Altered glial expression of the cannabinoid 1 receptor in the subiculum of a mouse model of Alzheimer's disease

RESEARCH ARTICLE Al e ed glial exp ession o he cannabinoid 1 ecep o in he subiculum o a mouse model o Alzheime 's disease I zia Te adillos 1,2 | I zia Bonilla-Del Río 1,2 | Nago e Puen e 1,2 | Mai ane Se ano 1,2 | Amaia Mimenza 1,2 | Lei e Lekunbe i 1,2 | Ilazki Anau -Lusa 1,2 | Lei e Regue o 1,2 | Inmaculada Ge ikagoi ia 1,2 | Samuel Ruiz de Ma ín Es eban 3 | Cecilia J. Hilla d 4 | Ma ía T. G ande 3 | Julián Rome o 3 | Izaskun Elezga ai 1,2 | Ped o G andes 1,2 1 Depa men o Neu osciences, Facul y o Medicine and Nu sing, Uni e si y o he Basque Coun y UPV/EHU, Leioa, Spain 2 Achuca o Basque Cen e o Neu oscience, Leioa, Spain 3 Facul y o Expe imen al Sciences, Uni e sidad F ancisco de Vi o ia, Pozuelo de Ala c on, Spain 4 Depa men o Pha macology and Toxicology, Neu oscience Resea ch Cen e , Medical College o Wisconsin, Milwaukee, Wisconsin, USA Co espondence Ped o G andes, Depa men o Neu osciences, Facul y o Medicine and Nu sing, Uni e si y o he Basque Coun y UPV/EHU, Leioa, Spain. Email: [email p o ec ed]us Funding in o ma ion Eusko Jau la i za, G an /Awa d Numbe s: IT1230-19, IT1620-22; Minis e io de Ciencia e Inno aci on, G an /Awa d Numbe s: PID2019-107548RB-I00, PID2019-108992RB-I00; Resea ch and Educa ion Componen o he Ad ancing a Heal hie Wisconsin Endowmen a he Medical College o Wisconsin Abs ac The al e a ion o he endocannabinoid one usually associa es wi h changes in he exp ession and/o unc ion o he cannabinoid CB 1 ecep o . In Alzheime 's disease (AD), amyloid be a (Aβ)-con aining agg ega es induce a ch onic in lamma o y esponse leading o eac i i y o bo h mic oglia and as ocy es. Howe e , how his glial esponse impac s on he glial CB 1 ecep o exp ession in he subiculum o a mouse model o AD, a b ain egion pa icula ly a ec ed by la ge accumula ion o plaques and concomi an subcellula changes in mic oglia and as ocy es, is unknown. The CB 1 ecep o localiza ion in bo h glial cells was in es iga ed in he subiculum o male 5xFAD/CB 2EGFP/ / (AD model) and CB 2EGFP/ / mice by immuno-elec on mic oscopy. The indings e ealed ha glial CB 1 ecep o s su e ema kable changes in he AD mouse. Thus, CB 1 ecep o exp ession inc eases in eac i e mic oglia in 5xFAD/CB 2EGFP/ / , bu emains cons an in as ocy es wi h CB 1 ecep o labeling ising p opo ionally o he pe ime e o he eac i e as ocy es. No leas , he CB 1 ecep o localiza ion in mic oglial p ocesses in he subiculum o con ols and closely su ounding amyloid plaques and dys ophic neu i es o he AD model, suppo s p e- ious sugges ions o he p esence o he CB 1 ecep o in mic oglia. These indings on he co ela ion be ween glial eac i i y and he CB 1 ecep o exp ession in mic oglial cells and as ocy es, con ibu e o he unde s anding o he ole o he endocannabi- noid sys em in he pa hophysiology o Alzheime 's disease. KEYWORDS as oglia, endocannabinoid sys em, immuno-elec on mic oscopy, mic oglia, neu odegene a ion I zia Te adillos and I zia Bonilla-Del Río sha e i s au ho ship. Recei ed: 2 Augus 2022 Re ised: 23 Oc obe 2022 Accep ed: 18 No embe 2022 DOI: 10.1002/glia.24312 This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion-NonComme cial-NoDe i s License, which pe mi s use and dis ibu ion in any medium, p o ided he o iginal wo k is p ope ly ci ed, he use is non-comme cial and no modi ica ions o adap a ions a e made. © 2022 The Au ho s. GLIA published by Wiley Pe iodicals LLC. 866 Glia. 2023;71:866–879. wileyonlinelib a y.com/jou nal/glia 1|INTRODUCTION As ocy es and mic oglia ac i i y associa ed wi h ch onic in lamma ion induced by Aβ-con aining agg ega es esul s in abno mal mo phology and p oli e a ion o bo h glial cells in AD (Beni o e al., 2003;McAlpine e al., 2021; Smi e al., 2021). The po en ial o cannabinoids o a ge se e al p ocesses in ol ed in AD pa hogenesis is ega ded as a he a- peu ic s a egy (Casa ejos e al., 2013; Chen e al., 2012;Eljaschewi sch e al., 2006; Tala ico e al., 2019), in pa because he main cannabinoid CB 1 ecep o is localized in b ain egions and