www.aging-us.com AGING 2019, Vol. 11, Advance Priority Research Paper Telomerase gene therapy ameliorates the effects of neurodegeneration associated to short telomeres in mice Kurt Whittemore1, Aksinya Derevyanko1, Paula Martinez1, Rosa Serrano1, Martí Pumarola2,3, Fàtima Bosch4,5, Maria A. Blasco1 1Telomeres and Telomerase Group, Molecular Oncology Program, Spanish National Cancer Centre (CNIO), Madrid 28029, Spain 2Unit of Murine and Comparative Pathology (UPMiC), Department of Animal Medicine and Surgery, Veterinary Faculty, Universitat Autònoma de Barcelona, 08193 Bellaterra (Cerdanyola del Vallès), Barcelona, Spain 3Networking Research Center on Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN), Universitat Autònoma de Barcelona, 08193 Bellaterra (Cerdanyola del Vallès), Barcelona, Spain 4Center of Animal Biotechnology and Gene Therapy, Department of Animal Medicine and Surgery, Universitat Autònoma de Barcelona, Bellaterra 08193, Spain 5Center of Animal Biotechnology and Gene Therapy, Department of Biochemistry and Molecular Biology, School of Veterinary Medicine, Universitat Autònoma de Barcelona, Bellaterra 08193, Spain Correspondence to: Maria A. Blasco; email: mblasco@cnio.es Keywords: telomerase, gene therapy, neurodegeneration, TERT Received: April 26, 2019 Accepted: May 17, 2019 Published: May 28, 2019 Copyright: Whittemore et al. This is an open‐access article distributed under the terms of the Creative Commons Attribution License (CC BY 3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. ABSTRACT Neurodegenerative diseases associated with old age such as Alzheimer’s disease present major problems for society, and they currently have no cure. The telomere protective caps at the ends of chromosomes shorten with age, and when they become critically short, they can induce a persistent DNA damage response at chromosome ends, triggering secondary cellular responses such as cell death and cellular senescence. Mice and humans with very short telomeres owing to telomerase deficiencies have an earlier onset of pathologies associated with loss of the regenerative capacity of tissues. However, the effects of short telomeres in very low proliferative tissues such as the brain have not been thoroughly investigated. Here, we describe a mouse model of neurodegeneration owing to presence of short telomeres in the brain as the consequence of telomerase deficiency. Interestingly, we find similar signs of neurodegeneration in very old mice as the consequence of physiological mouse aging. Next, we demonstrate that delivery of telomerase gene therapy to the brain of these mice results in amelioration of some of these neurodegeneration phenotypes. These findings suggest that short telomeres contribute to neurodegeneration diseases with aging and that telomerase activation may have a therapeutic value in these diseases. INTRODUCTION treatments for any of these diseases owing to the fact that their molecular origins are still poorly understood. In- Neurodegenerative diseases, such as Alzheimer’s and stead, palliative treatments are directed to alleviate down- Parkinson’s, are associated with aging, and their pre- stream events, such as beta-amyloid deposition in the case valence is increasing as there are more individuals that of Alzheimer’s or lack of dopamine generation in the case reach older ages [1–3]. To date, there are no curative of Parkinson’s due to the loss of dopaminergic neurons. www.aging-us.com 2916 AGING Telomeres are heterochromatic protective structures at showed that increased TERT expression in the context the ends of chromosomes that consist of TTAGGG of cancer resistant transgenic mice was sufficient to repeats bound by a six-protein complex known as delay aging and extend mouse longevity by 40% [22]. shelterin [4,5]. A minimum length of telomeric repeats We further demonstrated that telomerase reactivation is necessary for shelterin binding and protection [4,5]. in adult tissues by using adeno-associated viruses Telomerase is a reverse transcriptase which can (AAV9-Tert) was able to significantly delay age- elongate telomeres de novo by the addition of telomeric related diseases and increase longevity in wild-type repeats onto chromosome ends [6], in this manner mice [23]. In particular, AAV9-Tert treatment re- compensating progressive telomere attrition as a conse- sulted in reduced age-related osteoporosis, reduced quence of cell division. Telomerase is composed of two glucose intolerance, increased neuromuscular coor- essential subunits, the telomerase reverse transcriptase dination, enhanced memory in an object recognition (TERT) and the telomerase RNA component (Terc), test, improved mitochondrial fitness, and delayed which is used as a template for the synthesis of cancer, thus demonstrating that telomere shortening is telomeric repeats [6]. Adult tissues, including the stem causative of aging and is at the origin of a wide range of cell compartments, do not have sufficient telomerase age-associated diseases, including cognitive decline activity to compensate for telomere shortening [23]. More recently, we have also shown that AAV9- associated with tissue regeneration and cell division [7– mediated telomerase activation has therapeutic effects 9]. When telomeres reach a critically short length, this in pre-clinical mouse models for diseases associated triggers activation of a persistent DNA damage with short telomeres such as aplastic anemia [24], response at telomeres and the subsequent induction of myocardial infarction [25], and pulmonary fibrosis cellular senescence or apoptosis [5,10]. Short/dysfunc- [26]. tional telomeres are considered one of the primary hallmarks of aging both in mice and humans, as they In spite of the fact that the brain is a low-proliferative lead to other hallmarks of aging, such as genomic tissue, there are regenerative areas within the brain such instability, cellular senescence, and loss of the as the hippocampus, the subventricular zone (SVZ), and regenerative capacity of tissues [11]. In particular, the olfactory bulb. Interestingly, several studies suggest critically short telomeres in the stem cell compartments the presence of short telomeres in patients with lead to impaired tissue renewal and homeostasis [12– advanced Alzheimer’s disease [27–32]. In the case of 14]. Interestingly, the rate of telomere shortening Parkinson’s disease, extensive research has not been throughout lifespan is influenced by both genetic factors performed. A few studies have found a correlation (i.e. mutations in genes necessary for telomere between short telomere length and Parkinson’s disease maintenance) and environmental factors (i.e. cigarette [33,34], while other studies found no correlation smoke has a negative effect) [15,16]. Interestingly, [35,36]. These studies were performed with peripheral there are a number of diseases associated to mutations blood leukocytes, and telomere length was not in telomerase and other telomere maintenance genes measured in the brain cells implicated in Parkinson’s known as “telomere syndromes”, which include disease. In mice, we have previously shown that telo- dyskeratosis congenita, aplastic anemia and pulmonary mere attrition in the context of the Terc-deficient mouse fibrosis, among others (for a review see [17]). These model [19] impairs in vitro proliferation of adult neural syndromes are characterized by the presence of stem cells from the SVZ [37], and that telomere extremely short telomeres, which prematurely impair shortening in mice disrupts neuronal differentiation and the regenerative capacity of tissues, affecting both high neurogenesis [38]. These findings pose the interesting and low proliferative tissues [17,18]. Prior to the hypothesis that telomerase reactivation in the brain may discovery of human “telomere syndromes”, similar have significant therapeutic effects. Indeed, both in our findings were made by studying mice genetically AAV9-Tert mediated telomerase reactivation model modified to lack the telomerase RNA component [23] and in a genetic Tert reactivation mouse model (Terc-/-) [19]. Terc-deficient mice have shorter telo- [39], there were beneficial effects associated to meres with increasing mouse generations and this telomerase reactivation in the brain. In particular, results in a progressive decrease of both median and telomerase re-expression in a telomerase-deficient maximum longevity [20,21]. These mice show pre- mouse model with short telomeres, resulted in a larger mature appearance of different pathologies affecting brain weight, a thicker myelin sheath, better perfor- both proliferative and non-proliferative tissues, which mance in an innate avoidance test as a measure of the are accompanied by an impaired regenerative capacity health of the olfactory bulb, and increased molecular (reviewed in [10,13]). markers of Ki67, Sox2, doublecortin, and Olig2 [39]. Similarly, telomerase re-expression in the brain using In support of critically short telomeres being a AAV9 gene therapy in adult wild-type mice was determinant of aging and longevity, we previously sufficient to improve cognitive function [23]. www.aging-us.com 2917 AGING Here, we set to evaluate whether mice with short weeks old), young G3 Tert-/- (14-weeks old), and old G3 telomeres in the brain owing to telomerase deficiency Tert-/- mice (69-72-weeks old). We found that telomeres could be used as a bona fide neurodegeneration model. were significantly shorter in the hippocampus, dentate To this end, we characterized the molecular conse- gyrus, subventricular zone, and neocortex of G3 Tert-/- quences of the presence of short telomeres in the mice compared to the same regions in wild-type mice neurogenic areas of the brain as well as the potential (Figure 1C-F). Accordingly, we also found that the cognitive defects associated to short telomeres in these percentage of short telomeres in G3 Tert-/- in these brain mice. We also explored whether similar signs of neuro- regions was higher (Figure 1C-F). Short telomeres degeneration can be found in wild-type mice associated were defined as telomeres with a fluorescence intensity to physiological mouse aging. Finally, we demonstrate less than the 15th percentile of the intensity values of a that telomerase activation using AAV9-Tert gene control. therapy can ameliorate the brain phenotypes in these mice. Next, we determined whether shorter telomeres in these brain regions of Tert-/- mice were associated to reduced RESULTS proliferation and increased DNA damage [54]. To this end, we used immunohistochemistry to determine the Histological and molecular defects associated to number of Ki67-positive cells as a marker of cycling shorter telomeres in the brain of telomerase- cells and the number of cells positive for γH2AX as a deficient mice marker for DNA damage. In the case of wild-type mice, we observed significantly fewer Ki67-positive cells in First, we set to characterize brain phenotypes in the older mice (71-weeks old) compared to younger ages telomerase-deficient Tert-/- mouse model used in this (8-weeks old) in the hippocampus, the dentate gyrus, study [40–43]. We first weighed the brains of age- subventricular zone (SVZ) of the lateral ventricle matched wild-type and third generation (G3) Tert-/- anterior to the hippocampus level, and the neocortex mice, which have shorter telomeres owing to telomerase (Figure 2A-C). Interestingly, young G3 Tert-/- mice deficiency for three generations. We observed that the (14-weeks old) also showed lower numbers of Ki67- brains of G3 Tert-/- mice showed a tendency to be positive cells than age-matched wild-type controls in smaller than those of age-matched wild-type controls these brain regions, and this was further reduced in the (Figure 1A), in agreement with a previous report [39]. old G3 Tert-/- mice (Figure 2A-C). Regarding DNA A reduced brain size may be indicative of neuro- damage, we found increased numbers of γH2AX- degeneration as this phenotype is also observed in positive cells in the hippocampus, dentate gyrus, and patients with advanced Alzheimer’s disease [44]. In this SVZ of old wild-type mice (71-weeks old) compared to regard, we quantified the size of the hippocampus and young mice (8-weeks old) (Figure 2D,E). Young G3 dentate gyrus (DG) of the different mouse cohorts Tert-/- mice also showed significantly lower numbers of (Figure 1B). We found that G3 Tert-/- mice also show γH2AX-positive cells in the hippocampus, dentate smaller hippocampus and dentate gyrus regions than gyrus, and subventricular zone compared to older G3 wild-type mice (Figure 1B). Tert-/- mice (Figure 2D,E). In the neocortex, more γH2AX associated DNA damage was only observed in We next set to characterize the molecular