 lM) as indicated in the table. Data areexpressed as mean ± SEM for the number of experiments using Student’s t-test for paired data.

Fig. 2. Insulin release in batch incubated islets isolated from 9 to 10-month-old ob/ob (white columns) and lean mice (black columns). Thebatch incubation was performed as described in Fig. 1. Data areexpressed as mean ± SEM. *, P < 0.05; ***, P < 0.001 vs. control(11.1 mM glucose alone). Number of experiments is indicated in thecolumns.

286 S. Persson-Sjogren et al. / Neuropeptides 40 (2006) 283–290

VIP and ACh (1.42 ± 0.21 ng/lg n = 7) in isletsfrom 24-month-old BALB/c mice in the absence ofbicarbonate.

PACAP strongly increased insulin release in isletsfrom 9 to 10-month-old ob/ob and lean mice in KRH(Fig. 2) but PACAP had no effect in islets from 17-month-old ob/ob mice (Table 4). There was also no effectof PACAP in islets from 16-month-old BALB/c mice inKRH (data not shown). The combination of PACAPand VIP had no additive effect in any group of islets(Fig. 2, Tables 3 and 4).

3.3. Effect of VIP on glucose-induced insulin release in

perifused islets from 5-week-old mice

VIP (10 nM) enhanced glucose-induced insulinrelease in islets from 5-week-old ob/ob and lean miceperifused in KRBH (Figs. 3 and 4). In the ob/ob mouseislets the total amount of insulin released during20 min stimulation with VIP was 2196.01 ± 453.67 ng/mg dry islet weight (ng/mg) vs. 1001.22 ± 96.92 ng/mg, n = 5, P < 0.05 with 11.1 mM glucose alone. Thecorresponding amounts of insulin were 433.84 ±95.38 ng/mg and 183.43 ± 37.40 ng/mg, n = 6, P <0.02 in islets from lean mice. Islets from ob/ob micereleased more insulin than islets from lean micein response to both glucose (P < 0.02) and VIP (P <0.001; ANOVA). VIP had the same potentiating effecton glucose induced insulin release in both types ofislets (ob/ob mouse islets: 119 ± 38% and lean mouseislets: 140 ± 39%).

Table 3Effect of VIP, ACh and carbachol on insulin release in KRBH buffer

Glucose Test substance Insulin release (ng/lg dry islet weight · h�1) Compared with number

1 3 0.15 ± 0.02 (8)2 11.1 4.17 ± 0.40 (8) 1 P < 0.0013 11.1 VIP 8.53 ± 0.78 (8) 2 P < 0.014 11.1 ACh 10.46 ± 1.18 (7) 2 P < 0.0015 11.1 Carbachol 11.08 ± 0.66 (8) 2 P < 0.0016 11.1 VIP, ACh 14.30 ± 1.85 (8) 3 P < 0.05, 4 P < 0.057 11.1 VIP, carbachol 14.54 ± 1.15 (8) 3 P < 0.01, 5 P < 0.05

Islets from 9 to 10-month-old ob/ob mice were prepared and preincubated for 30 min in the presence of 3 mM glucose as described in Section 2.3 andthen incubated in 3 or 11.1 mM glucose for 60 min at 37 �C in the presence of ACh (10 lM), carbachol (10 lM) and VIP (10 nM) as indicated in thetable. Data are expressed as mean ± SEM for the number of experiments indicated in parentheses. n.s. denotes P > 0.05 for difference from controlby using Student’s t-test for paired data.

Table 4Effect of PACAP and VIP on insulin release in KRH buffer

Glucose Test substance Insulin release (ng/lg dry islet weight · h�1) Compared with number

1 3 0.06 ± 0.01 (8)2 11.1 1.60 ± 0.15 (8) 1 P< 0.0013 11.1 VIP 2.56 ± 0.16 (8) 2 P < 0.014 11.1 PACAP 2.18 ± 0.30 (8) 2 n.s.5 11.1 VIP, PACAP 2.34 ± 0.12 (7) 3, 4 n.s.