cells, including glia, a ec ed by he disease. We ha e p e iously es ima ed ha abou 56% o he CB 1 ecep o labeling localizes o GABAe gic e minals, 12% o glu ama e gic e minals, 6% o as ocy es, 15% o mi o- chond ia and he es o o he cells and compa men s o be de e - mined (Bonilla-Del Rίoe al.,2019;Bonilla-DelRíoe al.,2021). The exp ession o CB 1 ecep o s in as ocy es seems o be eg- ula ed by mul iple ac o s and can a y unde di e en b ain condi- ions, o example, ansien ecep o po en ial anilloid 1 knock ou mice show a signi ican dec ease in CB 1 densi y in as ocy es (Egaña-Hugue e al., 2021), acu e Δ-9- e ahyd ocannabinol expo- su e causes CB 1 inc ease in hese glial cells (Bonilla-Del Río e al., 2021), o adolescen binge d inking signi ican ly dec eases CB 1 in as ocy es in he adul b ain (Bonilla-Del Rίoe al.,2019). Al hough he impac ha as oglial dys unc ion may ha e in AD is s ill poo ly unde s ood (Smi e al., 2021; Ve kh a sky & Nede gaa d, 2018), pieces o e idence sugges ha changes in he endocannabinoid sys em occu in as ocy es nea AD lesions. Thus, as ocy es closely associa ed wi h Aβagg ega es show mo e in e - media e ilamen p o eins and hype ophy o cell bodies (Esca in e al., 2019;Smi e al.,2021). These as ocy es clea as well as deg ade Aβagg ega es and elease p o-in lamma o y molecules (Fa ina e al., 2007) which can be diminished by endocannabinoids ac ing on CB 1 ecep o s localized in hese glial cells (Me na- Lau en & Ma sicano, 2015). Also, high a y acid amide hyd olase (FAAH) le els, he main deg ading enzyme o he endocannabinoid anandamide, ha e been ound in as ocy es a ound neu i ic plaques (Aba e e al., 2021;Beni oe al.,2003). Howe e , he eal impac o he p og ession o Alzheime 's disease on he exp ession o CB 1 ecep o s in as ocy es su ounding he lesions is unknown. Mic oglia is ano he impo an playe in he pa hogenesis o Alzheime 's disease. Cannabinoids p e en Aβ-induced neu ode- gene a ion by educing mic oglial ac i i y. Bo h CB 1 and CB 2 ecep- o sexp essedinmic ogliaa ein ol edin hisac ionas hey inhibi neu oin lamma ion by p e en ing eac i e oxygen species (ROS) o ma ion and cy okines elease by hese cells (Casa ejos e al., 2013; Ma ín-Mo eno e al., 2011; Ramí ez e al., 2005; Tala ico e al., 2019). Mic oglia also elici s a signi ican inc ease in endocannabinoid p oduc ion ha , in u n, ac i a es mo e CB 1 and CB 2 ecep o s and hei signaling cascades, ampli ying he an i- in lamma o y and p o ec i e mic oglial pheno ype (Du y e al., 2021; Mecha e al., 2016). In ac , he numbe and Aβphago- cy ic capaci y o mic oglial cells dec ease in mice lacking CB 2 ecep- o s (de Ma ín e al., 2022). The e a e pieces o e idence indica ing ha mic oglia cons i u- i ely exp esses CB 1 ecep o s ha media e some o he cannabi- noid e ec s in es ing mic oglial cells (Kaplan, 2013; Na a o e al., 2018; Ribei o e al., 2013;S ella,2009;Thione al.,2018), and egula e neu oin lamma ion in a sex-dependen manne (De Meij e al., 2021). CB 1 ecep o s ha e been no iced in cul u ed mic oglia (Ca lisle e al., 2002; Facchine i e al., 2003;Klege is e al., 2003; Molina-Holgado e al., 2002; Sinha e al., 1998; S e ano e al., 1996; Waksman e al., 1999;Wal e e al.,2003), and speci ic an i-CB 1 an ibodies de ec ed some sca e ed CB 1 sig- nal in mic oglia in he hypo halamic a cua e nucleus o emales (De Meij e al., 2021). CB 1 ecep o exp ession inc eases in many in lamma o y and neu odegene a i e diseases like AD (Bisogno & Di Ma zo, 2010; Ribei o e al., 2013); howe e , e y li le is known abou he localiza- ion and exp ession o cannabinoid ecep o s in glial cells in AD. We hypo hesize in his s udy ha CB 1 ecep o exp ession in glia is al e ed in he subiculum o a mouse model o AD, a b ain egion pa - icula ly a ec ed by la ge accumula ion o plaques, as a esul o concomi an subcellula changes in mic oglia and as ocy es. Ou indings show ha CB 1 ecep o s in mic oglial cells su e ema kable modi ica ions in 5xFAD/CB 2EGFP/ / mice, a mu ine model o AD ecen ly epo ed o ha e an inc ease in CB 2 ecep o s in mic oglia ela ed o dys ophic neu i es (Ruiz de Ma ín Es eban e al., 2022), simila ly o he CB 2 ise obse ed in plaque-associa ed mic oglia (Beni o e al., 2003,2007). 