UBC Faculty Research and Publications

Decreased cortical FADD protein is associated with clinical dementia and cognitive decline in an elderly… Ramos-Miguel, Alfredo; García-Sevilla, Jesús A; Barr, Alasdair M; Bayer, Thomas A; Falkai, Peter; Leurgans, Sue E; Schneider, Julie A; Bennett, David A; Honer, William G; García-Fuster, M. J Mar 20, 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-13024_2017_Article_168.pdf [ 1.81MB ]
JSON: 52383-1.0343277.json
JSON-LD: 52383-1.0343277-ld.json
RDF/XML (Pretty): 52383-1.0343277-rdf.xml
RDF/JSON: 52383-1.0343277-rdf.json
Turtle: 52383-1.0343277-turtle.txt
N-Triples: 52383-1.0343277-rdf-ntriples.txt
Original Record: 52383-1.0343277-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessDecreased cortical FADD protein isassociated with clinical dementia andcognitive decline in an elderly communitysampleAlfredo Ramos-Miguel1,2, Jesús A. García-Sevilla3,4, Alasdair M. Barr1,5, Thomas A. Bayer6, Peter Falkai7,Sue E. Leurgans8, Julie A. Schneider8, David A. Bennett8, William G. Honer1,2 and M. Julia García-Fuster3,4*AbstractBackground: FADD (Fas-associated death domain) adaptor is a crucial protein involved in the induction of celldeath but also mediates non-apoptotic actions via a phosphorylated form (p-Ser194-FADD). This study investigatedthe possible association of FADD forms with age-related neuropathologies, cognitive function, and the odds ofdementia in an elderly community sample.Methods: FADD forms were quantified by western blot analysis in dorsolateral prefrontal cortex (DLPFC) samplesfrom a large cohort of participants in a community-based aging study (Memory and Aging Project, MAP),experiencing no-(NCI, n = 51) or mild-(MCI, n = 42) cognitive impairment, or dementia (n = 57).Results: Cortical FADD was lower in subjects with dementia and lower FADD was associated with a greater load ofamyloid-β pathology, fewer presynaptic terminal markers, poorer cognitive function and increased odds of dementia.Together with the observations of FADD redistribution into tangles and dystrophic neurites within plaques inAlzheimer’s disease brains, and its reduction in APP23 mouse cortex, the results suggest this multifunctional proteinmight participate in the mechanisms linking amyloid and tau pathologies during the course of the illness.Conclusions: The present data suggests FADD as a putative biomarker for pathological processes associated with thecourse of clinical dementia.Keywords: Alzheimer’s disease, Aging, Neurotoxicity, Neuroplasticity, ApoptosisBackgroundAging is a relevant risk factor for Alzheimer’s disease(AD) [1], which is the main cause of dementia and ischaracterized by deposition of amyloid β in neuritic pla-ques, accumulation of tau in intracellular neurofibrillarytangles, and neuronal loss (see review in [2]). Cognitivedecline is a prerequisite for the clinical diagnosis ofdementia associated with AD, and usually correlatesbetter with neurofibrillary tangles of hyperphosphorylatedtau than with amyloid β plaques (reviewed in [3, 4]).Although nearly all brains in old age contain some patho-logical signs of AD, only some individuals develop thedisease [5–9], which seems to be greatly influenced by dif-ferences in cognitive reserve [10–12]. This premise,together with the time-course of the pathophysiology ofthe disease (i.e., preclinical stage initiated 15–20 yearsprior to emergence of clinical signs; [4]) suggests thatthese pathological markers may not be sufficient ornecessary to initiate cognitive decline in humans. Conse-quently, the identification of new biomarkers for diagnosisand for a sensitive assessment of the progression of AD(i.e., cognitive decline) is an important area of currentresearch.In line with this, cell death mediated via apoptosis hasbeen thought to be one of the possible underlying* Correspondence: j.garcia@uib.es3IUNICS, University of the Balearic Islands, Ctra. de Valldemossa km 7.5,E-07122 Palma de Mallorca, Spain4Instituto de Investigación Sanitaria de Baleares, Palma de Mallorca, SpainFull list of author information is available at the end of the article© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Ramos-Miguel et al. Molecular Neurodegeneration  (2017) 12:26 DOI 10.1186/s13024-017-0168-xmechanisms of neuronal cell loss in AD (see review in[13]). In fact, several studies have been performed in post-mortem brain tissue from AD patients on some of thecellular components of the apoptotic pathway [13]. Forexample, the expression of some components of theextrinsic pathway, like the cell death receptor Fas (e.g.,[14–16]), its ligand Fas-L [17] and some effector caspases[18, 19] showed evidence of apoptosis (i.e., DNA fragmen-tation determined by TUNEL method, [18]) in the brainsof AD patients. However, several contradictory resultshave also been described, suggesting that the apoptoticmechanism of neuronal death in AD might be infrequentor undetectable (e.g., [20, 21]), and that in fact, even cellsurvival mechanisms might take place as an adaptiveresponse to a prior insult (see revision in [13]).In terms of cell fate regulation (i.e., balance betweencell death and survival), a key adaptor molecule of theapoptotic Fas receptor [22], the FADD (Fas-associateddeath domain) is a crucial protein involved in the induc-tion of cell death but also able to mediate non-apoptoticactions (cell survival, differentiation and neuroplasticity)via a phosphorylated form (p-Ser194-FADD) and itsnuclear localization [23, 24]. Multifunctional FADDworks in vivo as a common and major signaling step inthe initial activation of structurally different receptors(e.g. neurotransmitter G protein-coupled receptors andreceptor channels; see [25–27]). Although FADD has acrucial role during embryogenesis/development [28], lit-tle is known about its expression or functions as thebrain ages [29] and in age-related neurophatologies. InAD one prior study showed that although FADD signal-ing pathway was triggered within the basal forebraincholinergic neurons, as FADD-positive tangle-like struc-tures were localized in neurons, there was no apoptoticcell death as measured by DNA fragmentation [30].Another study showed that the induction of neuralapoptosis produced by amyloid β in hippocampal neuroncultures was mediated via FADD and caspase-8 activity[31]. However, none of these studies explored thepotential link between brain FADD (or other apoptoticmarkers) variations and cognitive decline or AD-relatedpathology.Against this background, this study investigated thepossible associations of FADD forms (i.e., speculated in-crease of pro-apoptotic FADD) with the presence andseverity of multiple age-related neuropathologies, as wellas with cognitive function and the risk of clinical demen-tia in an elderly large community sample. Corticalexpression of FADD forms (i.e., pro-apoptotic FADDand anti-apoptotic p-FADD) were measured in postmor-tem tissue samples from a large cohort of community-dwelling participants of the Memory and Aging Project(MAP) [32], with or without clinical diagnosis of demen-tia, and representative of the broad range of cognitiveimpairment in the overall elderly, aging population.Moreover, to further ascertain the possible role of FADDregulation in AD, we utilized APP23 transgenic mice,which overexpress mutant human amyloid precursorprotein (APP) and develop brain amyloid β depositsin brain progressively with age [33]. A preliminaryreport of a portion of this work was presented at the55th Annual Meeting of the American College ofNeuropsychopharmacology [34].MethodsSelection of MAP participants: cognitive andneuropathological evaluationsThe Memory and Aging Project (MAP) recruits elderlyvolunteers (more than 1,800 since 1997) without knowndementia at enrollment, living in the metropolitan areaof Chicago (IL, USA) [9, 32]. All participants signed aninformed consent and an Anatomic Gift Act for organdonation upon death. The Institutional Review Board ofRush University Medical Center approved this study.The overall follow-up rate was 95% and the autopsy rateexceeded 80% resulting in more than 650 autopsies. Alarge number of MAP participants (n = 426) from con-secutive autopsies were used in a recent study (see [35]).From those, n = 150 participants were pseudorandomlyselected using a random sampling tool in JMP software(version 12.1; SAS Institute, NC, USA), and included inthe present study. See Table 1 for a summary of theirdemographic, cognitive (no cognitive impairment, NCI;mild cognitive impairment, MCI; clinical