The Preventive Effects And Underlying Mechanisms Of Phosphodiesterase-4 Inhibitor FFPM In Alzheimer’s Disease

Background and objective:Alzheimer’s disease (AD) ocurred in old age is a common neurodegenerative disease that is characterized by progressive cognitive impairment, which was firstly reported by German neuropathology doctor Alois Alzheimer in 1907 and was named after alzheimer. The major clinical symptoms of AD present as memory delcine, abnormal behavior, linguistic barrier, changes of personality and athletic ability. The β-amyloid hypothesis in AD pathology is generally accepted and scientists attempted to explore new drugs for AD by inhibiting Aβ production or promoting Aβ clearance. However, none of these strategies were successful, indicating that Aβ might not be an ideal target for AD. Recently, along with the deep research in AD, most scholars thought that the simple theory of neurotransmitters and receptors could not fully explain the pathogenesis of AD and correctly guide for the exploration of new anti-AD drugs and disease treatment. Therefore, researches focused on the theory of neurotransmitters and receptors was gradually transfered to genetic and molecular level, such as the second messengers system in brain, transcriptional factors and synaptic protein.Phosphodiesterase 4 (PDE4) is one of specific enzymes that hydrolyzed cyclic adenosine monophosphate (cAMP), which includes PDE4A, PDE4B, PDE4C and PDE4D subtypes and can futher subdivide into 25 splice variants. PDE4 was studied as a potential therapeutic target for depression and Alzheimer’s disease for more than twenty years. However, so far, no PDE4 inhibitors were approved for psychiatric disorders due to their high incidence of nausea and emesis. PDE4 inhibitors increased intracelluar cAMP level via inhibiting PDE4 activity and activated cAMP dependent protein kinase A (PKA) followed by phosphorylation of cAMP response element binding protein (CREB) at serine 133, which exerted anti-depressant and cognition enhancement effect through cAMP/PKA/CREB signal pathway. Based on the chemical structures of PDE4 inhibitors, crystal structure and spatial conformation of PDE4 protein as well as the possible mechanisms of emetic responses by Rolipram, we thought a brain-penetrant and good selective for PDE4D compound was essential for a safe and effective PDE4 inhibitor. Previous studies demonstrated that FFPM had a good selectivity and inhibitory activity to PDE4D2, which was approximately 2000 times better than PDE2A3, PDE5A1, PDE7A1 and PDE9A2. In addition, FFPM could impove the acute lung injury and inflammatory responses induced by lipopolysaccharide (LPS). However, the exact role of FFPM in central nervous system, especially AD, was not poorly determined. Therefore, in the present study, we investigated the effects of FFPM on PDE4 enzyme activity, pharmacodynamics, pharmacokinetics, the possible mechanisms involved and emetic potential. We expected to discover a safe and effective PDE4 inhibitor to prevent and cure AD.Methods:(1) In order to confirm FFPM was a PDE4 inhibitor, partially purified human recombinant PDE4CAT, PDE4A4, PDE4B2, PDE4C1, PDE4D4 cDNA were used to determine the inhibitory effects of FFPM on PDE4 enzyme activities and the classic PDE4 inhibitor Rolipram was used as a positive control. The IC50 values of FFPM were calculated according to enzyme activity inhibition curve. AD mice model induced by scoloplamine hydrobromide that was used to evaluate the effect of FFPM (0.06,0.12,0.24,0.48 mg/kg) on the learning and memory deficits in acquistion trial and probe trial of Morris water maze. To demonstrate whether FFPM was effective to pass through the blood-brain barrier, ultra fast liquid chromatography-tandem mass spectrometry (UFLC-MS/MS) was carried out to detect the concentration of FFPM in plasma and brain and to study the pharmacokinetics of FFPM in C57 mice to attain the main parameters such as Tmax, Cmax, t1/2, MRTo-t, AUC0-t and AUC0-∞. Combined injection of ketamine and xylazine to mice alternative model was performed to observe the duration of anesthesia, which was regarded as an endpoint to reflect the emetic effect of FFPM. Mice were orally administered with FFPM (0.50 mg/kg), Rolipram (0.50 mg/kg) and vehicle. Beagle bogs were treated with different doses of FFPM (0.25 mg/kg,0.50 mg/kg,1.0 mg/kg) and Rolipram (1.0 mg/kg) to further validate whether FFPM led to nausea or emesis side effects.