2|MATERIAL AND METHODS 2.1 |E hics s a emen The p o ocols o animal ca e and use we e app o ed by he Com- mi ee o E hics o Animal Wel a e o he Uni e si y o he Basque Coun y (M20/2020/109) and we e in acco dance o he Eu opean Communi ies Council Di ec i e o Sep embe 22, 2010 (2010/63/ EU) and Spanish egula ions (Real Dec e o 53/2013, BOE 08-02-2013). E o s we e made o minimize he numbe and su - e ing o animals. 2.2 |Expe imen al animals Expe imen s we e done in 6.5–7.5-mon h-old male CB 2EGFP/ / mice, con ols and co-exp essing i e AD mu a ions (5xFAD; Oakley e al., 2006) p e iously used in ou labo a o y o he localiza ion o CB 2 ecep o s (Ruiz de Ma ín Es eban e al., 2022). The CB 2EGFP/ / mice we e gene a ed a he Genoway acili ies (Lyon, F ance) by designing a a ge ing s a egy consis ing o he inse ion o an enhanced g een luo escen p o ein (EGFP) epo e gene, p eceded by an in e nal ibosomal en y si e sequence (IRES) in o he 30 un ansla ed egion (UTR) o he mouse cn 2 gene. This esul ed in exp ession o he epo e gene (EGFP) unde he con ol o he TERRADILLOS ET AL.867 10981136, 2023, 4, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/glia.24312 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [06/11/2023]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License mouse endogenous cn 2 p omo e , and ansc ip ion o he same bicis onic mRNA as he CB 2 ecep o p o ein. In addi ion, hese mice co-exp essed 5xFAD mu a ions a he same ime. The 5xFAD mice wi h a C57BL/6J backg ound we e pu chased om Jackson Labo a- o y (Ba Ha bo , Main, USA). To ob ain he co-exp ession, 5xFAD mice we e ma ed wi h CB 2EGFP/ / mice o a leas i e gene a ions o gene a e 5xFAD/CB 2EGFP/ / mice (L opez e al., 2018). This 5xFAD model does no seem o exp ess he mu a ion ha causes he neu o i- b illa y degene a ion (Oblak e al., 2021), bu i may occu indi ec ly h ough neu onal degene a ion and Aβ1–42 deposi s. 2.3 |B ain issue p ocessing Mice we e anes he ized wi h ke amine/xylazine (100 mg/10 mg/kg body weigh , in ape i oneal injec ion) and subsequen ly pe used ansca dially a oom empe a u e (RT) h ough he le en icle. Fi s wi h phospha e bu e ed saline (PBS) 0.1 M (pH 7.4) o 20 s, and hen wi h he ixa i e solu ion composed o 4% o maldehyde, 0.2% pic ic acid and 0.1% glu a aldehyde in PBS 0.1 M (pH 7.4) o 10–15 min, wi h a ixa i e solu ion olume o 80 ml pe mouse. The b ains we e hen emo ed om he skull and pos - ixed in he ixa i e solu ion o app oxima ely 1 week a 4C. Subsequen ly, hey we e s o ed in 1:10 dilu ed ixa i e solu ion a 4C wi h 0.025% sodium azide. B ain ib o- sec ions we e cu co onally a 50 μm and s o ed wi h 1 ml o phos- pha e bu e (PB) 0.1 M (pH 7.4) wi h 0.025% sodium azide a 4C. 2.4 |Immunohis ochemis y o ligh mic oscopy B ain sec ions con aining he subiculum we e collec ed in PB 0.1 M (pH 7.4) a RT, p e-incuba ed wi h a blocking solu ion o 10% bo ine se um albumin (BSA), 0.1% sodium azide and 0.5% i on X-100 in 1x is-bu e ed saline (TBS 1x) (pH 7.4) o 30 min a RT, and incuba ed wi h he ollowing p ima y polyclonal an ibodies: guinea pig an i-CB 1 ecep o (1:100; CB 1 -GP-A 530; AB_2571593, F on ie Ins i u e Co., l d), abbi an i-ionized calcium-binding adap e molecule 1 (Iba1, 1:500; 019-19741; AB_839504, FUJIFILM Wako Pu e Chemical Co - po a ion), abbi an i-glu ama e aspa a e anspo e 1 (An i-A522 [GLAST] EAAT1, 0.3 μg/ml; Ab#314; AB_231456, kindly gi ed by P o . Niels Ch is ian Danbol , Uni e si y o Oslo). They we e p epa ed in blocking solu ion and gen ly shaken o 2 days a 4C o 1 day a RT. Then, sec ions we e washed wi h 1% BSA and 0.5% i on X-100 in TBS 1x o 30 min, and incuba ed wi h a bio inyla ed an i- guinea pig (1:200, Bio in-SP-A iniPu e Goa An i-Guinea Pig IgG; AB_2337394, Jackson Immuno Resea ch), o bio inyla ed an i- abbi seconda y an ibody (1:200, Bio in-SP-A iniPu e Donkey An i-Rabbi IgG; AB_2340593, Jackson Immuno Resea ch) dilu ed in he washing solu ion o 1 h on a shake a RT. They we e washed wi h 1% BSA and 0.5% i on X-100 