Received: 3 June 2021 | Revised: 9 February 2022 | Accepted: 20 February 2022 DOI: 10.1111/acel.13581 RESEARCH ARTICLE AAV-­mediated expression of secreted and transmembrane αKlotho isoforms rescues relevant aging hallmarks in senescent SAMP8 mice J. Roig-­Soriano1 | C. Griñán-­Ferré2 | J. F. Espinosa-­Parrilla1 | C. R. Abraham3 | A. Bosch1,4,5,6 | M. Pallàs2 | Miguel Chillón1,4,5,7 1 Institut de Neurociènces (INc), Department of Biochemistry and Molecular Biology, Universitat Autònoma Barcelona, Bellaterra, Spain 2 Pharmacology Section, Department of Pharmacology, Toxicology, and Therapeutic Chemistry, Faculty of Pharmacy and Food Sciences, Institut de Neurosciències-­Universitat de Barcelona (NeuroUB), Barcelona, Spain 3 Department of Pharmacology and Experimental Therapeutics, Boston University School of Medicine, Boston, Massachusetts, USA 4 Vall d'Hebron Institut de Recerca (VHIR), Barcelona, Spain 5 Unitat producció de Vectors (UPV), Universitat Autònoma Barcelona, Bellaterra, Spain 6 Centro de Investigación Biomédica en Red sobre Enfermedades Neurodegenerativas (CIBERNED), Instituto de Salud Carlos III, Madrid, Spain 7 Institució Catalana de Recerca i Estudis Avançats (ICREA), Passeig Lluis Companys, Barcelona, Spain Correspondence Miguel Chillón, INc, Universitat Autònoma Abstract Barcelona, Spain. Senescence represents a stage in life associated with elevated incidence of morbidity Email: miguel.chillon@uab.es and increased risk of mortality due to the accumulation of molecular alterations and Funding information tissue dysfunction, promoting a decrease in the organism's protective systems. Thus, Generalitat de Catalunya, Grant/ Award Number: 2017 SGR 106; aging presents molecular and biological hallmarks, which include chronic inflammation, Agència de Gestió d'Ajuts Universitaris epigenetic alterations, neuronal dysfunction, and worsening of physical status. In this i de Recerca, Grant/Award Number: 2018FI_B00657; Ministerio Ciencia context, we explored the AAV9-­mediated expression of the two main isoforms of the Innovación, Govierno de España, Grant/ aging-­protective factor Klotho (KL) as a strategy to prevent these general age-­related Award Number: PID2019-­ 104034RB-­ I00; Ministerio de Economía, Industria features using the senescence-­accelerated mouse prone 8 (SAMP8) model. Both se- y Competitividad, Gobierno de creted and transmembrane KL isoforms improved cognitive performance, physical España, Grant/Award Number: PCIN-­ 2015-­229, PID2020-­116735RB-­I 00/ state parameters, and different molecular variables associated with aging. Epigenetic AEI/10.13039/501100011033 and landscape was recovered for the analyzed global markers DNA methylation (5-­mC), SAF2016-­33307 hydroxymethylation (5-­hmC), and restoration occurred in the acetylation levels of H3 and H4. Gene expression of pro-­ and anti-­inflammatory mediators in central nervous system such as TNF-­α and IL-­10, respectively, had improved levels, which were com- parable to the senescence-­accelerated-­mouse resistant 1 (SAMR1) healthy control. Additionally, this improvement in neuroinflammation was supported by changes in the histological markers Iba1, GFAP, and SA β-­gal. Furthermore, bone tissue structural variables, especially altered during senescence, recovered in SAMP8 mice to SAMR1 control values after treatment with both KL isoforms. This work presents evidence J. Roig-­Soriano and C. Griñán-­Ferré contributed equally to this work. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2022 The Authors. Aging Cell published by Anatomical Society and John Wiley & Sons Ltd. Aging Cell. 2022;21:e13581.  wileyonlinelibrary.com/journal/acel | 1 of 15 https://doi.org/10.1111/acel.13581 2 of 15 | ROIG-­SORIANO et al. of the beneficial pleiotropic role of Klotho as an anti-­aging therapy as well as new specific functions of the KL isoforms for the epigenetic regulation and aged bone structure alteration in an aging mouse model. KEYWORDS AAV9, anti-­aging, epigenetics, Klotho, neurodegeneration, osteoporosis, SAMP8, senescence 1 | I NTRO D U C TI O N previously mentioned aging hallmarks. Among them, the chronokine Klotho (KL) is an exciting candidate due to its pleiotropic anti-­aging Due to the increasing proportion of the aged population in modern protection (Abraham et al., 2016; Dubal et al., 2015; Kuro-­o et al., societies, research in senescence hallmarks, and pleiotropic meth- 1997). KL is involved in pathways that drive age-­related chronic ods to reverse them are a very active field of research (Lopez-­Otin disorders such as kidney disease, tissue dysfunction, neurode- et al., 2013). Senescence is understood as a gradual degenerative generative diseases, and cancer, among others (Moos et al., 2020; process that after adulthood leads to tissue dysfunction due to the Semba et al., 2014). The Klotho gene is predominantly expressed in deregulation of critical cellular events such as epigenetic mecha- the kidney and choroid plexus in the brain as two main transcripts nisms, mitochondrial function, oxidative stress (OS), inflammation, (Shiraki-­Iida et al., 1998). Full-­length mRNA translates into a single-­ intercellular communication, among others (Delgado-­Morales et al., pass transmembrane protein (m-­KL), composed of two extracellular 2017; McHugh et al., 2018). Those alterations are due to both ge- domains (KL1 and KL2). The extracellular domains can be released netic and environmental factors, which generate long-­term accumu- from the membrane by protease-­ mediated shedding, generating lative damage and increase vulnerability to death (Delgado-­Morales soluble, circulating processed Klotho (p-­KL) (Chen et al., 2014). The et al., 2017; Lopez-­Otin et al., 2013). alternatively spliced mRNA possesses a premature STOP codon and One of the most critically affected systems is the epigenetic-­ generates a secreted protein (s-­KL) containing just the KL1 domain mediated gene regulation, a well-­known hallmark of aging that leads and an additional 15 amino acids at its C-­terminus (Shiraki-­Iida et al., to genomic instability and gene expression deregulation (Delgado-­ 1998). Specific functions of s-­KL isoform have not been extensively Morales et al., 2017). Epigenetic modulation includes diverse molecu- studied. However, our previous studies indicate that s-­KL is active lar processes, including DNA methylation (5-­mC), hydroxymethylation and possesses potent anti-­aging properties in the CNS (Masso et al., (5-­hmC), histone modifications, and non-­coding RNA (Grinan-­Ferre 2018). et al., 2018). It is well established that variations in global DNA methyl- Among other functions, KL isoforms are key regulators of Ca+2 ation (5-­mC) and hydroxymethylation (5-­hmC) epigenetic patterns can and PO 4−3 ion mineral homeostasis (Chen et al., 2018) and generate influence gene expression and alter neuronal function, promoting the protection against oxidative and inflammatory stress (Yamamoto progression of age-­related chronic and neurodegenerative diseases et al., 2005), both pathways being involved in health and senescence in mouse models and humans (Coppieters et al., 2014; Grinan-­Ferre protection (Lopez-­Otin et al., 2013). Interestingly, several studies et al., 2018). Similarly, the inflammatory response is also exacerbated have demonstrated that high levels of KL are associated with en- during aging, resulting in chronic damage in the central nervous sys- hanced cognition (Dubal et al., 2015; Masso et al., 2018), health tem (CNS) (Kinney et al., 2018). Cellular alterations such as microglio- status, and longevity in both mice and humans (Arking et al., 2002; sis and astrocytosis, as well as dysregulation of inflammatory markers Kurosu et al., 2005). However, the KL promoter is susceptible to age-­ such as tumor necrosis factor-­alpha (TNF-­α), interleukin-­1beta (IL-­1β), dependent methylation, reducing its expression during aging pro- or interleukin-­6 (IL-­6), among others, have also been described during gression (King et al., 2012). This hypermethylation of the promoter normal aging (Zoller et al., 2018). These changes, together