Adenosinergic System and BDNF Signaling Changes as a Cross-Sectional Feature of RTT: Characterization of Mecp2 Heterozygous Mouse Females

Rett Syndrome is an X-linked neurodevelopmental disorder (RTT; OMIM#312750) associated to MECP2 mutations. MeCP2 dysfunction is seen as one cause for the deficiencies found in brain-derived neurotrophic factor (BDNF) signaling, since BDNF is one of the genes under MeCP2 jurisdiction. BDNF signaling is also dependent on the proper function of the adenosinergic system. Indeed, both BDNF signaling and the adenosinergic system are altered in Mecp2-null mice (Mecp2−/y), a representative model of severe manifestation of RTT. Considering that symptoms severity largely differs among RTT patients, we set out to investigate the BDNF and ADO signaling modifications in Mecp2 heterozygous female mice (Mecp2+/−) presenting a less severe phenotype. Symptomatic Mecp2+/− mice have lower BDNF levels in the cortex and hippocampus. This is accompanied by a loss of BDNF-induced facilitation of hippocampal long-term potentiation (LTP), which could be restored upon selective activation of adenosine A2A receptors (A2AR). While no differences were observed in the amount of adenosine in the cortex and hippocampus of Mecp2+/− mice compared with healthy littermates, the density of the A1R and A2AR subtype receptors was, respectively, upregulated and downregulated in the hippocampus. Data suggest that significant changes in BDNF and adenosine signaling pathways are present in an RTT model with a milder disease phenotype: Mecp2+/− female animals. These features strengthen the theory that boosting adenosinergic activity may be a valid therapeutic strategy for RTT patients, regardless of their genetic penetrance.


Introduction
Rett Syndrome (RTT; OMIM#312750) is a neurodevelopmental syndrome that affects mostly girls.In the last 20 years, it has presented a prevalence of 5-10:100.000cases [1,2].RTT patients are clinically described as presenting an apparently normal development, reaching most expected development milestones until 6-18 months after birth, when some signs and symptoms of the disease become apparent [3][4][5].As the disease progresses, the following symptoms may arise: stereotypic hand movement and loss of purposeful hand movements, motor coordination features, epileptic seizures, language and learning deficits, and cognitive impairments ranging from mild to severe [6].
The vast majority of RTT cases are caused by mutations in the Methyl-CpG-binding protein 2 (MECP2) gene, present in the X chromosome [1].MECP2 encodes the homonymous protein MeCP2, which displays a wide range of functions, including a critical role in the regulation of genetic transcription [7,8].One of the genes under MeCP2 control is the brain-derived neurotrophic factor (BDNF) gene [9].BDNF is an important neurotrophin, with impactful functions during neuronal development and maturation but also in the modulation of synaptic plasticity of mature neuronal circuits [10] via the activation of the high-affinity tropomyosin receptor kinase B full-length (TrkB-FL) receptors [11,12].In addition, BDNF can also bind, though with less affinity, to TrkB truncated isoforms (TrkB-Tc), which act as a negative effectors of TrkB-FL receptors [13,14].
Previous findings from our group and others have shown that the brain of Mecp2 −/y male mice possess lower levels of the BDNF protein [15][16][17][18].Moreover, there is also a reduction in the levels of TrkB-FL receptor in the cortex and hippocampus of Mecp2 −/y male mice, which may also contribute to BDNF-associated functional deficits found in this RTT model [15].These findings strengthen the theory that BDNF signaling is downregulated in Mecp2 −/y animals and may be paramount to RTT pathophysiology.On top of that, we also showed that extracts of the hippocampus and cortex from symptomatic Mecp2 −/y mice display lower levels of adenosine and its membrane receptor A 2A R when compared with wild-type (WT) littermates.
Mounting evidence indicates that synergism between adenosine A 2A R and BDNF receptors is crucial for a suitable neuronal function and synaptic transmission.This includes the maintenance of adequate levels of BDNF and TrkB-FL receptors and consequently its mediated actions, such as the preservation of long-term potentiation (LTP), considered an electrophysiological model for the basic mechanisms involved in learning and memory formation [19][20][21][22][23][24][25][26][27].Impairments in the A 2A R/BDNF crosstalk were observed in symptomatic Mecp2 −/y mice.And it was proved, for the first time, that some of BDNF functional deficits in Mecp2 −/y mice can be overcome by selective A 2A R activation, leading us to hypothesize that boosting BDNF signaling by increasing the availability of endogenous adenosine and/or the activity of A 2A R could be a useful therapeutic strategy for RTT patients [15].
Although the use of symptomatic RTT Mecp2-null animals in research has the advantage of reproducibility and relatively low variability in the results produced [28,29], RTT is a disease characterized by a broad spectrum of clinical manifestations with variable severity, as predicted by the existence of different patterns of MeCP2 expression levels [30][31][32].This prompted the need for investigating Mecp2 heterozygous female mice (Mecp2 +/− ) presenting variable phenotypes, which are less severe when compared with Mecp2-null male mice due to X-chromosome inactivation (XCI) as a proxy of the majority of RTT female clinical patients [33].Herein, we set out to investigate alterations in adenosinergic/BDNF signaling in the cortex and hippocampus of Mecp2 +/− heterozygous female mice.Furthermore, we used a pharmacological approach to manipulate adenosine receptor signaling in an attempt to reverse BDNF signaling deficits in synaptic plasticity in this disease-relevant RTT mouse model.