events the old G3 Tert-/- mice (Figure 2F). These results associated to telomerase deficiency in the brain of G3 indicate that physiological aging in wild-type mice, as Tert-/- mice. We first determined telomere length in well as accelerated telomere shortening as a conse- several brain regions of age-matched wild-type and G3 quence of telomerase-deficiency in the Tert-/- mice, lead Tert-/- mice, including the hippocampus since this to decreased proliferation and increased DNA damage in structure is critical for learning, memory, memory for the hippocampus, dentate gyrus, SVZ, and neocortex. episodic events, and neurogenesis [45,46], as well as the DG area within the hippocampus since this region is Next, we set to address the impact of shorter telomeres involved in neurogenesis [47–51] (see scheme in Figure on neurogenesis, inflammation, and formation of tau 1C). In addition, we also studied telomere length in the protein aggregations. In this regard, neurogenesis has subventricular zone which is important for neurogenesis been suggested to act as a brain repair mechanism which in the adult brain [52], and the neocortex which is could mitigate the effects of neurodegeneration that important for higher-order brain functions such as occurs with Alzheimer’s disease and possibly aging sensory perception, cognition, generation of motor [54,55]. Doublecortin is expressed in developing neurons commands, spatial reasoning, and language [53]. To this and it is considered a bona fide marker of neurogenesis end, we performed quantitative telomere FISH (Q- [39,56]. To this end, we set to study the expression of FISH) directly on coronal paraffin brain sections from doublecortin in various brain regions of both wild-type young wild-type (8-10-weeks old), old wild-type (71- and G3 Tert-/- mice. We found that the number of cells www.aging-us.com 2918 AGING Figure 1. Mice deficient for telomerase have smaller brains, shorter telomeres, more proliferation, more DNA damage, and less neurogenesis. (A) Brain weight and representative images of young and old wild-type and G3 Tert-/- mice. (B) Area of hippocampus and dentate gyrus in untreated mice quantified from representative images of brain sections stained with hematoxylin and eosin. (C) Q- FISH for telomere spot fluorescence measured in the hippocampus, (D) the dentate gyrus specifically, (E) the subventricular zone, and (F) the neocortex. The mean telomere spot fluorescence is shown. The percentage of short telomeres is also shown with “short” being defined as a fluorescence intensity less than the 15th percentile of the fluorescence intensity values of a control sample. Cartoon diagrams label the different regions of the brain. In part (C), A scan of a coronal brain cross-section and G3 Tert-/- mice.without We found fluorescence is shown with the that the number of cells hippocampus region highlighted in yellow. Representative images show the telomere spots labeled with Cy3-Tel probe (in red), and nuclei expressing doublecortin was significantly decreased in stained with DAPI (blue). www.aging-us.com 2919 AGING Figure 2. Immunohistochemistry of Ki67 and γH2AX in the brain. (A-F) The quantification and representative images of the immunohistochemistry for positive cells per field of view for (A-C) Ki67, and (D-F) γH2AX in brain regions such as the hippocampus, dentate gyrus, subventricular zone (SVZ) of the lateral ventricle anterior to the hippocampus level, and the neocortex. The data is shown for young and old wild-type and G3 Tert-/- mice. Data represent the mean ±SE of analyzed mice within each group. For the histopathology results, the number of mice analyzed per group is indicated, as well as the number of fields of view, and the number of positive cells. The t-test was used for statistical analysis. *p<0.05; **p<0.01; ***p<0.001. www.aging-us.com 2920 AGING expressing doublecortin was significantly decreased in the number of cells positive for tau phosphorylated at old wild-type mice (71-weeks old) compared to young serine 396 [66,67]. We observed more cells positive for (8-weeks old) wild-type mice in the hippocampus and phosphorylated tau in the hippocampus and dentate gyrus the SVZ, and showed the same tendency in the dentate of old wild-type mice, and this effect was further gyrus and the neocortex (Figure 3A-C). Young G3 increased in old G3 Tert-/- mice (Figure 3F). Tert-/- mice also showed lower numbers of cells ex- pressing doublecortin in the hippocampus, dentate In summary, these findings indicate that short telomeres gyrus, subventricular zone, and neocortex compared to in the brain contribute to impair brain neurogenesis and age-matched wild-type controls, and this was further to increase neuroinflammation and abnormal tau phos- aggravated with increasing age (Figure 3A-C). The phorylation. Interestingly, similar phenotypes were same trends were observed for doublecortin expression observed associated to physiological aging in very old in the parietotemporal region of the cerebral cortex wild-type mice. (Figure S1A), and the occipital region of the cerebral cortex (Figure S1B). Neurogenic niches were also Telomerase-deficient mice are more susceptible to identified in hematoxylin and eosin stains in the parietal MPTP neurotoxin subventricular zone area, and the young wild-type mice had higher levels of neurogenic niches than old wild- To determine whether mice deficient for telomerase type mice (Figure S1C). In agreement with this finding, were more susceptible to cellular damage, we per- young G3 Tert-/- mice had more neurogenic niches than formed an experiment with the MPTP (1-methyl-4- old G3 Tert-/- mice, but less than young wild-type mice phenyl-1,2,3,6-tetrahydropyridine) neurotoxin. MPTP (Figure S1C). These findings indicate that shorter specifically damages dopaminergic neurons in the telomeres associated with aging and to telomerase- substantia nigra region of the brain [68]. Inside these deficiency correlate with impaired neurogenesis in neurons, MPTP is metabolized to MPP+ which then different regions of the brain. interferes with complex I of the electron transport chain in mitochondria, ultimately resulting in oxidative A sign of brain aging is neuroinflammation [57]. In damage. The MPTP neurotoxin model is a common particular, the expression of glial fibrillary acidic model used to study Parkinson’s disease [69]. In our protein (GFAP) by astrocytes increases with aging as experiment, 12-16-week old male wild-type and G3 astrogliosis and neuroinflammation occurs, and such an Tert-/- mice were injected IP (intraperitoneally) with 0.5, increase is also observed in mouse models with in- 5, 10, or 20 mg/kg MPTP and sacrificed 7 days later creased neuroinflammation [58–60]. The increase in (scheme in Figure S2A). An additional injection was GFAP expression accompanies increased expression of administered 2 hours later since multiple injections have inflammatory cytokines, accumulation of proteotoxic been suggested by previous protocols [69]. A footprint aggregates, and senescence [58]. To address, whether test resulted in a trend showing that wild-type and G3 physiological aging and/or short telomeres in the Tert-/- mice injected with the MPTP had a shorter hind context of telomerase deficiency lead to increased paw stride length (Figure S2B). In addition, we per- inflammation in the brain, we measured the number of formed a tail suspension test to measure mobility. In immune astrocyte cells with strong expression of the particular, we hung the mice from the tail for a period of GFAP marker [61]. We observed that young wild-type 5 min and measured their immobility time. Mice that mice have low levels of GFAP in the hippocampus and are old, unhealthy, or more depressed, move less in the dentate gyrus, whereas old wild-type, young G3 Tert-/-, tail suspension test [70]. We found that wild-type mice and old G3 Tert-/- mice have more GFAP-positive exhibited a shorter immobility time than MPTP-treated astrocytes in these areas (Figure 3D). We did not find wild-type and G3 Tert-/- mice (Figure S2C). After 7 significantly more GFAP in the neocortex of G3 Tert-/- days, the mice were sacrificed, and the brain was mice compared to wild-type, but did find an increase of preserved with formalin. In order to investigate the GFAP with age in both generations (Figure 3E). dopaminergic neurons, we performed an immuno- fluorescence experiment for the tyrosine hydroxylase Another sign of brain aging is the accumulation of tau or marker. Tyrosine hydroxylase is important in dopami- abnormal phosphorylation of tau protein located nergic neurons since it converts the amino acid L- throughout the brain in various cell types such as neu- tyrosine to L-DOPA which is a precursor for dopamine rons, astrocytes, and oligodendrocytes, ultimately [71]. We found that 12-16-week old male mice injected resulting in aggregates and neurofibrillary tangles [62– with higher doses of MPTP had lower levels of tyrosine 64] . Indeed, tau has been associated with Alzheimer’s hydroxylase (Figure S2D). Interestingly, saline-treated because hyperphosphorylation of tau results in loss of its control G3 Tert-/- mice showed less tyrosine hydro- biological activity [65]. To address this in our mouse xylase than wild-type mice (Figure S2D), and G3 Tert-/- models, we measured protein aggregation by determining mice treated with a low dose of MPTP showed less ty- www.aging-us.com 2921 AGING Figure 3. Immunohistochemistry of doublecortin, GFAP, and tau in the brain. (A-F) The quantification and representative images of the immunohistochemistry for positive cells per field of view for (A-C) doublecortin, (D-E) glial fibrillary acidic protein (GFAP), and (F) p-Tau(Ser396) in brain regions such as the hippocampus, dentate gyrus, and the neocortex. The data is shown for young and old wild- type and G3 Tert-/- mice. Data represent the mean ±SE of analyzed mice within each group. For the histopathology results, the number of mice analyzed per group is indicated, as well as the number of fields of view, and the number of positive cells. The t-test was used for statistical analysis. *p<0.05; **p<0.01; ***p<0.001. www.aging-us.com 2922 AGING rosine hydroxylase than similarly treated wild-type mice type and G3 Tert-/- mice by using positron emission (Figure S2D). tomography (PET) to detect fluorodeoxyglucose (FDG) in the brain after injection. Brain glucose metabolism Additionally, old wild-type and old G3 Tert-/- mice had has been found to decrease in human patients with lower levels of tyrosine hydroxylase than young mice Alzheimer’s disease and during senile dementia [73– (Figure S2E). Together these results suggest that G3 75]. We found that G3 Tert-/- mice had a lower standard Tert-/- mice are more susceptible to damage to glucose uptake value (SUV) in the brain compared to dopaminergic neurons in the substantia nigra due to wild-type brains, which is an indication of lower MPTP neurotoxin, and this damage results in decreased metabolic activity (Figure S3E). The weights of the mobility. mice were taken into account when calculating the SUV. Thus, telomerase deficiency and shorter telomeres Behavioral defects associated to shorter telomeres in resulted in decreased glucose metabolism in the brain, the brain of telomerase-deficient mice which is associated with poorer cognitive performance and increased neurodegeneration. We previously showed that old wild-type mice perform more poorly than young mice in tests such as the object Together, these findings show that Tert-/- mice with recognition test, rotarod test, and tightrope test [23], short telomeres exhibit more neurodegeneration than suggesting a negative impact of telomere shortening wild-type mice, suggesting an impact of telomere length associated with aging in these tests. Thus, we here set on brain aging in mice. to address whether telomerase deficient mice performed more poorly than age-matched wild-type controls in AAV9-mediated telomerase transduction of mouse different behavioral and cognitive tests. The tightrope brains test measures the ability of mice to balance on a tightrope, and the rotarod test measures neuromuscular Next, we set to study whether expression of telomerase coordination while running on a rotating rod. In in these mouse models could ameliorate the signs of addition, we performed a tail suspension test to measure neurodegeneration described above. Previous reports mobility as described previously in this text. We found have shown that adeno associated viruses of the that late generation G1, G2, G3, and G4 Tert-/- mice serotype 9 (AAV9) are able to cross the blood brain performed worse than age-matched wild-type mice in barrier and transduce a significant percentage of cells the tightrope