Introduction

Alzheimer’s disease (AD) is the most prevalent neurodegenerative disease associated with dementia, with more than 26 million cases worldwide in 2006 and an expected 106 million cases by 20501. Little is known about the etiology of AD and the identification of a cure has remained elusive. Recent studies have shown that cocaine- and amphetamine-regulated transcripts (CART) can mitigate cell death and neurological impairments induced by ischemia2,3,4. Thus, the present study tests whether CART shows promise as a novel therapeutic candidate for the treatment of AD.

CART is an anorexigenic neuropeptide produced by the pituitary gland, adrenal gland and pancreas and is widely distributed throughout the central and peripheral nervous systems5. CART regulates food consumption, body weight, stress control, reward and pain transmission. Recently, we reported that icariin-mediated protection against amyloid-β (Aβ)-induced neuronal death was modulated by upregulation of CART6. Furthermore, CART treatment attenuated neuronal loss in the substantia nigra and mitigated behavioral deficits in the MPTP model of Parkinson’s disease7,8. These studies suggest that this neuropeptide may alleviate the pathology associated with neurodegenerative diseases.

The mechanisms underlying the neuroprotective effects of CART are still under investigation. CART treatment is known to promote the in vitro survival of cultured hippocampal neurons by upregulating brain-derived neurotrophic factor9. In an ischemic injury model, protection by estrogen was mediated by an increase in endogenous CART and activation of the ERK signal transduction cascade3,4. Similarly, CART-mediated ERK signaling mitigated cortical neuronal demise in response to Aβ toxicity6. Further mechanistic examinations have indicated that ischemia-induced suppression of CART expression was modulated by the transcriptional repressor known as neuron-restrictive silencer element10. In addition, it has been proposed that attenuation of dopaminergic neuronal cell loss after CART treatment in MPTP-treated mice could be attributed to the anti-oxidative properties of CART7.

In the present study, the protective properties of CART peptide in in vitro and in vivo models of AD are characterized. It is demonstrated, using immunostaining, that CART expression is naturally elevated in the brains of APP/PS1 transgenic mice and in AD patients in the Aβ plaques, perhaps as a compensatory defense mechanism. It is also observed that CART treatment preserves synapse and dendritic structure in cellular and animal models of AD. Results suggest that the mechanisms underlying CART-mediated neuroprotection include attenuation of oxidative damage and improvement of mitochondrial function. These findings support the notion that CART-based therapies may be beneficial in AD patients.

Results

Regional distribution of CART in the brains of APP/PS1 transgenic mice and AD patients

In an effort to test whether CART expression is altered in AD brain, cortical tissues from three AD patients and three age-matched non-demented individuals were subjected to immunohistochemistry for CART. In controls, weak immunoreactivity for CART was barely detectable in the cortex (Fig.1a, arrow within panel i). However, in the cortex of AD patients, there was increased CART immunoreactivity with punctuate distribution surrounding structures that were speculative of Aβ plaques (Fig. 1a inset and arrow within panel ii). Similarly, increased CART immunoreactivity pattern was also found in the cortex of 10-month-old APP/PS1 mice (Fig. 1a inset and arrow within panel iii). To determine the possible association between elevated CART immunoreactivity and Aβ plaques, double staining of CART and Aβ was performed in 8-month-old APP/PS1 mice. As shown in Fig. 1b, punctuate CART staining was found in clusters that surrounded and/or were within Aβ plaques.

Figure 1
figure 1

CART immunoreactivity is prominent in cortical Aβ plaques in AD.

(a) CART immunostaining is distributed uniformly in cell bodies of cortical neuron-like cells in a control non-AD patient (i) and concentrated around the cores of Aβ plaque-like structures in AD brain (representative of three non-AD and three AD brains) (ii) and in 10-month-old APP/PS1 transgenic mice (iii). Scale bar =100 μm. (b) Similar staining patterns for CART (green) and Aβ (red) are seen in 8-month-old APP/PS1 mice, where the CART signal was most prominent in the regions surrounding the plaques but not at the core of the plaques (indicated by white arrows). Scale bar =100 μm. The images are representatives of 3 mice per phenotype with similar results.