The Magnitude of Hippocampal LTP Is Maintained in Mecp2 +/− Female Mice but BDNF Loses Its Effect on LTP Magnitude
From previous works, it is known that LTP is impaired in Mecp2 −/y animals [15], and there is a strong impairment in the facilitation effect of BDNF on LTP [19,24].However, data on alterations in LTP and BDNF signaling in Mecp2 +/− female animals are scarcer [29,34,35].Therefore, we set out to investigate alterations in basal LTP and the effect of BDNF on this form of synaptic plasticity in a relevant RTT female mouse model.
2.2.BDNF Protein Levels Are Decreased in the Hippocampus and Cortex of Symtomatic Heterozygous Mecp2 +/− Female Mice BDNF protein levels were characterized in the cortex and hippocampus of Mecp2 +/− female mice by Western blot alongside Mecp2 levels.

BDNF Protein Levels
Are Decreased in the Hippocampus and Cortex of Symtomatic Heterozygous Mecp2 +/− Female Mice BDNF protein levels were characterized in the cortex and hippocampus of Mecp2 +/− female mice by Western blot alongside Mecp2 levels.

TrkB-FL Protein Levels Are Not Changed in the Hippocampus of Mecp2 +/− Mice
Since differences were found in BDNF protein levels, we next investigated if TrkB receptor protein levels were also affected in the hippocampus and cortex of this RTT female mouse model.

Adenosine (Plus Inosine) Levels Remain Unaltered in the Cortex and Hippocampus of Heterozygous Mecp2 +/− Female Mice
The differences found in BDNF-TrkB-FL signaling could be aggravated by changes in the adenosinergic system, since adenosine has an important role in the modulation of BDNF signaling [36].In this way, adenosine + inosine (an adenosine metabolite) amounts were measured in order to evaluate if adenosinergic system changes could also be involved in the aforementioned BDNF signaling deficits previously described.
Heterozygous Mecp2 +/− females had no significant (p > 0.05) differences in adenosine plus inosine content normalized to the wet tissue weight when compared with WT The differences found in BDNF-TrkB-FL signaling could be aggravated by changes in the adenosinergic system, since adenosine has an important role in the modulation of BDNF signaling [36].In this way, adenosine + inosine (an adenosine metabolite) amounts were measured in order to evaluate if adenosinergic system changes could also be involved in the aforementioned BDNF signaling deficits previously described.Heterozygous Mecp2 +/− females had no significant (p > 0.05) differences in adenosine plus inosine content normalized to the wet tissue weight when compared with WT controls, both in the cortex (WT: 382 ± 71 nmol/g, n = 7 vs.Mecp2 +/− : 382 ± 79 nmol, n = 8; Figure 4A) and in the hippocampus (WT: 225 ± 72 nmol, n = 5 vs. Mecp2 +/− : 259 ± 94 nmol, n = 6; Figure 4B).Interestingly, cortical levels of adenosine + inosine (ADO + INO) were higher (about threefold) than those detected in the hippocampus (Figure 4A,B).

Mecp2 +/− Females Display Alterations in Hippocampal A 2A R and A 1 R Adenosine Receptor Levels
We also addressed putative changes in A 1 R and A 2A R adenosine receptor levels in hippocampal homogenates of Mecp2 +/− and WT female animals.