in TBS 1x (30 min). Tissue was incuba ed wi h he a idin-bio in pe oxidase complex (ABC; 1:50, Eli e, Vec o Labo a- o ies, Bu lingame, CA, USA) p epa ed in he washing solu ion, o 1 h a RT. Samples we e washed wi h 1% BSA and 0.5% i on X-100 in TBS 1x (3 x 1 min) and las ly wi h PB 0.1 M (pH 7.4) and 0.5% i on X-100 (2 x 10 min). Labeling was e ealed wi h 0.05% diaminobenzi- dine (DAB) in PB 0.1 M (pH 7.4) con aining 0.5% i on X-100 and 0.01% hyd ogen pe oxide o 3.5 min a RT. This was ollowed by washes in PB 0.1 M (pH 7.4) wi h 0.5% i on X-100 (3 x 1 min, 2 x 10 min). Tissue sec ions we e moun ed on gela inized slides, d ied and dehyd a ed in g aded e hanol o 5 min each. A e insing wi h xylene (3 x 5 min), he slides we e co e slipped wi h DPX. The subi- culum was examined and pho og aphed wi h a Zeiss AxioCam ligh mic oscope coupled o a Zeiss AxioCam HRc came a. 2.5 |Immunohis ochemis y o elec on mic oscopy The p o ocol is al eady published (Puen e e al., 2019). Fou o i e sec- ions pe b ain con aining he subiculum we e selec ed. They we e p e- incuba ed in a blocking solu ion (1 ml/well) o 10% BSA, 0.02% saponin and 0.1% sodium azide in TBS 1x (pH 7.4), o 30 min on he shake a RT. Tissue was hen incuba ed wi h a p ima y guinea pig an i-CB 1 ecep o an ibody (1:100) in combina ion wi h a abbi an i-GLAST an i- body (0.3 μg/ml) o a abbi an i-Iba1 an ibody (1:500). The solu ion con ained 10% BSA in TBS 1x, 0.1% sodium azide and 0.004% saponin. Incuba ion was pe o med on a shake o 2 days a 4C ollowedby washes in 1% BSA/TBS 1x. Then, sec ions we e incuba ed wi h 1.4 nm gold-conjuga ed goa an i-guinea pig IgG an ibody (Fab agmen , 1:100, #2055, Nanop obes, Inc., Yaphank, NY, USA). They we e also incuba ed wi h bio inyla ed an i- abbi IgG an ibody (1:200) dilu ed in 1% BSA/TBS 1x wi h 0.004% saponin on a shake o 4 h a RT. Tissue was washed in 1% BSA/TBS 1x on a shake a RT and incuba ed wi h ABC (1:50) p epa ed in washing solu ion o 1.5 h a RT. Sec ions we e washed in 1% BSA/TBS 1x, kep o e nigh a 4C and pos - ixed wi h 1% glu a aldehyde in TBS 1x (1 ml/well) o 12 min a RT. Then, hey we e washed in double dis illed wa e and gold pa icles we e sil e - in ensi ied wi h he HQ Sil e ki (#2012, Nanop obes, Inc., Yaphank, NY, USA) in he da k o 12 min a RT. A e in ensi ica ion, he sec- ions we e washed i s in double dis illed wa e and hen wi h PB 0.1 M (pH 7.4) o 30 min. The bio inyla ed an ibody was e ealed wi h 0.05% DAB p epa ed in PB 0.1 M (pH 7.4) con aining 0.5% i on X-100 and 0.01% hyd ogen pe oxide o 3.5 min a RT, ollowed by se e al washes in PB 0.1 M (pH 7.4). They we e osmica ed (1% osmium e oxide in PB 0.1 M, pH 7.4) in he da k o 20 min, washed in PB 0.1 M (pH 7.4), dehyd a ed in g aded e hanol, clea ed in p opylene oxide, p e-embedded in 1:1 p opylene oxide/Epon 812 esin on a shake o e - nigh a RT and embedded in pu e Epon 812 esin. Immunogold labeling was isualized wi h a ligh mic oscope in sec ions con aining he subicu- lum,and issuepo ionswi hgoodandconsis en CB 1 ecep o labeling we e iden i ied and immed down o ul a hin sec ioning. The p oce- du e has al eady been desc ibed in de ail (Bonilla-Del Rίoe al.,2019; Bonilla-Del Río e al., 2021; Gu ié ez-Rod íguez e al., 2018; Puen e e al., 2019). Th ee o ou semi- hin sec ions (0.7 μm- hick) we e ob ained wi h a his o-diamond kni e (Dia ome USA) and s ained wi h 1% oluidine blue. To u he s anda dize he condi ions, only he 868 TERRADILLOS ET AL. 10981136, 2023, 4, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/glia.24312 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [06/11/2023]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License i s i e ul a hin sec ions we e cu (50 nm- hick) wi h an ul a- diamond kni e (Dia ome USA), collec ed on o nickel mesh g ids and coun e s ained wi h 2.5% lead ci a e o 20 min a RT. Elec on mic og aphs we e andomly aken wi h a Hamama su FLASH digi al came a inse ed in a ansmission elec on mic oscope (JEOL JEM 1400 Plus). Sampling was always ca e ully and accu a ely done using he ol- lowing ana omical coo dina es o delimi he subiculum: in e au al, 0.40/0.00 mm; b egma, 3.40/3.80 mm (F anklin & Paxinos, 2008). 