dementia) andpathological characteristics. Apolipoprotein E (APOE)genotyping was performed with PCR assays by Agen-court (Beckman Coulter Genomics, Brea, CA, USA).Prior reports extensively reported the methodologicalapproaches to perform systematic cognitive, clinical andneuropathological evaluations [9, 36]. Annual cognitiveevaluations included a series of 21 standard tests, 19 ofwhich were used for summary measures of episodic,semantic and working memory, perceptual speed, andvisuospatial ability, and finally summarized into onesingle variable to derive a global cognitive function score[9, 37]. The mini mental state examination (MMSE) isalso reported for comparison to other studies (seeTable 1). A board-certified neuropsychologist blind to allpathological data reviewed test results and rated the levelof cognitive impairment. A study clinician evaluated eachparticipant and diagnosed dementia and AD following theNational Institute of Neurological and CommunicativeDisorders and Stroke and the Alzheimer’s Disease andRelated Disorders Association criteria [38] implementedas described [39]. Cognitive impairment not meeting thecriteria for dementia was diagnosed as mild cognitive im-pairment (MCI) as described [40]. NCI refers to thosewithout MCI or dementia [41].Ramos-Miguel et al. Molecular Neurodegeneration  (2017) 12:26 Page 2 of 14The pathological examinations were made by a board-certified neuropathologist, blind to all clinical data. ADpathology (i.e. plaques and tangles) was evaluated informalin-fixed, paraffin-embedded sections frommultiple key brain regions in the frontal, temporal,parietal, and occipital lobes, as previously described [42],although only data from the dorsolateral prefrontal cor-tex (DLPFC) was used for statistical modeling, unlessotherwise specified. Briefly, sections from all subjectsand brain areas were assessed using both a modifiedBielschowsky silver staining for counts of diffuse andneuritic plaques, and neurofibrillary tangle (NFTs), asdescribed [43]. Immunocytochemistry with amyloid-β(clones 10D5 or 4G8) to quantify the percent area occu-pied by amyloid-β by image analysis, and phosphotau(clone AT8) antibodies – to quantify the density of tautangles by stereology [42]. The severity and/or stage ofAD in each participant was later addressed following theNational Institute on Aging (NIA)-Reagan criteria, whichincorporates the Consortium to Establish a Registry ofAlzheimer’s Disease (CERAD) scale [44], and Braakstaging [45]. Other neuropathologies, including cerebro-vascular diseases (macroscopic and microscopic infarcts,arteriolosclerosis and atherosclerosis), Lewy bodies, andhippocampal sclerosis, were also examined as describedelsewhere [9, 36]. Stereological approaches to accountfor resting, activated or total microglial cells in theDLPFC were detailed earlier [46].AnimalsAPP23 transgenic mice, overexpressing a variant ofhuman APP carrying the ‘Swedish double mutation’KM670/671NL [33], and wild-type (WT) littermateswere provided by Novartis Pharma (Basel, Switzerland)at different ages (3, 12 and 22 months old; n = 5–6 pergenotype and age). Mice were killed by decapitation, andthe frontal cortex was dissected and prepared for furtherWestern blot (WB) analysis [47].Tissue collection, immunoassays and quantification oftarget proteinsAt the time of autopsy of MAP participants, tissue slabsfrom the middle-frontal gyrus (Brodmann’s area 46/9) ofthe DLPFC were dissected following a standard humanbrain atlas [48], and stored at −80 °C. The DLPFC wasselected for its central role in complex cognitive tasksand contribution to age-related cognitive decline [49].For further immunoassays, grey matter tissue was care-fully sampled from each of the slabs avoiding thawing.Total homogenates from DLPFC samples (40–80 mg)were prepared in ice-cold PBS pH 7.4 following usualprocedures [11]. Then, protein concentrations wereTable 1 Demographic, cognitive and pathological characteristicsa of MAP participants included in the present studyVariable All participants (n = 150) NCI (n = 51) MCI (n = 42) Dementia (n = 57)DemographicFemale, no. (%) 105 (70%) 40 (78%) 27 (64%) 38 (67%)Age at death, years 88.7 ± 6.3 86.7 ± 6.7 88.1 ± 6.8 90.9 ± 4.7Education, years 14.6 ± 2.9 13.9 ± 2.5 15.0 ± 2.8 15.0 ± 3.3Race, no. W:AA 146:4 49:2 41:1 56:1APOE ε4 carrier, no. (%) 33 (22%) 7 (14%) 11 (26%) 15 (26%)PMI, hours 6.6 ± 3.8 6.2 ± 2.1 7.6 ± 5.6 6.2 ± 3.1Cognitive function proximate to deathGlobal cognition score −0.77 ± 1.01 0.10 ± 0.39 −0.47 ± 0.43 −1.80 ± 0.78MMSE 21.9 ± 8.2 27.7 ± 1.9 24.9 ± 4.6 14.4 ± 8.1PathologicalMacroinfarcts, no (%)b 45 (30%) 10 (20%) 15 (36%) 20 (35%)Lewy bodies, no (%)b 25 (17%) 4 (8%) 1 (2%) 20 (35%)Hippocampal sclerosis, no (%)b 14 (9%) 0 (0%) 4 (10%) 10 (18%)Arteriolosclerosisb 1.39 ± 0.87 1.24 ± 0.84 1.38 ± 0.82 1.53 ± 0.93Amyloid plaquesc 5.3 ± 5.9 2.9 ± 3.3 5.5 ± 6.5 7.3 ± 6.5Tanglesc 1.1 ± 3.0 0.4 ± 1.7 1.1 ± 3.0 1.8 ± 3.7Abbreviations: AA Afro-American, AD Alzheimer’s disease, MAP Memory and Aging Project, MCI mild cognitive impairment, MMSE mini mental state examination,NCI, no cognitive impairment, PMI postmortem interval, SD standard deviation, W WhiteaValues are mean ± SD unless noted otherwisebGlobal valuescValues obtained in the contralateral dorsolateral prefrontal cortex (DLPFC) by immunohistochemistry with specific antibodiesRamos-Miguel et al. Molecular Neurodegeneration  (2017) 12:26 Page 3 of 14determined by DC assay (Bio-Rad, Hercules, CA, USA)and samples were adjusted to equal concentrations withhomogenization buffer. A greater amount of a referenceMAP cortical sample (~1 g) was homogenized andprepared following the same conditions to be used as aninternal control (i.e., standard sample) in the immuno-blot assays.Cortical samples from MAP participants (40 μg) orAPP23/WT mice (10 μg) were resolved by electrophor-esis on 10% SDS–PAGE minigels (Bio-Rad Laboratories,Hercules, CA, USA). Every gel was run with 14 brainsamples, including 11 MAP participants (or 11 APP23/WT mice samples) selected randomly, and the triplicatestandard samples (the reference MAP sample or a poolof WT mice) space-loaded across the gel, and a molecu-lar weight ladder (Bio-Rad). Each cortical sample wasassessed at least three times in different gels (i.e., withrandomly allocated brain samples) on different days.Following electrophoretic separation, proteins weretransferred to nitrocellulose membranes, then incubatedovernight at 4 °C in blocking solution with anti-FADD(H-181) (1:5000; Santa Cruz Biotechnology, CA, USA)(see [50, 51] for labeling in post-mortem human braintissue). The secondary antibody was incubated for 1 h atroom temperature (1:5000 dilution; Cell Signaling).Immunoreactivity of target proteins was detected withECL reagents (Amersham, Buckinghamshire, UK) andsignal of bound antibody was visualized by exposure toautoradiographic film (Amersham ECL Hyperfilm) for 1to 60 min, then quantified by densitometric scanning(GS-800 Imaging Calibrated Densitometer, Bio-Rad).Quantification of p-FADD and ß-actin protein contentswere performed by sequentially stripping and reprobingall membranes, first for anti-phospho-Ser194 FADD(1:1000; Santa Cruz Biotechnology, CA, USA), and thenfor anti-ß-actin (clone AC-15) (1:10000; Sigma-Aldrich,MO, USA). For each sample, immunoreactivity of FADDor p-FADD was first divided by that of ß-actin (i.e., pro-tein content data normalization) within the same gel,and then calculated as a percentage of in-gel standards.The mean value for each sample from at least threeseparate gels was used as a final estimate. During theabove procedures, the experimenter was blind to thedemographic, cognitive and pathological characteristicsof MAP participants.Quantification of the presynaptic proteins syntaxin-1,synaptosomal-associated protein of 25 kDa (SNAP25),vesicle-associated membrane protein (VAMP) andsyntaxin-binding protein-1 (STXBP1) was performedpreviously by either enzyme-linked immunosorbentassay [11] or Western blotting [35], in the same brainsamples. The DLPFC or overall brain immunodensitiesof the three SNARE proteins (i.e. syntaxin-1, SNAP25,VAMP) were z-scored and averaged to obtain a variableaccounting for cortical or global synapse density, directlyrelated to synaptopathy [52].Immunofluorescence analysisIn an attempt to better characterize the potentialrelationship between FADD (i.e., cellular and anatomicallocalization), cognitive decline and AD pathology, 6MAP subjects (3 pathology-free NCI subjects and 