Islets from 17-month-old ob/ob mice were prepared and preincubated for 30 min in the presence of 3 mM glucose as described in Section 2.3 and thenincubated in 3 or 11.1 mM glucose in the presence or absence of PACAP (10 nM) and VIP (10 nM) for 60 min at 37 �C as indicated in the table. Dataare expressed as mean values ± SEM for the number of experiments indicated in parentheses. n.s. denotes P > 0.05 for difference from control usingStudent’s t-test for paired data.

Fig. 3. Dynamics of insulin release in islets isolated from 5-week-oldob/ob mice. Islets were preperifused in KRBH buffer with 3 mMglucose for 30 min. The thick black bar indicates that the islets wereexposed to medium containing 11.1 mM glucose (s) or 11.1 mMglucose + 10 nM VIP (d) for 20 min. Medium with 3 mM was thenreintroduced. Data are expressed as mean ± SEM for five perifusions.

Fig. 4. Dynamics of insulin release in islets isolated from 5-week-oldlean mice, and perifused in KRBH buffer. The perifusions wereperformed as described in Fig. 3. Data are expressed as mean ± SEMfor six perifusions.

S. Persson-Sjogren et al. / Neuropeptides 40 (2006) 283–290 287

Table 5Effect of theophylline on insulin release in the presence of glucose in KRBH buffer

Test substance Insulin release (ng/lg dry islet weight · h�1)

ob/ob mouse (5 weeks) Compared with Lean mouse (5 weeks) Compared with

2.78 ± 0.32 (6) a 0.35 ± 0.12 (7) b0.05 T 3.45 ± 0.58 (6) a n.s. 1.00 ± 0.61 (7) b n.s.0.5 T 10.06 ± 0.82 (6) a P < 0.001 1.12 ± 0.46 (6) b n.s.1 T 13.47 ± 1.44 (6) a P < 0.001 1.83 ± 0.57 (7) b P < 0.055 T 16.37 ± 1.37 (6) a P < 0.001 11.44 ± 3.72 (7) b P < 0.05

Islets from 5-week-old mice were prepared and preincubated for 30 min in the presence of 11.1 mM glucose ± different concentrations (mM) oftheophylline (T) for 60 min at 37 �C as indicated in the table. Data are expressed as mean values ± SEM for the number of experiments indicated inparentheses. n.s. denotes P > 0.05 for difference from control using Student’s t-test for paired data.

Table 6Effect of theophylline on insulin release in the presence of KRBH or KRH buffer

Test substance Insulin release (ng/lg dry islet weight · h�1)

ob/ob mouse(15 months)

Compared with Lean mouse(15 months)

Compared with Lean mouse(15 months; KRH)

Compared with

4.11 ± 0.45 a 0.80 ± 0.14 (11) b 1.15 ± 0.28 (14) c0.05 T 4.45 ± 0.69 a n.s. 1.12 ± 0.38 (11) b n.s. 1.10 ± 0.12 (14) c n.s.0.5 T 7.18 ± 0.75 a P < 0.01 3.66 ± 0.53 (11) b P < 0.001 2.24 ± 0.33 (14) c P < 0.011 T 9.49 ± 0.56 a P < 0.001 4.44 ± 0.46 (11) b P < 0.001 4.19 ± 0.23 (14) c P < 0.0015 T 12.09 ± 1.62 a P < 0.01 9.05 ± 0.99 (11) b P < 0.001 7.56 ± 0.61 (14) c P < 0.001

Islets from 15-month-old mice were prepared and preincubated for 30 min in the presence of 3 mM glucose as described in Section 2.3 and thenincubated in 11.1 mM glucose ± different concentrations (mM) of theophylline (T) for 60 min at 37 �C as indicated in the table. Data are expressed asmean values ± SEM for the number of experiments indicated in parentheses. n.s. denotes P > 0.05 for difference from control using Student’s t-testfor paired data.

