Hippocampal acetylcholine receptor activation-dependent long-term depression in streptozotocin-induced diabetic rats.

Cholinergic innervation of the hippocampus correlates with memory formation. In a well-established animal model of type 1 diabetes mellitus, obtained by injecting young adult rats with streptozotocin (STZ), reductions have been reported in the expression of acetylcholine receptors and choline acetyltransferase. In this study, we showed that long-term synaptic depression (LTD) induced by carbachol (CCh), a nonselective cholinergic receptor agonist, at Schaffer collateral-CA1 synapses in hippocampal slices was significantly weaker in streptozotocin-induced diabetic rats (STZ rats) than in age-matched control rats. No significant change was observed in the paired-pulse ratio between before and 80 min after the application of CCh in control and STZ rats. Moreover, CCh-induced LTD in control and STZ rats was not affected by an NMDA receptor antagonist. Although the application of CCh down-regulated the surface expression of GluA2 in the hippocampus of control rats, but not STZ rats. Therefore, the present results suggest that acetylcholine receptor-mediated LTD in STZ rats requires the internalization of AMPA receptors on the postsynaptic surface and their intracellular effects in the hippocampus.

Intracerebroventricular administration of oxytocin and intranasal administration of the oxytocin derivative improve ß-amyloid peptide (25-35)-induced memory impairment in mice.

AIM: We previously reported that oxytocin, a peptide hormone, can reverse the ß-amyloid peptide (25-35) (Aß25-35 )-induced impairments of the murine hippocampal synaptic plasticity. In this study, we examined the effects of oxytocin on the Aß25-35 -induced impairment of cognitive behavior in murine in order to investigate the potential of oxytocin as a clinical treatment tool for Alzheimer's disease (AD). METHODS: The Y-maze and Morris water maze (MWM) tests were performed. Since the intracerebroventricular (ICV) administration is both invasive and impractical, we further utilized intranasal (IN) delivery to the brain. For this purpose, we prepared an oxytocin derivative containing cell-penetrating peptides and a penetration accelerating sequence, which was subsequently used in our IN administration experiments. RESULTS: We herein showed that the ICV administration of oxytocin in mice exerted memory-improving effects on the Aß25-35 -induced amnesia in both the Y-maze and MWM tests. The IN administration of the oxytocin derivative exhibited memory-improving effects in the Y-maze test. Moreover, we acquired evidence that the fluorescein isothiocyanate-labeled oxytocin derivative was distributed throughout the mouse brain following its IN administration. CONCLUSION: Our results suggest that the oxytocin derivative is effective for its IN delivery to the brain and may be particularly useful in the clinical treatment of cognitive impairment, such as that characterizing AD.

Long-term exposure to high glucose induces changes in the expression of AMPA receptor subunits and glutamate transmission in primary cultured cortical neurons.

Hyperglycemia, which occurs under the diabetic conditions, induces serious diabetic complications. Diabetic encephalopathy has been defined as one of the major complications of diabetes, and is characterized by neurochemical and neurodegenerative changes. However, little is known about the effect of long-term exposure to high glucose on neuronal cells. In the present study, we showed that exposure to glutamate (100 mM) for 7 days induced toxicity in primary cortical neurons using the MTT assay. Additionally, high glucose increased the sensitivity of AMPA- or NMDA-induced neurotoxicity, and decreased extracellular glutamate levels in primary cortical neurons. In Western blot analyses, the protein levels of the GluA1 and GluA2 subunits of the AMPA receptor as well as synaptophysin in neurons treated with high glucose were significantly increased compared with the control (25 mM glucose). Therefore, long-term exposure to high glucose induced neuronal death through the disruption of glutamate homeostasis.

Neuromedin U modulates neuronal excitability in rat hippocampal slices.