3dementia cases with confirmed cortical AD pathology)were selected for further immunofluorescence analysis.Tissue blocks from the DLPFC (BA9) were cut coronallywith a vibrating microtome (Leica, Nussloch, Germany)to a thickness of 40 μm, and floating sections were cryo-preserved in solution at −20 °C until further use.Antigen retrieval was done in 20 mM citrate buffer(pH 6.0, 80 °C, 20 min) and was followed by blockingsections and overnight incubation at 4 °C with one ofthe following primary antibodies: anti-FADD (H-181,Santa Cruz, 1:50), anti-NeuN (clone A60) (1:250; Chemi-con, CA, USA), anti-syntaxin-1 (clone SP7) (1:1000;[53]), anti-amyloid β (clone 6 F/3D) (1:100; Dako,Glostrup, Denmark), anti-mis-folded pathologic tau(clone Alz-50) (1:500; [54]). The next day, sections wereincubated with Alexa Fluor® 488-, 555- or 647-conju-gated anti-mouse or anti-rabbit secondary antibodies(1:500; Southern Biotech, AL, USA). Auto-fluorescencewas eliminated by incubation in 0.1% Sudan Black B and70% ethanol solution for 15 min. Sections were mountedin gelatin-coated slides and protected with an anti-fademounting reagent. A series of orthogonal images werecaptured at a resolution of 1024×1024 pixels using aLSM 5 Pascal confocal microscope (Zeiss; Jena,Germany) and were visualized using a 63x/1.2 N.A.water-immersion objective. Co-staining of FADD withNeuN and syntaxin-1 was assessed to determine pres-ence in neurons and/or synapses, while co-staining ofFADD with amyloid β or mis-folded pathologic taudetermined its presence within plaques and/or tangles,respectively.Statistical analysesData were analyzed and plotted with GraphPad Prism,Version 6 (GraphPad Software, CA, USA) and/or JMP.The level of significance was p ≤ 0.05. Initial compari-sons of the cortical pathologic burden between clinicallydiagnosed groups were done by Kruskal-Wallis followedby Dunn’s test. Following WB assays, FADD andp-FADD immunodensities were normalized to corre-sponding β-actin values, and calculated as a percentageof in-gel standards (see [35]). While these values wereused for plots, later statistical modeling procedures re-quired a logarithmic transformation and standardizationin order to obtain normally distributed measures, asconfirmed by Shapiro-Wilk test. Multivariate analysesRamos-Miguel et al. Molecular Neurodegeneration  (2017) 12:26 Page 4 of 14were performed to detect potentially confounding fac-tors (e.g. demographic variables, APOE genotype,tobacco or alcohol consumption, psychotropic drug pre-scription, etc.) influencing cortical FADD/p-FADDlevels, as well as other interesting associations of thesemolecules with multiple clinical, pathologic or neuro-chemical variables measured along the study. Amongthe confounding factors, only postmortem interval(PMI) significantly correlated with FADD (r = −0.278; p< 0.001) and p-FADD (r = 0.240; p < 0.003) brain values,and therefore was included as a covariate (among othervariables) in all follow-up models. Differences in FADDand p-FADD immunodensities between clinically diag-nosed (i.e., NCI, MCI, dementia) or pathologicallygraded (i.e. CERAD scaled or Braak staged) groups wereassessed by analysis of covariance (ANCOVA), control-ling for sex, age, years of education and PMI, followedby Tukey’s HSD test. Given the potential effects of cor-tical FADD levels on cognitive performance, logistic orlinear regression models (controlled for sex, age, educa-tion, PMI, and APOE genotype) were evaluated withclinical dementia or cognitive function proximal to deathas respective outcomes, and pathological and neuro-chemical variables as predictors (see [35]). Additionally,we constructed univariate random-effects models,controlled for demographics and neuropathologies asabove, to study the potential influence of FADD corticalimmunodensites (measured postmortem) on the cogni-tive decline rates of MAP participants, as previouslydescribed [12]. Note that these models assume fixedvalues of cortical FADD levels longitudinally, a limitationthat must be considered when interpreting the results.For WB experiments involving transgenic mice, datawas analyzed with two-way ANOVA, in which genotype(WT vs. APP23) and age (3, 12 and 22 months old) weretreated as independent variables, followed by multiplet-tests for two-group comparisons at each age.ImageJ 2.0 (NIH, MA, USA) was used to determineand quantify the extent of colocalization between twoimmunofluorescent dyes in confocal imaging using anunbiased built-in method [55, 56].ResultsCharacteristics of MAP participantsDescriptive statistics for demographic, cognitive andpathological characteristics of MAP participants aresummarized in Table 1. Out of the total of 150 MAPparticipants, 51 subjects presented no cognitive impair-ment (NCI), 42 displayed mild cognitive impairment(MCI), while 57 were clinically diagnosed with dementia(see Table 1, cognitive function proximate to death). Asexpected, common AD disease pathology (i.e. amyloid-βload and tau tangle density) in the DLPFC was moreabundant in dementia cases, as compared to NCIparticipants (2.5–5.1 fold, p < 0.001) (see Fig. 1). None ofthese markers could separate MCI from NCIparticipants, while dementia cases showed greater dens-ity of phosphotau tangles than MCI participants (1.6fold, p = 0.003) (Fig. 1b). Notably, and as previouslyreported in larger epidemiologic studies [7, 9] there isample variability in all clinically diagnosed groups.Fig. 1 Measures of Alzheimer’s disease pathology in the DLPFC ofMAP participants in relation to their clinical diagnoses. Followingstandard immunohistochemical assays with antibodies against (a)amyloid β (clones 10D5 or 4G8) and (b) phosphotau (clone AT8), theacquired images were thresholded and the percent areaimmunostained was estimated for each participant and plotted byclinical diagnosis criteria into no cognitive impairment (NCI, n = 51),mild-cognitive impairment (NCI, n = 42) or dementia (DEM, n = 57).Whiskers represent 10th and 90th percentiles and boxes encloseinterquartile ranges crossed by the median of AD pathology scoreswithin groups. Differences among groups were assessed byKruskal-Wallis followed by Dunn’s multiple comparison test. **p < 0.01and ***p < 0.001Ramos-Miguel et al. Molecular Neurodegeneration  (2017) 12:26 Page 5 of 14Association of cortical FADD with common age-relatedpathologyThe total amounts of FADD protein forms (FADD andp-FADD) were quantified in the DLPFC of MAP partici-pants and normalized by β–actin protein content. Ourresearch group, which has been working with these spe-cific antibodies for over a decade, characterized FADDand p-FADD specific bands with several antibodies andenzymatic dephosphorylation assays in brain tissue [50,51, 57, 58] (see detailed revisions on these antibodiescharacterization in [59, 60]). As reviewed in [60], andshown in the present results, total FADD is mainlyimmunodetected as a 51-kDa dimeric form, while p-FADD is immunodetected as a 116-kDa oligomeric formboth in rodent and human brains. The results revealedlarge variability among MAP participants for FADDimmunodensities (interquartile range = 20–83%), andp-FADD (57–102%) as compared to that of β-actin(93–105%).Protein expression levels of FADD species did notdiffer in the DLPFC of MAP participants displayingcerebrovascular diseases (including infarcts), Lewy bod-ies, cortical atrophy, or hippocampal sclerosis from thosewho did not (data not shown). By contrast, FADD levels(but not those of p-FADD) negatively correlated withcortical amyloid-β load (r = −0.197; p = 0.038), withoutapparent association with phosphotau density (r =−0.124; p = 0.118) (Fig. 2). The association betweenFADD and plaque pathology was stronger when usingneuritic (r = −0.266; p < 0.001) and diffuse (r = −0.258; p= 0.006) plaque counts on the silver-stained sections,rather than amyloid-β immunodensities, which are moresensitive (Fig. 2). Consequently, subjects graded withdefinite AD pathology by CERAD had lower corticalFADD (but not p-FADD) immunodensities (−45%, p =0.004) compared to plaque pathology-free participants(Fig. 3a). These differences in FADD contents were notfound in association with Braak stage (Fig. 3b). Of note,the above neuropathological assessments were part ofprevious studies, where detailed information on imageprocessing and quantitative approaches was reported[36, 41–43].Additionally, greater FADD cortical immunodensity wasassociated with lower number of activated (r = −0.224; p =0.018, n = 110), but not resting or total, microglial