with the makes the recovery of KL expression a thought-­provoking strategy previously mentioned deregulations, directly affect cognition and for addressing age-­associated conditions. memory formation (Grinan-­Ferre et al., 2016, 2018). The senescence-­accelerated prone mouse 8 (SAMP8) and its Tissue degeneration is another feature associated with aging. healthy control, the senescence-­ accelerated mouse resistant 1 Skeletal tissue is especially altered during this period, characterized (SAMR1), are well-­established murine models for the study of age-­ by a progressive loss in bone mineral density (BMD) and alterations related cognitive decline (Akiguchi et al., 2017). SAMP8 mice start in the shape and structure of bone components (Ferguson et al., developing age-­related deficits since they are 2-­month-­old, becom- 2003; Riggs et al., 2004). All these changes result in reduced bone ing clearly different from SAMR1 animals at the age of 5 months. quality, including a reduction in bone strength and flexibility, leading Features seen during SAMP8 mice aging resemble human senes- to an increased prevalence of fractures (Ammann & Rizzoli, 2003; cence progression, including the appearance of memory and cogni- Silva et al., 2002). tive deficits (Grinan-­Ferre et al., 2018), elevated reactive oxidative Besides stress factors that could enhance senescence, some species (ROS) levels (Zhou et al., 2018), altered gene expression molecules with anti-­aging properties can decelerate or even rescue due to epigenetic control deregulation (Grinan-­Ferre et al., 2016), ROIG-­SORIANO et al. | 3 of 15 osteoporosis (Chen et al., 2009), and histological features associated All animals went through the following tests allowing one day of re- with neuronal damage and inflammation (Akiguchi et al., 2017). covery between tests. Specifically, the present study assessed the physical, behavior, and cognitive conditions after the individual expression of the two main KL transcripts in SAMP8 mice to rise exogenously KL’s protec- 2.3 | Horizontal wooden bar tive properties in this mouse model. To this end, we analyzed age-­ related features, including cognitive and physical decline, epigenetic A circular wooden bar with 1 cm of diameter was horizontally markers, histological state, and long bone integrity. Furthermore, placed 40 cm above a soft floor made of expanded polystyrene. to reveal isoform-­specific properties in aged mice, we separately Mice were carefully suspended on the bar by their upper limbs, a expressed the two main KL transcripts allowing for the first time a maximum test length of 40 s. Time spent on the bar (resistance) comparative analysis between them. and distance walked along the bar (coordination) were recorded. Two trials per animal were done, and the best performance was selected. 2 | E X PE R I M E NTA L PRO C E D U R E S 2.1 | Animal housing 2.4 | Open field test (OFT) SAMR1 (n = 9) and SAMP8 (n = 29) male mice (7-­month-­old) were The OFT was performed as previously described (Archer, 1973). The used to perform behavioral and molecular analyses. The animals floor was divided into two areas defined as the center and peripheral were divided randomly into four groups: SAMR1 Control (SR1 Null) zone. Behavior was scored with SMART® ver.3.0 software (Panlab), (n = 9), SAMP8 Control (SP8 Null) (n = 9), both injected with and each trial was recorded for later analysis using a camera situated AAV9 Null, SAMP8 injected with AAV9 s-­KL (SP8 s-­KL) (n = 11), above the apparatus. Mice were placed at the center and allowed and SAMP8 injected with AAV9 m-­KL (SP8 m-­KL) (n = 9). Klotho to explore the white polywood box (50 × 50 × 25 cm) for 5 min. constructs (s-­KL and m-­KL) derived from the Klotho mouse gene. Afterward, the mice were returned to their home cages, and the OFT Mice had free access to food and water and were kept under stand- apparatus was cleaned with 70% ethanol (EtOH). The parameters ard temperature conditions (22±2°C) and a 12-­h light/dark cycle scored included center staying duration, rearings, defecations, uri- (300 lux/0 lux). We worked with 7-­month-­old mice because com- nations, and the distance travelled, calculated as the sum of global pared with SAMR1 mice, SAMP8 mice show clear age-­related defi- distance travelled in the open field arena for 5 min. cits resembling human senescence progression. Animals were deeply anesthetized by intraperitoneal injection of 10 mg/kg of ketamine (Imalgene 500, Rhone-­Merieux) and 1 mg/kg of xylazine (Rompun, 2.5 | Object location test (OLT) Bayer) diluted in NaCl 0.9%. Intracerebroventricular stereotaxic in- jections of AAV vectors were performed in the right hemisphere at The object location test (OLT) is a well-­established task based on coordinates, −0.2 mm Antero-­posterior, −2 mm Dorso-­ventral, and the spontaneous tendency of rodents to spend more time explor- +1 mm medio-­lateral from bregma. The vector dose was 1 × 1011 ing a novel object location than a familiar object location, as well viral genomes per animal in 6 μl, administered at a 0.5 μl/min speed as to recognize when an object has been relocated. The test was using an ultramicropump (WorldPrecision Instruments). carried out for 3 days in a wooden box (50 × 50 × 25 cm), in which Adeno-­associated viruses (AAV) serotype 9 were generated in three walls were white except one that was black. The first day, the HEK293 cells by the triple transfection method at the Unitat de box was empty, and the animals just habituated to the open field Producció de Vectors (UPV) at Universitat Autònoma de Barcelona arena for 10 min. The second day, two objects were placed in front following a protocol described previously (Piedra et al., 2015). of the black wall, equidistant from each other and the wall. The All experimental procedures involving animals were performed objects were 10-­cm high and identical. The animals were placed following standard ethical guidelines of the European Communities into the open field arena and allowed to explore both objects and Council Directive 86/609/EEC and by the Institutional Animal Care surroundings, for 10 min. Afterward, animals were returned to and Use Committee of the University of Barcelona (670/14/8102, their home cages, and the OLT apparatus was cleaned with 70% approved at 11/14/2014) and by Generalitat de Catalunya (10291, ethanol. On the third day, one object was moved in front of the approved 1/28/2018). white wall to test the spatial memory. Trials were recorded using a camera mounted above the open field area, and the total explora- tion time was determined by scoring the amount of time (seconds) 2.2 | Behavioral tests spent sniffing the object in the new location (TN) and the object in the old location (TO). To evaluate the cognitive performance, Before the behavioral tests procedure, brief examination was done the discrimination index (DI) was calculated, which is defined as to ensure animals did not presented vision or locomotor problems. (TN − TO)/(TN + TO). 