ßA₁₋₄₀ (Polypeptide Group, France) was dissolved in PBS (1.72 mg/ml), aliquoted and stored at −80°C until used. WIN was purchased from Sigma, JWH was from Tocris (Cookson Ltd., UK), SR141716 (SR1; Rinaldi-Carmona et al., 1994) and SR144528 (SR2; Rinaldi-Carmona et al., 1998) were kindly donated by Sanofi-Synthelabo (Montpellier, France). For in vitro experiments each of these compounds was dissolved in DMSO at 10 mM, aliquoted and stored at −80°C. Before their use, drugs were diluted in appropriate solvent and DMSO never exceeded 0.1% in pharmacological experiments. For in vivo experiments, WIN and JWH were initially dissolved in chloroform (on ice), quickly aliquoted to prevent evaporation, dried under a stream of N₂, and aliquots stored desiccated. Before their use, drugs were diluted in ethanol and added to the drinking water. Salts and other reagents were analytical grade from Merck.

For immunocytochemistry, cryoprotected and fixed frozen frontal cortex samples were obtained from the Neurologic Tissue Bank, Hospital Clinic, Barcelona, Spain, and processed as previously described (Ramírez et al., 2005). Human brains were obtained by the Neurologic Tissue Bank following written consent. Controls consisted of 3 males and 2 females (median 70.0, range 38.0–0.0 years of age; median 17.0, range 3.5–21.0 h of post-mortem interval), and clinically diagnosed and neuropathologically defined AD patients consisted of 3 females and 3 males (median 74.0, range 66.0–88.0 years; median 5.5, range 4.0–9.0 h).

Tg APP transgenic mice were obtained via heterozygous breeding of mice expressing the 695 aa long isoform of the human APP containing a double mutation Lys 670-Asn, Met 671-Leu (swedish mutation) under transcriptional control of the hamster prion promoter on a C57BL/6 breeding background (Hsiao et al., 1996). Male Tg APP, and wild type (wt) littermates, used as controls, were 7 months old at the beginning of the experiments. Mice were group-housed (4–5 animals per cage) under controlled temperature (23 ± 2°C), with a 12:12 h light/dark cycle and with ad libitum access to food and water. All of the experiments were performed according to ethical regulations on the use and welfare of experimental animals of the European Union and the Spanish Ministry of Agriculture, and the procedures were approved by the bioethical committee of the CSIC.

WIN and JWH were administered in the drinking water at a dose of 0.2 mg/Kg/day using ethanol (0.1%) as vehicle (Martín-Moreno et al., 2012). The amount of water drank by the animals was assessed every other day and the treatment was adjusted to their weight. There was no difference in the body weight or the ingested water between groups, all along the experiment, discarding a possible reinforcing effect of cannabinoids.

Animals were sacrificed by cervical dislocation followed by decapitation at 11 months of age after 4 months chronic treatment. The brain was sagittally divided. One brain hemisphere was rapidly dissected on a cold plate, frozen on dry ice and stored at −80°C until assayed. The other hemisphere was immersion fixed in 4% paraformaldehyde (4% PF) in sodium phosphate buffer (PB) 0.1 M for 24 h, cryoprotected in sucrose 15% (24 h) and 30% (24 h) in PB, snap frozen in hexane (−60°C), and stored at −20°C until cut with a sliding microtome.

For pharmacological studies male mice, wt used as control, or Tg APP mice (line 2576) (25–30 g, 12 months of age) were sacrificed by decapitation following cervical dislocation. The thoracic aorta was removed, cleaned and cut into segments of 2 mm length. Rings were mounted in Multy Myograph System 610M (Danish Myo Technology, Denmark) at 37 ± 0.5°C and gassed continuously with a mixture of 95% O₂-5% CO₂, in a solution of the following composition: PSS (mM): NaCl 140, KCl 5, MgCl₂ 1, CaCl₂ 1.5, HEPES 5, and glucose 10. After equilibration, arterial rings were mounted between two parallel tungsten wires under a resting tension of 2 g. The isometric force was digitalized by Myodaq 2.01 program (Danish Myo Technology, Denmark) and displayed on a personal computer.

Arteries were preconstricted with 123.5 mM K⁺ (KPSS) for 4 min, washed, and then a) concentration-response curves for phenylephrine (Phe, 0.1-10 μM) and the thromboxane analog U46619 (0.01-0.1 μM) were performed; b) in a different set of experiments arterial rings were preconstricted with a submaximal concentration of U46619 (0.03 μM) for 15 min and then vasodilation to ACh 10 μM was assessed. After 2 washes vessels were incubated with Aβ (1 μM) for 15 min, and stimulated with U46619 (0.03 μM) for 15 min followed by relaxation with ACh 10 μM. To assess the vasodilatory effects of WIN and JWH concentration-response curves (1nM-10 μM) were performed in segments preconstricted with U46619. Concentration-response curves for WIN and JWH were performed in arterial segments treated with the selective CB₁ or CB₂ antagonists (SR1 and SR2), added 5 min before preconstriction with U46619. To investigate the effect of cannabinoid agonists on ACh-induced vasodilation, agonists were added to arterial rings at 0.5 μM after 15 min incubation with Aβ (1 μM). Given that the effect of Aβ is irreversible (Thomas et al., 1996) different arterial rings were used for each experiment. Tension was expressed as mN/mm artery length, or as a percentage of initial preconstriction (either with K⁺ or U46619). Indeed each ring was its own control, avoiding the variance between the responses of different rings from the same animal (decreased responsivity as the rings approached the abdominal aortic region).