(2) The DNA from tail tissues of APPswe/PSlde9 (APP/PS1) transgenic mice were extracted followed by RT-PCR amplification reaction and agarose electrophoresis to genetype identification. Behavioral tests included novel object recognition test, Morris water maze test and step down passive avoidance test were carried out to evaluate the cognitive function of APP/PS1 transgenic mice and wild type mice in same background at age of 4,6,8 months old, which could provide a reliable timepoint for FFPM. The expression of CD68 (activated microglial marker) congo red plaques and co-localization of two in brain of mice were examined by immunohistochemical and congo red staining analysis. ELISA analysis was used to determine soluble Aβ40, Aβ42 and insoluble Aβ40, Aβ42 in brain of mice at age of 6, 8 months old, and release of TNF-a and IL-1β in brain of mice at age of 8 months old.(3) 7 months old APP/PS1 transgenic mice were orally administered with FFPM (0.125,0.25,0.50 mg/kg) for consecutive 21 days, Rolipram as a positive control, using locomotor activity test, novel object recognition test, Morris water maze test and step down passive avoidance to determine the effect of FFPM on spontaneous movement and memory ability in 8 months old APP/PS1 transgenic mice. cAMP content in hippocampus of mice were detected by ELISA. The protein expression of p-PKA, p-CREB, CREB, BDNF, Synapsin-1, PSD95, NF-κB p65, Histone H3, iNOS, COX-2, TNF-α,IL-1β and GAPDH were investigated using western blotting. The BDNF mRNA level was examined using Real time PCR. The immunoreactivity of BDNF, NeuN and co-localization of two in hippocampus were detected by immunofluorescence analysis.(4) The effects of FFPM (0.625～80 μM) and Aβ42 (1.25-20 μM) on viablity of BV-2 cells were detected using MTT analysis. FFPM (2.5,5.0,10 μM) and Rolipram (20 μM) against Aβ42 (2.5 μM) treated BV-2 cells followed by PKA inhibitor H89 were conducted.24 h later, the secretion of TNF-a, IL-1β and IL-6 in cell supernatants were examined by ELISA analysis. BV-2 cells treated with Aβ42 at different time points (0.5,1,2,4 h),10 μM FFPM pretreatment 1 h followed by 1 h Aβ42 and (or) PKA inhibtor H89 treatment, then p-PKA, p-CREB and GAPDH expression were investigated by western blotting. FFPM and (or) Aβ42 co-incubated for 1 h, nuclear translocation of NF-κB p65 was examined by immunocytochemistry technique. The NF-κB p65 expression in cytoplasm and nucleus, p-ERKl/2 expression were also examined using western blotting. The supetnants of Aβ42 and (or) FFPM and (or) H89 treated BV-2 cells were collected as conditioned mediums, which were used to co-incubated with SH-SY5Y cells for 24 h, then the expression of Synapsin-1 and PSD95 were detected by western blotting.Results:(1) Inhibitory value as IC50 of FFPM against PDE4CAT, PDE4A4, PDE4B2, PDE4C1, PDE4D4 is 56.2 nM,31 nM,42.9 nM,230 nM and 11.8 nM respectively. In contrast, inhibitory IC5o value of FFPM against PDE4CAT, PDE4A4, PDE4B2, PDE4C1, PDE4D4 is 2480 nM,159 nM,783 nM,118 nM and 260 nM. FFPM ameliorated scopolamine-induced learning and memory deficits in mice as demonstrated by decreasing the escape latency to reach hidden platform in acquistion trial, and increasing duration in target quadrant in probe trial which displayed significant difference at doses of 0.12 mg/kg,0.24 mg/kg and 0.48 mg/kg compared with normal mice. Pharmacokinetic studies were conducted that mice were orally administered with FFPM (0.50 mg/kg). The main paramters obtained were as followed:Tmax= 2h, Cmax= 9.723±2.994 ng/ml, t1/2=1.487±0.198 h, MRT0-t = 2.816±0.121 h, AUC0-t= 32.5±0.959 ng/ml · min, AUC0-∞= 33.91±10.675 ng/ml · min. As showed in the plasma-brain/concentration-time curves, FFPM was able to be detected in brain of mouse 15 min after drug administration, indicated that FFPM permeated into brain effectively. In the experiment of ketamine/xylazine induced anesthesia, Rolipram (0.50 mg/kg) significantly shorten the duration of anesthesia in mice as compared with vehicle treated one. In contrast, FFPM exhibited the same anesthetic effect with vehicle, which suggested that FFPM had little emetic response to mouse. The results of beagle dogs further showed that Rolipram-treated dogs had evident emetic time as 9.60 ± 1.373 min after drug adminstration. However, FFPM at different doses (0.25,0.50,1.0 mg/kg) and vehicle had no significant emetic effects during 120 min observation period. Both the results of mice and beagle dogs confirmed that FFPM displayed low emetic potency.(2) All of the mice after genotype identification with APP and PS1 mutation gene amplification bands are the APP/PS1 transgenic mice or else are wild type mice. The result of novel object recognition showed that there were no obvious difference in recognition index for the two same objects between APP/PS1 trangenic mice and wild type mice at 4,6,8 months old. Compared with wild type mice, only 8-month-old APP/PS1 transgenic mice displayed significant low preference for new objects. In the Morris water maze test, all of the mice showed a tendency to find the hidden platfrom in acquistion trial.8-month-old APP/PS1 transgenic mice took more time to reach the platform than wild type mice did. Furthermore,6,8 months old APP/PS1 spent less exploratory time in platform quadrant than wild type mice did in