4 of 15 | ROIG-­SORIANO et al. 2.6 | Novel object recognition test (NORT) Advanced cDNA Synthesis Kit (Bio-­rad). Gene expression was ana- lyzed by real-­time quantitative PCR (RT-­qPCR) on a Bio-­Rad CFX-­ In brief, mice were placed in a 90°, two-­arm, (25 × 20 × 5 cm) black 384 PCR machine at the Analysis and Photodocumentation Service maze. The walls could be removed for easy cleaning. Light intensity in of the Universitat Autonòma de Barcelona. Each reaction contained mid-­field was 30 lux. Before the memory test, a 3-­day-­habituation was 25 ng of cDNA, 7.5 μl of iTaqTM Universal SYBR Green Supermix performed in which the mice were placed individually in the apparatus (Bio-­Rad), and a primer concentration of 0.2 nM, with a final reaction for 10 min. On Day 4, the familiarization phase took place, in which volume of 15 μl. Primers used are listed in Table S2. the animals were placed in the maze in the presence of two identical, The analysis of qPCR data was done following the ΔΔCt method. novel objects (A + A) or (B + B) located at the end of each maze. During Cycle threshold (Ct) were normalized subtracting from each experi- this 10-­min acquisition trial, the mice were allowed to explore the two mental Ct, the difference of its housekeeping (HK) Ct in regard to the identical objects. After 2 and 24 h from that trial, the animals were HK’s average value. HK genes used were β-­actin and GAPDH. Melting submitted to a 10-­min retention trial in which one of the two objects curves were also analyzed to ensure unique amplificon generation. was replaced by a novel one. The behavior of the animals was recorded Each sample (n = 6 per group) was tested at least in duplicates, and during the 2 and 24 h retention trials using a camera fixed to the ceiling Cts higher than 37 cycles were considered as not amplified. and situated above the apparatus. The amount of object exploration was defined as sniffing or touching the objects with nose and/or fore- paws. The time spent exploring the new object (TN) and the time spent 2.10 | Global epigenetic changes analysis exploring the old one (TO) were measured manually, and the discrimi- nation index (DI) was calculated as (TN − TO)/(TN + TO) (Ennaceur Isolation of genomic DNA from the hippocampus tissue (n = 3 per & Delacour, 1988). For the elimination of olfactory cues, 70% ethanol group) was conducted using the FitAmpTM Blood and Cultured Cell was used to clean the removable arms and objects after each trial. DNA Extraction Kit according to the manufacturer's instructions. Then, Methylflash Methylated DNA Quantification Kit (Epigentek) and MethylFlash HydroxyMethylated DNA Quantification Kit were used 2.7 | Elevated plus maze (EPM) in order to detect epigenetic markers. Briefly, these kits are based on specific antibody detection of 5-­mC and 5-­hmC residues, which trig- The elevated plus maze (EPM) consisted of two open arms ger an ELISA-­like reaction that allows colorimetric quantification by (30 × 5 × 15 cm); two enclosed arms (30 × 5 × 15 cm) positioned 40 cm reading absorbance at 450 nm using a Microplate Photometer. The above the floor. The junction of four arms formed a central square plat- absolute amount of methylated or hydroxymethylated DNA (propor- form (5 × 5 cm). Each animal was placed on the central platform facing tional to the Optical Density [OD] intensity) was measured and quanti- one of the open arms and was allowed to explore freely for 5 min. The fied using a standard curve plotting OD values vs. five serial dilutions behavior parameters evaluated were the number of entries in the open of a control methylated and hydroxymethylated DNA (0.5–­10 ng). arms and the percentage of time spent in the open and closed arms, time spent in the center zone, rearings, defecations, urination, among others, scored with SMART® vers.3.0 software (Panlab). 2.11 | Global histone acetylation H3 and H4 quantification 2.8 | Brain processing Histone extracts from the hippocampus tissue (n = 3 per group) were prepared by using a total histone extraction kit (Epigentek) ac- Animals were euthanized by cervical dislocation 3 days after the last cording to the manufacturer's protocol. Detection of global histone behavioral test. Brains were immediately removed from the skull. H3/H4 acetylation levels was measured using the EpiQuik™ global The hippocampus and cortex were then isolated and frozen in pow- histone H3/H4 acetylation assay kit (Epigentek). Briefly, histone dered dry ice. They were maintained at −80°C until use for gene proteins (1–­2 μg) were added to the strip wells. Acetylated histone expression and epigenetic changes analysis described below. H3/H4 was detected with a high-­affinity antibody, and the ratios and amounts of acetylated histone H3/H4 were displayed with a horseradish peroxidase-­conjugated secondary antibody color devel- 2.9 | RNA extraction and gene expression analysis opment system. The color was measured by reading absorbance at 450 nm using a Microplate Photometer. Total RNA isolation was carried out using TRIsureTM reagent follow- ing the manufacturer's instructions (Bioline Reagent). The liver, hip- pocampus, or cortex samples were homogenized using TissueLyser 2.12 | Histology LT sample disruption apparatus (QIAGEN). RNA quantity and pu- rity were measured with NanoDrop™ 1000 Spectrophotometer Animals were perfused by intracardiac injection with cold PBS solu- (Thermo Scientific). RNA retrotranscription was done using iScript™ tion followed with a solution of PFA 4%. After fixation, brains were ROIG-­SORIANO et al. | 5 of 15 dissected and submerged in PFA 4% for 24 h. Samples were washed 1272 (Bruker) computerized microtomography imaging system using PBS and then incubated in 30% sucrose at 4°C for cryoprotec- at the Centre de Recerca en Ciència i Enginyeria Multiescala de tion. Brains were embedded in OCT (Tissue-­Tek) and frozen on dry Barcelona (CRCEMB) at Universitat Politècnica de Catalunya ice. Slices of 10 μm thickness were obtained with a cryostat Leica (UPC). CM3050S (Leica). Images were reconstructed with the NRecon v1.6 (Bruker) Staining procedures consisted of permeabilization with 0.3% program and analyzed with the CT-­Analyser v1.13 image program Tris-­Tween buffer solution for 20 min or with methanol (Panreac) (Bruker). Cortical bone analysis was done with 100 slices by manu- for 10 s, followed by 1h blocking with 0.5% bovine serum albumin ally selecting volumes of interest (VOIs), with a binary threshold of (BSA) in TBST. Primary antibody incubation was done at 4°C over- 50–­255. Analyzed image intervals started from peronei insertion to- night (ON). Next, three rounds of 10 min long wash with TBST at ward upper tibial epiphysis. For trabecular bone analysis, 150 slices room temperature (RT) were done, followed by 1h incubation at RT were analyzed with a binary threshold of 20–­255. All selected inter- with the secondary antibody. Samples were subsequently washed val of images started 100 VOIs from the femoral distal growth plate with other three rounds of 10 min and a 5 min long incubation with toward tibial diaphysis. Finally, 3D representations of the bones 1:500 Hoescht dilution to stain nuclei. Finally, slides were washed were obtained with the CTVox v3.3 program (Bruker). and mounted with Fluoromount (Thermo Scientific). A list of anti- Mineral density was calculated with the CT-­Analyzer v1.13 pro- bodies used is shown in Table S3. gram calibrating bone absorbance with two 2mm diameter hydroxy- Images were obtained with an optical epifluorescence micro- apatite phantoms (Bruker-­MicroCT) of known density of 0.25 and scope Nikon Eclipse 90i or confocal laser scanning microscope (Zeiss 0.75 g/cm3. LSM 700). Pictures were analyzed using ImageJ software to obtain optical density (OD), corrected total cell fluorescence (CTCF), cell number, and mean cell size values. 2.14 | Statistical analysis Statistical analysis and graphic representation were done with 2.13 | Bone study GraphPad Prism ver.6 (GraphPad Software). Statistical differences between groups were analyzed with a two-­t ailed unpaired Student's Right legs of perfused animals were isolated and placed in a PFA t-­test when comparing two groups, and one-­way analysis of vari- 4% solution for preservation. Tibias were dissected, muscle tis- ance (ANOVA), followed by Tukey as a post-­hoc analysis when it sue was removed, and bones placed in PBS solution with 0.05% was necessary. Data are expressed as mean ± standard error of the sodium azide (NaN3). MicroCT analysis was done with a SkyScan mean (SEM). Statistical difference was accepted when p values were (a) (b) (d) (c) F I G U R E 1 (a) αKL gene structure, transcription and AAV9 constructs. (b) Experimental design. (c) qPCR analysis of KL isoforms expression levels in the liver (Lv), hippocampus (Hc), and cortex (Cx) of animals treated with AAV9 Null, s-­KL or m-­KL. Data represented as fold change compared to the SP8 Null group. Mean ± standard error of the mean (SEM), n = 6; *p < 0.05; **p < 0.01; ***p < 0.001. (d) Histological analysis of treated brains stained with KL (1:100) (green) and nuclei stained with Hoechst (1:500) (blue). Scale bar for immunohistochemical images is 100 μm (first