Exogenous Activation of A2AR Restores BDNF-Induced LTP Facilitation in the Hippocampus of Heterozygous Mecp2 +/− Female Mice
Adenosine A2AR activation is known to boost TrkB-FL receptor signaling [36].Therefore, we aimed to test if exogenous activation of A2AR with a potent specific agonist could compensate for the decreased A2AR levels found in the hippocampus of Mecp2 +/− female mice.In this way, we tested the effect of CGS21680 (10 nM) on LTP in hippocampal slices in the presence or absence of exogenously added BDNF (20 ng/mL) (see e.g., [38]).

Exogenous Activation of A 2A R Restores BDNF-Induced LTP Facilitation in the Hippocampus of Heterozygous Mecp2 +/− Female Mice
Adenosine A 2A R activation is known to boost TrkB-FL receptor signaling [36].Therefore, we aimed to test if exogenous activation of A 2A R with a potent specific agonist could compensate for the decreased A 2A R levels found in the hippocampus of Mecp2 +/− female mice.In this way, we tested the effect of CGS21680 (10 nM) on LTP in hippocampal slices in the presence or absence of exogenously added BDNF (20 ng/mL) (see e.g., [38]).

Discussion
Considering the genetic profile of RTT, heterozygous female mice are more representative of the genetic phenomena occurring in this syndrome when compared with Mecp2-null male mice [29,33].Some studies suggest that, as a consequence of XCI and differences in skewing of the mutant X chromosome, among other emerging factors, patients carrying similar mutations might reveal differences in the clinical severity of the manifestation of the syndrome [39][40][41].Despite this knowledge, the advantages associated with studying Mecp2-null mice, like high reproducibility, good characterization, and rapid establishment of symptomatology that mimics a great part of the symptoms present in patients [28,42], still turn female heterozygous mice into a backup model for RTT studies.The data presented on changes in BDNF signaling and in the adenosinergic system show, however, that some characteristics can cut across different phenotypic severities caused by total or partial loss of Mecp2 expression and/or function.
Even though regulation of the Bdnf gene by Mecp2 has been widely studied, the exact mechanisms behind this regulation are not clearly known [9,32].Most of the studies describe a decrease in BDNF protein levels in Mecp2-null male mice, detectable at the onset of the first behavioral impairments [16].Several studies demonstrate a significant decrease in BDNF levels in symptomatic Mecp2-null mice, not only in the cortex and hippocampus but also in other brain regions [15][16][17][18].The decrease in BDNF levels shows a progression from the caudal brain regions affecting all brains around 7 weeks of age [17,43].Studies in heterozygous female mice are scarce, but there is evidence for decreased levels of BDNF in the medulla, pons, sensory nodose ganglion, and cortex in symptomatic animals [35].