cells(Fig. 2), as characterized and quantified previously [46].Confocal microscopy studies labeling FADD and HLA-DR-positive (activated) microglia revealed significant colo-calization between both markers within microglial pro-cesses, but not in their nuclei (Additional file 1: FigureS1). This type of overlap might be attributed to the engulf-ment of neuronal-derived apoptotic material, and there-fore activated microglia might mediate pro-apoptoticFADD clearance. By contrast, higher FADDimmunodendities, but not p-FADD, were strongly associ-ated with greater amounts of most presynaptic markersquantified in the same cortical samples in prior studies[11, 35], including syntaxin-1 (r = 0.274; p = 0.004), synap-tosomal-associated protein of 25 kDa (SNAP-25; r= 0.290;p= 0.002), vesicle-associated membrane protein (VAMP; r=0.327; p < 0.001), and syntaxin-binding protein-1 (STXBP1; r= 0.520; p < 0.001) (Fig. 2). Of note, loss of these markers isrelated to synaptic pathology in aging and AD [61]. Anindex of synapse density was estimated by averagingsyntaxin-1, SNAP-25 and VAMP immunodensities (i.e.,the so-called SNARE proteins) in order to obtain a vari-able accounting for cortical synaptopathy in MAP par-ticipants in later statistical models. As expected, thisindex also correlated with FADD values (r = 0.312; p <0.001).Association of cortical FADD with clinical dementia andcognitive functionComparing MAP participants by clinical diagnosesrevealed that FADD was lower in the DLPFC of subjectswith dementia relative to NCI (−42%, p = 0.003) andMCI (−27%, p = 0.006), while p-FADD was not different(Fig. 3c,d). Given these associations, we addressed thehypothesis that lower cortical FADD levels may contrib-ute to increased likelihood of dementia and/or greaterFig. 2 Heatmap of Pearson’s r- (bottom-left) and P- (top-right) valuesfollowing multiple pairwise correlations between the pathologic,stereological and neurochemical variables indicated. Abbreviations:Aβ, amyloid β; AD, Alzheimer’s disease; DP, diffuse plaques; NFT,neurofibrillary tangles; NP, neuritic plaques; p-, phospho-; SNAP25,synaptosomal-associated protein of 25 kDa; STX1, syntaxin-1; STXBP1,syntaxin-binding protein-1; Syn dens, synaptic density; VAMP,vesicle-associated membrane proteinRamos-Miguel et al. Molecular Neurodegeneration  (2017) 12:26 Page 6 of 14cognitive impairment, either by mediating the effects ofage-related neuropathologies or independent of these in-dices. We therefore performed a series of logistic andlinear regression models taking into account demo-graphics, multiple age-related pathologies (i.e., amyloidplaques, tangles, Lewy bodies, cerebrovascular diseases,hippocampal sclerosis), and overall synaptic density,omitting or including cortical FADD levels in the models(see Table 2 and data not shown).As expected, age and the presence of Lewy bodieswere associated with both higher odds of dementia andpoorer cognitive performance prior to death. Surpris-ingly, in the present MAP subset, cerebrovasculardiseases and hippocampal sclerosis were related withneither dementia nor cognitive function. However, inprevious studies using a much larger MAP sample, theselatter associations were indeed observed (see e.g. [12]).Interestingly, the effect of DLPFC amyloid-β load on thelikelihood of dementia, which was marginally significantwhen the FADD data was not included in the model(data not shown), was not significant after addition ofFADD suggesting a mediation effect. Thus, participantswith lower FADD cortical density displayed a signifi-cantly greater likelihood of dementia (odds ratio = 0.433,p = 0.002). For example, a MAP participant with averagedemographic and pathologic characteristics had a 2.5fold-higher likelihood of dementia if the cortical densityof FADD is in the lower quartile versus the higher quar-tile. Likewise, higher FADD levels in the DLPFC wereassociated with better global cognitive function in MAPFig. 3 Regulation of FADD (upper plots) and p-FADD (bottom plots) protein forms in the DLPFC of MAP participants ranked either by: a CERADseverity scale according to plaque pathology, and displaying no (4, n = 40), sparse (3, n = 26), moderate (2, n = 42), or frequent (1, n = 42) plaqueload; b Braak staging according to the spread of the tauopathy, and displaying no or transentorhinal deposition (0–II, n = 26), limbic spread (III–IV,n = 92), or neocortical spread (V–VI, n = 32); or (c) clinical diagnoses with no cognitive impairment (NCI, n = 51), mild-cognitive impairment (NCI,n = 42) or dementia (DEM, n = 57). Whiskers represent 10th and 90th percentiles of FADD or p-FADD values (normalized by β-actin proteincontent), with boxed interquartile ranges crossed by the median for each experimental group and expressed as percentage of an in-gel-standard.Differences among groups were assessed (after log-transformation and standardization of the datasets) by ANCOVA controlling for age, sex,education and PMI followed by Tukey’s HSD post hoc test. **p < 0.01. d Representative immunoblots of FADD, p-FADD and β-actin, with variousparticipants and standard (ST) samples. The molecular masses of the various proteins are indicated in kDa. Full gel images are included inAdditional file 2: Figure S2Ramos-Miguel et al. Molecular Neurodegeneration  (2017) 12:26 Page 7 of 14participants (β = 0.499, p = 0.008), while the DLPFC loadof amyloid-β showed a marginal but significant effect oncognitive function (β = −0.119, p = 0.046) after addingFADD to the linear regression model. Cortical FADDimmunodensity did not mediate the large effects of theDLPFC tauopathy on cognitive function; and surpris-ingly had no impact on the risk of clinical dementia inthe current MAP sample (Table 2). Importantly, aftercontrolling for the demographic and pathologic indicesin Table 2, variations in DLPFC FADD immunodensityexplained 3.44% of the total variance in cognition amongMAP participants. Of note, the associations between theclinical outcomes and the DLPFC pathologic and neuro-chemical variables reported in Table 2 were very similarto those observed when using the overall brain measure-ments of amyloid and tau pathologies (data not shown).Given that clinical diagnoses were not performed lon-gitudinally, diagnoses of NCI/MCI/dementia were basedon cognitive evaluations nearest death. Therefore, estab-lishing an association between the present neurochem-ical data and the duration of the illness was not possible.Because the onset of cognitive decline was somewhatvariable (see left panel in the Fig. 4), we performedrandom-effects models to evaluate the potential associ-ation between longitudinally ascertained cognitivedecline rates and postmortem cortical immunodensitiesof FADD, adjusting for demographics and neuropathol-ogies. Although a trend of faster cognitive decline ratewas observed for those subjects within the lower tertileof FADD cortical values (see Fig. 4), in the statisticalapproaches using univariate random-effects models(controlling for all above confounders and pathologies)this neurochemical variable only showed a marginal,non-significant association with the annually evaluatedcognitive function (β = 0.266, p = 0.080), possibly becausethe current sample size is underpowered for this type ofanalysis.Differences in FADD cortical distribution in clinicaldementiaIn DLPFC sections from subjects (n = 3) with no cogni-tive impairment (NCI) and free from age-related neuro-pathology, most FADD intensity appeared in theneuronal cell body, especially within nuclei (see mergeFADD-NeuN), and neuropil (synapses) (see Fig. 5a). Inparticular, about 65% of FADD signal intensity was colo-calized with NeuN, while 4% of FADD signal intensitywas colocalized with syntaxin-1. By contrast, in samplesfrom MAP participants with definite AD (e.g. AD1 andAD2 subjects in Fig. 5a), there was no obvious colocali-zation between FADD and NeuN labelings in neuronalbodies. Interestingly, FADD appeared to accumulate inTable 2 Regression models showing the associations of FADD levels in the DLPFC of MAP participants (and relevant covariates)with clinical dementia and cognitive functionClinical dementiaa Cognitive functionbModel terms Odds ratio 95% CI P-value Estimate SD P-valueAge at death 1.104 1.025–1.197 0.0117* −0.0293 0.0118 0.0144*Sex 1.231 0.440-3.433 0.6887 0.0856 0.0818 0.2976Education years 1.133 0.966–1.339 0.1319 −0.0278 0.0260 0.2877PMI 0.928 0.804–1.039 0.2401 −0.0052 0.0203 0.7991APOE ε4 allele 0.752 0.235–2.267 0.6195 0.0369 0.0921 