Full size image

To further characterize the temporal profile of CART immunoreactivity and its association with Aβ plaque formation, a series of double-label immunostaining for CART and Aβ was performed in APP/PS1 mice at 4 (prior to the onset of β-amyloid plaques), 6 and 8 (after β-amyloid plaque formation) months old and in age-matched control C57BL/6 (B6) mice. In the cortex and hypothalamus of both APP/PS1 mice at 4 months of age and of control B6 mice, CART immunoreactivity exhibited a similar distribution. Regardless of genetic profile, CART expression was more prominent in the hypothalamus than in the cortex and hippocampus and was mainly localized in neuron-like cells (Fig. 2a,b and c, top panels). By contrast, in 6- and 8-month-old APP/PS1 mice, Aβ plaques were widely detectable in the cortex and hippocampus and increased clusters of CART immunoreactivity was found to surround and/or to be within β-amyloid plaques (Fig. 2a,b and c, bottom panels).

Figure 2
figure 2

CART and Aβ expression and distribution in the cortex (a), hippocampus (b) and hypothalamus (c) of APP/PS1 transgenic mice and B6 control mice.

CART (green) immunofluorescence is prominent in cell bodies and neurites in B6 control mice and lower in 4-month-old APP/PS1 transgenic mice in the absence of Aβ plaques. In 6- and 8-month-old APP/PS1 mice, where Aβ plaques (red fluorescence) had emerged in the cortex and hippocampus, CART was localized in Aβ plaques. CART immunoreactivity was also prominent in cell bodies in the hypothalamus. Arrows indicate the location of CART immunoreactive structures shown in the insets. Scale bar = 100 μm. Images are representatives of 3 mice per group with similar results.

Full size image

CART expression in the brains of APP/PS1 transgenic mice

CART mRNA and protein levels in the cortex, hippocampus and hypothalamus of 8-month-old APP/PS1 transgenic and age-matched B6 mice were next analyzed. In APP/PS1 mice, CART mRNA levels were lower by 23.2% and 33.5% (Fig. 3a) and CART protein levels were decreased by 13.5% and 20.9% in the cortex and hippocampus, compared to B6 control mice as measured by ELISA (Fig. 3b). In the hypothalamus, APP/PS1 mice had similar CART mRNA levels but lower protein levels compared to control mice (Fig. 3a,b).

Figure 3
figure 3

Expression of endogenous CART in 8-month-old APP/PS1 mice and the capacity of exogenous CART to cross the blood-brain-barrier.

(a) CART mRNA levels in the cortex, hippocampus and hypothalamus in 8-month-old APP/PS1 mice and B6 control mice. All values are expressed as mean ± SEM, n = 6 mice/group, *p < 0.05 versus B6 control mice. (b) CART protein levels were measured by ELISA in the cortex, hippocampus and hypothalamus in 8-month-old APP/PS1 mice and B6 control mice. All values are expressed as mean ± SEM, n = 6 mice/group, *p < 0.05; **p < 0.01 versus B6 control mice. The capacity of CART to cross the blood-brain barrier after intravenous injection of pre-radiolabelled 131I CART peptide was compared to a control 131I solution. The retention of radiolabelled CART in various brain regions is shown at 10 minutes (c), 30 minutes (d), 60 minutes (e) and 120 minutes (f) after injection. The 131I solution exhibited no retention 120 minutes after injection. On the other hand, CART peptide exhibited high retention in the thalamus, cortex and hippocampus after influx into the brain. All values are expressed as mean ± SEM, n = 6 mice/group, **p < 0.01 versus 131I solution.