test and rotarod test at older ages (Figure [76–84]. To verify whether in our experimental setting S3A,B). Furthermore, we found that older G2 Tert-/- an IV (intravenous) injection of the AAV9 vector could mice are immobile in the tail suspension test for longer cause effective expression of the Tert transgene in the periods of time than age-matched wild-type mice and brain, we injected 8-week old mice with the AAV9- this was further aggravated in the G3 and G4 Tert-/- CMV-Tert, AAV9-CMV-eGFP, or AAV9-CAG-eGFP mice (Figure S3C), suggesting a negative impact of vector, and the mice were sacrificed after 2 weeks of progressively shorter telomeres in this test. Thus, treatment to study expression of the eGFP or Tert telomerase deficiency and shorter telomeres lead to transgenes (scheme in Figure 4A). The “CMV” significantly impaired performance in several tests that abbreviation corresponds to the cytomegalovirus measure neuromuscular coordination and balance. promoter, and the “CAG” abbreviation corresponds to the high-expression synthetic CAG promoter [85]. We To further assess a potential impact of telomerase first determined whether AAV9 carrying either eGFP or deficiency on cognitive function, we subjected the Tert genes caused gene expression in the mouse brain different mouse cohorts to an object recognition test. upon intravenous (IV) injection in the tail with a dose of We found that G4 Tert-/- mice spent less time 2E12 viral genomes (vg) per mouse (see Methods). To investigating a novel object in their environment than this end, 8-week old mice were injected with AAV9 age-matched wild-type mice (Figure S3D), which is an virus particles containing either Tert or eGFP (Figure indication of poor memory in G4 Tert-/- mice with 4A) and under either the CMV promoter [23–25], or the shorter telomeres [72]. Together, these findings indi- CAG promoter. Mice treated with the different viral cate that mice with shorter telomeres show noticeable vectors were sacrificed 2 weeks after tail injection and behavioral phenotypes consisting of a poorer per- several different analyses were performed. We formance in behavioral and cognitive tests. performed quantitative qPCR to detect Tert and eGFP mRNA expression in the brain (Figure 4B). We found Finally, we set to analyze the brain metabolic activity in expression of the eGFP gene in the brains of AAV9- the different mouse cohorts as another indication of eGFP treated mice but not in brains from the AAV9- brain defects in these mice. To this end, we compared Tert treated mice (Figure 4B). In turn, the Tert mRNA differences in metabolic activity in the brains of wild- was only detected in the brain of AAV9-Tert treated www.aging-us.com 2923 AGING Figure 4. Experiment scheme and confirmation of virus transduction in the brain. (A) Scheme of the gene therapy experiment. Wild-type mice (8-week old) were injected IV with AAV9-CMV-Tert, AAV9-CMV-eGFP, or AAV9-CAG-eGFP, and sacrificed 2 weeks after injection. (B) Level of eGFP and Tert mRNA relative to actin in the brain as measured by qPCR 2 weeks post IV tail injection with 2E12 vg of AAV9-CMV-Tert, AAV9-CMV-eGFP, AAV9-CAG-eGFP, or no virus. (C) Quantification and representative images of fluorescence as measured by an IVIS instrument in the body and brain of wild-type mice and mice injected IV in the tail with AAV9-CMV-eGFP. (D) Quantification and representative images of eGFP positive cells in the brain as measured by immunohistochemistry in mice injected in the tail with 2E12 vg www.aging-us.com 2924 AGING AAV9-CAG-eGFP or no virus. The percentage of eGFP positive cells was calculated from the whole coronal brain cross-section. The representative images show multiple regions throughout the brain as labelled. (E) Quantification and representative images of eGFP positive cells in the brain as measured by immunofluorescence in mice injected IV in the tail with 2E12 vg AAV9-CMV-eGFP, AAV9-CAG- eGFP, or no virus. The percentage of eGFP positive cells was calculated from the whole coronal brain cross-section. The representative images show multiple regions throughout the brain as labelled with DAPI stained nuclei in blue, eGFP in green, and Iba1 for microglia in red. (F) Scheme of experiment with injection of AAV9-Tert into young mice. The mice (wild-type, G3 Tert-/-, and G4 Tert-/- mice) were treated at a young age (27-30 weeks) by the IV tail injection of 2E12 vg of AAV9-Tert or AAV9-null virus. (G) The mRNA level of Tert in the brain at the humane endpoint. The mRNA level was measured by qPCR relative to GAPDH. The ages of the mice as well as the number of weeks of treatment for each group are indicated above the graph. Data represent the mean ±SE of analyzed mice within each group. The number of mice analyzed per group is indicated. The t-test was used for statistical analysis. *p<0.05; **p<0.01. mice but not in those treated with AAV9-eGFP (Figure 0.4% as detected by immunofluorescence (Figure 4E). 4B). We also measured eGFP fluorescence in the whole This provides a range for the transduction efficiency mouse, including the brain, using an IVIS instrument which is in approximately the same range reported by (Figure 4C). AAV9-eGFP treated mice displayed other studies using different promoters [77,88]. We detectable eGFP fluorescence in the body as well as in also addressed whether AAV9 transduced microglia the brain, whereas untreated wild-type control mice which express the marker Iba1. However, we did not only displayed background levels of fluorescence detect significant eGFP expression in the Iba1-positive (Figure 4C). Thus, these results confirmed that AAV9 microglia cells, also in agreement with previous reports viruses carrying either the eGFP or Tert genes were [77,89]. able to cross the blood-brain barrier and infect cells in the brain. Next, to address whether telomerase over-expression in the adult brain of mice could ameliorate signs of brain Note that the AAV9-CMV-eGFP virus should infect the damage and neurodegeneration in mice, age-matched same number of cells as the AAV9-CAG-eGFP since (27-30-weeks of age) wild-type, G3 Tert-/-, and G4 the viral capsid is the same. However, the CMV Tert-/- mice were treated with an IV tail injection of promoter results in lower levels of expression, and 2E12 vg of AAV9-Tert or AAV9-null vectors (scheme therefore we are unable to detect the AAV9-CMV- in Figure 4F). The mice were then followed throughout eGFP infected cells by immunohistochemistry or immu- their lifespan until they reached the humane endpoint at nofluorescence. Therefore, the AAV9-CAG-eGFP which time we performed a number of molecular vector was utilized to locate infected cells and determine determinations. First, and in agreement with transduc- the number of cells that were infected. We could not use tion of mouse brain, qPCR analysis demonstrated an AAV9-CAG-Tert vector because the genetic material increased Tert mRNA expression in the brains of mice is too large to efficiently pack into the viral vector. transduced with the AAV9-Tert vectors compared to those transduced with the null-vectors at the time of Next, in order to visualize eGFP expression in different death (Figure 4G). Thus, increased Tert expression was brain areas by immunohistochemistry and immuno- maintained throughout the lifespan of the mice even fluorescence, we used the AAV9-CAG-eGFP vector, until the humane endpoint. A graph of the body weights previously shown to have a very high expression level of the mice is presented in Figure S4. in the brain ([86,87]; see also Figure 4C). Upon tail injection, of AAV9-CAG-eGFP vectors, we used AAV9-Tert gene therapy results in less signs of immunohistochemistry to detect eGFP positive cells molecular aging in the brain in mice throughout the entire brain, including the dentate gyrus, CA1 of the hippocampus, CA3 of the hippocampus, First, we assessed the presence of DNA damage at the subventricular zone (SVZ) of the lateral ventricle, the humane endpoint using the γH2AX marker. We neocortex, and the thalamus, which showed the highest confirmed more cells with DNA damage with increasing eGFP expression (Figure 4D). Transduction of brain Tert-/- generations compared to wild-type mice in the cells was also confirmed by immunofluorescence using hippocampus, the dentate gyrus, and the neocortex of an antibody against GFP, including all the brain areas mice treated with AAV9-null vectors (Figure 5A,B). studied (Figure 4E). We observed that the transduction Interestingly, we observed significantly lower numbers efficiency of all cells in a whole brain cross-section of cells with DNA damage for wild-type, G3 Tert-/-, and from a position in the brain at the hippocampus level G4 Tert-/- mice groups treated with AAV9-Tert in the was approximately 2.5% as detected by hippocampus, dentate gyrus, and neocortex compared to immunohistochemistry (Figure 4D) and approximately the controls treated with the AAV9-null vectors (Figure www.aging-us.com 2925 AGING 5A,B). We also observed reduced levels of the senes- ing less DNA damage in the telomerase-treated brains. cence and DNA damage marker Trp53 mRNA in the We also addressed whether telomerase treatment had AAV9-Tert treated G4 Tert-/- mice compared to the any impact on the regenerative neurons of the brain by AAV9-null treated controls (Figure 5C), again suggest- quantifying cells positive for the neurogenesis marker Figure 5. Treatment with AAV9-Tert results in less DNA damage in the brain. (A-B) Quantification and representative images of the histopathology for γH2AX in the (A) hippocampus, (A) dentate gyrus, and (B) neocortex for the cohort of mice (wild-type, G3 Tert-/-, and G4 Tert-/- mice) injected with 2E12 vg of AAV9-Tert or AAV9-null. (C) The level of Trp53 mRNA in the brain as measured by qPCR relative to GAPDH at the humane endpoint. Data represent the mean ±SE of analyzed mice within each group. The number of mice analyzed per group is indicated. The t-test was used for statistical analysis. *p<0.05; **p<0.01; ***p<0.001. www.aging-us.com 2926 AGING Figure 6. Treatment with AAV9-Tert results in more neurogenesis and less inflammation in the brain. (A-B) Quantification and representative images of the histopathology for doublecortin in the hippocampus (A), dentate gyrus (A), and neocortex (B). (C) Immunofluorescence of tyrosine hydroxylase in the substantia nigra. (D-E) Quantification and representative images of the histopathology for glial fibrillary acidic protein (GFAP) in the hippocampus (D), dentate gyrus (D), and (E) neocortex. Data represent the mean ±SE of analyzed mice within each group. The number of mice analyzed per group is indicated. The t-test was used for statistical analysis. *p<0.05; **p<0.01; ***p<0.001. www.aging-us.com 2927 AGING doublecortin (Figure 6A,B). As expected, in the AAV9- ment, the performance of most of the AAV9-Tert null treated controls, we found significantly lower groups was better than the performance of the corres- numbers of cells positive for doublecortin in the hippo- ponding AAV9-null groups (Figure 7). The number of campus and dentate gyrus of G4 Tert-/- mice compared to holes that the mice searched before finding the goal box the wild-type mice (Figure 6A,B). Interestingly, AAV9- is shown in Figure 7A. A chi-squared test showed that Tert treatment resulted in significantly increased numbers the mice treated with AAV9-Tert required less time to of doublecortin-positive cells in the hippocampus, the find the goal box hidden beneath one of the 20 holes on dentate gyrus, and the neocortex in G4 Tert-/- mice the outer edge of the table compared with the compared to those treated with AAV9-null vectors corresponding AAV9-null treated controls (Figure 7B). (Figure 6A,B). Differences were not observed between The graphs for the path length used to find the goal box the wild-type groups, which died at a very old age (Figure 7C), the percent of the holes searched in the (Figure 6A,B). target quadrant of the table (Figure 7D), and the percent time spent searching in the target quadrant (Figure 7E) We next evaluated dopaminergic neurons responsible are also shown. Additionally, G3 and G4 Tert-/- mice for motor control and important in Parkinson’s disease performed more poorly in this Barnes maze test than the by quantifying the tyrosine hydroxylase marker (Figure wild-type controls. We also constructed representative 6C). We found that G3 Tert-/- mice treated with AAV9- paths taken by the mice in different treatment groups Tert had a higher intensity of tyrosine hydroxylase (Figure 7F). These paths were obtained from videos of fluorescence than those treated with the AAV9-null the mice performing the test (Figure 7F). In order to vector (Figure 6C). address statistical significance in this test for behavior and memory, a chi-squared test was performed by Finally, we determined the number of GFAP-positive counting the number of AAV9-Tert test trials in which cells which is indicative of neuroinflammation. We the value was better than the average value for the found significantly lower levels of cells positive for corresponding AAV9-null group of the same generation. GFAP in the hippocampus, dentate gyrus, and neocortex These counts were compared to the number of trials that in G3 Tert-/- mice treated with the AAV9-Tert virus would be expected to be better by chance if the AAV9- compared to the AAV9-null treated controls (Figure Tert therapy had no effect, which is half of the total 6D,E). Thus, the AAV9-Tert therapy shows a trend number of trials (Table 1). The p-value from this chi- towards reducing the level of inflammation in the brain, squared test for the time variable was significant with a which reaches statistical significance between some of value of 0.00876, in agreement with the fact that nearly the groups. all the AAV9-Tert trials had a better value than the corresponding AAV9-null group average value, with a In summary, the fact that we observed fewer cells with few exception trials. A similar chi-squared test also DNA damage, higher levels of neurogenesis, less showed that younger AAV9-null mice (approximately inflammation, and more tyrosine hydroxylase in dopa- 30-week old mice after 4 weeks from the injection with minergic neurons indicates that the AAV9-Tert gene the AAV9-null virus) performed better than older therapy had an impact on these molecular markers of AAV9-null mice (approximately 60-week old mice after aging in the brain. 