2.6 |An ibodies speci ici y Expe imen s we e always pe o med unde he same condi ions. In addi ion, nega i e con ols omi ing he p ima y an ibodies we e done. Fu he mo e, he CB 1 ecep o an ibody was es ed in CB 1 /b ain issue (Ma sicano e al., 2002) by double p e-embedding immunogold (CB 1 ) and immunope oxidase (Iba1 o GLAST) me hod o elec on mic oscopy (Figu e 1). The an i-A522 (EAAT1 [GLAST]) an ibody (Ab#314) a ge ing he C- e minal esidues 522–541 o a EAAT1 (Hu e al., 2020) was used o iden i y as ocy ic compa men s. GLAST was es ic ed o as ocy es and localized in acellula ly wi h no de ec able labeling in ne e e minals, as p e iously desc ibed (Leh e e al., 1995; Schmi e al., 1997). Fu he mo e, GLAST dis ibu ion is e y simila in oden s and humans (Li e al., 2012), so he use o an i-GLAST an ibody is a good app oach o label as ocy es in he AD model. The speci ici y o he Iba1 an ibody has been con i med in p e- ious s udies (Delcamb e e al., 2016; Szabo & Gulya, 2013). GLAST can be exp essed in bo h mic oglia and as ocy es unde ce ain condi ions (Bescho ne e al., 2007). Expe imen s we e conduc ed o igu e his ou . S ikingly, GLAST and Iba1 we e no seen o co- localize in a subicula a ea o 3506 μm 2 analyzed in CB 2EGFP/ / and o 4743 μm 2 in 5xFAD/CB 2EGFP/ / (da a no shown). Taken oge he , we could easonably conclude ha GLAST and Iba1 a e selec i e ma ke s o as ocy e and mic oglia, espec i ely, in he 6.5–7.5 mon h-old- CB 2EGFP/ / and 5xFAD/CB 2EGFP/ / mice s udied. 2.7 |Quan i a i e and s a is ical assessmen To ensu e homogeneous labeling be ween all samples, only he i s 1.5 μm om he sec ion su ace o each specimen we e conside ed o he analysis. A ea, pe ime e , numbe o p ocesses and CB 1 ecep- o exp ession in as ocy es, we e s udied in 5596 μm 2 o i e CB 2EGFP/ / , and in 7681 μm 2 o se en 5xFAD/CB 2EGFP/ / mice. In addi ion, he a ea, pe ime e and numbe o mic oglial p ocesses we e measu ed in 7900 μm 2 o se en CB 2EGFP/ / , and in 10,793 μm 2 o 10 5xFAD/CB 2EGFP/ / mice. Fo he s udy o CB 1 ecep o s in mic o- glia, 6345 μm 2 in i e CB 2EGFP/ / and 7268 μm 2 in se en 5xFAD/ CB 2EGFP/ / mice, we e analyzed. CB 1 ecep o labeling in as ocy es and mic oglia was assessed in GLAST- and Iba1-immunoposi i e p ocesses, espec i ely. The p opo - ion o cell compa men s wi h CB 1 ecep o signal was hen abula ed. Posi i e labeling was conside ed when a leas one immunopa icle was wi hin 30 nm o he memb ane s udied. CB 1 ecep o densi y (pa i- cles/μm memb ane) was also de e mined by coun ing gold pa icles in he posi i e compa men s. Memb ane leng h (pe ime e ) was mea- su ed wi h he Image-J so wa e (NIH; SCR_003070). All alues we e gi en as mean ± S.E.M. using a s a is ical so wa e package (G aphPad P ism 8, SCR_002798, G aphPad So wa e, Inc., San Diego, USA). The no mali y es (Kolmogo o –Smi no no mali y es ) was always applied be o e s a is ics was done. Da a we e ana- lyzed by non-pa ame ic o pa ame ic es s: Mann–Whi ney U es o S uden 's Unpai ed - es (*p< .05). Mino con as and b igh ness adjus men s we e made o he ig- u es using Adobe Pho oshop (Adobe Pho oshop, SCR_014199, CS3, Adobe Sys ems, San Jose, CA, USA) and Gimp (GNU Image Manipula- ion P og am, SCR_003182). 3|RESULTS 3.1 |Glial mo phology in he subiculum o CB 2EGFP/ / and 5xFAD/CB 2EGFP/ / mice Changes in s aining densi y we e de ec ed in mic oglial and as ocy ic cells (Figu e 2a–d). Mic oglia iden i ied by Iba1 in CB 2EGFP/ / FIGURE 1 Subiculum o CB 1 knock ou mice (CB 1 /). Double immunogold and immunope oxidase me hod o elec on mic oscopy. An ibodies we e es ed in CB 1 /mice (a–d). Simul aneous labeling o CB 1 (gold) and Iba1 (a and b; DAB immunodeposi s in b own) o GLAST (c and d; DAB in pink). No CB 1 ecep o signal is obse ed. Scale ba s: 50 nm TERRADILLOS ET AL.869 10981136, 2023, 4, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/glia.24312 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [06/11/2023]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License (Figu e 