0.6892Macroinfarctsc 1.295 0.498–3.328 0.5913 −0.1040 0.1616 0.5208Lewy bodiesc 9.190 2.833–35.790 0.0005* −0.5993 0.1971 0.0028*Hipp. sclerosis 3.935 0.832–21.845 0.0930 −0.1160 0.2813 0.6808Arteriolosclerosisc 1.271 0.783–2.090 0.3350 −0.1022 0.0824 0.2166Amyloid plaquesd 1.366 0.974–1.948 0.0756 −0.1186 0.0588 0.0458*Tanglesd 1.040 0.906–1.208 0.5865 −0.0887 0.0265 0.0011*Synaptic densitye 1.798 0.895–3.891 0.1151 −0.1003 0.1065 0.3475FADDf 0.433 0.248–0.717 0.0019* 0.4993 0.1840 0.0075*Abbreviations: C.I. confidence intervals, DLPFC dorsolateral prefrontal cortex, FADD Fas-associated protein with death domain, Hipp., hippocampal, MAP Memoryand Aging Project, MCI mild cognitive impairment, NCI no cognitive impairment, PMI postmortem interval, S.D., standard deviation, SNAP-25,synaptosomal-associated protein of 25 kDa, VAMP, vesicle-associated membrane proteinaLogistic regression model of the estimated odds ratios of clinical dementia vs. non-dementia (i.e. NCI and MCI participants) per unit of regressorbLinear regression model predicting global cognitive function nearest to deathcGlobal valuesdValues obtained in the contralateral DLPFC by immunohistochemistry with specific antibodieseEstimated as the mean value of calculated densities of the presynaptic proteins syntaxin-1, SNAP-25 and VAMP in the same brain samplesfValues normalized by β-actin*Statistically significant P-value < 0.05Ramos-Miguel et al. Molecular Neurodegeneration  (2017) 12:26 Page 8 of 14dystrophic neurites and tangle-like structures. Notably,the same results were observed for all 3 AD cases, whilenone of the NCI subjects presented this anomalous dis-tribution of FADD. As these results suggested a possiblerole for FADD in the mechanisms of pathologic taudeposition, DLPFC sections from subjects with AD wereused to study the possible colocalization of FADD withpathological tau (Alz-50) and/or β-amyloid. The resultsshown in Fig. 5b demonstrated the presence of FADD intangles and in dystrophic neurites where FADD coloca-lized with Alz-50 immunoreactivity. FADD was alsopresent in the dystrophic neurites accumulating aroundneuritic plaques (see Fig. 5b).Decreased cortical FADD in APP23 mice with agingGiven the observed decrease in FADD content in MAPparticipants with clinical dementia, and the potential asso-ciation with common AD pathology, we utilized APP23transgenic mice to further explore the possible role ofFADD in a common animal model of AD-like syndrome(i.e., amyloid-β plaques accumulation with age). The re-sults showed, in parallel to the human data, decreasedimmunodensities of cortical FADD (normalized by β-actincontent) in APP23 mice as measured by a two-wayANOVA (interaction Genotype x Age: F2,28 = 4.43, p =0.021). Post-hoc multiple comparisons via t-tests revealedsignificant decreases for adult (12 months old, p = 0.027)and aged (22 months old, p = 0.030) APP23 mice as com-pared to age-matched WTcontrols (Fig. 6).DiscussionPro-apoptotic FADD was lower in an elderly community-based cohort of subjects with dementia, contrary to theinitial prediction of a higher level. Interestingly, lowerFADD in the DLPFC of MAP subjects was associated withgreater amyloid-β cortical accumulation, reduced synapticdensity, microglial activation, lower cognitive function, andhigher odds of dementia. In addition, subjects with AD de-mentia presented an anomalous cortical FADD distribution(i.e., presence in tangles and in dystrophic neurites),compared to NCI subjects (i.e., FADD labeling in neur-onal bodies). The redistribution of this adaptor protein intocellular compartments where phosphotau accumulatesduring the progression of AD may suggest a possible rolefor FADD in the mechanisms of pathologic tau deposition.Finally, in a common animal model of AD (i.e., APP23mice), cortical FADD was also decreased, indicating thatFADD loss may be caused by age-related amyloid path-ology, and suggesting that this multifunctional moleculemight be a key component in the amyloid cascade.Cell death signals mediated by the extrinsic apoptoticpathway are initiated through the interaction of Fasreceptor with FADD adaptor, which promotes activationof effector caspases and leads to cell death [22, 62–64].By contrast, when FADD is phosphorylated and translo-cated to the nucleus, it mediates anti-apoptotic actions[23, 59]. Thus FADD is a molecule key in controllingcell-fate as it has demonstrated great plasticity in itsactions (see [60]). Interestingly, this study found markedreductions in FADD (i.e., pro-apoptotic form), but notp-FADD (i.e., anti-apoptotic form), in the DLPFC ofMAP participants displaying clinical dementia and/or alarge burden of AD pathology. Similar results were ob-served in a preliminary postmortem study performed inan independent small subset of patients with AD (n = 5)in which cortical (BA9) FADD was decreased (by aboutFig. 4 Cognitive decline trajectories of MAP participants ranked by cortical FADD levels. Global cognitive function was evaluated annually afterenrolment, for 5.6 years in average (range 3–12 years), and the scores were standardized as described in Methods. Participants were then alignedlongitudinally by the last cognitive test before death. Participants were divided into groups according to their tertile of FADD cortical expressionlevels. The spaghetti plot on the left represents individual cognitive trajectories across the study. The panel on the right represents individualannual scores (points), with the best fit (solid line) and the 95% confidence interval (shaded area) of the cognitive trajectories for eachFADD-ranked group overlappedRamos-Miguel et al. Molecular Neurodegeneration  (2017) 12:26 Page 9 of 14Fig. 5 Immunofluorecence characterization of FADD in the DLPFC of neuropathology-free NCI (n = 3) and Alzheimer’s diseased (AD; n = 3; twodifferent subjects shown) MAP participants. Single-channel (in greys) or merged (in RGB) confocal images correspond to triple co-immunolabeledsections with antibodies against FADD (H181, Santa-Cruz, 1:50) combined with either (a) synatxin-1 (STX1, clone SP7, locally produced, 1:1000)and NeuN (Chemicon, clone A60, 1:250), or (b) beta-amyloid (Aβ, clone 6 F/3D, Dako, 1:100) and misfolded, pathologic tau (clone Alz-50, locallyproduced, 1:500). In merged images, colors were arbitrarily assigned (as indicated at the top) to maximize overlap visualization. Overlap panels onthe right are ImageJ-generated bitmaps highlighting those pixels where significant colocalization over an unbiased threshold of intensities betweenthe indicated channels were detected in the corresponding pairwise colocalization analyses. Note the change in FADD localization in NCI (mainly inneuronal nuclei and soma, and also some neuropil staining) compared to AD (redistributed to dystrophic neurites, tangles, and within amyloidplaques) brains. Scale bars: 30 μmRamos-Miguel et al. Molecular Neurodegeneration  (2017) 12:26 Page 10 of 1469%), while p-FADD was unaltered when compared tomatched controls (n = 4) [65]. In contrast, basal fore-brain cholinergic neurons in AD brains expressed highlevels of FADD but at the same time did not containfragmented DNA, a cellular marker of apoptosis [30]. Thepresent results suggest that at the time of death, therewere no signs of pro-apoptotic activation mediated byFADD, but a decrease in its content. A possible explan-ation for FADD decrease could be that initial FADD trans-location out of the nucleus (possibly towards the neurites)could have started apoptotic cascades leading to celldeath. This massive death of cortical neurons could havefurther explained the loss of FADD in these subjects. Infact, beyond the amyloid hypothesis biological cascade,the literature suggests a distinct timeline for the increasein biomarker expression (i.e., amyloid β deposits, andhyperphosphorylation of tau) leading to cell death, andthe start-point of cognitive decline that culminates in clin-ical dementia [2]. Another possible explanation wouldfavor the neuroplastic actions of this multifunctional pro-tein suggesting an adaptive response to a prior insult.Interestingly, cognitive decline during aging is due notonly to neuronal loss, but is the result of functionalchanges occurring over time (reviewed in [4]), includingsynaptic dysfunction (or dysplasticity). According to thismodel, FADD levels correlate positively with synapticdensity markers, such as the SNARE proteins, and nega-tively with microglia activation, which is thought to be re-sponsible for greater