row) and 50 μm (second row) 6 of 15 | ROIG-­SORIANO et al. (a) (b) (c) (d) (e) (f) (g) F I G U R E 2 (a) Body weight represented as the % weight change per week. Statistical analysis was done with the area under the curve value. (b) Percentage of the time spent in open/closed arms in the elevated plus maze test (EPM). Time spent in the central zone of OF presented as seconds (s). (c) Mean horizontal activity during OF test. (d) Coordination and strength of treated mice measured with the horizontal wooden bar test. (e) Moved object discrimination index measured in OLT. (f) Discrimination index of the new object in NORT. The first graph represents short-­term memory and second long-­term memory establishment. (g) Polygonal graph presenting a summary of relevant behavioral parameters. Mean ± standard error of the mean (SEM), n = 9–­11; &p = 0.05 *p < 0.05; **p < 0.01, ***p < 0.001, ****p < 0.0001 ROIG-­SORIANO et al. | 7 of 15 TA B L E 1 Parameters measured in the Open Field Test (OFT) SR1 Null SP8 Null SP8 s-­KL SP8 m-­KL ** # Locomotor activity (cm) 1122.81 ± 43.01 945.66 ± 35.65 1059.80 ± 39.91 962.39 ± 47.53 Time in zone-­Center (s) 20.76 ± 2.38 15.49 ± 4.07 13.62 ± 4.17 21.36 ± 3.70 Time in zone-­Periphery (s) 279.24 ± 2.38 284.51 ± 4.07 286.38 ± 4.17 278.64 ± 3.70 Rearings (n) 21.78 ± 2.73 23.89 ± 3.11 29.36 ± 2.75 25.11 ± 3.55 Groomings (n) 2.33 ± 0.50* 1.00 ± 0.17 1.18 ± 0.18 1.44 ± 0.18 Defecations (n) 2.00 ± 0.58 1.89 ± 0.42 1.27 ± 0.38 1.11 ± 0.45 Urinations (n) 0.22 ± 0.15 0.33 ± 0.17 0.45 ± 0.16 0.11 ± 0.11 Note: n, a number of events. Results are expressed as a mean ± standard error of the mean (SEM). *p < 0.05.; **p < 0.01 versus SR1 Null. #p = 0.05 versus SP8 Null. TA B L E 2 Parameters measured in the Elevated Plus Maze Test (EPM) SR1 Null SP8 Null SP8 s-­KL SP8 m-­KL Total distance (cm) 1023.72 ± 75.82 1052.37 ± 64.41 1048.05 ± 60.01 1041.68 ± 49.37 *** Time in center zone (% test) 26.25 ± 3.40 43.53 ± 2.45 38.49 ± 1.82 37.86 ± 1.56 Time in open arms (% test) 7.18 ± 2.70 16.40 ± 4.32 19.87 ± 3.15 19.47 ± 2.56 *** Time in closed arms (% test) 66.56 ± 5.64 40.06 ± 2.65 41.64 ± 2.74 42.67 ± 3.02 Rearings (n) 15.33 ± 1.64 13.11 ± 0.98 13.00 ± 1.09 12.78 ± 1.05 *** Defecations (n) 1.78 ± 0.40 0.00 ± 0.00 0.27 ± 0.19 0.00 ± 0.00 Urinations (n) 0.33 ± 0.17 0.00 ± 0.00 0.09 ± 0.09 0.11 ± 0.11 Note: n, a number of events. Results are expressed as a mean ± standard error of the mean (SEM). ***p < 0.001 versus SR1 Null. ≤0.05, and outliers were detected by Grubb's test and removed from 3.2 | KL expression improved physical the analysis. condition and cognitive performance without changes in age-­related behaviors in treated SAMP8 mice 3 | R E S U LT S The effect of AAV-­mediated KL isoforms expression on the physi- 3.1 | AAV administration efficiently expressed KL cal and cognitive state was evaluated by analyzing body weight, in SAMP8 animals behavioral indicators, and activity parameters. Briefly, a 5% body weight increase was detected in SAMP8 animals during the experi- Three different adeno-­associated virus (AAV) vectors were gener- ment length. In contrast, both control SAMR1 (SR1) and s-­KL treated ated containing the sequence of the full-­length transmembrane animals maintained the average body weight constant (Figure 2a). Klotho, the secreted isoform and a Null control vector. AAV9 se- No other significant differences in anxiety indicators were ob- rotype in known to extravasate to bloodstream after intra-­CNS served in the elevated plus maze (EPM) or the open field test (OFT) injection (Schuster et al., 2014); so, it was the serotype selected (Figure 2b) (Tables 1 and 2). Physical tests indicated a significant for its capacity to efficiently transduce neurons but also liver increase in horizontal activity of SAMP8 s-­KL treated group com- cells (Figure 1a). Intracerebroventricular administration was done pared with SAMP8 Null in the OFT (Figure 2c), as well as a significant in 7-­m onth-­old mice, and a therapeutic period of 10 weeks was increase in resistance and partial recovery in coordination in hori- set, finishing the experiment when animals were 10-­m onth-­old zontal wooden bar test after KL treatments (Figure 2d). After two (Figure 1b). The effectiveness of AAV9-­m ediated expression of months of treatment, memory was also analyzed using object loca- m-­K L and s-­K L was evaluated by RNA analysis from different tis- tion test (OLT) and novel object recognition test (NORT). A partial sues. As seen in Figure 1c, both AAV9 s-­K L and m-­K L mediated but significant improvement in OLT performance after KL treatment strong αKlotho expression two months after AAV injection in the was observed (Figure 2e). Notably, both s-­KL and m-­KL treatments different areas analyzed. Vector expression was associated with increased mice capacity to recognize the new object during NORT. an increase in KL protein in the hippocampus (Figure 1d), area of These differences were statistically significant when compared with particular interest due to its importance in learning and memory the obtained discrimination index (Figure 2f). Finally, differences formation. in physical, behavioral, and cognitive parameters among control 8 of 15 | ROIG-­SORIANO et al. SAMR1 and SAMP8 and SAMP8 KL treated mice were depicted in a in ten-­eleven translocase family (TETs) (Figure 3d). On the contrary, we polygonal graph (Figure 2g). found increased histone H3 (Figure 3e) and H4 (Figure 3f) acetyla- tion levels in both s-­KL and m-­KL SAMP8 groups in comparison with decreased levels in SAMP8 Null group, without significant changes 3.3 | KL treatment rescues global epigenetic in histone deacetylase enzymes (Hdac) 1–­5, despite a strong tendency landscape in SAMP8 mice that was observed for Hdac6 (Figure 3g). Dysregulation of global epigenetic markers provokes genomic in- stability and alterations in gene expression. This effect is increased 3.4 | KL treatment affects during aging as well, as seen in models resembling the aging phe- histological and biochemical markers of notypes. As expected, SAMP8 Null animals had global hypometh- SAMP8 animals ylation and increased hydroxymethylation. Remarkably, both s-­KL and m-­KL treatments rescued the global 5-­mC (Figure 3a) and 5-­ Effects of KL treatment in the brain were studied both by histol- hmC (Figure 3b) levels of SAMP8 mice to control SAMR1 values. ogy and gene expression. Secreted and transmembrane isoforms In contrast, those changes modestly correlate with modification presented similar effects in the histological variables studied. As in the transcription of chromatin-­modifying enzymes involved in shown in Figure 4a, the number of the inflammation marker Ionized these epigenetic markers. This is the case of the DNA (cytosine-­5)-­ calcium-­binding adaptor molecule 1 (Iba1) expressing cells was lower methyltransferase family (DNMTs), where we only find a significantly (or showed a tendency) in all the areas analyzed in SAMP8 mice reduced Dnmt1 expression in m-­KL treated animals. (Figure 3c). compared with control SAMR1. Interestingly, both s-­KL and m-­KL Furthermore, no significant changes of gene expression were found treatments statistically increased the size of Iba1+ cells and tended (a) (b) (c) (d) (e) (f) (g) F I G U R E 3 Global DNA methylation (a) and hydroxymethylation (b) levels measured using antibodies-­based colorimetric methods. The results are presented as the mean relative content of analyzed markers. Gene expression of enzymes involved in DNA methylation (c) and hydroxymethylation (d) expressed as fold change compared with SP8 Null. Global acetylation levels of histones H3 (e) and H4 (f), presented as % of SP8 Null acetylation levels. (g) Gene expression of enzymes involved in histone deacetylation expressed as fold change compared with SP8 Null. Mean ± standard error of the mean (SEM), n = 3–­6. &p = 0.052; *p < 0.05 **p < 0.01 F I G U R E 4 Immunofluorescence analysis of brain slices Iba1 of the Cortex (CX), CA1, CA3, and dentate gyrus (DG). (a) Scale bar for images 50 μm. Value measured as mean number of cells per field (first graph) and