30 weeks from the injection with the AAV9-null virus) as shown in Table 2, supporting the fact that neuro- AAV9-Tert gene therapy has beneficial effects on degeneration naturally occurs with age. memory DISCUSSION In order to assess whether telomerase treatment had any impact on brain function, we measured the memory of Mouse models of Parkinson’s and Alzheimer’s disease the mice using the Barnes maze test [90]. In the Barnes have been generated in the past, however, many of these maze test, mice are placed onto a table with 20 holes mouse models have failed to reproduce the complexity around the edge. Only one of the holes contains a box of the human pathologies [69,91–93], most likely which the mice can enter. The mice try to escape the because these diseases are normally associated with aversive stimulus of a wide-open space where they are aging, and the aging process may not have been fully exposed, and they try to find and enter the goal box for modeled in the tested mouse models. One of the safety. The mice are first guided to the location of the hallmarks of aging is the progressive shortening of box with a glass beaker on the first habituation day. telomeres, which are the end-capping structures of They are then allowed to find the box on their own for chromosomes and essential for chromosomal stability several trials for two training days. After this there is [11]. When telomeres reach a critically short length this one day of rest, and then the performance of the mice is is sufficient to trigger a persistent DNA damage measured on the final test day. At 30 weeks of treat- response, and to trigger cellular senescence and/or www.aging-us.com 2928 AGING apoptosis [94]. Short telomeres also impair the ability of (Senescence Accelerated Mouse-Prone 8) mouse model stem cells to regenerate tissues, thus contributing to [96], which exhibits features of accelerated aging, signs tissue aging [12,13]. Here, we set to test whether a of neurodegeneration at an early age, and poor perfor- mouse model of premature aging owing to the presence mance in memory tests [96], or in mice injected with of short telomeres showed signs of molecular brain amyloid beta peptide, which induces cognitive defects aging similar to those that occur with physiological [97]. In particular, the increased inflammation often aging. In this regard, we describe here that wild-type observed in these mouse models, may lead to the mice show signs of neurodegeneration with aging and proliferative exhaustion of immune cells in the brain that these signs are anticipated in the context of the such as microglia, and ultimately shorter telomeres. telomerase-deficient mouse model. These findings Here, we observed increased neurodegeneration with age suggest that telomere shortening may be one of the in both old wild-type mice and in G3 Tert-/- mice with determinants of brain aging. Future studies warrant short telomeres. Indeed, old age is the main risk factor analysis of telomere length in other models of Alzhei- for many diseases, including neurodegenerative diseases mer’s disease, such as including the APP/PS1 mouse [98–100]. Telomerase-deficient G3 Tert-/- mice had model, expressing mutated amyloid precursor protein smaller brains and shorter telomeres in the brain, (APP) and mutated presenilin (PS1) [95], the SAMP8 and this was coincidental with increased brain aging as Figure 7. Treatment with AAV9-Tert improves memory in the Barnes maze test. (A-E) Quantification of the performance in the Barnes maze test after 30 weeks of treatment for (A) number of holes searched before reaching the goal box, (B) time required to reach the goal box, (C) path length used to reach the goal box, (D) the percent of holes searched in the target quadrant, and (E) the percent time spent searching in the target quadrant. (F) Representative paths of the mice on the Barnes maze table after 30 weeks of treatment. Data represent the mean ±SE of analyzed mice within each group. The number of mice analyzed per group is indicated. www.aging-us.com 2929 AGING indicated by a number of molecular markers. In some tion in these markers already reached the maximum or instances, however, the differences between wild-type minimum level at older ages, and therefore a and the G3 Tert-/- mice were lost at old ages. A difference between the wild-type and G3 Tert-/- mice possible explanation for this finding is that the altera- was not observed. Table 1. Barnes maze chi-squared analysis: Comparison of AAV9-Tert and AAV9-null mice 30 weeks after treatment. Number of holes searched Trials with performance Trials Number Group Total trials better than corresponding expected of Mice AAV9-null group average by chance WT AAV9-Tert 5 15 5 7.5 Chi-squared G3 Tert-/- AAV9-Tert 5 14 12 7 p-value G4 Tert-/- AAV9-Tert 5 15 7 7.5 0.109 Time Trials with performance Trials Number Group Total trials better than corresponding expected of Mice AAV9-null group average by chance WT AAV9-Tert 5 15 6 7.5 **Chi-squared G3 Tert-/- AAV9-Tert 5 14 13 7 p-value G4 Tert-/- AAV9-Tert 5 15 13 7.5 0.00876 Path length Trials with performance Trials Number Group Total trials better than corresponding expected of Mice AAV9-null group average by chance WT AAV9-Tert 5 15 9 7.5 Chi-squared G3 Tert-/- AAV9-Tert 5 14 12 7 p-value G4 Tert-/- AAV9-Tert 5 15 10 7.5 0.0951 Percent holes in target quadrant Trials with performance Trials Number Group Total trials better than corresponding expected of Mice AAV9-null group average by chance WT AAV9-Tert 5 15 3 7.5 Chi-squared G3 Tert-/- AAV9-Tert 5 14 6 7 p-value G4 Tert-/- AAV9-Tert 5 15 7 7.5 0.237 Percent time in target quadrant Trials with performance Trials Number Group Total trials better than corresponding expected of Mice AAV9-null group average by chance WT AAV9-Tert 5 15 4 7.5 Chi-squared G3 Tert-/- AAV9-Tert 5 14 8 7 p-value G4 Tert-/- AAV9-Tert 5 15 6 7.5 0.354 Chi-squared tables to calculate statistical significance for the mice after 30 weeks of treatment comparing AAV9-Tert treated mice with their corresponding controls of the same treatment age and generation www.aging-us.com 2930 AGING Table 2. Barnes maze chi-squared analysis: comparison of young and old AAV9-null mice 30 weeks after treatment. Number of holes searched Trials with performance better Number Total Trials expected Group than corresponding of Mice trials by chance AAV9-null group average WT AAV9-Tert 5 15 12 7.5 Chi-squared G3 Tert-/- AAV9-Tert 4 12 9 6 p-value G4 Tert-/- AAV9-Tert 5 15 11 7.5 0.0541 Time Trials with performance better Number Total Trials expected Group than corresponding of Mice trials by chance AAV9-null group average WT AAV9-Tert 5 15 13 7.5 **Chi- G3 Tert-/- AAV9-Tert 4 12 11 6 squared G4 Tert-/- AAV9-Tert 5 15 11 7.5 p-value 0.00732 Path length Trials with performance better Number Total Trials expected Group than corresponding of Mice trials by chance AAV9-null group average WT AAV9-Tert 5 15 12 7.5 *Chi-squared G3 Tert-/- AAV9-Tert 4 12 11 6 p-value G4 Tert-/- AAV9-Tert 5 15 9 7.5 0.0278 Percent holes in target quadrant Trials with performance better Number Total Trials expected Group than corresponding of Mice trials by chance AAV9-null group average WT AAV9-Tert 5 15 12 7.5 Chi-squared G3 Tert-/- AAV9-Tert 4 12 9 6 p-value G4 Tert-/- AAV9-Tert 5 15 10 7.5 0.0807 Percent time in target quadrant Trials with performance better Number Total Trials expected Group than corresponding of Mice trials by chance AAV9-null group average WT AAV9-Tert 5 15 10 7.5 **Chi- G3 Tert-/- AAV9-Tert 4 12 12 6 squared G4 Tert-/- AAV9-Tert 5 15 12 7.5 p-value 0.00851 www.aging-us.com 2931 AGING Preventing accumulation of short telomeres may also treatments with specific factors in young blood such prevent or ameliorate brain aging by allowing stem cells as the GDF11 protein [114,115]. Nevertheless, even to proliferate and regenerate damaged tissue. We have more benefits from telomerase gene therapy may be previously demonstrated that preventing accumulation observed if higher transduction efficiencies are obtained. of short telomeres through telomerase gene therapy can ameliorate the symptoms of cardiovascular disease [25], MATERIALS AND METHODS pulmonary fibrosis [26], aplastic anemia [24], and aging in general [23]. Mice Thus, to demonstrate that telomere shortening may be For all the treatment groups, mice were injected IV in one of the causes of brain aging, here we studied the the tail with 2E12 vg of AAV9 virus particles potential therapeutic effects of a telomerase gene containing the genetic material of interest in a volume therapy in ameliorating molecular signs of neuro- of 100 μL 0.001% Pluronic F-68 in PBS. The Pluronic degeneration associated with physiological mouse aging F-68 serves as a surfactant to prevent aggregations of as well as in the context of the telomerase-deficient the virus particles from forming [77]. In the experiment mouse model. There have been many successes for gene to confirm infection of mouse brain cells, mice were therapy in the last decade [101–107], and the first gene injected with either AAV9-CMV-eGFP, AAV9-CAG- therapy was licensed for human use in Europe in 2012 eGFP, or AAV9-CMV-Tert. In the neurodegeneration (Glybera) [108]. Viral vectors have also been used to aging study, the mice were injected with either AAV9- transduce a wide variety of tissues from liver, skeletal Tert or AAV9-null. The mice were injected at an age of muscle, heart, brain, lung, pancreas, to tumors [109]. 