288 S. Persson-Sjogren et al. / Neuropeptides 40 (2006) 283–290

3.4. Effect of theophylline on glucose-induced insulin

release in isolated islets from 5-week to 15-month-old

mice

Theophylline, dose dependently potentiated glucose-induced insulin release (Tables 5 and 6). The effect oftheophylline was stronger in islets from 5-week-old ob/ob mice than in islets from 5-week-old lean mice (Table5).

4. Discussion

We now show that there is an age-related pattern forthe effect of VIP and PACAP on glucose-induced insulinrelease. Islets isolated from young lean mice were lesssensitive to VIP and PACAP stimulation than isletsfrom young ob/ob mice. With increasing age there is areduced sensitivity for cholinergic stimulation in isletsfrom ob/ob mice (Persson-Sjogren and Lindstrom,2004). The present results suggest an age-dependentinterplay between VIP and PACAP and cholinergicagonists.

4.1. Effects of VIP and PACAP on insulin release

VIP increased glucose-induced insulin release inbatch-incubated islets from 4-week-old obese mice but

had no effect on insulin release in islets from 4-week-old lean mice. PACAP enhanced insulin secretion stim-ulated by glucose in islets from 4-week-old ob/ob andlean mice in the presence of bicarbonate. In the absenceof bicarbonate, PACAP stimulated insulin release in ob/ob mouse islets but not in islets from lean mice. Omis-sion of extracellular bicarbonate creates an unphysiolog-ical environment for the islets, but makes it possible tostudy the dependence of buffering capacity on cell func-tion. Islets from young normoglycemic mice may bemore sensitive to changes in intracellular buffering whencompared with islets from ob/ob mice.

VIP stimulated glucose-induced insulin release inislets from 9 to 17-month-old obese and lean mice, buthad no effect in islets from 24-month-old lean mice.PACAP also potently enhanced insulin secretion in isletsfrom the 9 to 10-month-old animals but did not affectsecretion in islets from 16 to 17-month-old animals inthe absence of bicarbonate.

VIP and PACAP are co-localized in islet nerves(Hannibal and Fahrenkrug, 2000) and in the enteropan-creatic nerves (Kirchgessner and Liu, 2001). This co-localization may be of functional significance butcombining PACAP and VIP had no additional effecton insulin release in any type of mouse or age groupwhen compared with the effect of PACAP or VIP alone.PACAP and VIP bind to the same receptor, the VPAC2-receptor, and PACAP also binds to the PAC1 receptor

S. Persson-Sjogren et al. / Neuropeptides 40 (2006) 283–290 289

(Filipsson et al., 1998). When the neuropeptides arecombined they may therefore compete for receptor bind-ing on the b-cell.

4.2. Effects of cholinergic agonists and neuropeptides on

insulin release

Cholinergic agents stimulate insulin release in isletsfrom old mice but only in the presence of bicarbonate(Persson-Sjogren and Lindstrom, 2004). Gagermanet al. (1980) also found a lower response to ACh inbicarbonate free buffer. The present results show thatthe response to ACh is weaker in islets from older ani-mals when compared with islets from young mice alsowhen bicarbonate is present. We suggest that long-termhyperglycemia makes the islets more sensitive tochanges in buffering capacity. The effect of the phos-phodiesterase inhibitor theophylline was larger in isletsfrom 5-week-old ob/ob mice than in islets from 5-week-old lean mice. These results imply that stimuli thatraise islet cAMP levels participate in the developmentof hyperinsulinemia in the young ob/ob mouse. We alsofind that the effects of cAMP signaling does not changewith age to the same extent as cholinergic regulation ofthe islets.

VIP and ACh are co-localized in the submandibulargland and both are released upon parasympathetic ner-vous stimulation (Lundberg et al., 1982). In addition,VIP potentiates the stimulatory effect of ACh on sali-vary secretion (Lundberg et al., 1982). ACh and VIPare co-localized also in islet neurons (Persson-Sjogrenet al., 1996). When islets were incubated in the presenceof both cholinergic agonists and VIP in KRBH buffer,insulin release was larger than with either of the twosubstances alone. The effect was additive and no poten-tiation was observed which suggests that VIP and AChact independently of each other to stimulate insulinrelease. A reduced ACh response could therefore becompensated for by an increased VIP stimulation.