Neuromedin U (NMU) is a neuropeptide that was initially isolated from the porcine spinal cord and later from several species. Although NMU receptors exist in the CA1 region of the hippocampus, the role of NMU in hippocampal synaptic transmission remains unknown. In the present study, we demonstrated that the colocalization ratio of NMU type 1 (NMUR1) or type 2 (NMUR2) receptors was higher with neuronal nuclei (a neuronal marker) than with glial fibrillary acidic protein (an astrocyte marker) in the CA1 region of rats. Moreover, we revealed that the bath application of NMU (1 µM) enhanced extracellular field excitatory postsynaptic potentials at Schaffer collateral-CA1 synapses in rat hippocampal slices (+28.9 ± 1.3%; P < 0.05). After extracellular recordings, we examined the pattern of neuronal activation induced by NMU using c-Fos immunohistochemistry (Fos-IR). Histological analyses revealed that NMU increased Fos-IR in the CA1 region, but reduced the proportion of Fos-IR colocalized with glutamic acid decarboxylase (a GABA neuron marker). These results suggest that the activation of NMU receptors contributes to GABAergic neuronal activity in the CA1 region of the hippocampus.

GLP-2 restores impairments in spatial working memory and hippocampal LTD via the MEK/ERK pathway in juvenile-onset diabetes rats.

Type 1 diabetic animal models, generated by injecting streptozotocin (STZ), have been widely used in research. We previously reported that juvenile-onset diabetes mellitus (JDM) rats, which were prepared by administering STZ to 17-day-old rats, developed cognitive impairments and hippocampal synaptic plasticity deficiencies, which were restored by glucagon-like peptide-1 (GLP-1). GLP-1 and GLP-2 are simultaneously derived from proglucagon and act through their own specific receptors. The present study was performed to investigate the potential of GLP-2 in JDM rats. The results obtained demonstrated that GLP-2 restored impairments in spatial working memory and hippocampal long-term depression (LTD) in JDM rats, and that the MEK1/2 inhibitor, U0126, inhibited this recovery. Therefore, GLP-2 has potential in the treatment of cognitive deficits in childhood-onset diabetes.

Cognitive and hippocampal synaptic profiles in monosodium glutamate-induced obese mice.

Obesity is a growing worldwide public health issue and is associated with a range of comorbidities, including cognitive deficits. The present study investigated synaptic changes in the hippocampus during the development of obesity. The treatment of newborn mice with monosodium-L-glutamate (MSG, 2 mg/g) induced obesity and recognition memory deficits in the novel object recognition (NOR) test at 16-17 weeks, but not at 8-9 weeks. Hippocampal synaptic plasticity, including long-term potentiation (LTP) and long-term depression (LTD), and excitatory synaptic transmission at Schaffer collateral-CA1 (SC-CA1) synapses were compared between MSG-treated mice and age-matched control mice. LTP and fiber volley amplitudes were enhanced in MSG-treated mice at 16-17 weeks, but not at 8-9 weeks. Furthermore, the strength of paired-pulse facilitation (PPF) changed in MSG-treated mice at 16-17 weeks, but not at 8-9 weeks. These results suggest that enhanced LTP in the SC-CA1 synapses of MSG-induced obese mice involves presynaptic rather than postsynaptic mechanisms.

Glucagon-like peptide-2 rescues memory impairments and neuropathological changes in a mouse model of dementia induced by the intracerebroventricular administration of streptozotocin.

Glucagon-like peptide 2 (GLP-2) is derived from the proglucagon gene expressed in the intestines, pancreas and brain. Our previous study showed that GLP-2 improved lipopolysaccharide-induced memory impairments. The current study was designed to further investigated the potential of GLP-2 in memory impairment induced by intracerebroventricular administration of streptozotocin (ICV-STZ) in mice, which have been used as an animal model of sporadic Alzheimer's disease (AD). STZ was administered on alternate days (Day-1 and Day-3) in order to induce dementia in male ddY mice. ICV-STZ-treated mice were administered GLP-2 (0.6 µg/mouse, ICV) for 5 days from 14 days after the first ICV administration of STZ. In these mice, we examined spatial working memory, the biochemical parameters of oxidative stress, or neurogenesis. The GLP-2 treatment restored spatial working memory in ICV-STZ-treated mice. ICV-STZ-treated mice showed markedly increased thiobarbituric acid reactive species (TBARS) and decreased glutathione (GSH) levels, and GLP-2 significantly restored these ICV-STZ-induced changes. GLP-2 also significantly restored neurogenesis in the subgranular zone of the dentate gyrus in ICV-STZ-treated mice. We herein demonstrated that GLP-2 significantly restored ICV-STZ-induced memory impairments as well as biochemical and histopathological alterations, and accordingly, propose that the memory restorative ability of GLP-2 is due to its potential to reduce oxidative stress.

Excitatory effect of bradykinin on intrinsic neurons of the rat heart.