synaptic pruning in AD [66].Following linear and logistic regression models controlledfor multiple confounders, lower FADD levels were associ-ated with increased likelihood of clinical dementia and re-duced global cognitive function, partially mediating theeffects of amyloid-β accumulation, but not those of phos-photau deposition. Interestingly, these FADD-amyloid-βmediation effects were observed in models where the out-come was clinical dementia (which compared NCI/MCIversus dementia subjects), but not in those predicting cog-nitive function. Perhaps, this type of interaction betweenFADD and amyloid-β may play a role in the transitionfrom non-dementia (either NCI or MCI) to dementia.Immunofluorescence assays were performed tocharacterize FADD at the cellular and anatomical levels,and to evaluate possible changes in FADD expressionpatterns in subjects with clinical dementia as comparedto NCI controls. Prior studies have suggested that FADDis expressed in human neurons [67, 68] as well as inneuron-enriched cultures from human brain cortex [60].Other studies have also suggested that FADD is expressedin glial cells (e.g., glioblastoma cell lines; [69]). At thesubcellular level, FADD forms are expressed (rodent andhuman brains) in cytosol and nucleus and to a lesserextent in membranes (e.g., [58]; see revision in [60]). In linewith this, the present results showed that NCI subjects ac-cumulated most FADD intensity in the neuronal cell body,especially within nuclei. In contrast, subjects with clinicaldementia presented an anomalous FADD distribution, withFADD presence in tangles and in dystrophic neurites, rep-resented by the colocalization observed between FADD andpathological tau (i.e., Alz-50 immunoreacctivity). A reportof increased FADD in AD (i.e., within the basal forebraincholinergic neurons), demonstrated FADD colocalized withphosphorylated tau immunoreactive tangles but not withdense-core amyloid β plaques [30]. These results, in linewith ours, suggest a possible role for FADD in the mecha-nisms of pathologic tau deposition.Previous observations in transgenic mouse models sug-gested that cerebral amyloidosis in APP23 mice caused aFig. 6 Decreased cortical FADD in APP23 mice with aging. aImmunodensity of FADD protein (normalized by β-actin proteincontent) were quantified by Western blotting in cortical homogenatesfrom APP23 transgenic and wild-type (WT) mice at 3, 12 and 22 monthsof age. Group of treatment: WT-3 months (n = 6), APP23-3 months(n = 6), WT-12 months (n = 6), APP23-12 months (n = 5), WT-22 months(n = 6), APP23-22 months (n = 5). Columns represent mean values ±SEM per group and expressed as percentage of an in-gel standard.Two-way ANOVA detected an interaction Genotype x Age (F2,28 = 4.43,p < 0.05). Post-hoc multiple comparison t-tests revealed significantdecreases for adult (12 months old) and aged (22 months old) APP23mice as compared to age-matched WT controls. *p < 0.05. bRepresentative immunoblots of FADD and β-actin, with one sampleper group and age. The molecular masses of the various proteins areindicated in kDa. Full gel images are included in Additional file 2:Figure S2Ramos-Miguel et al. Molecular Neurodegeneration  (2017) 12:26 Page 11 of 14modest neuronal loss in neocortex at early ages, followedby more neurons with necrotic-apoptotic phenotype inthe neocortex at 24 months of age [70]. However, to thebest of our knowledge no prior reports evaluated the regu-lation of apoptotic markers (i.e., FADD) in APP23 mice.In the present study, aged APP23 mice displayed reducedlevels of cortical FADD, suggesting FADD loss may bedependent on age-related amyloid pathology. Interestingly,cognitive decline in AD seems to correlate better withneurofibrillary tangles of hyperphosphorylated tau thanwith amyloid β plaques (see review in [4]). Interestingly,the present results suggest a functional interaction be-tween FADD and pathological tau, but at the same timeshows that FADD is sensitive to the accumulation ofamyloid β. Therefore, it is reasonable to speculate thatFADD might participate in the process of connectingthese two classical pathological markers in the progress ofclinical dementia, opening room for further studies.ConclusionsThe present results demonstrate that cortical FADD was de-creased in an elderly, community-based cohort subjects withdementia. Interestingly, loss of FADD in the DLPFC wasassociated with a greater load of amyloid pathology, loss ofpresynaptic terminal markers, poorer cognitive function andincreased risk of dementia. Moreover, subjects with ADpresented an anomalous cortical FADD distribution, (i.e.,presence in tangles and in dystrophic neurites) as comparedto NCI subjects (i.e., FADD labeling in neuronal bodies) sug-gesting a possible role for FADD in the mechanisms ofpathologic tau deposition. Moreover, the decrease inFADD content was consistent with findings in a trans-genic mouse model of AD. Overall, the present datasuggests FADD as a putative biomarker of the cognitivedecline associated with the course of clinical dementia.Future studies should investigate the precise role of thismultifunctional adaptor protein within the amyloid cas-cade, possibly linking plaque-mediated synaptotoxicityand tauopathy in AD.Additional filesAdditional file 1: Figure S1. Colocalization of FADD and HLA-DRpositive (activated) microglia in the DLPFC of neuropathology-free NCI(n = 3) MAP participants. Single-channel (in greys) or merged confocalimages correspond to double co-immunolabeled sections with antibodiesagainst FADD (H181, Santa-Cruz, 1:50; magenta) and HLA-DR (clone CR3/43,Dako, 1:100; green). In merged image, colors were arbitrarily assigned tomaximize overlap visualization. Overlap panel is an ImageJ-generatedbitmap highlighting those pixels where significant colocalization over anunbiased threshold of intensities between the indicated channels wasdetected in pairwise colocalization analyses. Unlike its neuronal localizationpattern, FADD seems absent from the microglial nuclei, and mayorcolocalization between these markers appears in activated microglial processes(see yellow arrows). Possibly, FADD microglial inclusions might derive frompost-apoptotic neurons. Scale bar: 20 μm. (PDF 292 kb)Additional file 2: Figure S2. (a) Representative full gel immunoblots ofFADD, p-FADD and ß-actin proteins in the DLPFC of MAP participants, withvarious participants and standard (ST) samples. The red square representsthe portion selected for Fig. 3d. (b) Representative full gel immunoblots ofFADD and ß-actin proteins in cortical homogenates from APP23transgenic mice. The red square represents the portion selected forFig. 5b. The apparent molecular masses of the various proteins weredetermined by calibrating the blots with prestained molecular weightmarkers as shown on the left-hand side. (PDF 2893 kb)AbbreviationsAD: Alzheimer’s disease; ANCOVA: Analysis of covariance; ANOVA: Analysis ofvariance; APOE: Apolipoprotein E; APP: Amyloid precursor protein;BA: Brodmann’s area; BSA: Bovine serum albumin; CERAD: Consortium toEstablish a Registry of Alzheimer’s Disease; CV: Coefficient of variation;DEM: Dementia; DLPFC: Dorsolateral prefrontal cortex; ECL: Enhancedchemiluminescence; FADD: Fas-associated death domain; MAP: Memory andAging Project; MCI: Mild cognitive impairment; MMSE: Mini mental stateexamination; NCI: No cognitive impairment; NFTs: Neurofibrillary tangles;NIA: National Institute on Aging; PAGE: Polyacrylamide gel electrophoresis;PBS: Phosphate-buffered saline; PMI: Postmortem interval; SD: Standarddeviation; SDS: Sodium dodecyl sulfate; SNAP-25: Synaptosomal-associatedprotein of 25 kDa; STXBP1: Syntaxin-binding protein-1; VAMP:Vesicle-associated membrane protein; WB: Western blot; WT: Wild-typeAcknowledgmentsThe authors would like to express our gratitude to all participants in MAP,and to the staff in Rush Alzheimer’s Disease Center. We also thank AntonioCrespo for his skillful technical assistance. JAG-S is a member of the Institutde Estudis Catalans (Barcelona, Catalonia, Spain). MJG-F is a ‘Ramón y Cajal’Researcher (MINECO-UIB).FundingThis study was supported by SAF2014-55903-R to MJG-F from Ministerio deEconomía y Competitividad (MINECO, Spain), and by Grants MT-14037 andMOP-81112 to WGH from the Canadian Institutes of Health Research. TheMemory and Aging Project is a collaborative, multidisciplinary and translationalresearch project subsidized by the National Institute on Aging (GrantsR01AG42210).Availability of data and materialsThe