Discussion
Considering the genetic profile of RTT, heterozygous female mice are more representative of the genetic phenomena occurring in this syndrome when compared with Mecp2-null male mice [29,33].Some studies suggest that, as a consequence of XCI and differences in skewing of the mutant X chromosome, among other emerging factors, patients carrying similar mutations might reveal differences in the clinical severity of the manifestation of the syndrome [39][40][41].Despite this knowledge, the advantages associated with studying Mecp2-null mice, like high reproducibility, good characterization, and rapid establishment of symptomatology that mimics a great part of the symptoms present in patients [28,42], still turn female heterozygous mice into a backup model for RTT studies.The data presented on changes in BDNF signaling and in the adenosinergic system show, however, that some characteristics can cut across different phenotypic severities caused by total or partial loss of Mecp2 expression and/or function.
Even though regulation of the Bdnf gene by Mecp2 has been widely studied, the exact mechanisms behind this regulation are not clearly known [9,32].Most of the studies describe a decrease in BDNF protein levels in Mecp2-null male mice, detectable at the onset of the first behavioral impairments [16].Several studies demonstrate a significant decrease in BDNF levels in symptomatic Mecp2-null mice, not only in the cortex and hippocampus but also in other brain regions [15][16][17][18].The decrease in BDNF levels shows a progression from the caudal brain regions affecting all brains around 7 weeks of age [17,43].Studies in heterozygous female mice are scarce, but there is evidence for decreased levels of BDNF in the medulla, pons, sensory nodose ganglion, and cortex in symptomatic animals [35].
In the present work, the data obtained using Mecp2 heterozygous females, Mecp2 +/− , show a decrease in BDNF protein levels, both in the cortex and hippocampus of symp-9 of 15 tomatic animals, with a simultaneous loss of the facilitatory effect of BDNF upon LTP in the hippocampus.These results are in line with the previously mentioned studies showing a significate decrease in BDNF levels not only in the cortex and hippocampus of Mecp2 −/y males but also in other brain regions of symptomatic male Mecp2 −/y and Mecp2 +/− female models of RTT [15][16][17][18] In contrast with the observed reduction in BDNF levels, no alteration was found in TrkB-FL levels in the hippocampus and cortex of Mecp2 +/− females.These two brain regions are highly affected in Mecp2 −/y animals and are associated with multiple functions altered in RTT, namely cognition [15,16,44].Our results in Mecp2 +/− females differ from those previously obtained in Mecp2 −/y males, where a reduction in TrkB-FL in the cortex and hippocampus could potentially explain the inability of exogenous BDNF to potentiate synaptic plasticity.These arguments could also be used to justify the inefficacy of some therapeutic strategies grounded in the increase in BDNF levels [9,15].
Contrary to what has been observed in the hippocampus of Mecp2 −/y animals [13,37], heterozygous females do not show alterations in basal LTP.However, a recent study already described alterations in the hippocampal LTP of a symptomatic RTT female mouse model, heterozygous for Mecp2 (Mecp2 tm1.1Jae ) [34].Although these changes were described in an Mecp2-null model, the and the strain differ from the one studied in the present study, suggesting that LTP impairments in Mecp2 heterozygous models might also be dependent on the age and strain of the model used.Nevertheless, it is also important to take into account that different protocols could be used to study LTP.Indeed, the mentioned study used a stronger θ-burst protocol, which could also explain the discrepancy between both studies.Moreover, we found that the exogenous BDNF was unable to induce any potentiation of hippocampal LTP, as previously observed in Mecp2 −/y animals [15].Importantly, the analysis of I/O curves demonstrated that deficits in the facilitation of hippocampal LTP were not due to alterations in synaptic efficacy but rather probably due to changes in BDNF signaling.Interestingly, our data showed a BDNF effect on PTP facilitation in Mecp2 +/− female animals.These data suggest that the BDNF facilitatory effect was transient and, therefore, insufficient to be translated into LTP in Mecp2 +/− female animals, in opposition to WT animals, where the BDNF faciliatory effect is just observed in long-term plasticity forms.In fact, diminished PTP is described in one study performed in Mecp2 −/y male animals [45].The few characterizations of short-term forms of plasticity made, so far, in Mecp2-null mice models could be a missing piece that is important to understand if any mechanisms during short-term neuronal plasticity are changed and that could justify the alterations described in long-term plasticity forms in RTT [34,[45][46][47].
Still, although not entirely superimposed to the alterations found in previous studies in Mecp2 −/y animals, in the model now studied, which displays a milder symptomatic phenotype, the dysregulation of BDNF signaling was also observed.
In addition, a dysregulation of the adenosinergic system in an RTT female mouse model was also presented in this study.Adenosine and BDNF signaling are tightly linked, since A 2A R activation is known to increase BDNF and TrkB-FL levels; promote transactivation of TrkB-FL receptors and downstream signaling through the Gs-cAMP-PKA pathway; and promote translocation of TrkB-FL levels to lipid rafts in the cell membrane [21,24,36,48].Therefore, deciphering the dysregulation in this system could bring new insights into the mechanisms contributing to BDNF signaling impairments in RTT.
Here, we observed alterations in the expression of adenosine receptors subtypes A 1 and A 2A , in which downstream signaling pathways play different roles in cell physiology, as A 1 R and A 2A R are canonically coupled to Gi and Gs, respectively [49].In the hippocampus of Mecp2 +/− female, we observed an increase in A 1 R and a decrease in A 2A R protein levels.These results are in line with our previous study using Mecp2 −/y male animals, showing an increase in A 1 R and a reduction in A 2A R levels both in the cortex and hippocampus [15].The reduction in A 2A R protein levels in the hippocampus of Mecp2 +/− female mice may, at least in part, be responsible for the lack facilitatory effect of BDNF upon LTP observed in this study.
However, in contrast with our observations in Mecp2 −/y male animals, where a reduction in overall adenosine levels was observed in the cortex and hippocampus, in the Mecp2 +/− female mouse model, adenosine levels remained unaltered [15].This maintenance of the adenosine levels may contribute to a milder RTT phenotype compared with homozygous Mecp2 −/y males.
The mechanism underlying the overall alterations in A 1 R and A 2A R levels in both RTT models remains elusive.However, we hypothesize the overexcitability associated with this pathology could be a triggering factor for these changes which, at some extent, could compensate for the excitatory-inhibitory imbalance, mainly described in Mecp2-null mice [44].Remarkably, although lower levels of A 2A R were detected, the available receptors were enough to respond to exogenous activation with a selective agonist, since CGS21680 was able to rescue the action of BDNF on LTP.
In this context, we observed that pharmacological activation of A 2A R allowed the recovery of the facilitatory action of BDNF on LTP.These changes were congruent with those described previously in Mecp2 −/y mice, suggesting that the dysregulation of BDNF signaling and the adenosinergic system are a hallmark of RTT and transversal to different mouse models presenting different disease severities.Furthermore, these data also support the importance of adenosinergic system regulation for effective BDNF functions [15].
Taken together, the results presented here reinforce the hypothesis that adenosineenhancing therapies could be a useful strategy in RTT where there is a decrease in BDNF signaling together with impairments in the adenosinergic system.
The animals were housed in a 12 h light/dark cycle, with food and water provided ad libitum.Experiments involving animals were taken into careful consideration in order to reduce the number of animals sacrificed.All animals were handled according to the Portuguese law on Animal Care and European Union guidelines (86/609/EEC).