27-30 weeks. The mice were genetically wild-type or Our findings demonstrate that AAV9-Tert treatment can knockout for telomerase. Throughout these studies, the ameliorate signs of neurodegeneration with aging in telomerase knockout mice belonged to different wild-type mice as well as in the context of the generations: either G1, G2, G3, or G4 Tert-/- genera- telomerase-deficient mouse model with the presence of tions. The mouse strain was >95% C57BL/6 back- short telomeres. Our treatment was applied through an ground, and all of the mice used in these experiments IV tail injection, and therefore, many other cell types were male. throughout the body would be infected in addition to the cells in the brain. Improvements of health in other All mice were produced and housed in the specific organs may have an impact on the brain and invest- pathogen-free barrier of the CNIO institution in Madrid tigating the nature of this relationship could be Spain. After weaning, five mice were housed per cage interesting for future studies. Note also that we did not and fed ad libitum with a non-purified sterilizable observe any increased incidence of cancer in the mice Teklad 2018 18 percent protein rodent diet (Harlan treated with AAV9-Tert, which matched our expecta- TD.2018S). All animal procedures were approved by tions since several other articles have demonstrated that the CNIO-ISCIII Ethics Committee for Research and telomerase reactivation alone does not lead to Animal Welfare (CEIyBA) and conducted in accor- tumorigenesis in vivo [23,39,110,111]. dance to the recommendations of the Federation of European Laboratory Animal Science Associations. Of note, the AAV9 serotype used here to express telomerase in the brain primarily transfects neurons and MPTP Neurotoxin treatment astrocytes but fails to transduce microglia [77]. In our experimental setting, we found that less than 5% of the Wild-type or G3 Tert-/- 12-16-week old male mice were cells in the brain received the transgene using our vector injected IP with saline or MPTP neurotoxin at a and delivery method. Interestingly, in spite of the low concentration of 0.5, 5, 10, or 20 mg/kg body weight. transduction efficiency, we observed significant effects of Behavior tests such as the footprint test and tail AAV9-Tert gene therapy in decreasing DNA damage, suspension test were then performed. The mice were increasing neurogenesis as indicated by increased double- sacrificed in a CO2 chamber 7 days after injection, and cortin expression, as well as decreasing neuroinflam- the brain was preserved in formalin for subsequent mation (decreased GFAP expression). These findings molecular tests. suggest that even a small number of neurons transduced with Tert may increase the health of the environment and Recombinant AAV vectors benefit cells that were not infected, for instance, through changing the secretory profile of cells. In an analogous Vectors were generated by triple transfection of manner, factors found in young blood induce a younger HEK293 cells as described previously [116]. Cells were phenotype in the recipient cells, as observed from cultured in roller bottles (Corning) in DMEM 10% FBS parabiosis experiments with young blood [112,113] and to 80% confluence and co-transfected with a plasmid www.aging-us.com 2932 AGING carrying the expression cassette flanked by the AAV2 AAV9-null, and 5 G4 Tert-/-AAV9-Tert mice. After the viral ITRs, a helper plasmid carrying the AAV rep2 and injection, all the mice were allowed to age until the cap9 genes, and a plasmid carrying the adenovirus humane endpoint. Behavioral and cognitive tests such helper functions (kindly provided by K.A. High, as the Barnes maze test, object recognition test, rotarod Children’s Hospital of Philadelphia). The expression test, tightrope test, and tail suspension test were cassettes used were one of the following: eGFP under performed at various timepoints throughout the lifespan the control of the CMV promoter, eGFP under the of the mice. At the humane endpoint, the mice were control of the CAG promoter, or murine Tert under the sacrificed in a CO2 chamber. Half of the brain was control of the CMV promoter. AAV9 vectors were preserved in nitrogen for freezing, and half the brain purified with an optimized method based on two was preserved in formalin for paraffin sections. The consecutive cesium chloride gradients, dialyzed against paraffin sections cut in a coronal orientation were used PBS, filtered, and stored at -80 °C until use [117]. The for q-FISH to measure telomere length, and for titer of the viral genome particles was determined by immunohistochemistry to measure the markers γH2AX, quantitative real time PCR. doublecortin, and GFAP. Confirmation of virus infection and measurement of Telomere Q-FISH analysis of paraffin sections transduction efficiency Paraffin-embedded tissue sections were hybridized with Groups of 8-week old mice were IV tail injected with a PNA-telomeric Cy3 probe, and fluorescence intensity 2E12 vg of AAV9 vectors containing different pieces of of telomeres was determined as described previously genetic material in 100 μL of 0.001% Pluronic F-68 in [118,119]. After deparaffinization, tissues were post PBS. The mouse group size and identity of the injected fixed in 4% formaldehyde for 5 min, washed 3X5 min vectors were as follows: 3 mice AAV9-CMV-Tert, 3 in PBS and incubated at 37 °C for 15 min in pepsin mice AAV9-CMV-eGFP, 2 mice AAV9-CAG-eGFP, solution (0.1% Porcine Pepsin, Sigma; 0.01 M HCl, and 3 control mice receiving no vector. At 2 weeks Merck). After repeating another round of washes and post-injection, the mice were sacrificed, and IVIS was fixation, slides were dehydrated in a 70%–90%–100% used to measure the fluorescence in the AAV9-CMV- ethanol series (5 min each). After air drying the slides, eGFP mice and controls. Then half the brain was 30 μL of telomere probe mix (10 mM TrisCl pH 7, 25 preserved in liquid nitrogen for freezing, and half the mM MgCl2, 9 mM citric acid, 82 mM Na2HPO4, 70% brain was preserved in formalin for paraffin sections. deionized formamide (Sigma), 0.25% blocking reagent RNA was extracted from frozen tissue, and a qPCR was (Roche) and 0.5 μg/mL Telomeric PNA probe performed to check the level of the Tert and eGFP (Panagene)) were added to each slide. A cover slip was applied, and the slides were incubated for 3 min at 85 expression. The paraffin sections were used in an °C. The slides were then incubated 2 hr at room immunohistochemistry test for GFP, and for an immu- temperature in a wet chamber in the dark. Slides were nofluorescence experiment for GFP and the microglia washed 2X15 min in 10 mM TrisCl pH 7, 0.1% BSA in marker Iba1. The GFP antibody used in immuno- 70% formamide with shaking, then 3X5 min in TBS histochemistry was mouse monoclonal anti-GFP (Roche 0.08% Tween 20, and then incubated in a 4 μg/mL 4',6- Cat No 11 814 460 001) as described in more detail in diamidino-2-phenylindole DAPI (Sigma) bath in PBS the immunohistochemistry section. The GFP antibody before mounting samples in Vectashield (VectorTM). used in immunofluorescence was chicken anti-GFP Confocal images were acquired as stacks using an SP5- (Aves Labs Inc. Cat No GFP-1020), and the Iba1 WLL confocal microscope (Leica) and maximum antibody used was rabbit anti-Iba1 (Wako Cat No 019- projection images were created using the Fiji version of 19741). the ImageJ software (NIH) [120,121]. Telomere signal intensity was quantified using Definiens Developer Cell Neurodegeneration aging experiment software (version XD 64 2.5; Definiens AG). To assess whether AAV9-Tert gene therapy could Immunohistochemistry reduce neurodegeneration with age, mice were injected with either AAV9-null or AAV9-Tert virus. The mice Brains were fixed in 10% neutral buffered formalin with were injected IV in the tail with 2E12 vg of virus in a 4% formaldehyde, paraffin-embedded, cut at 3 µm, volume of 100 μL 0.001% Pluronic F-68 in PBS. The mounted in Superfrost Plus slides (Thermo Fisher mice were injected at an age of 27-30 weeks. The Scientific), and dried overnight. The paraffin brain cohort consisted of the following mice: 5 wild-type sections were cut in a coronal orientation and a series of AAV9-null, 5 wild-type AAV9-Tert, 4 G3 Tert-/- sections from the posterior part of the brain to the AAV9-null, 5 G3 Tert-/- AAV9-Tert, 5 G4 Tert-/- anterior part of the brain were placed onto a slide. For www.aging-us.com 2933 AGING different staining methods, slides were deparaffinized in Cat No 019-19741). Next the slides were washed 4X7 xylene and re-hydrated through a series of graded min with 1XPBS 0.1% Tween 20 at RT, washed with ethanol washes until a final wash with water. 1XPBS 5 min, and incubated with secondary antibody 1 Consecutive sections were stained with an automated hr at RT in a humidity chamber. The secondary Ventana Discovery XT immunostaining platform antibodies used were one of the following: goat anti- (Roche). Antigen retrieval was first performed with rabbit Cy3 (Jackson Cat No 4134) or goat anti-chicken high pH CC1m buffer (Roche) and endogenous pero- AF488 (Life Technologies Cat No A11039). Next the xidase was blocked with 3% hydrogen peroxide. Next slides were washed 3X10 min with PBS, incubated with the slides were incubated with one of the following DAPI 5 min at RT, washed 5 min with PBS, 30 μL of primary antibodies: pre-diluted rabbit monoclonal SP6 Vectashield (VectorTM) was added, and the slides were anti-Ki67 (Master Diagnostica Cat No 000311OQD), sealed with nail polish. Confocal images were acquired mouse monoclonal JBV301 anti-γH2AX at a 1/25,000 as stacks using an SP5-WLL confocal microscope dilution (Millipore Cat No 05-636), mouse monoclonal (Leica) and maximum projection images were created 7.1+13.1 anti-GFP at a 1/500 dilution (Roche Cat No using the Fiji version of the ImageJ software (NIH) 11.814.460.001), rabbit polyclonal anti-doublecortin at [120,121]. Fluorescence intensity was quantified using a 1/500 dilution (Abcam Cat No ab18723), rabbit Definiens Developer Cell software (version XD 64 2.5; polyclonal anti-glial fibrillary acidic protein (GFAP) at Definiens AG). a 1/750 dilution (DAKO Z0334), or rabbit monoclonal anti-tau (phospho S396) at a 1/2000 dilution (Abcam Quantitative real-time PCR Cat No ab10930). After the primary antibody step, slides were incubated with either the rabbit anti-mouse Total RNA from nitrogen frozen tissues was extracted secondary antibody (Epitomics) or the OmniMap anti- with Qiagen’s RNeasy mini kit (Qiagen Cat No 74106), rabbit detection system (Ventana Roche) conjugated according to the manufacturer’s instructions. Before with horseradish peroxidase. The immunohisto- processing, RNA samples were DNaseI treated using chemistry (IHC) reaction was developed using the RNase-free DNase Set (Qiagen Cat No 79254) ChromoMap 3,3'-Diaminobenzidine tetrahydrochloride according to the manufacturer’s instructions. First- DAB (Ventana Roche) and nuclei were counterstained strand cDNA was synthesized from this RNA using an with Carazzi’s hematoxylin. Finally, the slides were iScript cDNA synthesis kit (Bio-RAD Laboratories, Inc. dehydrated, cleared, and mounted with a permanent Cat No 1708891) according to the manufacturer’s mounting medium for microscopic evaluation. Positive instructions. The reaction consisted of 15 μL of 1 μg control sections known to be positive for the primary RNA, 4 μL of 5X iScript reaction mix, and 1 μL iScript antibody were included for each staining procedure. reverse transcriptase. A Primus 96 Plus thermocycle Images of whole slides were acquired with an AxioScan instrument was used with the following settings: 5 min Z1 slide scanner (Zeiss) and visually checked with the 25 °C; 30 min 42 °C; 5 min 85 °C; 4 °C hold. The PCR Zen Blue software (Zeiss). The digital images from the reactions were performed using the GoTaq qPCR scan were then quantified by counting the number of Master Mix (Promega Cat No A6002) according to the positive cells in different fields of view. Either Zen manufacturer’s instructions. The reaction mix consisted Blue software (Zeiss) or the Fiji version of the ImageJ of 6 μL GoTaq PCR master mix, 0.2 μL 10 uM forward software (NIH) [120,121] was used for this quan- primer, 0.2 μL 10 μM reverse primer, 2 μL of 1/10 tification. dilution of first-strand cDNA, and 3.6 μL H20. Quantitative real-time PCR was performed for these Immunofluorescence reactions using an ABI PRISM 7700 384 well plate reader (Applied Biosystems) according to the manu- For immunofluorescence (IF), the slides were first facturer’s instructions. The thermocycle program was prepared and antigen retrieval was performed as as follows: Hold stage (1.6 °C/s ramp; 50 °C 2 min; 1.6 described previously in the immunohistochemistry °C/s ramp; 95 °C 10 min); PCR Stage with 40 cycles section. Next the slides were washed 2X5 min in PBS. (95 °C 0:15 min; 1.6 °C/s ramp; 60 °C 1 min; 1.6 °C/s The slides were then permeabilized in 0.5% Triton X- ramp; 68 °C 1 min); Melt Curve Stage (1.6 °C/s ramp; 100 in PBS for 3 hr at room temperature (RT), washed 5 95 °C 0:15 min; 1.6 °C/s ramp; 60 °C 1 min; 0.05 °C/s min in 1X PBS, blocked with fetal bovine serum (FBS) ramp; 95 °C 0:15 min). The following primers were for 2 hr, and then the slides were incubated with used for PCR reactions: primary antibody overnight at 4 °C in a humidity chamber. The primary antibodies used were one of the TERTFor: 5’-GGATTGCCACTGGCTCCG-3’, following: rabbit polyclonal anti-tyrosine hydroxylase TERTRev: 5’-TGCCTGACCTCCTCTTGTGAC-3’, (Millipore Cat No AB152), chicken anti-GFP (Aves GFPFor: 5’-ACCCTGAAGTTCATCTGCA-3’, Labs Inc. Cat No GFP-1020), or rabbit anti-Iba1 (Wako GFPRev: 5’-GGACTTGAAGAAGTCGTGC-3’, www.aging-us.com 2934 AGING Trp53For: 5’-GTCACAGCACATGACGGAGG-3’, The object recognition test measures the ability of Trp53Rev: 5’-TCTTCCAGATGCTCGGGATAC-3’, mice to remember objects in their environment and mActFor: 5’-GGCTGTATTCCCCTCCATCG-3’, provides an index of recognition memory [72,126]. mActRev: 5’-CCAGTTGGTAACAATGCCATGT-3’, The mice were first presented with two identical GAPDHFor: 5’-AATGGCAGCCCTGGTGAC-3’, objects for 10 min. Then after 24 hours, the mice were GAPDHRev: 5’-AGACGGCCGCATCTTCTT-3’. presented with one object that was identical to the previous objects and