datasets used and/or analyzed during the current study are availablefrom the corresponding author on reasonable request.Author’s contributionsMJG-F, JAG-S, AR-M and WGH designed the study. MJG-F performed allcharacterization and quantification experiments of FADD protein forms inhuman brain tissues. AR-M, with the participation of AMB, quantified FADDprotein in mouse brain tissues and performed the immunofluorescence as-says in human brain sections. TAB and PF contributed to the APP23 micestudy. DAB and JAS conceived the Memory and Aging Project, performed allclinical and pathological exams, and procured human tissue samples. SELcomplied all participants’ demographic, clinical and pathological data. MJG-F,JAG-S, AR-M and WGH wrote the first draft of the manuscript. All authorscritically contributed to the discussion of the results and approved the finalversion of the manuscript.Competing interestsWGH has received consulting fees or sat on paid advisory boards for: InSilico, Lundbeck/Otsuka, Eli Lilly, and Roche. AMB is on the advisory board orreceived consulting fees from Roche Canada, and received educational grantsupport from BMS Canada. The Organizations cited above had no role in(and therefore did not influence) the design of the present study, theinterpretation of results, and/or preparation of the manuscript. All otherauthors have no financial interest on the reported data and declare that nocompeting interests exist.Consent for publicationNot applicable.Ramos-Miguel et al. Molecular Neurodegeneration  (2017) 12:26 Page 12 of 14Ethics approval and consent to participateAll participants signed an informed consent and an Anatomic Gift Act fororgan donation upon death. The Institutional Review Board of RushUniversity Medical Center approved this study.All experiments utilizing mice were performed in accordance with Germananimal protection law, aiming to minimize the number of mice used andtheir suffering.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1BC Mental Health and Addictions Research Institute, Vancouver, Canada.2Department of Psychiatry, University of British Columbia, Vancouver, Canada.3IUNICS, University of the Balearic Islands, Ctra. de Valldemossa km 7.5,E-07122 Palma de Mallorca, Spain. 4Instituto de Investigación Sanitaria deBaleares, Palma de Mallorca, Spain. 5Department of Anesthesiology,Pharmacology and Therapeutics, University of British Columbia, Vancouver,Canada. 6Department of Psychiatry, University Medicine Goettingen,Goettingen, Germany. 7Department of Psychiatry and Psychotherapy,Ludwig-Maximilians-University Munich, Munich, Germany. 8Rush Alzheimer’sDisease Center, Rush University Medical Center, Chicago, USA.Received: 15 December 2016 Accepted: 9 March 2017References1. Reitz C, Mayeux R. Alzheimer disease: epidemiology, diagnostic criteria, riskfactors and biomarkers. Biochem Pharmacol. 2014;88:640–51.2. Scheltens P, Blennow K, Breteler MM, de Strooper B, Frisoni GB, Salloway S,Van der Flier WM. Alzheimer’s disease. Lancet. 2016;388:505–17.3. Spires-Jones TL, Hyman BT. The intersection of amyloid beta and tau atsynapses in Alzheimer’s disease. Neuron. 2014;82:756–71.4. Mufson EJ, Ikonomovic MD, Counts SE, Perez SE, Malek-Ahmadi M, ScheffSW, Ginsberg SD. Molecular and cellular pathophysiology of preclinicalAlzheimer’s disease. Behav Brain Res. 2016;311:54–69.5. Katzman R, Terry R, DeTeresa R, Brown T, Davies P, Fuld P, Renbing X, PeckA. Clinical, pathological, and neurochemical changes in dementia: Asubgroup with preserved mental status and numerous neocortical plaques.Ann Neurol. 1988;23:138–44.6. Snowdon DA, Greiner LH, Mortimer JA, Riley KP, Greiner PA, Markesbery WR.Brain infarction and the clinical expression of Alzheimer disease. The NunStudy. JAMA. 1997;277:813–7.7. Bryne C, Matthews FE, Xuereb JH, Broome JC, McKenzie J, Rossi M, Ince PG,McKeith IG, Lowe J, Esiri MM, Morris JH. Pathological correlates of late-onsetdementia in a multicentre, community-based population in England andWales. Lancet. 2001;357:169–75.8. White L, Small BJ, Petrovitch H, Ross GW, Masaki K, Abbott RD, Hardman J,Davis D, Nelson J, Markesbery W. Recent clinical-pathologic research on thecauses of dementia in late life: update from the Honolulu-Asia Aging Study.J Geriatr Psychiatry Neurol. 2005;18:224–7.9. Bennett DA, Schneider JA, Buchman AS, Barnes LL, Boyle PA, Wilson RS.Overview and findings form the Rush Memory and Aging Project. CurrAlzheimer Res. 2012;9:646–63.10. Stern Y. What is cognitive reserve? Theory and research application of thereserve concept. J Int Neuropsychol Soc. 2002;8:448–60.11. Honer WG, Barr AM, Sawada K, Thornton AE, Morris MC, Leurgans SE, et al.Cognitive reserve, presynaptic proteins and dementia in the elderly. TranslPsychiatry. 2012;2:e114.12. Boyle PA, Wilson RS, Yu L, Barr AM, Honer WG, Schneider JA, et al. Much oflate life cognitive decline is not due to common neurodegenerativepathologies. Ann Neurol. 2013;74:478–89.13. Ankarcrona M, Winblad B. Biomarkers for apoptosis in Alzheimer’s disease.Int J Geriatr Psychiatry. 2005;20:101–5.14. de la Monte SM, Sohn YK, Wands JR. Correlates of p53- and Fas (CD95)-mediated apoptosis in Alzheimer’s disease. J Neurol Sci. 1997;152:73–83.15. Ferrer I, Puig B, Krupinski J, Carmona M, Blanco R. Fas and Fas ligandexpression in Alzheimer’s disease. Acta Neuropathol. 2001;102:121–31.16. Erten-Lyons D, Jacobson A, Kramer P, Grupe A, Kaye J. The FAS gene, brainvolume, and disease progression in Alzheimer’s disease. AlzheimersDement. 2010;6:118–24.17. Nishimura T, Akiyama H, Yonehara S, Kondo H, Ikeda K, Kato M, Iseki E,Kosaka K. Fas antigen expression in brains of patients with Alzheimer-typedementia. Brain Res. 1995;695:137–45.18. Masliah E, Mallory M, Alford M, Tanaka S, Hansen L. Caspase dependentDNA fragmentation might be associated with excitotoxicity in Alzheimerdisease. J Neuropathol Exp Neurol. 1998;57:1041–52.19. Engidawork E, Gulesserian T, Yoo BC, Cairns N, Lubec G. Alteration ofcaspases and apoptosis-related proteins in brains of patients withAlzheimer’s disease. Biochem Biophys Res Commun. 2001;281:84–93.20. Jellinger KA, Stadelmann C. Problems of cell death in neurodegenerationand Alzheimer’s disease. J Alzheimers Dis. 2001;3:31–40.21. Raina AK, Hochman A, Zhu X, Rottkamp CA, Nunomura A, Siedlak SL, BouxH, Castellani RJ, Perry G, Smith MA. Abortive apoptosis in Alzheimer’sdisease. Acta Neuropathol. 2001;101:305–10.22. Chinnaiyan AM, O'Rourke K, Tewari M, Dixit VM, 1995. FADD, a novel deathdomain containing protein, interacts with the death domain of Fas andinitiates apoptosis. Cell. 1995;81:505–1223. Alappat E, Feig C, Boyerinas B, Volkland J, Samuels M, Murmann AE,Thorburn A, Kidd VJ, Slaughter CA, Osborn SL, Winoto A, Tang WJ, Peter ME.Phosphorylation of FADD at serine 194 by CKIalpha regulates itsnonapoptotic activities. Mol Cell. 2005;19:321–32.24. Park SM, Schickel R, Peter ME. Nonapoptotic functions of FADD-bindingdeath receptors and their signaling molecules. Curr Opin Cell Biol.2005;17:610–6.25. García-Fuster MJ, García-Sevilla JA. Monoamine receptor agonists, actingpreferentially at presynaptic autoreceptors and heteroreceptors,downregulate the cell fate adaptor FADD in rat brain cortex.Neuropharmacology. 2015;89:204–14.26. García-Fuster MJ, García-Sevilla JA. Effects of anti-depressant treatments onFADD and p-FADD protein in rat brain cortex: enhanced anti-apoptotic p-FADD/FADD ratio after chronic desipramine and fluoxetine administration.Psychopharmacology. 2016;233:2955–71.27. Keller B, García-Sevilla JA. Regulation of hippocampal Fas receptor anddeath-inducing signaling complex after kainic acid treatment in mice. ProgNeuropsychopharmacol Biol Psychiatry. 2015;63:54–62.28. Yeh WC, de la Pompa JL, McCurrach ME, Shu HB, Elia AJ, Shahinian A, Ng M,Wakeham A, Khoo W, Mitchell K, El-Deiry WS, Lowe SW, Goeddel DV, MakTW. FADD: essential for embryo development and signaling from some, butnot all, inducers of apoptosis. Science. 1998;279:1954–8.29. Cheng W, Zhang R, Yao C, He L, Jia K, Yang B, Du P, Zhuang H, Chen J, LiuZ, Ding X, Hua Z. A critical role of Fas-associated protein with death domainphosphorylation in intracellular reactive oxygen species homeostasis andaging. Antioxid Redox Signal. 2014;21:33–45.30. Wu CK, Thal L, Pizzo D, Hansen L, Masliah E, Geula C. Apoptotic signalswithin the basal forebrain cholinergic neurons in Alzheimer’s disease. ExpNeurol. 