Ex Vivo Electrophysiological Recordings
Hippocampus isolation and slice preparation: The animals were first anesthetized with isoflurane (1,2-Propylenglycol 50% (v/v)) in an anesthesia chamber.When animals showed the first signs of anesthetic state, like reduction in respiratory rate and lack of reflexes, they were sacrificed by decapitation.In order to have access to the brain, the skull was exposed by cutting the skin at the top of the head, and then the brain was carefully removed and placed in ice-cold artificial cerebrospinal fluid (aCSF-Krebs' solution) containing 124 mM NaCl; 3 mM KCl; 1.25 mM NaH 2 PO 4 , 26 mM NaHCO 3 , 1 mM MgSO 4 , 2 mM CaCl 2 , and 10 mM glucose previously gassed with 95% O 2 and 5% CO 2 , pH 7.4.The two brain hemispheres were separated through the midline, and the hippocampus was isolated, taking care not to damage it with the spatulas.When isolated, the hippocampus was cut perpendicularly to the long axis into slices with 400 µm thickness with the Mcllwain tissue chopper.Slices were then placed in a resting chamber in Krebs' solution and permanently oxygenated at room temperature for one hour in order to recover [15].
LTP induction and quantification: After functional and energetic recovery, slices were transferred to the recording chamber for submerged slices, being continuously superfused with bathing solution (Krebs' solution) gassed with 95% O 2 and 5% CO 2 at 32 • C. The flux of bathing solution was established at 3 mL/min and the drugs used were added to this superfusion solution.Recordings were obtained with an Axoclamp 2B amplifier and digitized (Axon Instruments, Foster City, CA, USA).Individual responses were monitored, and averages of eight consecutive responses were continuously stored on a personal computer with the LTP program [52].Field excitatory postsynaptic potentials (fEPSPs) were recorded through an extracellular microelectrode (2-8 MW resistance, Harvard apparatus LTD, Fircroft way, Edenbridge, Kent) placed in the stratum radiatum of the CA1 area.Stimulation (rectangular 0.1 ms pulses, once every 10 s) was delivered through a concentric electrode placed on the Schaffer collateral-commissural fibers in the stratum radiatum near CA3-CA1 border.The stimulus intensity (80-260 µA) was initially adjusted to obtain a large fEPSP with a minimum population spike contamination.Stimulation was delivered alternatively to two independent pathways (previously tested by paired pulsed facilitation protocol).LTP was induced by a θ-burst protocol consisting of three trains of 100 Hz, three stimuli, separated by 200 ms, as previously described [19].LTP was quantified as the % change in the average slope of the fEPSP taken 50 to 60 min after LTP induction in relation to the average slope of the fEPSP, measured during the 10 min that preceded the induction of LTP.In each individual experiment, the same LTP-inducing paradigm was delivered to each pathway.At 1 h after LTP induction, in one of the pathways (S1-Figure 1B), BDNF (20 ng/mL), CGS21680 (10 nM), or both were added to the superfusion solution, and LTP was induced in the second pathway (S2-Figure 1B) no less than 15 min after the drug perfusion and only after stability of fEPSP slope values was observed for at least 10 min.The effect of the drug tested upon LTP was evaluated by comparing the magnitude of LTP in the first pathway in the absence of BDNF (control pathway) with the magnitude of LTP in the second pathway in the presence of the drug (test pathway); each pathway was used as control or test on alternate days [15].
Input-output curve (I/O): After obtaining a stable baseline for at least 10 min, the stimulus delivered to the slice was decreased until no fEPSP was evoked and subsequently increased in 20 µA steps.For each stimulation intensity, data from three consecutive averages of 8 fEPSP were collected.Inputs delivered to slices typically ranged from 60 µA to a supramaximal stimulation of 320 µA.The input-output curve was plotted as the relationship of fEPSP slope vs. stimulus intensity, which provides a measure of synaptic efficiency, as previously described [53].