The heart receives sympathetic and parasympathetic innervation through the intrinsic cardiac nervous system. Although bradykinin (BK) has negative inotropic and chronotropic properties of cardiac contraction, the direct effect of BK on the intrinsic neural network of the heart is still unclear. In the present study, the effect of BK on the intracardiac ganglion neurons isolated from rats was investigated using the perforated patch-clamp technique. Under current-clamp conditions, application of 0.1 µM BK depolarized the membrane, accompanied by repetitive firing of action potentials. When BK was applied repeatedly, the second responses were considerably less intense than the first application. The BK action was fully inhibited by the B2 receptor antagonist Hoe-140, but not by the B1 receptor antagonist des-Arg9-[Leu8]-BK. The BK response was mimicked by the B2 agonist [Hyp3]-BK. The BK-induced depolarization was inhibited by the phospholipase C inhibitor U-73122. BK evoked inward currents under voltage-clamp conditions at a holding potential of -60 mV. Removal of extracellular Ca2+ markedly increased the BK-induced currents, suggesting an involvement of Ca2+-permeable non-selective cation channels. The muscarinic agonist oxotremorine-M (OxoM) also elicited the extracellular Ca2+-sensitive cationic currents. The OxoM response did not exhibit rundown with repeated agonist application. The amplitude of current evoked by 1 µM OxoM was comparable to that induced by 0.1 µM BK. Co-application of 0.1 µM BK and 1 µM OxoM elicited the current whose peak amplitude was almost the same as that elicited by OxoM alone, suggesting that BK and OxoM activate same cation channels. BK also reduced the amplitude of M-current, while the M-current inhibitor XE-991 affected neither resting membrane potential nor the BK-induced depolarization. From these results, we suggest that BK regulates excitability of intrinsic cardiac neurons by both an activation of non-selective cation channels and an inhibition of M-type K+ channels through B2 receptors.

Pharmacological evidence for the relationship between the NMDA receptor and nitric oxide pathway and the antidepressant-like effects of glucagon-like peptide-2 in the mouse forced-swim test.

We previously demonstrated that glucagon-like peptide-2 (GLP-2) exerted antidepressant-like effects in mice. The aim of the present study was to investigate the relationship between N-methyl-D-aspartate (NMDA) receptor-nitric oxide-cyclic guanosine monophosphate (NO-cGMP) pathway and the antidepressant-like effects of GLP-2 in the forced-swim test (FST) in mice. Intracerebroventricularly administered GLP-2 (3 µg/mouse) decreased the immobility time in the FST. The pretreatment of mice with l-arginine (750 mg/kg, i.p.), a substrate for nitric oxide synthase, sildenafil (5 mg/kg, i.p.), a phosphodiesterase 5 inhibitor, or d-serine (300 mg/kg, i.p.), a NMDA receptor co-agonist, inhibited the antidepressant-like effects of GLP-2 (3 µg/mouse) in the FST. Meanwhile, l-nitroarginine methyl ester (10 mg/kg, i.p.), a non-specific nitric oxide synthase (NOS) inhibitor, 7-nitroindazole (30 mg/kg, i.p.), a neuronal NOS inhibitor, methylene blue (10 mg/kg, i.p.), an inhibitor of both NOS and soluble guanylate cyclase (sGC), ODQ (30 pmol/site, i.c.v.), a sGC inhibitor, or MK-801 (0.05 mg/kg, i.p.), an NMDA receptor antagonist, in combination with a sub-effective dose of GLP-2 (1.5 µg/mouse) also decreased the immobility time in the FST. The present study provided evidence for the synergistic antidepressant-like effects of GLP-2 and inhibition of the NMDA receptor-l-arginine-NO-cGMP pathway in the FST, thereby contributing to our understanding of the mechanisms underlying the antidepressant-like effects of GLP-2.

Effects of centrally administered glucagon-like peptide-2 on blood pressure and barosensitive neurons in spontaneously hypertensive rats.