Full size image

Exogenous CART attenuates memory deficits in APP/PS1 mice

Next, the effects of exogenous CART in 8-month-old APP/PS1 mice were assessed. The delivery method of CART peptide into the mice was dependent on its inherent capacity to cross the blood-brain barrier. Using radiolabeled CART peptide (131I-CART) and iodine-131 solution as a positive control, the radiolabeled peptide or solution was injected intravenously followed by determination of distribution in various brain regions of B6 mice at 10, 30, 60 and 120 minutes post-injection. As expected, the radiolabeled iodine solution crossed the blood-brain barrier and distributed in all parts of the brain with low retention time (Fig. 3c-f). At 120 minutes, more than 70% of the iodine solution had exited the brain.

CART crossed the blood-brain barrier with a distribution pattern similar to the iodine solution at 10 minutes post-injection (Fig. 3c). However, CART exhibited a much higher retention time than the iodine solution. After 120 minutes, the CART peptide was preferentially retained in the thalamus and cortex. Intravenous injection was subsequently selected over stereotaxic injection as the delivery method to minimize surgical injury. In the determination of the minimum effective dosage, the effects of CART on 8-month-old APP/PS1 and age-matched B6 mice via injection at 0.5, 1.0 and 2.0 μg/kg for 30 days with saline as vehicle were investigated. At these doses, there was no difference in average body weight and movement speed of the transgenic mice before and after CART injections (data not shown). As all three doses of CART ameliorated behavioral deficits in APP/PS1 mice (data not shown), the lowest dose (0.5 μg/kg) was used in all subsequent experiments.

The Morris water maze (MWM) test was administered before CART treatment (to confirm behavioral deficits occurring in APP/PS1 mice) and 72 hours after the final CART injection. The MWM tests demonstrated that escape latencies (time needed to locate the platform) were comparable in APP/PS1 mice, with and without CART peptides, but were longer than in B6 control mice with and without CART injections during the first trial (p < 0.05, Fig. 4a). However, APP/PS1 mice treated with CART exhibited significantly lower escape latencies (p < 0.01) than APP/PS1 mice without CART in three subsequent consecutive trials (Fig. 4a). These results indicated that CART treatment improved cognition and memory in APP/PS1 mice. Upon submerging the platform, APP/PS1 mice with CART treatment crossed the area of the platform more frequently (2.46 ± 0.40 vs. 1.25 ± 0.45) and spent more time in the target quadrant (35.08 ± 3.09% vs. 21.39 ± 3.87%) (Fig. 4b, c).

Figure 4
figure 4

Effects of exogenous CART treatment on memory impairments and Aβ plaque load in APP/PS1 mice.

(a) CART treatment of APP/PS1 transgenic mice significantly lowered the latency to escape a submerged platform in the Morris water maze test at 8 months of age. All data are expressed as mean ± SEM, n = 15 mice/group, *p < 0.05, **p < 0.01 versus B6 control mice. #p < 0.05 versus APP/PS1 transgenic mice without CART treatment. CART-treated APP/PS1 mice also made higher numbers of passes over the submerged platform (b) and spent more time in the target quadrant (c) compared to APP/PS1 mice without CART treatment. All data are expressed as mean ± SEM, n = 15 mice/group, **p < 0.01 versus B6 control mice, ##p < 0.01 versus APP/PS1 transgenic mice without CART treatment. (d) Aβ-immunoreactive plaques in 8-month-old APP/PS1 transgenic mice in the absence or presence of CART, Scale bar: 100 μm. Counts of total Aβ plaque numbers (e) and Aβ plaque load (f) in the cortex and hippocampus of 8-month-old APP/PS1 mice (6–12 sections/mouse). All values are expressed as mean ± SEM, n = 3 mice/group.

Full size image

CART dose not reduce Aβ plaques in the brains of APP/PS1 mice

Aβ plaque numbers in the cortex and hippocampus of APP/PS1 mice were then quantified (Fig. 4d-f), which indicated that there was no significant difference in Aβ plaque formation after CART treatment (p > 0.05).