Extraction and Analysis of Adenosine and Inosine by Liquid Chromatography with Diode Array Detection (HPLC/DAD)
Adenine nucleosides (adenosine and inosine) extracted from the cortex and hippocampus of WT and Mecp 2−/y female mice were measured by high-performance liquid chromatography with diode array detection (HPLC/DAD), as previously described [15].Snap-frozen cortex and hippocampal tissue samples were stored at −80 • C until use.For extraction, the samples were defrosted (250 µL) in round-bottom microcentrifuge tubes, followed by thorough tissue homogenization using a mixture of ice-cold acetonitrile: methanol: water (1:2:2) solution containing 2-chloro-adenosine (5 µM) as internal standard; the obtained mixture was centrifuged at 16,000× g for 20 min at 4 • C. Tissue homogenization and centrifugation were repeated twice.The two recovered supernatant extracts (~250 µL each) were mixed together and then centrifuged again at 16,000× g (for 20 min at 4 • C) using a 50-kDa cutoff filter (Amicon Ultra-0.5 50 K Filter Device; Merck KGaA, Darmstadt, Germany).After filtration, supernatant extracts were divided into 15 µL aliquots and stored at −80 • C until analysis.Using this procedure, recovery of adenine nucleosides was higher than 95%, as determined by adding 2-chloro-adenosine (5 µM) as internal standard before extraction (see e.g., ref. [15]).
Extraction media containing purines were 1/10 diluted with ultrapure water before HPLC/DAD analysis.The chromatographic separation of nucleosides was carried out us-ing an elution gradient composed of ammonium acetate (5 mM, with a pH of 6 adjusted with acetic acid) and methanol [54][55][56] (see Supplementary Figure S1).Separation of adenosine and inosine was achieved by reversed-phase liquid chromatography through a Hypersil GOLD C18 column (5 µM, 2.1 mm × 150 mm) equipped with a guard column (5 µm, 2.1 mm × 1 mm) and assayed using a Finigan Thermo Fisher Scientific System LC/DAD, equipped with an Accela Pump (version 1.04.0016)coupled to an Accela Autosample (version 2.2.1), a diode array detector, and an Accela PDA (version 2.3.0)running the Xcalibur software (version 2.1.0.1139) chromatography manager (RRID:SCR_014593;Thermo Fisher Scientific Inc., Thousand Oaks, CA, USA).During the procedure, the flow rate was set at 200 µL•min −1 , and the column temperature was maintained at 20 • C. The autosampler was set at 4 • C, and 50 µL of standard or sample was injected, in duplicate, for each HPLC analysis.Quantification of adenine nucleosides was carried out using calibration curves made of high-purity external standards, namely adenosine and inosine.