The central administration of glucagon-like peptide-2 (GLP-2) decreases blood pressure in rats. In the present study, we investigated the hypotensive effects of GLP-2 using spontaneously hypertensive rats (SHRs), an animal model of hypertension. The central administration of GLP-2 (0.6 µg) decreased mean arterial pressure (MAP) in SHRs (-24.1 ±â€¯4.5%; P < 0.05), but not in normotensive Wistar-Kyoto (WKY) rats (-10.6 ±â€¯7.4%; P > 0.05), whereas GLP-2 (6 µg) decreased MAP in WKY rats (-23.5 ±â€¯4.2%; P < 0.05) and SHRs (-46.7 ±â€¯11.6%; P < 0.01) under anesthesia with urethane and α-chloralose. Histological analyses revealed that the central administration of GLP-2 (6 µg) induced Fos immunoreactivity (Fos-IR) in the hypothalamic and medullary areas in WKY rats and SHRs. However, the distribution of Fos-IR in GABAergic neurons in the rostral ventrolateral medulla (RVLM) differed between WKY rats and SHRs. GLP-2 directly modulated the excitability of RVLM neurons in brainstem slices from SHRs, but not WKY rats. These results suggest that neuronal activity through the activation of GLP-2 receptors in the RVLM contributes to lowering blood pressure in SHRs.

Intranasal administration of neuromedin U derivatives containing cell-penetrating peptides and a penetration-accelerating sequence induced memory improvements in mice.

Neuromedin U (NMU) is a neuropeptide that is expressed and secreted in the brain and gut. We previously demonstrated that the intracerebroventricular (i.c.v.) administration of NMU inhibited inflammation-mediated memory impairment in mice. In order to utilize NMU as a clinical treatment tool for inflammation-mediated amnesia, we herein focused on non-invasive intranasal delivery because the i.c.v. administration route is invasive and impractical. In the present study, we prepared two NMU derivatives containing cell-penetrating peptides (CPPs), octaarginine (R8), and each penetration-accelerating sequence, namely FFLIPKG (PASR8-NMU) and FFFFG (F4R8-NMU), for intranasal (i.n.) administration. In the Y-maze test, the i.c.v. administration of lipopolysaccharide (LPS) (10µg/mouse) significantly decreased spontaneous alternation behavior, and this was prevented by the prior administration of PASR8-NMU or F4R8-NMU (5.6µg/mouse, i.n.). Moreover, the administration of PASR8-NMU or F4R8-NMU (5.6µg/mouse, i.n.) just before the Y-maze test also improved LPS-induced memory impairment. Indocyanine green (ICG)-labeled PASR8-NMU (i.n.) was significantly observed in the hippocampus and paraventricular hypothalamic nucleus 30min after its i.n. administration. PASR8-NMU, but not F4R8-NMU guaranteed the stability of the administration liquid for 24h. These results suggest that PASR8-NMU is effective for i.n. delivery to the brain, and may be useful in the clinical treatment of inflammation-mediated amnesia.

Neuroprotection by chotosan, a Kampo formula, against glutamate excitotoxicity involves the inhibition of GluN2B-, but not GluN2A-containing NMDA receptor-mediated responses in primary cultured cortical neurons.

Chotosan (CTS), a traditional herbal formula called Kampo medicine, was shown to be effective in the treatment of vascular dementia in a clinical study, and exerted protective effects against transient cerebral ischemia-induced cognitive impairment in mice. In the present study, we investigated the neuroprotective effects of CTS using primary cultured rat cortical neurons. CTS (250-1000 µg/mL) inhibited neuronal death induced by 100 µM glutamate. This glutamate-induced neuronal death was blocked by a GluN2B-, but not GluN2A-containing NMDA receptor antagonist. Therefore, the neuroprotective effects of CTS were related to an inhibition of GluN2B-containing NMDA receptor-mediated responses.

Serotonergic and dopaminergic systems are implicated in antidepressant-like effects of chotosan, a Kampo formula, in mice.

We previously demonstrated that chotosan (CTS), a traditional herbal formula called Kampo medicine, improves diabetes-induced cognitive deficits. In the present study, we investigated the antidepressant-like effects of CTS in mice. The administration of CTS (1.0 g/kg, for 3 days) decreased the immobility time in the forced-swim test, and this decrease was prevented by the prior administration of sulpiride (an antagonist of D2/3 receptors) and WAY100635 (an antagonist of 5-HT1A receptors). None of the treatments tested altered the locomotor activity of mice. These results suggest that CTS exerts antidepressant-like effects through changes in the serotonergic and dopaminergic systems.

Antidepressant-like effects exerted by the intranasal administration of a glucagon-like peptide-2 derivative containing cell-penetrating peptides and a penetration-accelerating sequence in mice.