2005;195:484–96.31. Ivins KJ, Thornton PL, Rohn TT, Cotman CW. Neuronal apoptosis induced bybeta-amyloid is mediated by caspase-8. Neurobiol Dis. 1999;6:440–9.32. Bennett DA, Schneider JA, Buchman AS, MendesdeLeon C, Bienias JL,Wilson RS. The Rush and Memory and Aging Project: study design andbaseline characteristics of the study cohort. Neuroepidemiology.2005;25:163–75.33. Sturchler-Pierrat C, Abramowski D, Duke M, Wiederhold KH, Mistl C, Rothacher S,et al. Two amyloid precursor protein transgenic mouse models with Alzheimerdisease-like pathology. Proc Natl Acad Sci U S A. 1997;94:13287–92.34. García-Fuster MJ, Ramos-Miguel A, Barr AM, Leurgans SE, Schneider JA,Bennett DA, Honer WG, García-Sevilla JA. Decreased FADD protein isassociated with clinical dementia and cognitive decline in a communitysample. 55th Annual Meeting of the American College ofNeuropsychopharmacology. Poster Session I, M4. 2016.35. Ramos-Miguel A, Hercher C, Beasley CL, Barr AM, Bayer TA, Falkai P,Leurgans SE, Schneider JA, Bennett DA, Honer WG. Loss of Munc18-1 longsplice variant in GABAergic terminals is associated with cognitive declineand increased risk of dementia in a community sample. Mol Neurodegener.2015;10:65.36. Schneider JA, Arvanitakis Z, Bang W, Bennett DA. Mixed brain pathologiesaccount for most dementia cases in community-dwelling older persons.Neurology. 2007;69:2197–204.Ramos-Miguel et al. Molecular Neurodegeneration  (2017) 12:26 Page 13 of 1437. Bennett DA, Schneider JA, Tang Y, Arnold SE, Wilson RS. The effect of socialnetworks on the relation between Alzheimer’s disease pathology and level ofcognitive function in old people: a longitudinal cohort study. Lancet Neurol.2006;5:406–12.38. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM.Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA WorkGroup under the auspices of Department of Health and Human ServicesTask Force on Alzheimer's Disease. Neurology. 1984;34:939–44.39. Bennett DA, Schneider JA, Aggarwal NT, Arvanitakis Z, Shah R, Kelly JF, FoxJH, Cochran EJ, Arends D, Treinkman A, Wilson RS. Decision rules guidingthe clinical diagnosis of Alzheimer’s disease in two community-basedcohort studies compared to standard practice in a clinic-based cohortstudy. Neuroepidemiology. 2006;27:169–76.40. Bennett DA, Wilson RS, Schneider JA, Evans DA, Beckett LA, Aggarwal NT,Barnes LL, Fox JH, Bach J. Natural history of mild cognitive impairment inolder persons. Neurology. 2002;59:198–205.41. Bennett DA, Schneider JA, Arvanitakis Z, Kelly JF, Aggarwal NT, Shah R,Wilson RS. Neuropathology of older persons without cognitive impairmentfrom two community-based studies. Neurology. 2006;66:1837–44.42. Bennett DA, Schneider JA, Wilson RS, Bienias JL, Arnold SE. Neurofibrillarytangles mediate the association of amyloid load with clinical Alzheimerdisease and level of cognitive function. Arch Neurol. 2004;61:378–84.43. Bennett DA, Wilson RS, Schneider JA, Evans DA, Aggarwal NT, Arnold SE,Cochran EJ, Berry-Kravis E, Bienias JL. Apolipoprotein E4 allele, Alzheimer’sdisease pathology, and the clinical expression of Alzheimer’s disease.Neurology. 2003;60:246–52.44. Mirra SS, Hart MN, Terry RD. Making the diagnosis of Alzheimer's disease. Aprimer for practicing pathologists. Arch Pathol Lab Med. 1993;117:132–44.45. Braak H, Braak E. Neuropathological stageing of Alzheimer-related changes.Acta Neuropathol. 1991;82:239–59.46. Bradshaw EM, Chibnik LB, Keenan BT, Ottoboni L, Raj T, Tang A, et al. CD33Alzheimer’s disease locus: altered monocyte function and amyloid biology.Nat Neurosci. 2013;16:848–50.47. Bayer TA, Schäfer S, Simons A, Kemmling A, Kamer T, Tepest R, Eckert A,Schüssel K, Eikenberg O, Sturchler-Pierrat C, Abramowski D, Staufenbiel M,Multhaup G. Dietary Cu stabilizes brain superoxide dismutase 1 activity andreduces amyloid Abeta production in APP23 transgenic mice. Proc NatlAcad Sci U S A. 2003;100:14187–92.48. Mai JK, Assheuer J, Paxinos G. Atlas of the Human Brain. 3rd ed. San Diego,CA: Academic; 1997.49. Morrison JH, Baxter MG. The ageing cortical synapse: hallmarks andimplications for cognitive decline. Nat Rev Neurosci. 2012;13:240–50.50. García-Fuster MJ, Ramos-Miguel A, Rivero G, La Harpe R, Meana JJ, García-Sevilla JA. Regulation of the extrinsic and intrinsic apoptotic pathways inthe prefrontal cortex of short- and long-term human opiate abusers.Neuroscience. 2008;157:105–19.51. García-Fuster MJ, Díez-Alarcia R, Ferrer-Alcón M, La Harpe R, Meana JJ,García-Sevilla JA. FADD adaptor and PEA-15/ERK1/2 partners in majordepression and schizophrenia postmortem brains: basal contents andeffects of psychotropic treatments. Neuroscience. 2014;277:541–51.52. Ramos-Miguel A, Sawada K, Jones AA, Thornton AE, Barr AM, Leurgans SE,et al., Presynaptic proteins complexin-I and complexin-II differentiallyinfluence cognitive function in early and late stages of Alzheimer’s disease.Acta Neuropathol. 2016 (in press) doi: 10.1007/s00401-016-1647-9.53. Honer WG, Hu L, Davies P. Human synaptic proteins with a heterogeneousdistribution in cerebellum and visual cortex. Brain Res. 1993;609:9–20.54. Wolozin BL, Pruchnicki A, Dickson DW, Davies P. A neuronal antigen in thebrains of Alzheimer patients. Science. 1986;232:648–50.55. Costes SV, Daelemans D, Cho EH, Dobbin Z, Pavlakis G, Lockett S. Automaticand quantitative measurement of protein-protein colocalization in live cells.Biophys J. 2004;86:3993–4003.56. Ramos-Miguel A, Honer WG, Boyda HN, Sawada K, Beasley CL, Procyshyn RM,et al. Exercise prevents downregulation of hippocampal presynaptic proteinsfollowing olanzapine-elicited metabolic dysregulation in rats: Distinct roles ofinhibitory and excitatory terminals. Neuroscience. 2015;301:298–311.57. García-Fuster MJ, Miralles A, García-Sevilla JA. Effects of opiate drugs on Fas-associated protein with death domain (FADD) and effector caspases in therat brain: Regulation by the ERK1/2 MAP kinase pathway.Neuropsychopharmacology. 2007;32:399–411.58. García-Fuster MJ, Ramos-Miguel A, Miralles A, García-Sevilla JA. Opioid receptoragonists enhance the phosphorylation state of Fas-associated death domain(FADD) protein in the rat brain: Functional interactions with casein kinase Iα,Gαi proteins, and ERK1/2 signaling. Neuropharmacology. 2008;55:886–99.59. Ramos-Miguel A, Álvaro-Bartolomé M, García-Fuster MJ, García-Sevilla JA.Role of multifunctional FADD (Fas-associated death domain) adaptor indrug addiction. In Addictions-From Pathophysiology to Treatment (Ed.David Belin). In Tech-Open Access Publisher. ISBN 978-953-51-0783-5. 2012.Chapter 7, pp. 201–26.60. García-Fuster MJ, Álvaro-Bartolomé M, García-Sevilla JA. The Fas receptor/Fas-associated protein and cocaine. Neuropathology of Drug Addictionsand Substance Misuse, Volume 2, Chapter 6 pp. 63–73. Editor: Dr. Victor R.Preedy. Academic Press (Elsevier). 2016.61. Honer WG. Pathology of presynaptic proteins in Alzheimer's disease: morethan simple loss of terminals. Neurobiol Aging. 2003;24:1047–62.62. Kumar S. Caspase function in programmed cell death. Cell Death Differ.2007;14:32–43.63. Sastry PS, Rao KS. Apoptosis and the nervous system. J Neurochem. 2000;74:1–20.64. Burke RE. Programmed cell death and new discoveries in the genetics ofparkinsonism. J Neurochem. 2008;104:875–90.65. García-Fuster MJ, Callado LF, Sastre M, Meana JJ, García-Sevilla JA. FADDadaptor in Alzheimer’s disease: A preliminary study utilizing postmortemhuman brains and a transgenic mouse model. 10th FENS Forum ofNeuroscience, abstract number FENS-0646. 2015.66. Hong S, Dissing-Olesen L, Stevens B. New insights on the role of microgliain synaptic pruning in health and disease. Curr Opin Neurobiol.2016;36:128–34.67. Hartmann A, Mouatt-Prigent A, Faucheux BA, Agid Y, Hirsch EC. FADD: a linkbetween TNF family receptors and caspases in Parkinson’s disease.Neurology. 2002;58:308–10.68. Bi FF, Xiao B, Hu YQ, Tian FF, Wu ZG, Ding L, Zhou XF. Expression andlocalization of Fas-associated proteins following focal cerebral ischemia inrats. Brain Res. 2008;1191:30–8.69. Tewari R, Sharma V, Koul N, Sen E. Involvement of miltefosine-mediated ERKactivation in glioma cell apoptosis through Fas regulation. J Neurochem.2008;107:616–27.70. Bondolfi L, Calhoun M, Ermini F, Kuhn HG, Wiederhold KH, Walker L,Staufenbiel M, Jucker M. Amyloid-associated neuron loss and gliogenesis inthe neocortex of amyloid precursor protein transgenic mice. J Neurosci.2002;22:515–22.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Ramos-Miguel et al. Molecular Neurodegeneration  (2017) 12:26 Page 14 of 14


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items