mean cell size in μm2 (second graph). Mean ± standard error of the mean (SEM), n = 3. (b) SA β-­gal activity staining of CX, CA1, CA2, and subiculum (SUB). Scale bar for images 50 μm. The value measured as optical density (OD) per image (arbitrary unit). Mean ± Standard error of the mean (SEM), n = 3. qPCR analysis of gene expression of inflammatory (c) and senescence markers (d) in the hippocampus area. Mean ± standard error of the mean (SEM), n = 5–­6; *p < 0.05 ROIG-­SORIANO et al. | 9 of 15 (a) (b) (c) (d) 10 of 15 | ROIG-­SORIANO et al. F I G U R E 5 Bone microstructure analysis using MicroCT. (a) 3D Cortical bone representation of control and Klotho-­treated mice. (b) 3D Trabecular bone representation of tibial epiphysis. Scale bar = 0.5 mm TA B L E 3 2D Structural analysis of tibial cortical and trabecular bone using MicroCT SR1 Null SP8 Null SP8 s-­KL SP8 m-­KL a b Periosteal perimeter (mm) 6.34 ± 0.24 6.65 ± 0.10 6.11 ± 0.40 6.35 ± 0.20 2 a b Cortical area (mm ) 0.75 ± 0.05 0.87 ± 0.08 0.75 ± 0.06 0.76 ± 0.06b Cortical thickness 0.24 ± 0.01 0.25 ± 0.01 0.25 ± 0.01 0.24 ± 0.02 a b Endocortical perimeter (mm) 2.15 ± 0.10 2.28 ± 0.05 2.00 ± 0.16 2.18 ± 0.11c 2 a b Medullary area (mm ) 0.30 ± 0.03 0.35 ± 0.02 0.27 ± 0.04 0.32 ± 0.03 Cortical BMD 1.22 ± 0.07 1.23 ± 0.07 1.21 ± 0.10 1.18 ± 0.05 Tibial length (mm) 16.14 ± 0.23 16.24 ± 0.13 16.17 ± 0.20 16.26 ± 0.08 BV/TV (%) 3.66 ± 0.80 2.49 ± 0.83 2.16 ± 0.91 2.36 ± 0.86 Trabecular thickness (mm) 0.029 ± 0.002 0.030 ± 0.003 0.027 ± 0.004 0.031 ± 0.004 Trabecular number (1/mm) 1.23 ± 0.18a 0.82 ± 0.24 0.78 ± 0.27 0.76 ± 0.22 Trabecular space (mm) 0.79 ± 0.13 1.30 ± 0.50 1.35 ± 0.39 1.39 ± 0.45 Trabecular BMD 0.69 ± 0.02 0.77 ± 0.07 0.70 ± 0.08 0.71 ± 0.07 Note: Data expressed as mean ± SD. a Statistical differences p < 0.05 between SAMR1 and SAMP8 Null. b Statistical differences p < 0.05 between SAMP8 Null and treated SAMP8. c Statistical differences p < 0.05 between SAMP8 s-­KL and SAMP8 m-­KL. to present similar cell number to SAMR1. A similar effect was ob- decreased, while the expression of the anti-­inflammatory cytokine served for the astrocyte marker glial fibrillary acidic protein (GFAP) IL-­10 showed a clear tendency to increase. (Figure S1), where the number of cells in SAMR1 was similar to Additionally, the activity of senescence-­associated β-­ SAMP8 Null, but the average size was reduced between 20% and galactosidase (SA β-­gal) was assessed, and a reduction in global 25%. After treatment, the average size of GFAP+ cells reached β-­gal enzyme activity was observed in subiculum and cortex after SAMR1 mice values, being more evident in CA1 and DG than in CA3. treatment with s-­KL (Figure 4b). Similar effects were observed for In accordance with these results, gene expression of Il-­1β and Tnf-­ β-­galactosidase enzyme accumulation; both KL isoforms tended to α in SAMP8 Klotho-­treated animals mimicked the healthy control recover enzyme levels in SP8 mice presenting levels equivalent to SAMR1 group (Figure 4c). Interestingly, after both KL treatments, SAMR1 mice (Figure S2). Further senescence-­associated markers expression of pro-­inflammatory proteins monocyte chemoattract- were analyzed, and KL treated groups tended to present lower p16 ant protein-­1 (MCP-­1) and cyclooxygenase 2 (COX-­2) was significantly and p21 expression levels, similarly to SAMR1 mice (Figure 4d). ROIG-­SORIANO et al. | 11 of 15 3.5 | KL treatment rescues structural changes memory in both working memory evaluated by NORT, and spatial observed in aged SAMP8 tibia memory tested by OLT. However, we did not find modifications in behavior and emotional assays in SAMP8 mice after KL treatment. Effects of KL treatment over the microstructure of the long bone KL molecular mechanisms are not entirely understood, but it has tibia (Figure 5) are quantified in Table 3. Mid-­tibial cortical bone pre- been described that KL increases synaptic plasticity and cognitive sented a diameter expansion in SAMP8 Null mice, corresponding to a performance, modifying several structures, and brain functions 5% perimeter and 16% cortical area expansion of SAMP8 compared during aging (Abraham et al., 2016). with SAMR1 cortical area. Likewise, SAMP8 medullary space was In accordance with these results, a study developed by Zhou and also increased by 6% and 17% perimeter and endocortical area, re- collaborators demonstrated neuronal and cognitive beneficial ef- spectively. Of note, Klotho treatment significantly prevented cortical fects of lentiviral-­mediated m-­KL expression on SAMP8 mice (Zhou expansion, showing parameter values equivalent to those of SAMR1 et al., 2018). Furthermore, our group previously described that a sin- animals (Table 3). Interestingly, a higher reduction in periosteal and gle injection into the CNS of AAV9 expressing s-­KL improved learn- endocortical perimeter was detected with s-­KL compared with m-­KL ing and memory performance in 12-­month-­old C57BL/6J males, treatment, although it was not statistically different from SAMR1 suggesting that s-­KL isoform activity and not just m-­KL, might be control group. In addition, other important parameters for determin- particularly important in the CNS (Masso et al., 2018). After com- ing bone strength and elasticity such as bone length, cortical thick- paring both isoforms, we obtained similar effects over memory and ness, or bone mineral density were not altered between SAMP8 physical state behavioral tests. However, the biosafety profile of the and SAMR1 mice, or after the treatment. Percentage of trabecular s-­KL protein is better than that of m-­KL, due to the lack of KL2 do- bone volume (BV/TV) was reduced in SAMP8 model compared with main, which is involved in bone mineral homeostasis, being this is SAMR1 (Table 3), which is compatible with the bone loss associated a relevant consideration in the field of Klotho-­based therapies for to the strain, a feature not affected by the KL treatment (Figure 5b). aging associated diseases. The effect of secreted KL treatment was only detectable in During aging, general epigenetic alterations are observed, result- SAMP8 mice, as SAMR1 animals treated with s-­KL did not show ing in a progressive global hypomethylation (in non-­CpG islands) to- alterations in the analyzed cortical or trabecular parameters com- gether with increased local hypermethylation (primarily CpG islands) pared with control SAMR1 animals (Table S1). Relevance in bone (Fraga et al., 2007; Lopez-­Otin et al., 2013), while hydroxymethyla- strength of the detected structural alteration was assessed with a tion is increased in non-­pathological aging, as well as in brains of aged 3-­points bending fracture test (Figure S3). No differences were ob- transgenic AD mice (Cadena-­del-­C astilloa et al., 2014). Accordingly, served comparing SAMP8 treatment groups in the strength needed in this study, that trend was replicated in the SAMP8 strain, previ- for bone fracture, although a tendency was observed to increased ously associated with accelerated epigenetic changes compared with bone plasticity under stress with KL treatments, mainly with the SAMR1 controls (Grinan-­Ferre et al., 2018). Strikingly, this typical m-­KL isoform. SAMP8 phenotype was rescued after KL treatments, paralleling the 5-­mC levels of control SAMR1 mice. Although Dnmt1 gene expres- sion was reduced, expression of Dnmt3a/b genes was not modified. 