CART preserves neurite and synapse structure in APP/PS1 mice

Neurite structures were examined using MAP-2 immunostaining and synapses were examined using synaptophysin immunostaining. In Aβ1-42-treated cortical neurons, an amount lower than or comparable to previous studies11,12,13 for 24 hours elicited significant loss of neurite structures and synaptophysin protein. However, pretreatment with CART attenuated neurite degeneration and increased synaptophysin protein levels (Fig. 5a). Neurite length was increased from 65.7 ± 12.7 to 191.9 ± 24.4 μm/cell with CART pretreatment (Fig. 5b). In addition, the synaptophysin-positive area within the neurite was about 2 times higher, increasing from 5.8 ± 1.3 to 12.3 ± 2.3 μm2/μm neurite in the presence of CART (Fig. 5c).

Figure 5
figure 5

CART treatment protects neurites against Aβ toxicity in cultured cortical neurons and preserves synaptic structures in APP/PS1 mouse brain.

(a) CART treatment prevents loss of neurite structures (MAP2, green) and enhances synaptophysin (red) expression in cultured cortical neurons. Scale bar =40 μm. CART treatment increased the average neurite length compared to no treatment (b) and increased synaptophysin-positive areas within neurites (c). All values are mean ± SEM of data from at least 45 randomly selected images for each assay, based on 3 independent experiments using different primary culture preparations. **p < 0.01 versus B6 control, #p < 0.05 versus cortical neurons treated with Aβ1-42 only. (d) CART treatment preserves synaptic ultrastructure in the cortex and hippocampus of APP/PS1 mice (8 months), as visualized by electron microscopy. Electron micrographs illustrate synaptic ultrastructure in the cortex (upper three panels) and hippocampus (lower three panels) of B6 control mice, untreated APP/PS1 mice and APP/PS1 mice with CART treatment. The black arrows indicate synaptic clefts; the red arrows indicate perforated synapses. PSD, postsynaptic density; SV, synaptic vesicles. Scale bar =0.5 μm. (e) CART treatment elevated synaptophysin levels in the hippocampus, as determined by Western blot analysis followed by quantification. The blot shown here is a representative of 3 independent experiments. Blot images were cropped for comparison. All values are expressed as mean ±SEM of the band intensity of synaptophysin normalized to GAPDH from 3 independent experiments, **p < 0.01 versus B6 control mice, #p < 0.05 versus APP/PS1 mice without CART treatment.

Full size image

Next, the effect of CART on synapses in the cortex and hippocampus in APP/PS1 mice was evaluated by transmission electron microscopy. In the cortex and CA1 area of the hippocampus of 8-month-old APP/PS1 mice, the number of synapses was much lower than that in B6 control mice. By contrast, B6 mice exhibited more synaptic vesicles in the presynaptic compartment and showed the typical asymmetric appearance of Gray’s type I synapses (Red arrow, Fig. 5d). Synaptic ultrastructure also appeared severely compromised in both regions in APP/PS1 mice. However, degeneration of synaptic architecture in cortex and hippocampus was partially reversed by CART treatment (Fig. 5d). This increase in synapses was accompanied by a 39.5% increase in synaptophysin protein levels in the hippocampus of APP/PS1 mice treated with CART (Fig. 5e).

Whether CART treatment enabled neurons to establish more effective long-term potentiation (LTP) was examined. Using brain slices from B6 control mice and APP/PS1 mice with and without CART treatment, it was found that the slope of the I/O curve obtained from APP/PS1 mice (B = 0.56; n = 12 slices/6 mice) was significantly less than in control B6 mice (B = 1.14; P < 0.01, n = 12 slices/6 mice) (Fig. 6a). APP/PS1 mice with CART treatment showed significantly improved slope of the I/O curves (B = 0.75; n = 12 slices/6 mice; p < 0.05 versus APP/PS1 mice without CART treatment), but still less than in B6 mice (Fig. 6a). In B6 mice, LTP was able to be stably induced (151.4 ± 7.15% at 55-60 min post-HFS; n = 12 slices/6 mice; Fig. 6b) by a high-frequency stimulus (HFS; 100 pulses at 100 Hz). However, the same HFS was unable to induce any increase in EPSP slope, with no LTP induction in APP/PS1 mice (100.43 ± 6.18%; n = 12 slices/6 mice; Fig. 6c). CART-treated APP/PS1 mice exhibited an increase in EPSP slope upon stimulation, resulting in the induction of LTP (129.35 ± 3.95%; n = 12 slices/6 mice; Fig. 6d).