The combination of PACAP and carbachol potenti-ated the stimulatory effect of either of the two sub-stances alone in islets from 4-week-old lean mice. Nosuch potentiating effect was observed in islets from obesemice of the same age. Possibly, the very strong stimula-tory effect of carbachol makes it difficult to detect anyfurther stimulation by PACAP. When islets from 9 to10-month-old ob/ob mice were challenged with a combi-nation of PACAP and carbachol an additive effect wasobserved. PACAP is one of the most insulinotrophicneuropeptides and has effects on glucose-induced insulinrelease at concentrations as low as 10�15 M (Yada et al.,1994). Because of this high potency it is possible that theeffects of the neuropeptide are strictly regulated to avoidhyperinsulinemia in older normoglycemic subjects. Thismay be why PACAP potentiated insulin release only inislets from young lean mice. ACh had no effect on

cAMP content in ob/ob mouse islets (Gagerman et al.,1978). However, cAMP synthesis was increased in iso-lated rat islets in the presence of the M-receptor agonistcarbamylcholine chloride (Tian and Laychock, 2001)indicating a cross-talk between PKC activation and gen-eration of cAMP.

4.3. Transient stimulatory effect of VIP on insulin release

The perifusion experiments confirm that the VIPreceptor system is functional in islets from young ob/ob and lean mice. VIP appears to have a large buttransient stimulatory effect on glucose-induced insulinrelease in islets from young mice. The potentiating effectof VIP on glucose-induced insulin release was the samein islets from both obese and lean mice but islets fromob/ob mice release more insulin than islets from normo-glycemic mice. VIP may therefore play a role in thehypersecretion of insulin in ob/ob mice.

4.4. Conclusions

Neuropeptides are stored in larger vesicles than clas-sical neurotransmitters in the neuronal varicosities, andit is believed that stronger nerve impulses are needed torelease neuropeptides (Hokfelt et al., 1987). It is possiblethat the increased parasympathetic tone observed inobese subjects (Tassava et al., 1992; Weyer et al.,2001) causes a subsequent increased release of VIP andPACAP. Since these neuropeptides are very potent insu-lin secretagogus this can lead to hyperinsulinemia. Per-haps, the strictly regulated effects on insulin release inislets from young normoglycemic mice are explainedby a lower parasympathetic tone. In the hyperinsuline-mic ob/ob mouse (Chen and Romsos, 1995; Edvell andLindstrom, 1999) this precise adjustment of insulinsecretion by the present neuropeptides may be blunted.We suggest that VIP and PACAP contribute to thedevelopment of the metabolic syndrome by amplifyinghyperinsulinemia.

Acknowledgements

This work was financially supported by the SwedishDiabetes Association and the Sahlberg Foundation.We thank Gunilla Forsgren and Eva Bostrom for skilfultechnical assistance. This work was supported by grantsfrom the Medical Faculty, Umea University and theSwedish Diabetes Association.

References

af Klinteberg, K., Karlsson, S., Ahren, B., 1996. Signaling mecha-nisms underlying the insulinotrophic effect of pituitary adenylate

290 S. Persson-Sjogren et al. / Neuropeptides 40 (2006) 283–290

cyclase-activating polypeptide in HIT-T15 cells. Endocrinology199, 2791–2798.

Ahren, B., 2000. Autonomic regulation of islet hormone secretion –implications for health and disease. Diabetologia 43, 393–410.

Ahren, B., Lundquist, I., 1981. Effects of vasoactive intestinalpolypeptide (VIP), secretin and gastrin on insulin secretion in themouse. Diabetologia 20, 54–59.

Ahren, B., Karlsson, S., Lindskog, S., 1990. Cholinergic regulation ofthe endocrine pancreas. In: Aquilonius, S.-M., Gillberg, P.-G.(Eds.), Progress in Brain Research, vol. 84. Elsevier SciencePublishers B.V. (Biomedical Division), Amsterdam, pp. 209–218.