4 | DISCUSSION This is, however, in line with previous results in SAMP8 mice after environmental enrichment intervention, where increased methyla- The anti-­aging and cognition-­e nhancing gene KL can act through tion levels were associated with no change or even a reduction in the its pleiotropic protective functions to counteract the diverse sign- levels of different members of the DNMT and TET families (Grinan-­ aling cascades and mechanisms underlying senescence (Moos Ferre et al., 2016). Similarly, continuous s-­KL and m-­KL expression et al., 2020). Therapeutic approaches increasing KL gene expres- reduced age-­associated 5-­hmC levels, albeit without promoting any sion can prevent further neurodegeneration and memory loss as- change in enzymes’ gene expression from the TETs family. These sociated with aging and Alzheimer's disease (AD) (Abraham et al., data are consistent with results reported recently by Irier and co-­ 2016; Dubal et al., 2015). Although KL is mainly expressed in the workers in mice, where no modifications in TETs family enzymes kidney and does not cross the blood-­b rain barrier (BBB) (Leon gene expression were found after an intervention that promoted et al., 2017), it is also expressed in the brain (Masso et al., 2015; cognitive improvement (Irier et al., 2014). Moos et al., 2020). Our study used intracerebroventricular admin- Although the modulation of histone acetylases and HDACs in istration of AAVs as a delivery approach to increase KL expression AD remains controversial, histone acetylation is another crucial epi- in the CNS as a promising therapeutic strategy to normalize some genetic modification in the AD pathology. It has been reported that relevant aging hallmarks present in SAMP8 mice. This procedure HDAC inhibition increased histone acetylation H3/H4 and rescued also allowed liver transduction due to viral vector extravasation learning and memory deficits in AD animal models (Grinan-­Ferre from cerebroventricular fluid to the bloodstream after the injec- et al., 2016; Lu et al., 2015). Here, we observed a significant increase tion (Schuster et al., 2014). in H3 and H4 acetylation levels after s-­KL or m-­KL single treatment Our results revealed the beneficial effects of sustained KL in SAMP8 mice, whereas an evident, though no significant, reduction expression by improving the physical condition and preserving in HDACs family gene expression was found. In agreement with our 12 of 15 | ROIG-­SORIANO et al. results, the SAMP8 strain showed reduced H3/H4 acetylation levels due to endocortical bone absorption and periosteal bone apposition when it was compared with age-­matched SAMR1, and the acetyla- (Chen et al., 2009; Riggs et al., 2004). This cortical bone expansion has tion levels were higher after interventions that improved the cogni- been described in long bones of human and animal models for oste- tive performance of the SAMP8 mouse (Grinan-­Ferre et al., 2016). oporosis research, and it is explained as a compensatory mechanism Previous studies linked genetic KL variants, which produce lower to increase bone strength under bone loss, alterations in trabecular circulating KL levels, with advanced epigenetic age, suggesting a di- connectivity, and low protein intake circumstances (Ammann & Rizzoli, rect or indirect role of KL over age-­related epigenetic status (Wolf 2003). Similarly, cortical expansion was detected in male C57BL/6 and et al., 2020). However, although many studies have associated the senescence-­accelerated-­mouse prone strains (SAMP), although it did beneficial effects of KL in animal models (Dubal et al., 2015; Masso not counteract increased bone weakness in aged individuals (Ferguson et al., 2018; Zhou et al., 2018), this is, to the best of our knowledge, et al., 2003; Silva et al., 2002). In our study, this tendency is reproduced the first time that KL has been linked to an intervention affecting in SAMP8 Null animals, which exhibit a tibial cortical bone expansion. global epigenetic mechanisms with resulting neuroprotective ef- Interestingly, this effect is abolished after two months expression of fects, including better physical condition and cognitive performance. the KL isoforms. The cortical expansion was only observed in SAMP8 In this context, we hypothesize that Klotho´s pleiotropic activities bones, as SAMR1 animals treated with AAV9 s-­KL did not exhibit al- can finally lead to changes in the epigenetic landscape through dif- tered bone parameters. Silva et al. reported that SAMP6 tibial cortical ferent mechanisms either directly, by increasing epigenetic markers and medullar cavity expansion correlated with decreased resistance as 5-­mC levels; or indirectly, by altering gene expression of several to fracture, with no BMD parameter changes (Silva et al., 2002). That chromatin-­modifying enzymes, such as Dnmt1 and Hdac6. correlation was not reproduced in our study using SAMP8 mice as It is well established that persistent inflammation plays a promi- no differences in breaking force were observed after the treatment. nent role in aged and AD brain, exacerbating Aβ pathology, and tau Interestingly, Klotho-­treated animals presented a tendency to increase hyperphosphorylation (Kinney et al., 2018). Previous studies have bone displacement prior to fracture. In this regard, previous studies also shown that KL suppresses several inflammatory markers such have reported a positive effect of Klotho in maintaining bone structure as TNF-­α (Maekawa et al., 2009), while KL depletion contributes to (Maruyama et al., 2015) and composition protection under normal and increased inflammation in a mouse model (Zhao et al., 2011), con- disease conditions (Lin et al., 2017), directly through FGF23 signaling firming the modulation induced by KL and suggesting that an inter- and, indirectly, by modulating the function of ion transporters and re- action occurs between KL and inflammation. Here, we observed ducing cellular damage. In contrast, low KL levels generate an osteo- a tendency in increased gene expression of the anti-­inflammatory porotic phenotype characterized by blood ion deregulation, low bone cytokine Il-­10 and a significant reduction in the expression of sev- turnover, and reduced fracture resistance (Maruyama et al., 2015; eral pro-­inflammatory markers such as Mcp-­1, Cox-­2, and a clear Sasaki et al., 2013). More specific analyses are needed to fully under- tendency to a reduced Tnf-­α and Il-­6 in s-­KL or m-­KL treated stand the molecular mechanisms used by KL isoforms to affect bone SAMP8 mice. Moreover, a recovery in levels of inflammation-­related structure and the other parameters analyzed. markers was observed after an immunofluorescence study of dif- From a comparative point of view, both KL isoforms resulted in ferent brain areas in treated mice. Previous studies in aged mice efficient improvement of cognitive and physical deficits associated reported decreased Iba1+ microglial cell number and arborization, with aged SAMP8 mice. Interestingly, just s-­KL isoform resulted especially in the cortex (Zoller et al., 2018). Similarly, progressive in body weight maintenance during the experiment and increased Iba1+ cell arborization loss was observed in human post-­mortem horizontal activity on the OF test. Both isoforms efficiently res- samples during aging and AD progression (Davies et al., 2017). In cued global epigenetic levels associated with younger animals, a this study, both s-­KL and m-­KL treatments statistically increased the property not previously described before for this protein. The anti-­ size of Iba1+ cells while the average size of GFAP+ cells resembled inflammatory properties were also comparable, although just s-­KL the size in SAMR1 mice. Regarding the cell morphology, the changes efficiently reduced SA β-­Gal activity in CX and SUB areas. These were much more evident in GFAP+ cells, were s-­KL and m-­KL treat- observed differences could be related to an increased availability ments were associated with greater cell processes thickness includ- or diffusion of s-­KL isoform due to its putative secreted form and ing those cells surrounding the vessels. In addition, SA β-­Gal has reduced molecular weight compared to m-­KL isoform. Finally, the been widely used to address neuronal wellness during aging since effect over bone structure was also comparable after treatment with an accumulation and increased activity of this enzyme have been both isoforms. described in senescent cells (Geng et al., 2010). Strikingly, after the In summary, our in vivo results after a single treatment of AAVs two months treatment, both isoforms of KL reduced the accumula- expressing s-­K L in SAMP8 mice revealed new potential therapeu- tion of the enzyme, although only the s-­KL isoform reduced SA β-­Gal tic properties of KL