Figure 6
figure 6

CART treatment facilitated functional recovery of synapses in APP/PS1 mice.

Electrophysiological determination of deficits in Schaffer collateral-CA1 synaptic transmission and the induction of long-term potentiation. (a) A plot showing input/output (I/O) curves in response to stimulus (0.1-1.1 mA) at the hippocampal CA1 region of B6 control mice, APP/PS1 mice and APP/PS1 mice receiving CART treatment (8 months old). The I/O curve was measured by averaging the slope of EPSPs against stimulus intensity at 0.1-1.1 mA. The slope of the regression line for the I/O curve in APP/PS1 mice was markedly reduced. With CART treatment, the reduction in I/O curve slope was significantly improved for APP/PS1 mice. All values are expressed as mean ± SEM from 6 mice in each group. *p < 0.05 versus B6 control mice, #p < 0.05 versus APP/PS1+CART. The capacity to establish long-term potentiation (LTP) from high-frequency stimulus (HFS, 100 pulses at 100 Hz) with the same pre-HFS intensity at hippocampal CA1 region of B6 control mice (b), APP/PS1 transgenic mice (c) and APP/PS1 mice with CART treatment (d).

Full size image

CART protects against mitochondrial dysfunction by neutralizing oxidative stress and mitochondrial DNA damage in vitro

We next examined the effects of CART treatment on the functional status of mitochondria in 8-month-old APP/PS1 mice compared to age-matched B6 control mice. APP/PS1 mice had substantially decreased membrane potential (34.1% in cortex and 34.0% in hippocampus) compared to B6 mice. CART treatment attenuated the reduction in mitochondrial membrane potential in APP/PS1 mice (improved by 35.0% in cortex, p < 0.05; and 31.1% in hippocampus, p < 0.05) (Fig. 7a), suggesting that CART ameliorated mitochondrial dysfunction. Furthermore, the activities of mitochondrial complex I and II, but not complex IV, were increased after CART treatment (Fig. 7b). In addition, neuronal ATP levels in APP/PS1 mice were ameliorated by CART, which elevated neuronal ATP by 140% (p < 0.05) compared to APP/PS1 mice without CART treatment (Fig. 7c).

Figure 7
figure 7

CART prevented mitochondrial dysfunction.

(a) CART prevented the collapse of mitochondrial membrane potential (Δψm) in the hippocampus and cortex (increased by 31.1% and 35.0%). n = 3 mice/group. (b) CART stimulated mitochondrial complex-I and complex-II activities but not complex IV in the cortex of 8-month-old APP/PS1 mice. n = 3 mice/group. (c) CART attenuated the decrease in intracellular ATP levels in the cortex of APP/PS1 mice (increased by 140.3%). n = 3 mice/group. (d) CART attenuated intracellular ROS levels in the cortex (decreased by 44.6%) and hippocampus (decreased by 30.0%), as quantified by colorimetric spectrophotometry. n = 3 mice/group. (e) CART treatment reduced mitochondrial DNA damage, as reflected by a drop in 8-OHdG levels in the cortex (10.9%) and the hippocampus (19.2%). (f) CART treatment reduced lipid peroxidation, as indicated by 4-HNE levels (30.0%) in the cortex but not the hippocampus. All values are expressed as mean ± SEM from 3 mice in each group. *p < 0.05, **p < 0.01 versus B6 control mice, #p < 0.05, ##p < 0.01 versus APP/PS1 mice without CART treatment.

Full size image

In many neurological disorders, mitochondrial dysfunction is mainly due to excessive oxidative stress14,34.