Data Analysis
The data are expressed as mean ± SEM of the n number of independent experiments.The significance of differences between the means of 2 conditions was evaluated by paired or unpaired t-tests; Welch correction was used in unpaired t-test as appropriate.Nonlinear regressions were used to fit data pertaining to I/O curves and radioligand binding experi-

Figure 1 .Figure 1 .
Figure 1.The magnitude of hippocampal LTP is maintained in Mecp2 +/− female mice, but BDNF loses the facilitatory effect upon LTP in Mecp2 +/− female animals: (A) Schematic representation of a transverse hippocampal slice and the recording configuration used in the present study.(B) Schematic representation of the experimental protocol used to induce LTP and to test the effect of BDNF at a concentration of 20 ng/mL (for a detailed description, see Section 4.2).Panels (C,D) represent the time course of averaged fEPSP slopes induced after θ-burst stimulations in hippocampal slices isolated from WT (grey symbols, n = 6) or Mecp2 +/− (pink symbols, n = 8) mice, either in the absence (dark-filled circles bellow the bar, light bars) or in the presence (white-filled circles bellow the bar, dark bars) of BDNF.Recording traces from representative experiments are shown for WT (black trace-basal synaptic potential (1,3); gray trace-synaptic potential after LTP (2,4)) and Mecp2 +/− (pink) animals on top of each graph (1 and 3-baseline; 2 and 4-LTP, 1 h after the θ-burst).The arrow indicates LTP induction.In (E), the histogram shows the magnitude of LTP in the absence (dark-LTP S1) and the presence of BDNF (20 ng/mL, light bars-LTP S2) obtained in hippocampal slices from WT (gray) or Mecp2 +/− (pink) animals; the PTP magnitude is shown in (F).Panels (G,H) show the input/output (I/O) curves and the respective means of the top parameter Figure 1.The magnitude of hippocampal LTP is maintained in Mecp2 +/− female mice, but BDNF loses the facilitatory effect upon LTP in Mecp2 +/− female animals: (A) Schematic representation of a transverse hippocampal slice and the recording configuration used in the present study.(B) Schematic representation of the experimental protocol used to induce LTP and to test the effect of BDNF at a concentration of 20 ng/mL (for a detailed description, see Section 4.2).Panels (C,D) represent the time course of averaged fEPSP slopes induced after θ-burst stimulations in hippocampal slices isolated from WT (grey symbols, n = 6) or Mecp2 +/− (pink symbols, n = 8) mice, either in the absence (dark-filled circles bellow the bar, light bars) or in the presence (white-filled circles bellow the bar, dark bars) of BDNF.Recording traces from representative experiments are shown for WT (black trace-basal synaptic potential (1,3); gray trace-synaptic potential after LTP (2,4)) and Mecp2 +/− (pink) animals on top of each graph (1 and 3-baseline; 2 and 4-LTP, 1 h after the θ-burst).The arrow indicates LTP induction.In (E), the histogram shows the magnitude of LTP in the absence (dark-LTP S1) and the presence of BDNF (20 ng/mL, light bars-LTP S2) obtained in hippocampal slices from WT (gray) or Mecp2 +/− (pink) animals; the PTP magnitude is shown in (F).Panels (G,H) show the input/output (I/O) curves and the respective means of the top parameter corresponding to responses generated by various stimulation intensities (60-340 µA) in WT (gray circles, n = 8) and Mecp2 +/− (pink circles, n = 8) mice hippocampal slices.* p < 0.05; ** p < 0.01 (paired Student's t-test) as compared with the absence of BDNF in the same experiments.

Figure 4 .
Figure 4. Adenosine (ADO) plus inosine (INO) content in the cortex and hippocampus of W heterozygous Mecp2 +/− female mice: Panels (A,B) represent the amount of ADO+INO in nm wet tissue weight detected by HPLC/DAD from extracts of the cortex and hippoca respectively, of wild-type (WT, gray bars) and Mecp2 +/− (HET, pink bars) female mice.Panels represent the adenosine deaminase (ADA) activity given by the INO/ADO ratio in extracts cortex and hippocampus, respectively, of WT (gray bars) and Mecp2 +/− (HET, pink bars) femal Box-and-whiskers plots are represented with whiskers ranging from minimum to maximum v the horizontal lines inside boxes indicate the corresponding medians.Each data set represen to eight individuals (see text for details); duplicate measurements were performed fo individual experiment.. Typical chromatograms obtained using these experimental settin shown in Supplementary Figure S1.