Boschero, A.C., Szpak-Glasman, M., Carniero, E.M., Bordin, S., Paul,I., Rojas, E., Atwater, I., 1995. Oxotremorine-m potentiation ofglucose-induced insulin release from rat islets involves M3 musca-rinic receptors. Am. J. Physiol. 268, E336–E342.

Chen, N.-G., Romsos, D.R., 1995. Enhanced sensitivity of pancreaticislets from preobese 2-week-old ob/ob mice to neurohormonalstimulation of insulin secretion. Endocrinology 136, 505–511.

Edvell, A., Lindstrom, P., 1999. Development of insulin secretoryfunction in young obese hyperglycemic mice (Umea ob/ob).Metabolism 44, 906–913.

Filipsson, K., Tornøe, K., Holst, J., Ahren, B., 1997. Pituitaryadenylate cyclase-activating polypeptide stimulates insulin andglucagon secretion in humans. J. Clin. Endocrinol. Metab. 82,3093–3098.

Filipsson, K., Sundler, F., Hannibal, J., Ahren, B., 1998. PACAP andPACAP receptors in insulin producing tissues: localization andeffects. Regul. Pept. 74, 167–175.

Gagerman, E., Idahl, L.-A., Meissner, H.P., Taljedal, I.-B., 1978.Insulin release, cGMP, cAMP and membrane potential in acetyl-choline-stimulated islets. Am. J. Physiol. 235, E493–E500.

Gagerman, E., Sehlin, J., Taljedal, I.-B., 1980. Effects of acetylcholineon ion fluxes and chlorotetracycline fluorescence in pancreaticislets. J. Physiol. 300, 505–513.

Hannibal, J., Fahrenkrug, J., 2000. Pituitary adenylate cyclase-activating polypeptide in intrinsic and extrinsic nerves of the ratpancreas. Cell Tissue Res. 299, 59–70.

Havel, P.J., Dunning, B.E., Verchere, C.B., Baskin, D.G., O’Dorisio,T., Taborsky Jr., G.J., 1997. Evidence that vasoactive intestinalpolypeptide is a parasympathetic neurortansmitter in the endocrinepancreas in the dog. Regul. Pept. 71, 163–170.

Henquin, J.C., Nenquin, M., 1988. The muscarinic receptor subtype inmouse pancreatic B-cells. FEBS Lett. 236, 89–92.

Herzog, C.D., Nowak, K.A., Sarter, M., Bruno, J.P., 2003. Mikrodi-alysis without acetylcholinesterase inhibition reveals an age-relatedattenuation in stimulated cortical acetylcholine release. Neurobiol.Aging 24, 861–863.

Hokfelt, T., Millhorn, D., Seroogy, K., Tsuruo, Y., Ceccatelli, S.,Lindh, B., Meister, B., Melander, T., Schalling, M., Bartfai, T.,Terenius, L., 1987. Coexistence of peptides with classical neuro-transmitters. Experientia 43, 768–780.

Jamen, F., Puech, R., Bockaert, J., Brabet, P., Bertrand, G., 2002.Adenylate cyclase-activating polypeptide receptors mediating insu-lin secretion in rodent pancreatic islets are coupled to adenylatecyclase but not to PLC. Endocrinology 143, 1253–1259.

Kirchgessner, A.L., Liu, M.T., 2001. Pituitary adenylate cyclaseactivating peptide (PACAP) in the enteropancreatic innervation.Anat. Rec. 262, 91–100.

Lindstrom, P., Sehlin, J., 1982. 5-Hydroxytryptamine stimulates 86Rb+

efflux from pancreatic b-cells. Biochim. Biophys. Acta 720, 400–404.

Lundberg, J.M., Anggard, A., Fahrenkrug, J., 1982. Complementaryrole of vasoactive intestinal polypeptide (VIP) and acetylcholine forcat submandibular gland blood flow and secretion in. Acta Physiol.Scand. 114, 329–337.