The intracerebroventicular (i.c.v.) administration of glucagon-like peptide-2 (GLP-2) to rodents was shown to have antidepressant-like effects in imipramine-resistant depression-model mice. In order to utilize GLP-2 as a clinical treatment tool for depression, we herein focused on the intranasal delivery that is non-invasive approach, because the i.c.v. administration is invasive and impractical. In the present study, we prepared a GLP-2 derivative containing cell penetrating peptides (CPPs) and a penetration accelerating sequence (PAS) (PAS-CPPs-GLP-2) for the intranasal (i.n.) administration. PAS-CPPs-GLP-2 (i.n.) exhibited antidepressant-like effects in the forced-swim test (FST) and tail suspension test (TST) in naïve mice as well as adrenocorticotropic hormone (ACTH) treated-mice. However, PAS-CPPs-GLP-2 (i.v.) and the GLP-2 derivative containing CPPs without a PAS (CPPs-GLP-2) (i.n.) did not affect the immobility time in the mouse FST. Moreover, fluorescein isothiocyanate (FITC)-labeled PAS-CPPs-GLP-2 (i.n.), but not FITC-labeled CPPs-GLP-2 (i.n.) was distributed through the mouse brain after the FST session. These results suggest that PAS-CPPs-GLP-2 is effective for i.n. delivery to the brain, and may be useful in the clinical treatment of major depression.

LARETH-25 and ß-CD improve central transitivity and central pharmacological effect of the GLP-2 peptide.

Depression is a common mental disorder. More than 350 million people of all ages suffer from depression worldwide. Although a number of antidepressants are available, >20% of patients with major depressive disorder suffer from treatment-resistant depression. Therefore, development of novel therapeutics to overcome this condition is required. We reported that intracerebroventricular administration of glucagon-like peptide-2 (GLP-2) exerts antidepressant-like effects treated with or without adrenocorticotropic hormone. In the present study, we developed a nasal formulation of GLP-2 containing 5% polyoxyethylene (25) lauryl ether and 1% ß-cyclodextrin that enhanced the resistance of GLP-2 to inactivation by dipeptidyl peptidase-4. Intranasal administration of this formulation (60µg/kg) increased the delivery of GLP-2 to the brain and had antidepressant-like effects on rats. These results suggest the potential of the GLP-2 nasal formulation for use as a novel antidepressant.

Neoechinulin A induced memory improvements and antidepressant-like effects in mice.

Neoechinulin A is an isoprenyl indole alkaloid that exhibits scavenging, neurotrophic factor-like, and anti-apoptotic activities. However, the effectiveness of neoechinulin A in animal models of disease has not yet been explored. In the present study, we investigated the effects of neoechinulin A on memory impairment in lipopolysaccharide (LPS)-treated mice and its antidepressant-like effects in mice. In the Y-maze test, the intracerebroventicular (i.c.v.) administration of LPS (10µg/mouse) significantly decreased spontaneous alternation behavior, which was prevented by the prior administration of neoechinulin A (300ng/mouse, i.c.v.). None of the treatments altered the locomotor activity of mice. Moreover, the administration of neoechinulin A decreased the immobility time in the forced-swim test or tail suspension test, which was prevented by the prior administration of WAY100635 (an antagonist of 5-HT1A receptors) and parachlorophenylalanine (an inhibitor of tryptophan hydroxylase). These results suggest that neoechinulin A improves memory functions in LPS-treated mice, and also exerts antidepressant-like effects through changes in the 5-HT system.

Changes in hippocampal synaptic functions and protein expression in monosodium glutamate-treated obese mice during development of glucose intolerance.