Measurements of mitochondrial membrane potential

Changes in mitochondrial membrane potential (Δψ) of hippocampus and cortex (n = 3 per group) were measured as previously described35. The mitochondria of brains were isolated using the Tissue Mitochondria Isolation Kit (Beyotime, Nan**g, China). Hippocampus tissues were immediately removed and washed by phosphate buffered saline (PBS). Then the tissues were homogenized with ten times volume of isolation buffer on ice and centrifuged at 600 g for 5 minutes at 4 °C. The supernatant was centrifuged again at 11, 000 g for 10 minutes at 4 °C to obtain the cytosol (supernatant) and mitochondria (deposition) fraction. The purified mitochondria and JC-1 staining solution (Beyotime, China) were mixed at 1:9. To evaluate the OD of JC-1, wavelengths of excitation and emission were set at 490 nm and 530 nm.

Measurement of intracellular reactive oxygen species

For determination of intracellular reactive oxygen species (ROS) levels, samples of hippocampus and whole cortex from each group (n = 3 per group) were prepared according to the manufacturer’s instructions (Jianchen, Nan**g, China). After preparation, the samples were re-suspended in 10 mM 2, 7-dichlorodihydrofluorescein diacetate (DCFH-DA) solutions, followed by 20 minutes’ incubation at 37 °C. Then the samples were subjected to colorimetric spectrophotometry to measure intracellular ROS levels.

Analysis of mitochondria complex activity and ATP levels

Activities of mitochondria complex I (NADH-ubiquinone oxidoreductase), complex II (succinate-ubiquinone oxidoreductase), complex III (ubiquinol-cytochrome c oxidoreductase) and complex IV (cytochrome c oxidase) were determined by colorimetric spectrophotometry (n = 3 mice per group) using commercially available kits (Genmed Scientifics Inc., USA). Briefly, mitochondrial suspension was incubated with the corresponding substrate (NADH, succinate and ferrocytochrome c ) at 25 °C for 1 minute and the change in absorbance at 340 nm, 600 nm and 550 nm was measured using the spectrophotometer. Background levels were measured in the absence of mitochondrial suspensions and activity was expressed as micromoles of substrate oxidized per min–1 mg–1 protein. The ATP level was detected using the Enhanced ATP Assay Kit (Beyotime, Nan**g, China). The isolated mitochondria (n = 3 per group) were lysed in an equal volume of lysis buffer and the content of ATP was measured by comparing the relative luciferase unit (RLU) of the sample to the standard curve of known ATP concentrations.

Measurement of 4-HNE and 8-OHdG

Intracellular 4-hydroxy-2-nonenal (4-HNE) and 8-hydroxydeoxyguanosine (8-OHdG) levels in the hippocampus and cortex (n = 3 per group) were measured as previously described36. Briefly, the samples were added into the microplate pre-coated with mouse 4-HNE or 8-OHdG antibody. After incubation with 4-HNE or 8-OHdG antibody with labeled HRP at 4 °C overnight, TMB substrate solution and stop solution was then added. The absorbance was measured at a wavelength of 450 nm and the concentration of 4-HNE or 8-OHdG was determined by comparing the O.D. of the sample to the standard curve.

Western blot

Western blots were performed according to previously published techniques4. Equal amounts of total protein from each hippocampal sample (n = 3 per group) were resolved by SDS-PAGE and transferred onto PVDF membranes. Membranes were probed with primary antibodies recognizing synaptophysin (1:5000, BD Transduction Laboratories; 611800). The relative amount of synaptophysin in each sample was detected using a chemiluminescence ECL kit (Amersham Pharmacia Biotech, Piscataway, NJ, USA). Relative intensity of each band was measured using NIH ImageJ software. All experiments were performed at least in triplicate.

Statistical analysis

Data were presented as mean ± SEM. Differences between groups were analyzed with a t-test for two groups and one-way ANOVA with post hoc tests for multiple groups (SPSS 16.0 system, SPSS, Chicago, USA). The criterion for statistical significance was set at p < 0.05.

Additional Information

How to cite this article: **, J.-L. et al. CART treatment improves memory and synaptic structure in APP/PS1 mice. Sci. Rep. 5, 10224; doi: 10.1038/srep10224 (2015).