Miura, Y., Gilon, P., Henquin, J.-C., 1996. Muscarinic stimulationincreases Na+ entry in pancreatic B-cells by a mechanism otherthan the emtying of intracellular Ca2+ pools. Biochem. Biophys.Res. Commun. 224, 67–73.

Persson-Sjogren, S., 2001. Neuroinsular complex type I: Morphologyand frequency in lean and genetically obese mice. Pancreas 23, 40–48.

Persson-Sjogren, S., Lindstrom, P., 2004. Effects of cholinergic M-receptor agonists on insulin release in islets from obese and leanmice of different ages. The importance of bicarbonate. Pancreas 29,e90–e99.

Persson-Sjogren, S., Forsgren, S., Rooth, P., Taljedal, I.-B., 1996.Initial increase and subsequent loss of vasoactive intestinalpolypeptide immunoreactivity in acetylcholinesterase-positive neu-rons of mouse islets transplanted to the kidney. Cell Tissue Res.284, 391–400.

Persson-Sjogren, S., Forsgren, S., Kjorell, U., Taljedal, I.-B., 1998.Intrinsic and extrinsic NPY nerves in transplanted neuroinsularcomplexes. Peptides 19, 1233–1240.

Persson-Sjogren, S., Zashihin, A., Forsgren, S., 2001. Nerve cellsassociated with the endocrine pancreas in young mice: an ultra-structural analysis of the neuroinsular complex type I. Histochem.J. 33, 373–378.

Phillips, R.J., Kieffer, E.J., Powley, T.L., 2003. Aging of the myentericplexus: neuronal loss is specific to cholinergic neurons. Autonom.Neurosci. 106, 69–83.

Strubbe, J.H., Steffens, A.B., 1993. Neural control of insulin secretion.Horm. Metab. Res. 25, 507–512.

Tassava, T.M., Okuda, T., Romsos, D.R., 1992. Insulin secretion fromob/ob mouse pancreatic islets: effects of neurotransmitters. Am. J.Physiol. 262, E338–E343.

Tian, Y., Laychock, S.G., 2001. Protein kinase C and calciumregulation of adenyl cyclase in isolated rat pancreatic islets.Diabetes 50, 2505–2513.

Tornøe, K., Hannibal, J., Fahrenkrug, J., Holst, J.J., 1997. PACAP-(1-38) as neurotransmitter in pig pancreas: receptor activationrevealed by the antagonist PACAP-(6-38). Am. J. Physiol. 273,G436–G446.

Wahl, M.A., Straub, S.G., Ammon, H.P.T., 1993. Vasoactive intes-tinal polypeptide-augmented insulin release: actions on ionic fluxesand electrical activity of mouse islets. Diabetologia 36, 920–925.

Weyer, C., Salbe, A.D., Lindsay, R.S., Pratley, R.E., Bogardus, C.,Tataranni, P.A., 2001. Exaggerated pancreatic polypeptide secre-tion in Pima Indians: can an increased parasympathetic drive to thepancreas contribute to hyperinsulinemia, obesity, and diabetes inhumans? Metabolism 50, 223–230.

Yada, T., Sakurada, M., Ihida, K., Nakata, M., Murata, F., Arimura,A., Kikuchi, M., 1994. Exaggerated pancreatic polypeptide secre-tion in Pima Indians: can an increased parasympathetic drive to thepancreas contribute to hyperinsulinemia, obesity, and diabetes inhumans? J. Biol. Chem. 269, 1290–1293.

Zhang, Y., Proenca, R., Maffei, M., Barone, M., Leopold, L.,Friedman, . JM., 1994. Positional cloning of the mouse obese geneand its human homologue. Nature 372, 425–432.

Zawalich, W.S., Zawalich, K.C., 1996. Signal transduction in isolatedislets from the ob/ob mouse: enhanced sensitivity of protein kinaseC to stimulation. Biochem. Biophys. Res. Commun. 223, 618–623.