Immunostaining was performed on floating sections (30 μm thick) as described (Gómez Del Pulgar et al., 2002). Sections were incubated with the different antibodies overnight at 4°C. Dilutions of antibodies were as follow: polyclonal anti-CB1 (1:900, CC2, raised in our laboratory, De March et al., 2008), polyclonal anti-collagen IV (Col IV; 1: 400, ref. CR013X; Fitzgerald, MA, USA). The CC2 antibody was raised in rabbits using as immunogen the 15 aa N-terminal end of the CB₁ receptor protein coupled to keyhole limpet hemocyanin. The antiserum was affinity purified, and it was characterized in wt mice and in CB₁ KO mice brain. CB₁ immunoreactivity brain distribution was in agreement with previous studies (Tsou et al., 1997, 1999). The anti-collagen antibody has been raised in rabbits using as immunogen Col IV from human and bovine placenta. It shows negligible cross reaction with Col I, Col II, Col III, and Col V. Development was conducted by the Avidin-Biotin Complex (ABC) method (Pierce), and immunoreactivity was visualized by 3,3′-diaminobenzidine oxidation as chromogen, with (CB₁) or without nickel enhancement (Col IV). Omission of primary or secondary antibodies resulted in no immunostaining.

Images were acquired with a Zeiss Axiocam high resolution digital color camera, using the same settings and segmentation parameters (MCID software; InterFocus Imaging, UK) for a given marker and experiment. The mean value for each animal per region results from the analysis of 5–6 sections. The percentage of the brain area covered by Col IV positive vessels was assessed by image analysis with MCID software.

Total RNA from pooled aortae (n = 3–4) was extracted using TRIzol reagent according to the manufacturer's instructions (Invitrogen). To avoid interference with potential genomic DNA amplification 1 μg of total RNA was treated with 1 μl DNAse I (Invitrogen) plus 1 μl of 10X Buffer (Invitrogen) and incubated for 15 min at RT, then EDTA (25 mM) was added and incubated at 65°C for 15 min to inactivate DNAse I. For cDNA synthesis a total of 1 μg of RNA were reverse-transcribed for 75 min at 42°C using 5 U of avian myeloblastosis virus reverse transcriptase (Promega) in the presence of 20 U of RNAsin (Promega). The PCR reaction was performed using TaQ polymerase (TaQ DNA polymerase Sigma) and a mixture of reverse and forward primers (5 pmol). The primers used were CB₁ forward 5′-AGCTTTGTTGACTTCCAGTGT and CB₁ reverse 5′-CTGCCCACAGATGCTGTGAA, CB₂ forward 5′-AGGAGCTGTCAGCTCAGGGTAT and CB₂ reverse 5′-CTGCGCCCCTAAGGACCTA. The PCR reaction (final volume10 μl) was performed in a Veriti thermal cycler (Applied Biosystems) and the PCR program was as follows: initial denaturation for 10 min at 95°C, then 40 cycles of denaturation (15 s, 95°C), annealing (30 s, 60°C), and extension (30 s, 60°C). The PCR products were analyzed by standard agarose gel electrophoresis, and gene expression levels were detected by the use of ethidium bromide.

Aortae 1–2 mm long rings were fixed in 4% PF/ 2.5% glutaraldehyde in cold 0.1 M Na cacodylate buffer immediately after dissection, for 6 h. The segments were washed five times with cacodylate buffer every 30 min, and left overnight at 4°C. The segments were postfixed with 1% osmium tetroxide and potassium ferrocyanide in distilled water for 1.5 h, they were washed with distilled water (3 × 10 min washes) and dehydrated in increasing acetone solutions (50–100% each for 15 min). The segments were then gradually embedded in resin (1:3, 1:1, 3:1 acetone:pure resin) and finally left in pure resin (TAAB 812 mix) at 60°C overnight. The resin embedded samples were sectioned by diamond knife, and the 80 nm sections were collected onto copper grids and post-stained with 1% uranyl acetate and Reynolds lead citrate for 4 and 3 min, respectively. Electron micrographs were obtained using a Jeol JEM-1010 high resolution transmission electronic microscope (Jeol, Tokyo, Japan).