one novel object. The amount of Micro PET imaging time the mouse spent investigating the new object by sniffing or touching the new object indicated that the Images were acquired using an eXplore Vista PET-CT mouse may have recognized that this object was new. instrument (GE Healthcare). Mice were injected with The final results were presented as the ratio of time the 15 MBq (megabecquerel) of 18F-FDG (ITP mouse spent investigating the new object to the total Cyclotron) into the lateral tail vein in a volume of 0.1 time investigating both objects. cc. During imaging, mice were anaesthetized with a continuous flow of 1% to 3% isoflurane/oxygen The tail suspension test evaluates psychological mixture (2 L/min) 45 min after radiotracer injection. depression, motivation, and energy levels in mice by MicroPET scans were performed for 20 min per bed. measuring the amount of time the mice spend trying to Only one mouse was used per bed. PET images were escape by struggling and moving when hung from their reconstructed with a three-dimensional Ordered Subsets tail [70]. Mice typically struggle less and move less Expectation-Maximization (OSEM) reconstruction with old age. Video recordings were captured for each algorithm and were analyzed using MMWS software mouse, and the amount of movement was then (eXplore Vista, GEHC) [122]. The images were quantified. More movement indicates less depression, analyzed by drawing a region of interest (ROI) and more motivation, and higher energy levels. The mice calculating the 18F-FDG SUV (standardized uptake were hung by the tail using a length of tape of value) using the following formula: SUV = Tissue approximately 17 cm. A plastic tube cut from a 10 mL concentration ROI activity (Mbq/ml) / (Injected dose pipette was placed around the tail before attaching the (Mbq) / Body weight (g)). tape in order to prevent the mice from climbing their own tails during the test. Videos of the mice were IVIS captured using a standard PowerShot SX430 camera (Canon) held stable by a STAR 75 tripod (Hama). The fluorescence intensity of the mice was measured Analysis of the video was performed by first converting immediately after sacrifice in a CO2 chamber. An IVIS the video into a sequence of images using the VLC Spectrum (Xenogen) instrument was used to measure VideoLAN Client (VideoLAN Organization) [127]. fluorescence, and the Living Image version 4.3.1 This image sequence was then analyzed with the Fiji (Caliper Life Sciences, Inc.) software was used for version of the ImageJ software (NIH) [120,121]. quantification. The fluorescence intensity was dis- Briefly, the color was inverted so that the white tape played with log scale and rainbow coloring with a min became black and the black background became white, value of 511 and a max value of 1205. The fluo- a z-projection of the average intensity was created to rescence intensity of the desired region of interest (ROI) identify where the mouse was located the majority of was then quantified. the time throughout the video sequence, a rectangular region of interest (ROI) was created around the region Behavioral and cognitive tests of tape close to the mouse, the ROI was restored onto the original image sequence (instead of the single frame Several behavioral and cognitive tests were performed z-projection), the integrated intensity of the ROI in each with the mice to measure health. The rotarod test was image was measured and output to a spreadsheet, and used to evaluate motor coordination and balance in mice the number of image frames in which the integrated [123]. The mice were tested in a rotarod apparatus density was greater than the threshold was determined. (Panlab model LE 8200) using a continuous These frames indicate that the mouse was moving since acceleration protocol from 4 to 40 rpm in a period of more white background is detected in these frames 120 s. The time before the mice fell was recorded, and instead of the black tape which would be in the frame if the average of three trials was used in the quantification. the mouse was not moving. The amount of time that the In the tightrope test [124,125], the mice were placed mouse was immobile was then displayed in a bar graph. onto a rod (“the tightrope”) for 5 trials for 60 s. Each This method of analysis correctly yielded long trial that the mouse did not fall was counted as a immobility times for mice which moved very little, and success, and the percent success for each mouse was short immobility times for mice that were moving determined. frequently. www.aging-us.com 2935 AGING The footprint test measures the stride length of the mice refers to the time for the mouse to first find the goal box and has been used to quantify mobility in Parkinson’s- hole (approach the hole and look into it), not actually like mouse models [128,129]. The paws of the mice enter the hole into the goal box. Sometimes the mice were painted with watercolor paint. The front and hind would find the hole, and then wait a considerable limbs were painted with a different color. The mice amount of time before actually entering the hole. The were then placed into a hallway with paper on the floor. videos were watched with the SMPlayer media player The hallway was 8 cm wide and 60 cm long. The mice (https://www.smplayer.info/) so that we could play the then moved through the hallway into a dark container at videos forward or backward frame by frame and at the end which they could hide in. Three trials were custom speeds such as twice the speed to assist with the performed with each mouse, and the distance between analysis. The SMPlayer window was also made the footprint paint marks on the paper was then transparent and placed over a diagram of the table with measured with a ruler. The hind paw stride length was ID numbers for each hole made in the Inkscape vector the average of the lengths for both the right and left graphics editor (https://inkscape.org). To determine paws, and the front paw stride length was measured in other variables such as path length, and the percent time the same manner. in the target quadrant, each video was first converted into an image sequence using the VLC VideoLAN The Barnes maze test is used to evaluate the learning Client (VideoLAN Organization) [127]. The image ability and spatial memory of mice [130–132]. The sequence was imported into the Fiji version of the mice are trained to find a hole leading to an escape box ImageJ software (NIH) [120,121], and analyzed. on a table with many other holes. Visual cues such as Briefly, regions of interest (ROIs) for the quadrants simple colored shapes around the room and normal were created. Any part of the image scene which was objects in the room were used so that the mice could very dark was cleared so that the black mouse was the orient themselves on the table. The training consisted darkest object in the image. The min value in each of a habituation day in which the mice were guided to quadrant for each frame was then measured. Frames the target box with a glass beaker from the center of the with a min value below a certain threshold were frames table, the first training day in which the mice were which contained the mouse, and the percent time the allowed to explore the table for 3 trials for 2 min, a mouse spent in each quadrant was determined. To second training day in which the mice were allowed to quantify the path length, the brightness of the image explore the table for 2 trials for 2 min, a resting day, and was increased to the max value, the contrast was then the final test day in which 3 trials of 2 min were increased until most of the artifacts in the image performed. Aversive stimuli such as a bright light disappeared and only the mouse could be seen. A thres- above the table from a flashlight and a digital hold was applied to the image so that only pure white metronome (Seiko Cat No DM50SE) which makes a and black colors remained. Then the MTrack2 plugin repetitive loud noise set at 132 bpm were used to [133] for Fiji was used to track the movement of the motivate the mouse to search for and escape into the mouse, quantify the path length, and also produce a goal box. The wide-open space on the table also serves trace image of the path taken by the mouse. The as an aversive stimulus since the mice are exposed to parameters necessary for the plugin would vary any threat. Note that the Barnes maze test is often depending on the conditions of the video, but in our considered to be more humane than the Morris water videos the following parameters usually produced maze. The final output of the test was represented as correct path tracings: Min Particle Size: 300, Max the time the mice took to find the target box, the path Particle Size: 10,000, Max Speed: 10,000, and length, the number of holes searched, the number of Minimum Track Length (frames): 5. The resulting holes searched in the target quadrant of the table, and variables of "number of holes searched", "path length", the time spent in the target quadrant of the table. "time", "percent time in target quadrant", and "percent Performance on this test is related to the health of the holes in target quadrant" were then graphed. hippocampus which is responsible for spatial memory. Videos of the Barnes maze test trials were analyzed to Statistical and mathematical analyses quantify the results. The maintenance facility of the CNIO constructed a goal post-like structure, and a An unpaired Student’s t-test was used to compare the standard PowerShot SX430 camera (Canon) was groups of mice for the brain weights, brain area, attached to the top of this structure with a screw above telomere spot fluorescence, histopathology, immuno- the table to video record the tests. To determine the fluorescence, PCR, footprint test, tightrope test, rotarod time required for each mouse to find the goal box, and test, tail suspension test, object recognition test, and to determine the hole ids visited before finding the goal PET test results since each individual mouse cannot box, the videos were simply watched, and the in- belong to more than one group and the variables are formation was recorded. Note that the time recorded assumed to be normally distributed. A χ2 (chi-squared) www.aging-us.com 2936 AGING test was used for the Barnes maze test to determine 5. de Lange T. Shelterin: the protein complex that whether AAV9-Tert treated groups performed better shapes and safeguards human telomeres. Genes Dev. than their corresponding AAV9-null controls with a 2005; 19:2100–10. matching generation. This test was implemented since https://doi.org/10.1101/gad.1346005 we were comparing categorical variables, specifically PMID:16166375 whether a mouse performed better or worse than the 6. Greider CW, Blackburn EH. Identification of a specific AAV9-null group of the same generation. For statis- telomere terminal transferase activity in tical tests and mathematical analysis, we used the Excel Tetrahymena extracts. Cell. 1985; 43:405–13. 2016 software (Microsoft). https://doi.org/10.1016/0092-8674(85)90170-9 PMID:3907856 ACKNOWLEDGEMENTS 7. Canela A, Vera E, Klatt P, Blasco MA. High-throughput We would like to thank the CNIO (Centro Nacional de telomere length quantification by FISH and its Investigaciones Oncológicas or “Spanish National application to human population studies. Proc Natl Cancer Centre” in Madrid Spain) confocal microscope Acad Sci USA. 2007; 104:5300–05. core, histopathology core, particularly Alba de Martino https://doi.org/10.1073/pnas.0609367104 and Patricia González García, molecular imaging core, PMID:17369361 and animal facility for all of their help and assistance. 8. Flores I, Canela A, Vera E, Tejera A, Cotsarelis G, We also thank Manuel Valiente for suggestions and Blasco MA. The longest telomeres: a general advice. The viral vectors were produced by the lab of signature of adult stem cell compartments. Genes Fàtima Bosch. Fàtima Bosch is an ICREA Academia Dev. 2008; 22:654–67. recipient, Generalitat de Catalunya, Spain. https://doi.org/10.1101/gad.451008 PMID:18283121 CONFLICTS OF INTEREST 9. Harley CB, Futcher AB, Greider CW. Telomeres shorten during ageing of human fibroblasts. Nature. The authors have no conflicts of interest to declare. 1990; 345:458–60. https://doi.org/10.1038/345458a0 PMID:2342578 FUNDING 10. Blasco MA. Telomeres and human disease: ageing, cancer and beyond. Nat Rev Genet. 2005; 6:611–22. Research in the Blasco lab is funded by the Spanish https://doi.org/10.1038/nrg1656 PMID:16136653 Ministry of Economy and Competitiveness Projects 11. López-Otín C, Blasco MA, Partridge L, Serrano M, (SAF2013-45111-R and SAF2015-72455-EXP), the Kroemer G. The hallmarks of aging. Cell. 2013; Comunidad de Madrid Project (S2017/BMD-3770), the 153:1194–217. World Cancer Research (WCR) Project (16-1177), and https://doi.org/10.1016/j.cell.2013.05.039 the Fundación Botín (Spain). PMID:23746838 REFERENCES 12. Blasco MA. Telomere length, stem cells and aging. Nat Chem Biol. 2007; 3:640–49. 