Glucose is the sole neural fuel for the brain and is essential for cognitive function. Abnormalities in glucose tolerance may be associated with impairments in cognitive function. Experimental obese model mice can be generated by an intraperitoneal injection of monosodium glutamate (MSG; 2 mg/g) once a day for 5 days from 1 day after birth. MSG-treated mice have been shown to develop glucose intolerance and exhibit chronic neuroendocrine dysfunction associated with marked cognitive malfunctions at 28-29  weeks old. Although hippocampal synaptic plasticity is impaired in MSG-treated mice, changes in synaptic transmission remain unknown. Here, we investigated whether glucose intolerance influenced cognitive function, synaptic properties and protein expression in the hippocampus. We demonstrated that MSG-treated mice developed glucose intolerance due to an impairment in the effectiveness of insulin actions, and showed cognitive impairments in the Y-maze test. Moreover, long-term potentiation (LTP) at Schaffer collateral-CA1 pyramidal synapses in hippocampal slices was impaired, and the relationship between the slope of extracellular field excitatory postsynaptic potential and stimulus intensity of synaptic transmission was weaker in MSG-treated mice. The protein levels of vesicular glutamate transporter 1 and GluA1 glutamate receptor subunits decreased in the CA1 region of MSG-treated mice. These results suggest that deficits in glutamatergic presynapses as well as postsynapses lead to impaired synaptic plasticity in MSG-treated mice during the development of glucose intolerance, though it remains unknown whether impaired LTP is due to altered inhibitory transmission. It may be important to examine changes in glucose tolerance in order to prevent cognitive malfunctions associated with diabetes.

Chemical profiling with HPLC-FTMS of exogenous and endogenous chemicals susceptible to the administration of chotosan in an animal model of type 2 diabetes-induced dementia.

In our previous study, the daily administration of chotosan (CTS), a Kampo formula consisting of Uncaria and other 10 different crude drugs, ameliorated cognitive deficits in several animal models of dementia including type 2 diabetic db/db mice in a similar manner to tacrine, an acetylcholinesterase inhibitor. The present study investigated the metabonomics of CTS in db/db mice, a type 2 diabetes model, and m/m mice, a non-diabetes control strain, to identify the exogenous and endogenous chemicals susceptible to the administration of CTS using high performance liquid chromatography equipped with an orbitrap hybrid Fourier transform mass spectrometer. The results obtained revealed that the systemic administration of CTS for 20 days led to the distribution of Uncalia plant-derived alkaloids such as rhynchophylline, hirsuteine, and corynoxeine in the plasma and brains of db/db and m/m mice and induced alterations in four major metabolic pathways; i.e., (1) purine, (2) tryptophan, (3) cysteine and methionine, (4) glycerophospholipids in db/db mice. Moreover, glycerophosphocholine (GPC) levels in the plasma and brain were significantly higher in CTS-treated db/db mice than in vehicle-treated control animals. The results of the in vitro experiment using organotypic hippocampal slice cultures demonstrated that GPC (10-30 µM), as well as tacrine, protected hippocampal cells from N-methyl-d-aspartate-induced excitotoxicity in a manner that was reversible with the muscarinic receptor antagonist scopolamine, whereas GPC had no effect on the activity of acetylcholinesterase in vitro. Our results demonstrated that some CTS constituents with neuropharmacological activity were distributed in the plasma and brain tissue following the systemic administration of CTS and may subsequently have affected some metabolic pathways including glycerophospholipid metabolism and cognitive function in db/db mice. Moreover, the present metabonomic analysis suggested that GPC is a putative endogenous chemical that may be involved in the tacrine-like actions of CTS in the present diabetic animal model.

Glucagon-like peptide-2-induced memory improvement and anxiolytic effects in mice.

We investigated the effectiveness of glucagon-like peptide-2 (GLP-2) on memory impairment in lipopolysaccharide (LPS)-treated mice, and anxiety-like behavior in adrenocorticotropic hormone (ACTH)-treated mice. In the Y-maze test, LPS (10 µg/mouse, i.c.v.) significantly decreased spontaneous alternation, which was prevented by pretreatment with GLP-2 (0.01-0.3 µg/mouse, i.c.v.). The GLP-2 treatment just before the Y-maze test also improved LPS-induced memory impairment. Continuous treatment with GLP-2 (3 µg/mouse, i.c.v.) had no effect on the open-field test in saline-treated or ACTH-treated mice. Chronic ACTH treatment did not cause anxiogenic effects in the elevated plus-maze test. GLP-2 showed weak anxiolytic-like effects in the elevated plus-maze test in ACTH-treated, but not saline-treated mice. Moreover, GLP-2 increased 5-HT, but not 5-HIAA and tryptophan hydroxylase 2 levels in the amygdala of ACTH-treated mice. Pharmacological depletion of 5-HT prevented the anxiolytic effects of GLP-2. These results suggest that GLP-2 protected and improved memory function in LPS-treated mice, and also had anxiolytic effects due to changes in the 5-HT system.