2a,a1) occupied a much la ge a ea in 5xFAD/CB 2EGFP/ / (Figu e 2b,b1). This inc ease seemed o ela e o bo h he numbe o mic oglial cells and he hickness o hei p ocesses (Figu e 2a1,b1). In he elec on mic oscope, only sca e ed mic oglial p ocesses we e obse ed in con ols (Figu e 3a,b), while nume ous Iba1-posi i e p o- cesses su ounding plaques (Figu e 3c, ) and dys ophic neu i es (Figu e 3c1,c3- ) we e seen in 5xFAD/CB 2EGFP/ / mice. The a ea o he mic oglial p ocesses was signi ican ly inc eased in 5xFAD/ CB 2EGFP/ / (0.3229 ± 0.05282 μm 2 ; ***p< .0001) ela i e o CB 2EGFP/ / (0.1000 ± 0.01754 μm 2 ;Figu e5a), as i was he o al mic oglial a ea pe sample no malized o 100 μm 2 (5xFAD/CB 2EGFP/ / : 2.074 ± 0.5156 μm 2 ; ***p< .0001; CB 2EGFP/ / : 0.3485 ± 0.06955 μm 2 ; Figu e 5b). In addi ion, a signi ican inc ease in he pe ime e o he mic oglial p olonga ions was de ec ed in 5xFAD/CB 2EGFP/ / (2.200 ± 0.1248 μm; ***p< .0001) e sus CB 2EGFP/ / (1.260 ± 0.1036 μm; Figu e 5c) and e lec ed in he pe ime e o he o al mic oglial p ocesses pe sample no malized o 100 μm 2 (15.73 ± 1.929 μmin5xFAD/ CB 2EGFP/ / ; 4.334 ± 0.5345 μminCB 2EGFP/ / ; ***p< .0001; Figu e 5d). Finally, signi ican changes we e also no iced in he numbe o mic oglial p ocesses (7.260 ± 0.6304/100 μm 2 in 5xFAD/CB 2EGFP/ / ; 3.413 ± 0.4092/100 μm 2 in CB 2EGFP/ / ;***p< .0001; Figu e 5e). As o as ocy es, GLAST s aining seen in CB 2EGFP/ / (Figu e 2c,c1) was mo e in ense in 5xFAD/CB 2EGFP/ / (Figu e 2d,d1). By using GLAST- DAB, he a ea, pe ime e and numbe o as ocy ic elemen s we e ana- lyzed in he elec on mic oscope (Figu e 4). As ocy ic p ocesses su - ounded dys ophic neu i es and plaques in 5xFAD/CB 2EGFP/ / (Figu e 4b–d) and showed a signi ican inc ease in hei a ea (5xFAD/ CB 2EGFP/ / : 0.2598 ± 0.01853 μm 2 ;CB 2EGFP/ / : 0.1565 ± 0.006515 μm 2 ; ***p< .0001; Figu e 5a). Howe e , no di e ences we e obse ed in he o al a ea pe sample occupied by as ocy ic p ocesses no malized o 100 μm 2 (5xFAD/CB2 EGFP/ / : 8.993 ± 0.8664 μm 2 ;CB2 EGFP/ / : 7.415 ± 0.6552 μm 2 ;p: .1711; Figu e 5b). The e was also a g ea inc ease in he pe ime e o he as ocy ic p ocesses (5xFAD/CB 2EGFP/ / : 2.833 ± 0.08486 μm; CB 2EGFP/ / : 2.116 ± 0.04741 μm; ***p< .0001; Figu e 5c). Ne e heless, no di e ences in he o al pe ime e o as ocy ic p ocesses pe samplewe ede ec ed(5xFAD/CB 2EGFP/ / : 99.89 ± 8.087 μm; CB 2EGFP/ / : 100.1 ± 6.811 μm; p: .9820; Figu e 5d). Consis en wi h hese esul s, a signi ican dec ease in he numbe o as ocy ic p ocesses in 5xFAD/CB 2EGFP/ / (35.52 ± 2.661 pe 100 μm 2 ) ela i e oCB 2EGFP/ / (47.33 ± 2.709 pe 100 μm 2 ) was no iced (**p: .0036; Figu e 5e). O e all, ou analyses in he subiculum o he Alzheime 's model e eal ha he la ge a ea o he mic oglia co ela es wi h an inc ease in size and numbe o hei p ocesses, while as ocy ic p ojec ions a e ewe bu bigge (Figu e 5a–e). 3.2 |CB 1 ecep o s in CB 2EGFP/ / and 5XFAD/ CB 2EGFP/ / subiculum The CB 1 ecep o labeling obse ed in CB 2EGFP/ / (Figu e 2e,e1) was mo e pa chy in 5xFAD/CB 2EGFP/ / showing many delimi ed ound a eas wi h much lowe o insigni ican s aining, p obably co espond- ing o neu i ic plaques su ounded by a neu opil wi h mo e CB 1 ecep- o immuno eac i i y (Figu e 2 , 1). FIGURE 2 Subiculum o CB 2EGFP/ / and 5xFAD/CB 2EGFP/ / mice showing Iba1, GLAST and CB 1 immunos aining. A idin-bio in pe oxidase me hod o ligh mic oscopy. Iba1 in CB 2EGFP/ / (a, a1) is d as ically inc eased in 5xFAD/CB 2EGFP/ / mic oglia (b, b1). GLAST in CB 2EGFP/ / (c, c1) is weake han in 5xFAD/CB 2EGFP/ / (d, d1). Dense CB 1 ecep o s aining in CB 2EGFP/ / (e, e1) changes o a weake and mo e pa chy appea ance in 5xFAD/CB 2EGFP/ / ( , 1). Scale ba s: 200 μm(a– ) and 50 μm (a1– 1) 870 TERRADILLOS ET AL. 10981136, 2023, 4, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/glia.24312 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [06/11/2023]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License 3.3 |Mic oglial CB 1 ecep o localiza ion in CB 2EGFP/ / and 5XFAD/CB 2EGFP/ / subiculum CB 1 ecep o pa icles we e localized o memb anes o