1. Scully T. Demography: To the limit. Nature. 2012; https://doi.org/10.1038/nchembio.2007.38 492: S2-3. https://doi.org/10.1038/492S2a PMID:17876321 PMID: 23222671 13. Flores I, Cayuela ML, Blasco MA. Effects of telomerase 2. Olshansky SJ, Ault AB. The fourth stage of the and telomere length on epidermal stem cell behavior. epidemiologic transition: the age of delayed Science. 2005; 309:1253–56. degenerative diseases. Milbank Q. 1986; 64:355–91. https://doi.org/10.1126/science.1115025 https://doi.org/10.2307/3350025 PMID:3762504 PMID:16037417 3. Sarkar A, Irwin M, Singh A, Riccetti M, Singh A. 14. Povedano JM, Martinez P, Flores JM, Mulero F, Blasco Alzheimer’s disease: the silver tsunami of the 21(st) MA. Mice with pulmonary fibrosis driven by telomere century. Neural Regen Res. 2016; 11:693–97. dysfunction. Cell Reports. 2015; 12:286–99. https://doi.org/10.4103/1673-5374.182680 https://doi.org/10.1016/j.celrep.2015.06.028 PMID:27335537 PMID:26146081 4. Blackburn EH. Switching and signaling at the 15. Armanios M. Telomeres and age-related disease: how telomere. Cell. 2001; 106:661–73. telomere biology informs clinical paradigms. J Clin https://doi.org/10.1016/S0092-8674(01)00492-5 Invest. 2013; 123:996–1002. PMID:11572773 https://doi.org/10.1172/JCI66370 PMID:23454763 www.aging-us.com 2937 AGING 16. King TE Jr, Pardo A, Selman M. Idiopathic pulmonary E, Jimenez V, Formentini I, Bobadilla M, Mizrahi J, de fibrosis. Lancet. 2011; 378:1949–61. Martino A, Gomez G, Pisano D, Mulero F, et al. https://doi.org/10.1016/S0140-6736(11)60052-4 Telomerase expression confers cardioprotection in PMID:21719092 the adult mouse heart after acute myocardial 17. Armanios M, Blackburn EH. The telomere syndromes. infarction. Nat Commun. 2014; 5:5863. https://doi.org/10.1038/ncomms6863 Nat Rev Genet. 2012; 13:693–704. PMID:25519492 https://doi.org/10.1038/nrg3246 PMID: 22965356 26. Povedano JM, Martinez P, Serrano R, Tejera Á, 18. Holohan B, Wright WE, Shay JW. Cell biology of Gómez-López G, Bobadilla M, Flores JM, Bosch F, disease: Telomeropathies: an emerging spectrum Blasco MA. Therapeutic effects of telomerase in mice disorder. J Cell Biol. 2014; 205:289–99. with pulmonary fibrosis induced by damage to the https://doi.org/10.1083/jcb.201401012 lungs and short telomeres. eLife. 2018; 7:e31299. PMID:24821837 https://doi.org/10.7554/eLife.31299 PMID:29378675 19. Blasco MA, Lee HW, Hande MP, Samper E, Lansdorp 27. Lin J, Epel E, Blackburn E. Telomeres and lifestyle PM, DePinho RA, Greider CW. Telomere shortening factors: roles in cellular aging. Mutat Res. 2012; and tumor formation by mouse cells lacking 730:85–89. telomerase RNA. Cell. 1997; 91:25–34. https://doi.org/10.1016/j.mrfmmm.2011.08.003 https://doi.org/10.1016/S0092-8674(01)80006-4 PMID:21878343 PMID:9335332 28. Panossian LA, Porter VR, Valenzuela HF, Zhu X, 20. Lee HW, Blasco MA, Gottlieb GJ, Horner JW 2nd, Reback E, Masterman D, Cummings JL, Effros RB. Greider CW, DePinho RA. Essential role of mouse Telomere shortening in T cells correlates with telomerase in highly proliferative organs. Nature. Alzheimer’s disease status. Neurobiol Aging. 2003; 1998; 392:569–74. https://doi.org/10.1038/33345 24:77–84. https://doi.org/10.1016/S0197- PMID:9560153 4580(02)00043-X PMID:12493553 21. Herrera E, Samper E, Martín-Caballero J, Flores JM, 29. Thomas P, Callaghan NJ, Fenech M. Telomere length Lee HW, Blasco MA. Disease states associated with in white blood cells, buccal cells and brain tissue and telomerase deficiency appear earlier in mice with its variation with ageing and Alzheimer’s disease. short telomeres. EMBO J. 1999; 18:2950–60. Mech Ageing Dev. 2008; 129:183-90. https://doi.org/10.1093/emboj/18.11.2950 https://doi.org/10.1016/j.mad.2007.12.004 PMID:10357808 PMID: 18242664 22. Tomás-Loba A, Flores I, Fernández-Marcos PJ, Cayuela 30. Hochstrasser T, Marksteiner J, Humpel C. Telomere ML, Maraver A, Tejera A, Borrás C, Matheu A, Klatt P, length is age-dependent and reduced in monocytes of Flores JM, Viña J, Serrano M, Blasco MA. Telomerase Alzheimer patients. Exp Gerontol. 2012; 47:160–63. reverse transcriptase delays aging in cancer-resistant https://doi.org/10.1016/j.exger.2011.11.012 mice. Cell. 2008; 135:609–22. PMID:22178633 https://doi.org/10.1016/j.cell.2008.09.034 31. Franco S, Blasco MA, Siedlak SL, Harris PL, Moreira PI, PMID:19013273 Perry G, Smith MA. Telomeres and telomerase in 23. Bernardes de Jesus B, Vera E, Schneeberger K, Tejera Alzheimer’s disease: epiphenomena or a new focus AM, Ayuso E, Bosch F, Blasco MA. Telomerase gene for therapeutic strategy? Alzheimers Dement. 2006; therapy in adult and old mice delays aging and 2:164–68. https://doi.org/10.1016/j.jalz.2006.03.001 increases longevity without increasing cancer. EMBO PMID:19595878 Mol Med. 2012; 4:691–704. 32. Zhang J, Kong Q, Zhang Z, Ge P, Ba D, He W. Telomere https://doi.org/10.1002/emmm.201200245 dysfunction of lymphocytes in patients with Alzheimer PMID:22585399 disease. Cogn Behav Neurol. 2003; 16:170–76. 24. Bär C, Povedano JM, Serrano R, Benitez-Buelga C, https://doi.org/10.1097/00146965-200309000-00004 Popkes M, Formentini I, Bobadilla M, Bosch F, Blasco PMID:14501538 MA. Telomerase gene therapy rescues telomere 33. Guan JZ, Maeda T, Sugano M, Oyama J, Higuchi Y, length, bone marrow aplasia, and survival in mice Suzuki T, Makino N. A percentage analysis of the with aplastic anemia. Blood. 2016; 127:1770–79. telomere length in Parkinson’s disease patients. J https://doi.org/10.1182/blood-2015-08-667485 Gerontol A Biol Sci Med Sci. 2008; 63:467–73. PMID:26903545 https://doi.org/10.1093/gerona/63.5.467 25. Bär C, Bernardes de Jesus B, Serrano R, Tejera A, Ayuso PMID:18511749 www.aging-us.com 2938 AGING 34. Watfa G, Dragonas C, Brosche T, Dittrich R, Sieber CC, contains evolutionarily conserved catalytic and Alecu C, Benetos A, Nzietchueng R. Study of telomere structural subunits. Genes Dev. 1997; 11:3109–15. length and different markers of oxidative stress in https://doi.org/10.1101/gad.11.23.3109 patients with Parkinson’s disease. J Nutr Health PMID:9389643 Aging. 2011; 15:277–81. 43. Martín-Rivera L, Herrera E, Albar JP, Blasco MA. https://doi.org/10.1007/s12603-010-0275-7 Expression of mouse telomerase catalytic subunit in PMID:21437559 embryos and adult tissues. Proc Natl Acad Sci USA. 35. Wang H, Chen H, Gao X, McGrath M, Deer D, De Vivo 1998; 95:10471–76. I, Schwarzschild MA, Ascherio A. Telomere length and https://doi.org/10.1073/pnas.95.18.10471 risk of Parkinson’s disease. Mov Disord. 2008; PMID:9724727 23:302–05. https://doi.org/10.1002/mds.21867 44. Mori E, Hirono N, Yamashita H, Imamura T, Ikejiri Y, PMID:18044760 Ikeda M, Kitagaki H, Shimomura T, Yoneda Y. 36. Eerola J, Kananen L, Manninen K, Hellström O, Tienari Premorbid brain size as a determinant of reserve PJ, Hovatta I. No evidence for shorter leukocyte capacity against intellectual decline in Alzheimer’s telomere length in Parkinson’s disease patients. J disease. Am J Psychiatry. 1997; 154:18–24. Gerontol A Biol Sci Med Sci. 2010; 65:1181–84. https://doi.org/10.1176/ajp.154.1.18 PMID:8988953 https://doi.org/10.1093/gerona/glq125 45. Fortin NJ, Agster KL, Eichenbaum HB. Critical role of PMID:20639300 the hippocampus in memory for sequences of events. 37. Ferrón S, Mira H, Franco S, Cano-Jaimez M, Bellmunt Nat Neurosci. 2002; 5:458–62. E, Ramírez C, Fariñas I, Blasco MA, Blasco M. https://doi.org/10.1038/nn834 PMID:11976705 Telomere shortening and chromosomal instability 46. Winocur G, Wojtowicz JM, Sekeres M, Snyder JS, abrogates proliferation of adult but not embryonic Wang S. Inhibition of neurogenesis interferes with neural stem cells. Development. 2004; 131:4059–70. hippocampus-dependent memory function. https://doi.org/10.1242/dev.01215 PMID:15269166 Hippocampus. 2006; 16:296–304. 38. Ferrón SR, Marqués-Torrejón MA, Mira H, Flores I, https://doi.org/10.1002/hipo.20163 PMID:16411241 Taylor K, Blasco MA, Fariñas I. Telomere shortening in 47. Cameron HA, McKay RD. Adult neurogenesis neural stem cells disrupts neuronal differentiation produces a large pool of new granule cells in the and neuritogenesis. J Neurosci. 2009; 29:14394–407. dentate gyrus. J Comp Neurol. 2001; 435:406–17. https://doi.org/10.1523/JNEUROSCI.3836-09.2009 https://doi.org/10.1002/cne.1040 PMID:11406822 PMID:19923274 48. Boldrini M, Fulmore CA, Tartt AN, Simeon LR, Pavlova 39. Jaskelioff M, Muller FL, Paik JH, Thomas E, Jiang S, I, Poposka V, Rosoklija GB, Stankov A, Arango V, Adams AC, Sahin E, Kost-Alimova M, Protopopov A, Dwork AJ, Hen R, Mann JJ. Human Hippocampal Cadiñanos J, Horner JW, Maratos-Flier E, Depinho RA. Neurogenesis Persists throughout Aging. Cell Stem Telomerase reactivation reverses tissue degeneration Cell. 2018; 22:589–599.e5. in aged telomerase-deficient mice. Nature. 2011; https://doi.org/10.1016/j.stem.2018.03.015 469:102–06. https://doi.org/10.1038/nature09603 PMID:29625071 PMID:21113150 49. Christie BR, Cameron HA. Neurogenesis in the adult 40. Liu Y, Snow BE, Hande MP, Yeung D, Erdmann NJ, hippocampus. Hippocampus. 2006; 16:199–207. Wakeham A, Itie A, Siderovski DP, Lansdorp PM, https://doi.org/10.1002/hipo.20151 PMID:16411231 Robinson MO, Harrington L. The telomerase reverse transcriptase is limiting and necessary for telomerase 50. Pereira AC, Huddleston DE, Brickman AM, Sosunov function in vivo. Curr Biol. 2000; 10:1459–62. AA, Hen R, McKhann GM, Sloan R, Gage FH, Brown https://doi.org/10.1016/S0960-9822(00)00805-8 TR, Small SA. An in vivo correlate of exercise-induced PMID:11102810 neurogenesis in the adult dentate gyrus. Proc Natl Acad Sci USA. 2007; 104:5638–43. 41. Greenberg RA, Allsopp RC, Chin L, Morin GB, DePinho https://doi.org/10.1073/pnas.0611721104 RA. Expression of mouse telomerase reverse PMID:17374720 transcriptase during development, differentiation and proliferation. Oncogene. 1998; 16:1723–30. 51. Tatebayashi Y, Lee MH, Li L, Iqbal K, Grundke-Iqbal I. https://doi.org/10.1038/sj.onc.1201933 The dentate gyrus neurogenesis: a therapeutic target PMID:9582020 for Alzheimer’s disease. Acta Neuropathol. 2003; 105:225–32. PMID:12557008 42. Harrington L, Zhou W, McPhail T, Oulton R, Yeung DS, Mar V, Bass MB, Robinson MO. Human telomerase 52. Ming GL, Song H. Adult neurogenesis in the mamma- www.aging-us.com 2939 AGING lian brain: significant answers and significant https://doi.org/10.1016/0022-510X(78)90107-7 questions. Neuron. 2011; 70:687–702. PMID:624958 https://doi.org/10.1016/j.neuron.2011.05.001 62. Spillantini MG, Goedert M. Tau protein pathology PMID:21609825 in neurodegenerative diseases. Trends Neurosci. 53. Lodato S, Arlotta P. Generating neuronal diversity in 1998; 21:428–33. https://doi.org/10.1016/S0166- the mammalian cerebral cortex. Annu Rev Cell Dev 2236(98)01337-X PMID:9786340 Biol. 2015; 31:699–720. 63. Mohandas E, Rajmohan V, Raghunath B. Neurobiology https://doi.org/10.1146/annurev-cellbio-100814- of Alzheimer’s disease. Indian J Psychiatry. 2009; 125353 PMID:26359774 51:55–61. 54. Kuhn HG, Dickinson-Anson H, Gage FH. Neurogenesis https://doi.org/10.4103/0019-5545.44908 in the dentate gyrus of the adult rat: age-related PMID:19742193 decrease of neuronal progenitor proliferation. J 64. Schindowski K, Bretteville A, Leroy K, Bégard S, Brion Neurosci. 1996; 16:2027–33. JP, Hamdane M, Buée L. Alzheimer’s disease-like tau https://doi.org/10.1523/JNEUROSCI.16-06- neuropathology leads to memory deficits and loss of 02027.1996 PMID:8604047 functional synapses in a novel mutated tau transgenic 55. Mu Y, Gage FH. Adult hippocampal neurogenesis and mouse without any motor deficits. Am J Pathol. 2006; its role in Alzheimer’s disease. Mol Neurodegener. 169:599–616. 2011; 6:85. https://doi.org/10.1186/1750-1326-6-85 https://doi.org/10.2353/ajpath.2006.060002 PMID:22192775 PMID:16877359 56. Couillard-Despres S, Winner B, Schaubeck S, Aigner R, 65. Kolarova M, García-Sierra F, Bartos A, Ricny J, Ripova Vroemen M, Weidner N, Bogdahn U, Winkler J, Kuhn D. Structure and pathology of tau protein in HG, Aigner L. Doublecortin expression levels in adult Alzheimer disease. Int J Alzheimers Dis. 2012; brain reflect neurogenesis. Eur J Neurosci. 2005; 2012:731526. https://doi.org/10.1155/2012/731526 21:1–14. https://doi.org/10.1111/j.1460- PMID:22690349 9568.2004.03813.x PMID:15654838 66. Guo C, Zhang S, Li JY, Ding C, Yang ZH, Chai R, Wang X, 57. Sparkman NL, Johnson RW. Neuroinflammation Wang ZY. Chronic hyperglycemia induced via the associated with aging sensitizes the brain to the heterozygous knockout of Pdx1 worsens neuro- effects of infection or stress. Neuroimmuno- pathological lesion in an Alzheimer mouse model. Sci modulation. 