In pharmacological experiments we used one vessel per mouse, therefore n represents number of animals. In brief, each aorta was cut into 5 different rings and each ring was used for a given treatment to avoid artefactual results. Results are expressed as mean ± standard error mean (SEM) or as mean ± standard deviation (SD). Statistical analysis was assessed by using two-way or one-way analysis of variance (ANOVA) followed by Wilcoxon's test, if the data follow a Gaussian distribution (KS normality test), or by Kruskal-Wallis test, followed by Dunn's test (version 5.0, Prism software, GraphPad, USA). A value of p < 0.05 was considered significant.

Previous studies have reported vascular alterations in the brain of AD affected individuals, such as increased vessel density and greater collagen deposition at the structural level. As shown in Figures 1A,B) we found an increase (≈30%) in Col IV positive vessel density in the gray matter of frontal cortex from AD patients compared with control subjects. Vessel density was significantly lower in the white matter compared to the gray matter. No difference in vessel density in the white matter was found between the control and AD group (Figures 1A,B). Vessel density in Tg APP mice was much higher (≈50%) in cortical areas compared to wt mice, however it showed similar density in the hippocampus (Figures 1C,D). Interestingly, prolonged in vivo oral treatment (0.2 mg/Kg/day) with both WIN, a mixed CB₁/CB₂ agonist, and JWH, a CB₂ selective agonist, counteracted the increased Col IV vessel density. In summary, similar vessel alterations were found in the neurologic condition and in the experimental model of AD, where a prolonged oral treatment of a cannabinoid agonist prevented vascular changes.

We next examined whether mice aortae expressed CB₁ and CB₂ receptors. CB₁ immunoreactivity has been previously reported in brain vessels (Ashton et al., 200), but the presence of CB₂ receptors is uncertain. CB₁ receptors were expressed in endothelial cells, at the basal lamina, but not in smooth muscle cells (Figure 2A, representative image of n = 3), while Col IV immunoreactivity just stained the basal lamina of the aorta (Figure 2B, representative image of n = 3). The immunostaining was very reproducible for both antibodies. There is debate on the specificity of CB₂ receptors antibody (Cécyre et al., 2014; Li and Kim, 2015), therefore we used PCR to demonstrate CB₂ and CB₁ receptor expression in aorta extracts (Figure 2C).

Constriction of aorta rings with high potassium (123.5 mM K⁺) was decreased by 50% in Tg APP mice aortae compared to those from wt mice (1.07 ± 0.13 and 2.25 ± 0.30 mN/mm respectively; p < 0.01, Student's t-test). Next, we tested two pharmacologically distinct vasoconstrictors: phenylephrine and the thromboxane analog U46619 (Figure 3). The vasoconstrictor response to 0.1 μM phenylephrine was enhanced by 2 fold in Tg APP compared to wt mice (Figure 3A), and that of 0.1 μM U46619 around a 50% (Figure 3B). We did not find any differences in the vasodilation induced by ACh (100 μM) between groups (data not shown). Furthermore, the cannabinoid agonists (15 min preincubation) under study did not change ACh vasodilation either (data not shown).

Given that Tg APP mice are continuously exposed to circulating Aβ we wondered whether the peptide would mediate those responses. Aβ (1 μM) preincubation increased vasoconstriction to 0.03 μM U46619 (Figure 3C) and decreased vasodilation to 10 μM ACh (Figure 3D), in aortae from wt mice, although the peptide alone did not show any vasoactivity. However, incubation with Aβ did not alter arterial vasoconstriction or vasodilation in Tg APP mice aortae (Figures 3C,D). Interestingly, both cannabinoid agonists rescued ACh-induced vasodilation in the presence of Aβ (Figure 4) in wt mice.

WIN concentration-dependently induced vasodilation in control mice, with a maximal effect of 80% at 1 μM (Figure 5A). The vasodilatory effect of JWH was smaller than the one induced by WIN, with a maximal effect of 56% at 1 μM (Figure 5B). In Tg APP aortae the vasodilation induced by WIN was significantly decreased at all the concentrations tested (Figure 5A), but in the case of JWH the effect at lower concentrations (1 and 10 nM) was decreased and at higher concentrations was similar between wt and Tg APP mice (Figure 5B).

Taken together these results show that vascular function is markedly altered in Tg APP mice and that Aβ may play a role in those altered responses. Furthermore, cannabinoid agonists induce vasodilation in aortic rings, which is partially preserved in Tg APP mice.

Some reports have described changes in the structure of Tg APP vessels (Christie et al., 2001; Tong et al., 2005). Therefore, we sought to determine if changes at the ultrastructural level may explain the vessel dysfunction observed in Tg APP mice. Toluidine labeled vessels showed similar vessel structure (Figures 6A,B). Endothelial cells appeared unaltered in both strains (Figures 6C,D). Moreover, smooth muscle cells also appeared normal, with normal numbers of mitochondria (not shown). However, there was a great difference in basal lamina collagen that was markedly increased in Tg APP when compared with wt aortae (Figures 6E,F).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.