Iba1-posi i e mic oglial p ocesses in bo h CB 2EGFP/ / (Figu e 3b2) and 5xFAD/ CB 2EGFP/ / mice (Figu e 3c– ). The analysis e ealed a signi ican inc ease in CB 1 -posi i e mic oglial p ocesses in 5xFAD/CB 2EGFP/ / (0.9942 ± 0.1259 CB 1+ p ocesses/100 μm 2 ) compa ed o CB 2EGFP/ / (0.3254 ± 0.07758 CB 1+ p ocesses/100 μm 2 ; ***p< .0001; Figu e 6a, le ). In addi ion, a s ike inc ease in he p opo ion o CB 1 -posi i e FIGURE 3 Double p e-embedding immunogold (CB 1 ) and immunope oxidase (Iba1) me hod o elec on mic oscopy in he subiculum o CB 2EGFP/ / and 5xFAD/CB 2EGFP/ / mice. In CB 2EGFP/ / , a ew slende Iba1 posi i e mic oglial elemen s a e obse ed (a, b) (DAB immunodeposi s in b own). Howe e , hick p ocesses o eac i e mic oglia appea in 5xFAD/CB 2EGFP/ / (c– ) su ounding plaques (in pu ple; c, ) and dys ophic neu i es (in u quoise; c– ). No ice memb ane CB 1 pa icles (o ange a ows) in mic oglial p ocesses o CB 2EGFP/ / (b2) and 5xFAD/CB 2EGFP/ / (c– ), wi h pa icula abundance in 5xFAD/CB 2EGFP/ / .CB 1 ecep o labeling is also in memb anes o exci a o y e minals (g een a ows and p o iles in a1, b2 and d1), inhibi o y e minals (yellow a ows and p o iles in b1, d2) and mi ochond ia (blue a ows and p o iles in a1 and d2), in bo h mu an s. Scale ba s: 2 μm TERRADILLOS ET AL.871 10981136, 2023, 4, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/glia.24312 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [06/11/2023]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License elemen s was de ec ed in 5xFAD/CB 2EGFP/ / mice (6.27 ± 1.15%; CB 2EGFP/ / : 3.79 ± 2.10%; **p: .0033; Figu e 6b, le ). Howe e , CB 1 ecep o densi y in he posi i e mic oglial p ocesses was signi ican ly educed in 5xFAD/CB 2EGFP/ / (69.44 ± 7577 pa icles/100 μmo memb ane; CB 2EGFP/ / : 135.5 ± 24.78/100 μm o memb ane; **p: .0023; Figu e 6c, le ). Fu he mo e, signi ican di e ences in he o al numbe o mic oglial CB 1 pa icles pe 100 μm 2 we e obse ed be ween 5xFAD/CB 2EGFP/ / (1.31 ± 0.18 pa icles) and CB 2EGFP/ / FIGURE 4 Double p e-embedding immunogold (CB 1 ecep o ) and immunope oxidase (GLAST) me hod o elec on mic oscopy in he subiculum o CB 2EGFP/ / and 5xFAD/CB 2EGFP/ / mice. GLAST-posi i e as ocy ic p ocesses (DAB immunodeposi s in pink) seen in CB 2EGFP/ / (a) a e hicke in 5xFAD/CB 2EGFP/ / (b–d). Obse e as ocy ic elemen s su ounding dys ophic neu i es ( u quoise in b–d) and close o a plaque (pu ple in d) in he 5xFAD/CB 2EGFP/ / .CB 1 pa icles ( ed a ows) localize o plasma memb anes o GLAST-posi i e as ocy ic p ocesses in CB 2EGFP/ / (a) and 5xFAD/CB 2EGFP/ / (b–d). Typical CB 1 ecep o labeling is also obse ed in memb anes o exci a o y e minals (g een a ows and p o iles in a2, b1), inhibi o y e minals (yellow a ows and p o iles in a2, b2) and mi ochond ia (blue a ows and p o iles in a2), in bo h CB 2EGFP/ / and 5xFAD/CB 2EGFP/ / . Scale ba s: 2 μm 872 TERRADILLOS ET AL. 10981136, 2023, 4, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/glia.24312 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [06/11/2023]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License (0.48 ± 0.13 pa icles; ***p: .0009; Figu e 6d, le ). Finally, he e we e no di e ences in he numbe o CB 1 pa icles pe posi i e mic oglial p ocess be ween bo h mu an s (5xFAD/CB 2EGFP/ / : 1.333 ± 0.08347 pa icles/p ocess; CB 2EGFP/ / : 1.450 ± 0.2112 pa icles/p ocess; p: .7736; Figu e 6e, le ). 3.4 |As oglial CB 1 ecep o localiza ion in CB 2EGFP/ / and 5XFAD/CB 2EGFP/ / subiculum The CB 1 ecep o was localized o memb anes o GLAST-posi i e as ocy ic p ocesses in bo h mu an s (Figu e 4a–d), as p e iously epo ed (Bonilla-Del Rίo e al., 2019; Bonilla-Del Río e al., 2021; Bosie e al., 2013; Gu ié ez-Rod íguez e al., 2018; Han e al., 2012). No signi ican di e ences we e de ec ed in he numbe o CB 1 - posi i e as ocy ic p olonga ions be ween bo h mice (CB 2EGFP/ / : 8.661 ± 0.8977 CB 1+ p ocesses/100 μm 2 ;5xFAD/CB 2EGFP/ / :7.967± 1.224 CB 1+ p ocesses/100 μm 2 ;p: .3094; Figu e 6a, igh ). Likewise, he pe cen age o CB 1 -posi i e as ocy ic b anches was s a is ically simila