Figure 4 .
Figure 4. Adenosine (ADO) plus inosine (INO) content in the cortex and hippocampus of WT and heterozygous Mecp2 +/− female mice: Panels (A,B) represent the amount of ADO+INO in nmol/g of wet tissue weight detected by HPLC/DAD from extracts of the cortex and hippocampus, respectively, of wild-type (WT, gray bars) and Mecp2 +/− (HET, pink bars) female mice.Panels (C,D) represent the adenosine deaminase (ADA) activity given by the INO/ADO ratio in extracts of the cortex and hippocampus, respectively, of WT (gray bars) and Mecp2 +/− (HET, pink bars) female mice.Box-and-whiskers plots are represented with whiskers ranging from minimum to maximum values; the horizontal lines inside boxes indicate the corresponding medians.Each data set represents five to eight individuals (see text for details); duplicate measurements were performed for each individual experiment.. Typical chromatograms obtained using these experimental settings are shown in Supplementary Figure S1.

Figure 5 .
Figure 5.A 1 R and A 2A R protein levels in the hippocampus of Mecp2 +/− female mice: Histograms represent averages of A 1 R ((A), WT hip , n = 5 and Mecp2 +/− hip , n = 5) and A 2A R ((B), WT hip , n = 6 and Mecp2 +/− hip , n = 6) immunoreactivity in hippocampal homogenates from WT (gray bars) and Mecp2 +/− (pink bars) female mice at 26-30 weeks-old.In (C), representative immunoblots used for quantification are shown.The immunoreactivity against GAPDH was used for normalization purposes.All values are mean± SEM.* p < 0.05 (unpaired Student's t-test) against WT values.Each circle represents one independent n.

Figure 6 .
Figure 6.Exogenous activation of A2AR restores the facilitatory effect of BDNF on hippocampal LTP in heterozygous Mecp2 +/− female mice: Panels (A,B) show the time course of the averaged field excitatory postsynaptic potential (fEPSP) slope after a θ-burst stimulation of hippocampal slices from Mecp2 +/− female mice obtained in the absence (gray circles) and in the presence (orange circles) of the selective A2AR agonist, CGS21680 (10 nM), alone ((A), n = 5) or coapplied with BDNF (20 ng/mL, (B), pink circles, n = 6).The A2AR agonist was applied 60 min after induction of LTP in the first pathway (gray circles-control LTP) and at least 15 min before induction of LTP in the second pathway (orange, pink circles).Representative traces from representative experiments are shown (black trace-basal synaptic potential (1,3); pink trace-synaptic potential after LTP (2,4)).The bar graph in panel (C) shows the percentage changes in the slope of fEPSP computed from panels (A,B) where dark-filled circles bellow the bars represent the presence of the drug and white-filled circles the absence of the drug.Gray bars represent control LTP (light gray-control experience for CGS alone, dark gray-control experience for CGS + BDNF experience), orange bar represents CGS alone effect upon LTP and pink bar CGS + BDNF effect upon LTP.Data are mean ± SEM. * p < 0.05 (paired Student's t-test).Each circle on bars represents o one independent n.

Figure 6 .
Figure 6.Exogenous activation of A 2A R restores the facilitatory effect of BDNF on hippocampal LTP in heterozygous Mecp2 +/− female mice: Panels (A,B) show the time course of the averaged field excitatory postsynaptic potential (fEPSP) slope after a θ-burst stimulation of hippocampal slices from Mecp2 +/− female mice obtained in the absence (gray circles) and in the presence (orange circles) of the selective A 2A R agonist, CGS21680 (10 nM), alone ((A), n = 5) or coapplied with BDNF (20 ng/mL, (B), pink circles, n = 6).The A 2A R agonist was applied 60 min after induction of LTP in the first pathway (gray circles-control LTP) and at least 15 min before induction of LTP in the second pathway (orange, pink circles).Representative traces from representative experiments are shown (black trace-basal synaptic potential (1,3); pink trace-synaptic potential after LTP (2,4)).The bar graph in panel (C) shows the percentage changes in the slope of fEPSP computed from panels (A,B) where dark-filled circles bellow the bars represent the presence of the drug and white-filled circles the absence of the drug.Gray bars represent control LTP (light gray-control experience for CGS alone, dark gray-control experience for CGS + BDNF experience), orange bar represents CGS alone effect upon LTP and pink bar CGS + BDNF effect upon LTP.Data are mean ± SEM. * p < 0.05 (paired Student's t-test).Each circle on bars represents o one independent n.