probe trial. The step down passive avoidance test result suggested that the step down latencies of 6,8 months old APP/PS1 mice is much lesser than wild type mice at the same ages. In addition, A(β level in brain of APP/PS1 transgenic mice increased partly relative to wild mice at age of 6,8 months, and the insoluble Aβ40 and Aβ42 increased statistically. The immunohistochemical analysis and congo red staining indicated the microglial cells in 8 months old APP/PS1 mice were activated by stimulis, which reflected as increased CD68 (microglial marker) expression. The congo red plaques were surrounded by a large number of activated microglias in brain of APP/PS1 transgenic mice. Compared with wild type mice, the release of TNF-a and IL-1β in 8-month old mice were greatly increased.(3) Mice treated with PDE4 inhibitor FFPM or Rolipram or vehicle for 21 days displayed no difference in locomotor activity, indicating that long term treatment with FFPM did not affect the spontaneous movement in both APP/PS1 transgenic mice and wild type mice. FFPM (0.50 mg/kg) significantly increased recognition index of APP/PS1 transgenic mice for novel object in novel object recognition test. Results of Morris water maze suggested that consecutive treatment with FFPM shorten latency of APP/PS1 transgenic mice to reach the hidden platform. At day 4 and 5 of acquistion trial, the effects of FFPM in APP/PS1 transgenic mice relative to vehicle-treated one were evident. In probe trial, FFPM at doses of 0.25 mg/kg and 0.50 mg/kg statistically increased swimming distance and exploratory time in platform quadrant in APP/PS1 transgenic mice. When compared with APP/PS1 transgenic mice treated with vehicle, the retention time of APP/PS1 transgenic mice on the platform in step down passive avoidance was prolonged by FFPM (0.25 mg/kg and 0.50 mg/kg). The chronic treatment with FFPM reversed the decreased cAMP level in the hippocampus of APP/PS1 transgenic mice, then increased hippocampal p-PKA, p-CREB, BDNF, Synapsin-1 and PSD95 expression while did not affect CREB and GAPDH expression. Consistent with the results of western blotting, FFPM reversed the downregulation of BDNF and NeuN immunoreactivtiy in the hippocampus of APP/PS1 transgenic mice as examined by immunofluorescence analysis, which indicated that increased BDNF expression might further contribute to neurons survival. In addition, FFPM increased BNDF mRNA in hippocampus of APP/PS1 transgenic mice. Moreover, repeated administration with FFPM also decreased hippocampal nuclear NF-κB p65 and proinflammatory cytokines such as iNOS, COX-2, TNF-βα and IL-1β protein expression.(4) Compared with the normal control, FFPM exhibited no evident toxicity to BV-2 cells at concentrations ranged from 0.625 to 80 μM, whereas Aβ42 (5,10,20 μM) declined BV-2 cells viablity rate below 80% as examined using MTT analysis. BV-2 cells pretreatment with FFPM (2.5,5,10 μM) for 1 h, to some extent, inhibited the proinflammatory cytokines release induced by Aβ42 (2.5 μM). Importantly, FFPM at doses of 10 μM statistically reduced TNF-a, IL-1β and IL-6 secretion and the effects were better than Rolipram. However, FFPM mediated inflammatory inhibition in BV-2 cells were blocked by pretreatment with PKA inhibitor H89 (10 μM). Aβ42 incubated with BV-2 cells for 1 h significantly decreased the expression of p-PKA and p-CREB, whereas these effects were reversed by FFPM. Further, pretreatment with H89 blocked the reversal effects of FFPM. In addition,2.5μM Aβ42 also induced NF-κB p65 translocated from cytoplasm to nucleu as observed in immunocytochemical analysis. FFPM treated for 1 h effectively reduced it. The results of western blotting futher showed that FFPM significantly reversed Aβ42 induced decreased NF-κB p65 expression in cytoplasm and the increased NF-κB p65 expression in nucleu. Aβ42 decreased p-ERKl/2 protein expression in BV-2 cells following 1 h incubation. FFPM displayed revervsal effect on it but not significantly. Additionally, FFPM could reversed the decreased expression of Synapsin 1 and PSD95 in SH-SY5Y cells, which caused by Aβ42 activated BV-2 conditioned medium.Conclusion:FFPM is a specific PDE4 inhibitor, which could effectively pass through the blood brain barrier, has more inhibitory potency than Rolipram on PDE4CAT, PDE4A, PDE4B and PDE4D. In addition, FFPM improved learning and memory deficits in scopolamine-induced mice and APP/PS1 transgenic mice. Moreover, FFPM displayed little emetic response in mice and beagle bogs. FFPM mainly enhanced cognitive function through activating cAMP/PKA/CREB signal pathway, which further mediated the transcription and translation of BDNF and survival of neurons, increased the protein expression of synapsin-1 and PSD95. Besides, FFPM also inhibited nuclear translocation in Aβ-induced microglial cells, thus protected neurons and synapses from inflammatory responses.