be ween 5xFAD/CB 2EGFP/ / (21.24 ± 2.37%) and CB 2EGFP/ / (17.75 ± 1.21%; p: .2303; Figu e 6b, igh ). The e was nei he di e - ences in CB 1 ecep o densi y in he as ocy ic posi i e p ocesses (5xFAD/CB 2EGFP/ / : 29.15 ± 2.220 pa icles/100 μmo memb ane; CB 2EGFP/ / : 37.57 ± 2.970 pa icles/100 μm o memb ane; p: .2209; Figu e 6c, igh ), no in he numbe o as ocy ic CB 1 pa icles pe 100 μm 2 (5xFAD/CB 2EGFP/ / : 12.78 ± 2.174 pa icles; CB 2EGFP/ / : 11.63 ± 1.265; p: .6716; Figu e 6d, igh ). Howe e , he numbe o CB 1 pa icles pe posi i e as ocy ic p ocesses was signi ican ly highe in 5xFAD/CB 2EGFP/ / (1.603 ± 0.05081 pa icles/p ocess) han in CB 2EGFP/ / (1.343 ± 0.03909 pa icles/p ocess; ***p: .0005; Figu e 6e, igh ). Al oge he , he numbe o mic oglial p ocesses exp essing CB 1 ecep o s inc eases and he la ge as ocy ic p o iles ha e mo e CB 1 ecep o s in he subiculum o he Alzheime 's mouse model (Figu e 6a–e). FIGURE 5 Mo phological pa ame e s o mic oglia and as ocy es in he subiculum o CB 2EGFP/ / and 5xFAD/ CB 2EGFP/ / mice. (a) Mic oglial and as ocy ic p ocesses a ea. (b) Glial a ea (mic oglia and as ocy es) no malized o 100 μm 2 . (c) Mic oglial and as ocy ic p ocesses pe ime e . (d) Glial pe ime e (mic oglia and as ocy es) no malized o 100 μm 2 . (e) Numbe o glial p ocesses (mic oglia and as ocy es) in 100 μm 2 . Da a we e analyzed by non-pa ame ic o pa ame ic es s (Mann–Whi ney U- es o S uden 's - es ). Mann–Whi ney U- es o S uden 's - es . p< .05*; p< .01**; p< .001***; p< .0001****. All da a a e ep esen ed as mean ± SEM TERRADILLOS ET AL.873 10981136, 2023, 4, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/glia.24312 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [06/11/2023]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License 4|DISCUSSION We de ec ed in he elec on mic oscope he p esence o plaques and a mul i ude o dys ophic neu i es in he 5xFAD/CB 2EGFP/ / mice ha con i ms he use ulness o his animal model o s udying he pa hophysiology o AD (Ruiz de Ma ín Es eban e al., 2022). We also obse ed an o e mic oglial and as ocy ic eac i i y wi h an inc ease in he a ea and pe ime e o hei p ocesses, demon- s a ing he exis ence o signi ican al e a ions in he subicula cy oa chi ec u e. Then we s udied he exp ession o he majo can- nabinoid CB 1 ecep o in glial cells in he subiculum o 5xFAD/ CB 2EGFP/ / and CB 2EGFP/ / mice. The main indings we e ha CB 1 ecep o exp ession conspicuously changes in mic oglial cells bu ecep o densi y emains s eady in as ocy es despi e he eac i i y o he as ocy ic p ocesses in he AD mouse. No leas , he localiza- ion o CB 1 ecep o s in mic oglial p ocesses in he subiculum o con ols and closely su ounding amyloid plaques and dys ophic neu i es in he subiculum o he AD model, suppo s he p esence o CB 1 in mic oglia. The disc ee amoun o CB 1 ecep o s in as o- cy es has been e ealed accu a ely by immuno-elec on mic oscopy (Bonilla-Del Río e al., 2021; Gu ié ez-Rod íguez e al., 2018; Puen e e al., 2019), a echnique ha has also been p o en in his s udy o be op imal o he localiza ion o CB 1 ecep o s in mic oglia. FIGURE 6 S a is ical assessmen o he CB 1 ecep o localiza ion in subicula as ocy es and mic oglia o CB 2EGFP/ / and 5xFAD/CB 2EGFP/ / mice. (a) Numbe o mic oglial (le ) and as ocy ic ( igh ) CB 1 posi i e p ocesses pe 100 μm 2 . (b) Pe cen age o CB 1 posi i e mic oglial (le ) and as ocy ic ( igh ) p ocesses. (c) CB 1 densi y in posi i e mic oglial (le ) and as ocy ic ( igh ) elemen s pe 100 μm. (d) Mic oglial (le ) and as ocy ic ( igh ) CB 1 pa icles pe 100 μm 2 . (e) CB 1 ecep o labeling pe mic oglial (le ) and as ocy ic ( igh ) p ocess. Da a we e analyzed by non- pa ame ic o pa ame ic es s (Mann–Whi ney U- es o S uden 's - es ). Mann–Whi ney U- es o S uden 's - es . p< .05*; p< .01**; p< .001***; p< .0001****. All da a a e ep esen ed as mean ± SEM 874 TERRADILLOS ET AL. 10981136, 2023, 4, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/glia.24312 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [06/11/2023]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License