by improving age-­related deficits and rele- activity in a statistically significant manner. More studies are needed vant molecular aging hallmarks such as epigenetic, inflammatory, to depict if this effect is due to β-­Gal breakdown or inhibition of the and bone alterations. Moreover, it reemphasized the therapeutic protein accumulation due to reduced cellular stress. potential of the s-­K L isoform in CNS and physical status, as well Previous bone studies have reported age-­related geometric remod- as its isoform-­specific effects, probably due to increased protein eling, tending to a medullary space and cortical perimeter expansion, diffusion. ROIG-­SORIANO et al. | 13 of 15 Arking, D. E., Krebsova, A., Macek, M., Macek, M., Arking, A., Mian, I. S., AC K N OW L E D G M E N T S Fried, L., Hamosh, A., Dey, S., McIntosh, I., & Dietz, H. C. (2002). This project was supported by the Ministerio Ciencia Innovación, Association of human aging with a functional variant of klotho. Retos Sociedad (PID2019-­104034RB-­I 00), the Ministerio de Proceedings of the National Academy of Sciences of the United States of America, 99(2), 856–­861. https://doi.org/10.1073/pnas.02248​ Economía, Industria y Competitividad (PID2020-­116735RB-­I 00/ 4299 AEI/10.13039/501100011033) and Fondo Europeo de Desarrollo Cadena-­del-­C astillo, C., Valdes-­Quezada, C., Carmona-­Aldana, F., Arias, Regional (MINECO-­FEDER) (SAF2016-­33307 and PCIN-­2015-­229), C., Bermúdez-­Rattoni, F., Recillas-­Targa, F. (2014). Age-­dependent and Generalitat de Catalunya (2017 SGR 106). J.R is a recipient of an increment of hydroxymethylation in the brain cortex in the FI fellowship by the AGAUR (2018FI_B00657). triple-­transgenic mouse model of Alzheimer’s disease. Journal of Alzheimer’s disease : JAD, 41(3), 845–­854. Chen, C. D., Tung, T. Y., Liang, J., Zeldich, E., Tucker Zhou, T. B., Turk, B. E., C O N FL I C T O F I N T E R E S T & Abraham, C. R. (2014). Identification of cleavage sites leading to Portions of this work are the subject of a patent application held by the shed form of the anti-­aging protein klotho. Biochemistry, 53(34), the Universitat Autonoma de Barcelona (UAB, Spain); the Universitat 5579–­5587. https://doi.org/10.1021/bi500​4 09n Chen, G., Liu, Y., Goetz, R., Fu, L., Jayaraman, S., Hu, M. C., & Mohammadi, de Barcelona (UB, Spain); the Institucio Catalana de Recerca i Estudis M. (2018). alpha-­Klotho is a non-­enzymatic molecular scaffold for Avançats (ICREA, Spain); and the Vall d'Hebron Institute of Research FGF23 hormone signalling. Nature, 553(7689), 461–­466. https:// (VHIR, Spain). doi.org/10.1038/natur​e25451 Chen, H., Zhou, X., Emura, S., & Shoumura, S. (2009). Site-­specific bone loss in senescence-­accelerated mouse (SAMP6): A murine model AU T H O R C O N T R I B U T I O N S for senile osteoporosis. Experimental Gerontology, 44(12), 792–­798. J. Roig-­Soriano, C. Griñán-­Ferré, M. Pallàs, J. F. Espinosa-­Parrilla, A. https://doi.org/10.1016/j.exger.2009.09.009 Bosch, and Miguel Chillon designed research. J. Roig-­Soriano, and C. Coppieters, N., Dieriks, B. V., Lill, C., Faull, R. L., Curtis, M. A., & Griñán-­Ferré performed research. J. Roig-­Soriano, C. Griñán-­Ferré, Dragunow, M. (2014). Global changes in DNA methylation and hy- droxymethylation in Alzheimer's disease human brain. Neurobiology J. F. Espinosa-­Parrilla, A. Bosch, C. R. Abraham, and Miguel Chillon of Aging, 35(6), 1334–­1344. https://doi.org/10.1016/j.neuro​biola​ analyzed and interpreted the data. J. Roig-­Soriano and C. Griñán-­ ging.2013.11.031 Ferré wrote the paper. Miguel Chillon is the corresponding author. Davies, D. S., Ma, J., Jegathees, T., & Goldsbury, C. (2017). Microglia All authors read and approved the final manuscript. show altered morphology and reduced arborization in human brain during aging and Alzheimer's disease. Brain Pathology, 27(6), 795–­ 808. https://doi.org/10.1111/bpa.12456 DATA AVA I L A B I L I T Y S TAT E M E N T Delgado-­Morales, R., Agis-­Balboa, R. C., Esteller, M., & Berdasco, The data that support the findings of this study are available from M. (2017). Epigenetic mechanisms during ageing and neu- the corresponding author upon reasonable request. rogenesis as novel therapeutic avenues in human brain dis- orders. Clin Epigenetics, 9, 67. https://doi.org/10.1186/s1314​ 8-­017-­0365-­z ORCID Dubal, D. B., Zhu, L., Sanchez, P. E., Worden, K., Broestl, L., Johnson, J. Roig-­Soriano https://orcid.org/0000-0002-6134-3141 E., Ho, K., Yu, G.-­Q., Kim, D., Betourne, A., Kuro-­o, M., Masliah, E., C. Griñán-­Ferré https://orcid.org/0000-0002-5424-9130 Abraham, C. R., & Mucke, L. (2015). Life extension factor klotho pre- J. F. Espinosa-­Parrilla https://orcid.org/0000-0001-6631-3542 vents mortality and enhances cognition in hAPP transgenic mice. Journal of Neuroscience, 35(6), 2358–­2371. https://doi.org/10.1523/ C. R. Abraham https://orcid.org/0000-0002-6649-1960 JNEUR​OSCI.5791-­12.2015 A. Bosch https://orcid.org/0000-0002-7205-2796 Ennaceur, A., & Delacour, J. (1988). A new one-­trial test for neu- M. Pallàs https://orcid.org/0000-0003-3095-4254 robiological studies of memory in rats. 1: Behavioral Miguel Chillón https://orcid.org/0000-0003-0840-2111 data. Behavioral Brain Research, 31(1), 47–­59. https://doi. org/10.1016/0166-­4328(88)90157​-­x Ferguson, V. L., Ayers, R. A., Bateman, T. A., & Simske, S. J. (2003). REFERENCES Bone development and age-­related bone loss in male C57BL/6J Abraham, C. R., Mullen, P. C., Tucker-­Zhou, T., Chen, C. D., & Zeldich, mice. Bone, 33(3), 387–­398. https://doi.org/10.1016/s8756​ E. (2016). Klotho is a neuroprotective and cognition-­ enhancing -­3282(03)00199​-­6 protein. Vitamins and Hormones, 101, 215–­238. https://doi. Fraga, M. F., Agrelo, R., & Esteller, M. (2007). Cross-­t alk between org/10.1016/bs.vh.2016.02.004 aging and cancer: The epigenetic language. Annals of the New Akiguchi, I., Pallàs, M., Budka, H., Akiyama, H., Ueno, M., Han, J., Yagi, York Academy of Sciences, 1100, 60–­74. https://doi.org/10.1196/ H., Nishikawa, T., Chiba, Y., Sugiyama, H., Takahashi, R., Unno, annals.1395.005 K., Higuchi, K., & Hosokawa, M. (2017). SAMP8 mice as a neuro- Geng, Y. Q., Guan, J. T., Xu, X. H., & Fu, Y. C. (2010). Senescence-­ pathological model of accelerated brain aging and dementia: Toshio associated beta-­g alactosidase activity expression in aging Takeda's legacy and future directions. Neuropathology, 37(4), 293–­ hippocampal neurons. Biochemical and Biophysical Research 305. https://doi.org/10.1111/neup.12373 Communications, 396(4), 866–­8 69. https://doi.org/10.1016/j. Ammann, P., & Rizzoli, R. (2003). Bone strength and its determinants. bbrc.2010.05.011 Osteoporosis International, 14(Suppl 3), S13–­18. https://doi. Griñán-­Ferré, C., Izquierdo, V., Otero, E., Puigoriol-­Illamola, D., Corpas, R., org/10.1007/s0019​8-­0 02-­1345-­4 Sanfeliu, C., Ortuño-­Sahagún, D., & Pallàs, M. (2018). Environmental Archer, J. (1973). Tests for emotionality in rats and mice: A review. enrichment improves cognitive deficits, AD hallmarks and epigene- Animal Behavior, 21(2), 205–­235. https://doi.org/10.1016/s0003​ tic alterations presented in 5xFAD mouse model. Front Cell Neurosci, -­3 472(73)80065​-­x 12, 224. https://doi.org/10.3389/fncel.2018.00224 14 of 15 | ROIG-­SORIANO et al. Griñan-­Ferré, C., Puigoriol-­Illamola, D., Palomera-­Ávalos, V., Pérez-­ McHugh, D., & Gil, J. (2018). Senescence and aging: Causes, conse- Cáceres, D., Companys-­Alemany, J., Camins, A., Ortuño-­Sahagún, quences, and therapeutic avenues. Journal of Cell Biology, 217(1), D., Rodrigo, M. T., & Pallàs, M. (2016). Environmental enrichment 65–­77. https://doi.org/10.1083/jcb.20170​8 092 modified epigenetic mechanisms in SAMP8 mouse hippocampus Moos, W. H., Faller, D. V., Glavas, I. P., Harpp, D. N., Kanara, I., Mavrakis, by reducing oxidative stress and inflammaging and achieving neu- A. N., Pernokas, J., Pernokas, M., Pinkert, C. A., Powers, W. R., roprotection. Frontiers in Aging Neuroscience, 8, 241. https://doi. Sampani, K., Steliou, K., Vavvas, D. G., Zamboni, R. J., Kodukula, K., org/10.3389/fnagi.2016.00241 & Chen, X. (2020). Klotho pathways, myelination disorders, neu- Irier, H., Street, R. C., Dave, R., Lin, L. I., Cai, C., Davis, T. H., Yao, B., rodegenerative diseases, and epigenetic drugs. BioResearch Open Cheng, Y., & Jin, P. (2014). Environmental enrichment modulates Access, 9(1), 94–­105. https://doi.org/10.1089/biores.2020.0004 5-­hydroxymethylcytosine dynamics in hippocampus. Genomics, Piedra, J., Ontiveros, M., Miravet, S., Penalva, C., Monfar, M., & 104(5), 376–­382. https://doi.org/10.1016/j.ygeno.2014.08.019 Chillon, M. (2015). Development of a rapid, robust, and universal King, G. D., Rosene, D. L., & Abraham, C. R. (2012). Promoter methylation picogreen-­based method to titer adeno-­associated vectors. Human and age-­related downregulation of Klotho in rhesus monkey. Age, Gene Therapy Methods, 26(1), 35–­42. https://doi.org/10.1089/ 34(6), 1405–­1419. https://doi.org/10.1007/s1135​7-­011-­9315-­4 hgtb.2014.120 Kinney, J. W., Bemiller, S. M., Murtishaw, A. S., Leisgang, A. M., Salazar, Riggs, B. L., Melton, L. J., Robb, R. A., Camp, J. J., Atkinson, E. J., Peterson, A. M., & Lamb, B. T. (2018). Inflammation as a central mechanism in J. M., Rouleau, P. A., McCollough, C. H., Bouxsein, M. L., & Khosla, Alzheimer's disease. Alzheimer's & Dementia: Translational Research S. (2004). Population-­based study of age and sex differences in & Clinical Interventions, 4, 575–­590. https://doi.org/10.1016/j. bone volumetric density, size, geometry, and structure at different trci.2018.06.014 skeletal sites. Journal of Bone and Mineral Research, 19(12), 1945–­ Kuro-­o, M., Matsumura, Y., Aizawa, H., Kawaguchi, H., Suga, T., Utsugi, 1954. https://doi.org/10.1359/JBMR.040916 T., Ohyama, Y., Kurabayashi, M., Kaname, T., Kume, E., Iwasaki, Sasaki, M., Hasegawa, T., Yamada, T., Hongo, H., de Freitas, P. H. L., H., Iida, A., Shiraki-­Iida, T., Nishikawa, S., Nagai, R., & Nabeshima, Suzuki, R., Yamamoto, T., Tabata, C., Toyosawa, S., Yamamoto, Y. I. (1997). Mutation of the mouse klotho gene leads to a syn- T., Oda, K., Li, M., Inoue, N., & Amizuka, N. (2013). Altered dis- drome resembling ageing. Nature, 390(6655), 45–­51. https://doi. tribution of bone matrix proteins and defective bone mineraliza- org/10.1038/36285 tion in klotho-­ deficient mice. Bone, 57(1), 206–­219. https://doi. Kurosu, H., Yamamoto, M., Clark, J. D., Pastor, J. V., Nandi, A., Gurnani, org/10.1016/j.bone.2013.08.008 P., McGuinness, O. P., Chikuda, H., Yamaguchi, M., Kawaguchi, H., Schuster, D. J., Dykstra, J. A., Riedl, M. S., Kitto, K. F., Belur, L. R., Shimomura, I., Takayama, Y., Herz, J., Kahn, C. R., Rosenblatt, K. P., McIvor, R. S., Elde, R. P., Fairbanks, C. A., & Vulchanova, L. (2014). & Kuro-­o, M. (2005). Suppression of aging in mice by the hormone Biodistribution of adeno-­associated virus serotype 9 (AAV9) vec- Klotho. Science, 309(5742), 1829–­1833. https://doi.org/10.1126/ tor after intrathecal and intravenous delivery in mouse. Frontiers in scien​ce.1112766 Neuroanatomy, 8, 42. https://doi.org/10.3389/fnana.2014.00042 Leon, J., Moreno, A. J., Garay, B. I., Chalkley, R. J., Burlingame, A. L., Semba, R. D., Moghekar, A. R., Hu, J., Sun, K., Turner, R., Ferrucci, L., & Wang, D., & Dubal, D. B. (2017). Peripheral elevation of a klotho O'Brien, R. (2014). Klotho in the cerebrospinal fluid of adults with fragment enhances brain function and resilience in young, aging, and without Alzheimer's disease. Neuroscience Letters, 558, 37–­4 0. and α-­synuclein transgenic mice. Cell Reports, 20(6), 1360–­1371. https://doi.org/10.1016/j.neulet.2013.10.058 https://doi.org/10.1016/j.celrep.2017.07.024 Shiraki-­Iida, T., Aizawa, H., Matsumura, Y., Sekine, S., Iida, A., Anazawa, Lin, W., Li, Y., Chen, F., Yin, S., Liu, Z., & Cao, W. (2017). Klotho preser- H., & Nabeshima, Y. (1998). Structure of the mouse klotho gene vation via histone deacetylase inhibition attenuates chronic kidney and its two transcripts encoding membrane and secreted pro- disease-­associated bone injury in mice. Scientific Reports, 7, 46195. tein. FEBS Letters, 424(1–­2), 6–­10. https://doi.org/10.1016/s0014​ https://doi.org/10.1038/srep4​6195 -­5793(98)00127​-­6 Lopez-­Otin, C., Blasco, M. A., Partridge, L., Serrano, M., & Kroemer, G. Silva, M. J., Brodt, M. D., & Ettner, S. L. (2002). Long bones from the (2013). The hallmarks of aging. Cell, 153(6), 1194–­1217. https://doi. senescence accelerated mouse SAMP6 have increased size but org/10.1016/j.cell.2013.05.039 reduced whole-­bone strength and resistance to fracture. Journal Lu, X., Wang, L., Yu, C., Yu, D., & Yu, G. (2015). Histone acetylation modi- of Bone and Mineral Research, 17(9), 1597–­1603. https://doi. fiers in the pathogenesis of Alzheimer's disease. Front Cell Neurosci, org/10.1359/jbmr.2002.17.9.1597 9, 226. https://doi.org/10.3389/fncel.2015.00226 Wolf, E. J., Chen, C.-­D., Zhao, X., Zhou, Z., Morrison, F. G., Daskalakis, Maekawa, Y., Ishikawa, K., Yasuda, O., Oguro, R., Hanasaki, H., Kida, N. P., Stone, A., Schichman, S., Grenier, J. G., Fein-­Schaffer, I., Takemura, Y., Ohishi, M., Katsuya, T., Rakugi, H. (2009). Klotho D., Huber, B. R., Abraham, C. R., Miller, M. W., & Logue, M. W. suppresses TNF-­α-­induced expression of adhesion molecules in (2020). Klotho, PTSD, and advanced epigenetic age in cortical the endothelium and attenuates NF-­κB activation. Endocrine, 35(3), tissue. Neuropsychopharmacology, 46(4), 721–­730. https://doi. 341–­3 46. https://doi.org/10.1007/s1202​0 -­0 09-­9181-­3 org/10.1038/s4138​6-­020-­0 0884​-­5 Maruyama, N., Shibata, Y., Mochizuki, A., Yamada, A., Maki, K., Inoue, T., Yamamoto, M., Clark, J. D., Pastor, J. V., Gurnani, P., Nandi, A., Kurosu, Kamijo, R., Miyazaki, T. (2015). Bone micro-­fragility caused by the H., Miyoshi, M., Ogawa, Y., Castrillon, D. H., Rosenblatt, K. P., & mimetic aging processes in α-­klotho deficient mice: In situ nanoin- Kuro-­o, M. (2005). Regulation of oxidative stress by the anti-­aging dentation assessment of dilatational bands. Biomaterials, 47, 62–­71. hormone klotho. Journal of Biological Chemistry, 280(45), 38029–­ https://doi.org/10.1016/j.bioma​teria​ls.2015.01.004 38034. https://doi.org/10.1074/jbc.M5090​39200 Masso, A., Sanchez, A., Bosch, A., Gimenez-­Llort, L., & Chillon, M. (2018). Zhao, Y., Banerjee, S., Dey, N., LeJeune, W. S., Sarkar, P. S., Brobey, R., Secreted alphaKlotho isoform protects against age-­dependent & Choudhary, S. (2011). Klotho depletion contributes to increased memory deficits. Molecular Psychiatry, 23(9), 1937–­1947. https:// inflammation in kidney of the db/db mouse model of diabetes via doi.org/10.1038/mp.2017.211 RelA (serine)536 phosphorylation. Diabetes, 60(7), 1907–­1916. Masso, A., Sanchez, A., Gimenez-­Llort, L., Lizcano, J. M., Canete, M., https://doi.org/10.2337/db10-­1262 Garcia, B., & Chillon, M. (2015). Secreted and transmembrane Zhou, H. J., Zeng, C. Y., Yang, T. T., Long, F. Y., Kuang, X., & Du, J. R. alphaklotho isoforms have different spatio-­ temporal profiles (2018). Lentivirus-­mediated klotho up-­regulation improves aging-­ in the brain during aging and Alzheimer's disease progression. related memory deficits and oxidative stress in senescence-­ PLoS One, 10(11), e0143623. https://doi.org/10.1371/journ​ accelerated mouse prone-­8 mice. Life Sciences, 200, 56–­62. https:// al.pone.0143623 doi.org/10.1016/j.lfs.2018.03.027 ROIG-­SORIANO et al. | 15 of 15 Zoller, T., Attaai, A., Potru, P. S., Russ, T., & Spittau, B. (2018). Aged mouse cortical microglia display an activation profile suggesting immuno- How to cite this article: Roig-­Soriano, J., Griñán-­Ferré, C., tolerogenic functions. International Journal of Molecular Sciences, Espinosa-­Parrilla, J. F., Abraham, C. R., Bosch, A., Pallàs, M., & 19(3). https://doi.org/10.3390/ijms1​9030706 Chillón, M. (2022). AAV-­mediated expression of secreted and transmembrane αKlotho isoforms rescues relevant aging S U P P O R T I N G I N FO R M AT I O N hallmarks in senescent SAMP8 mice. Aging Cell, 21, e13581. Additional supporting information may be found in the online https://doi.org/10.1111/acel.13581 version of the article at the publisher’s website.