2008; 15:323–30. Rep. 2016; 6:29396. https://doi.org/10.1159/000156474 PMID:19047808 https://doi.org/10.1038/srep29396 PMID:27406855 58. Salminen A, Ojala J, Kaarniranta K, Haapasalo A, 67. Ando K, Leroy K, Heraud C, Kabova A, Yilmaz Z, Hiltunen M, Soininen H. Astrocytes in the aging brain Authelet M, Suain V, De Decker R, Brion JP. Deletion express characteristics of senescence-associated of murine tau gene increases tau aggregation in a secretory phenotype. Eur J Neurosci. 2011; 34:3–11. human mutant tau transgenic mouse model. Biochem https://doi.org/10.1111/j.1460-9568.2011.07738.x Soc Trans. 2010; 38:1001–05. PMID:21649759 https://doi.org/10.1042/BST0381001 PMID:20658993 59. Zussy C, Brureau A, Keller E, Marchal S, Blayo C, Delair 68. Visanji NP, Brotchie JM. MPTP-induced models of B, Ixart G, Maurice T, Givalois L. Alzheimer’s disease Parkinson’s disease in mice and non-human primates. related markers, cellular toxicity and behavioral Curr Protoc Pharmacol. 2005; Chapter 5:Unit5.42. deficits induced six weeks after oligomeric amyloid-β https://doi.org/10.1002/0471141755.ph0542s29 peptide injection in rats. PLoS One. 2013; 8:e53117. PMID:21953394 https://doi.org/10.1371/journal.pone.0053117 PMID:23301030 69. Jackson-Lewis V, Przedborski S. Protocol for the MPTP mouse model of Parkinson’s disease. Nat Protoc. 60. Fu H, Dirosario J, Killedar S, Zaraspe K, McCarty DM. 2007; 2:141–51. Correction of neurological disease of mucopolysac- https://doi.org/10.1038/nprot.2006.342 charidosis IIIB in adult mice by rAAV9 trans-blood-brain PMID:17401348 barrier gene delivery. Mol Ther. 2011; 19:1025–33. https://doi.org/10.1038/mt.2011.34 PMID:21386820 70. Can A, Dao DT, Terrillion CE, Piantadosi SC, Bhat S, Gould TD. The tail suspension test. J Vis Exp. 2012; 61. Jacque CM, Vinner C, Kujas M, Raoul M, Racadot J, 59:e3769. https://doi.org/10.3791/3769 Baumann NA. Determination of glial fibrillary acidic PMID:22315011 protein (GFAP) in human brain tumors. J Neurol Sci. 1978; 35:147–55 71. Nagatsu T. Tyrosine hydroxylase: human isoforms, www.aging-us.com 2940 AGING structure and regulation in physiology and pathology. Correction of pathological accumulation of glyco- Essays Biochem. 1995; 30:15–35. PMID:8822146 saminoglycans in central nervous system and peri- pheral tissues of MPSIIIA mice through systemic AAV9 72. Leger M, Quiedeville A, Bouet V, Haelewyn B, gene transfer. Hum Gene Ther. 2012; 23:1237–46. Boulouard M, Schumann-Bard P, Freret T. Object recognition test in mice. Nat Protoc. 2013 8:2531-37. https://doi.org/10.1089/hum.2012.029 https://doi.org/10.1038/nprot.2013.155 PMID:22909060 PMID: 24263092 81. Cearley CN, Wolfe JH. Transduction characteristics of 73. Mosconi L. Brain glucose metabolism in the early and adeno-associated virus vectors expressing cap specific diagnosis of Alzheimer’s disease. FDG-PET serotypes 7, 8, 9, and Rh10 in the mouse brain. Mol studies in MCI and AD. Eur J Nucl Med Mol Imaging. Ther. 2006; 13:528–37. 2005; 32:486–510. https://doi.org/10.1007/s00259- https://doi.org/10.1016/j.ymthe.2005.11.015 005-1762-7 PMID:15747152 PMID:16413228 74. Ferris SH, de Leon MJ, Wolf AP, Farkas T, Christman 82. Dayton RD, Wang DB, Klein RL. The advent of AAV9 DR, Reisberg B, Fowler JS, Macgregor R, Goldman A, expands applications for brain and spinal cord gene George AE, Rampal S. Positron emission tomography delivery. Expert Opin Biol Ther. 2012; 12:757–66. in the study of aging and senile dementia. Neurobiol https://doi.org/10.1517/14712598.2012.681463 Aging. 1980; 1:127–31. PMID:22519910 https://doi.org/10.1016/0197-4580(80)90005-6 83. Gray SJ, Matagne V, Bachaboina L, Yadav S, Ojeda SR, PMID:24279935 Samulski RJ. Preclinical differences of intravascular 75. Reiman EM, Chen K, Alexander GE, Caselli RJ, Bandy AAV9 delivery to neurons and glia: a comparative D, Osborne D, Saunders AM, Hardy J. Functional brain study of adult mice and nonhuman primates. Mol abnormalities in young adults at genetic risk for late- Ther. 2011; 19:1058–69. onset Alzheimer’s dementia. Proc Natl Acad Sci USA. https://doi.org/10.1038/mt.2011.72 PMID:21487395 2004; 101:284–89. 84. Zhang H, Yang B, Mu X, Ahmed SS, Su Q, He R, Wang https://doi.org/10.1073/pnas.2635903100 H, Mueller C, Sena-Esteves M, Brown R, Xu Z, Gao G. PMID:14688411 Several rAAV vectors efficiently cross the blood-brain 76. Bevan AK, Duque S, Foust KD, Morales PR, Braun L, barrier and transduce neurons and astrocytes in the Schmelzer L, Chan CM, McCrate M, Chicoine LG, neonatal mouse central nervous system. Mol Ther. Coley BD, Porensky PN, Kolb SJ, Mendell JR, et al. 2011; 19:1440–48. Systemic gene delivery in large species for targeting https://doi.org/10.1038/mt.2011.98 PMID:21610699 spinal cord, brain, and peripheral tissues for pediatric 85. Miyazaki J, Takaki S, Araki K, Tashiro F, Tominaga A, disorders. Mol Ther. 2011; 19:1971–80. Takatsu K, Yamamura K. Expression vector system https://doi.org/10.1038/mt.2011.157 based on the chicken beta-actin promoter directs PMID:21811247 efficient production of interleukin-5. Gene. 1989; 77. Foust KD, Nurre E, Montgomery CL, Hernandez A, 79:269–77. https://doi.org/10.1016/0378- Chan CM, Kaspar BK. Intravascular AAV9 1119(89)90209-6 PMID:2551778 preferentially targets neonatal neurons and adult 86. Bey K, Ciron C, Dubreil L, Deniaud J, Ledevin M, astrocytes. Nat Biotechnol. 2009; 27:59–65. Cristini J, Blouin V, Aubourg P, Colle MA. Efficient CNS https://doi.org/10.1038/nbt.1515 PMID:19098898 targeting in adult mice by intrathecal infusion of 78. Haurigot V, Marcó S, Ribera A, Garcia M, Ruzo A, single-stranded AAV9-GFP for gene therapy of Villacampa P, Ayuso E, Añor S, Andaluz A, Pineda M, neurological disorders. Gene Ther. 2017; 24:325–32. García-Fructuoso G, Molas M, Maggioni L, et al. https://doi.org/10.1038/gt.2017.18 PMID:28425480 Whole body correction of mucopolysaccharidosis IIIA 87. Alves S, Bode J, Bemelmans AP, von Kalle C, Cartier N, by intracerebrospinal fluid gene therapy. J Clin Invest. Tews B. Ultramicroscopy as a novel tool to unravel 2013; 123:3254–71. the tropism of AAV gene therapy vectors in the brain. https://doi.org/10.1172/JCI66778 PMID:23863627 Sci Rep. 2016; 6:28272. 79. Manfredsson FP, Rising AC, Mandel RJ. AAV9: a https://doi.org/10.1038/srep28272 PMID:27320056 potential blood-brain barrier buster. Mol Ther. 2009; 88. Aschauer DF, Kreuz S, Rumpel S. Analysis of 17:403–05. https://doi.org/10.1038/mt.2009.15 transduction efficiency, tropism and axonal transport PMID:19247366 of AAV serotypes 1, 2, 5, 6, 8 and 9 in the mouse 80. Ruzo A, Marcó S, García M, Villacampa P, Ribera A, brain. PLoS One. 2013; 8:e76310. Ayuso E, Maggioni L, Mingozzi F, Haurigot V, Bosch F. https://doi.org/10.1371/journal.pone.0076310 www.aging-us.com 2941 AGING PMID:24086725 https://doi.org/10.3233/JAD-2008-15408 PMID:19096160 89. Foust KD, Salazar DL, Likhite S, Ferraiuolo L, Ditsworth D, Ilieva H, Meyer K, Schmelzer L, Braun L, Cleveland 97. Maurice T, Lockhart BP, Privat A. Amnesia induced in DW, Kaspar BK. Therapeutic AAV9-mediated mice by centrally administered beta-amyloid peptides suppression of mutant SOD1 slows disease progres- involves cholinergic dysfunction. Brain Res. 1996; sion and extends survival in models of inherited ALS. 706:181–93. https://doi.org/10.1016/0006- Mol Ther. 2013; 21:2148–59. 8993(95)01032-7 PMID:8822355 https://doi.org/10.1038/mt.2013.211 98. Niccoli T, Partridge L. Ageing as a risk factor for PMID:24008656 disease. Curr Biol. 2012; 22:R741–52. 90. Paul CM, Magda G, Abel S. Spatial memory: https://doi.org/10.1016/j.cub.2012.07.024 theoretical basis and comparative review on PMID:22975005 experimental methods in rodents. Behav Brain Res. 99. Reeve A, Simcox E, Turnbull D. Ageing and Parkinson’s 2009; 203:151–64. disease: why is advancing age the biggest risk factor? https://doi.org/10.1016/j.bbr.2009.05.022 Ageing Res Rev. 2014; 14:19–30. PMID:19467271 https://doi.org/10.1016/j.arr.2014.01.004 91. Masliah E, Rockenstein E, Adame A, Alford M, Crews PMID:24503004 L, Hashimoto M, Seubert P, Lee M, Goldstein J, 100. Marzetti E, Leeuwenburgh C. Skeletal muscle Chilcote T, Games D, Schenk D. Effects of α-synuclein apoptosis, sarcopenia and frailty at old age. Exp immunization in a mouse model of Parkinson’s Gerontol. Pergamon; 2006; 41: 1234–8. disease. Neuron. 2005; 46:857–68. https://doi.org/10.1016/j.exger.2006.08.011 https://doi.org/10.1016/j.neuron.2005.05.010 PMID:17052879 PMID:15953415 101. Aiuti A, Cattaneo F, Galimberti S, Benninghoff U, 92. Bard F, Cannon C, Barbour R, Burke RL, Games D, Cassani B, Callegaro L, Scaramuzza S, Andolfi G, Grajeda H, Guido T, Hu K, Huang J, Johnson-Wood K, Mirolo M, Brigida I, Tabucchi A, Carlucci F, Eibl M, et Khan K, Kholodenko D, Lee M, et al. Peripherally al. Gene therapy for immunodeficiency due to administered antibodies against amyloid β-peptide adenosine deaminase deficiency. N Engl J Med. enter the central nervous system and reduce 2009; 360:447–58. pathology in a mouse model of Alzheimer disease. https://doi.org/10.1056/NEJMoa0805817 Nat Med. 2000; 6:916–19. PMID:19179314 https://doi.org/10.1038/78682 PMID:10932230 102. Cartier N, Hacein-Bey-Abina S, Bartholomae CC, 93. Lim GP, Yang F, Chu T, Chen P, Beech W, Teter B, Tran Veres G, Schmidt M, Kutschera I, Vidaud M, Abel U, T, Ubeda O, Ashe KH, Frautschy SA, Cole GM. Dal-Cortivo L, Caccavelli L, Mahlaoui N, Kiermer V, Ibuprofen suppresses plaque pathology and inflam- Mittelstaedt D, et al. Hematopoietic stem cell gene mation in a mouse model for Alzheimer’s disease. J therapy with a lentiviral vector in X-linked adreno- Neurosci. 2000; 20:5709–14. leukodystrophy. Science. 2009; 326: 818–23. https://doi.org/10.1523/JNEUROSCI.20-15- https://doi.org/10.1126/science.1171242 05709.2000 PMID:10908610 PMID:19892975 94. Blackburn EH. Telomere states and cell fates. Nature. 103. Cavazzana-Calvo M, Payen E, Negre O, Wang G, 2000; 408:53–56. https://doi.org/10.1038/35040500 Hehir K, Fusil F, Down J, Denaro M, Brady T, PMID:11081503 Westerman K, Cavallesco R, Gillet-Legrand B, 95. Radde R, Bolmont T, Kaeser SA, Coomaraswamy J, Caccavelli L, et al. Transfusion independence and Lindau D, Stoltze L, Calhoun ME, Jäggi F, Wolburg H, HMGA2 activation after gene therapy of human β- Gengler S, Haass C, Ghetti B, Czech C, et al. Abeta42- thalassaemia. Nature. 2010; 467:318–22. driven cerebral amyloidosis in transgenic mice reveals https://doi.org/10.1038/nature09328 early and robust pathology. EMBO Rep. 2006; 7:940– PMID:20844535 46. https://doi.org/10.1038/sj.embor.7400784 104. Nathwani AC, Tuddenham EGD, Rangarajan S, PMID:16906128 Rosales C, McIntosh J, Linch DC, Chowdary P, Riddell 96. Pallas M, Camins A, Smith MA, Perry G, Lee HG, A, Pie AJ, Harrington C, O’Beirne J, Smith K, Pasi J, et Casadesus G. From aging to Alzheimer’s disease: al. Adenovirus-associated virus vector–mediated unveiling “the switch” with the senescence- gene transfer in hemophilia B. N Engl J Med. 2011; accelerated mouse model (SAMP8). J Alzheimers Dis. 365: 2357–65. 2008; 15:615–24. https://doi.org/10.1056/NEJMoa1108046 www.aging-us.com 2942 AGING PMID:22149959 their niches. Cell Cycle. 2012; 11:2260–67. https://doi.org/10.4161/cc.20437 PMID:22617385 105. Bainbridge JW, Smith AJ, Barker SS, Robbie S, Henderson R, Balaggan K, Viswanathan A, Holder GE, 114. Sinha M, Jang YC, Oh J, Khong D, Wu EY, Manohar R, Stockman A, Tyler N, Petersen-Jones S, Bhattacharya Miller C, Regalado SG, Loffredo FS, Pancoast JR, SS, Thrasher AJ, et al. Effect of gene therapy on Hirshman MF, Lebowitz J, Shadrach JL, et al. visual function in Leber’s congenital amaurosis. N Restoring systemic GDF11 levels reverses age- Engl J Med. 2008; 358:2231–39. related dysfunction in mouse skeletal muscle. https://doi.org/10.1056/NEJMoa0802268 Science. 2014; 344:649–52. PMID:18441371 https://doi.org/10.1126/science.1251152 PMID:24797481 106. Maguire AM, High KA, Auricchio A, Wright JF, Pierce EA, Testa F, Mingozzi F, Bennicelli JL, Ying GS, Rossi S, 115. Katsimpardi L, Litterman NK, Schein PA, Miller CM, Fulton A, Marshall KA, Banfi S, et al. Age-dependent Loffredo FS, Wojtkiewicz GR, Chen JW, Lee RT, effects of RPE65 gene therapy for Leber’s congenital Wagers AJ, Rubin LL. Vascular and neurogenic amaurosis: a phase 1 dose-escalation trial. Lancet. rejuvenation of the aging mouse brain by young 2009; 374:1597–605. systemic factors. Science. 2014; 344:630–34. https://doi.org/10.1016/S0140-6736(09)61836-5 https://doi.org/10.1126/science.1251141 PMID:19854499 PMID:24797482 107. Maguire AM, Simonelli F, Pierce EA, Pugh EN Jr, 116. Matsushita T, Elliger S, Elliger C, Podsakoff G, Mingozzi F, Bennicelli J, Banfi S, Marshall KA, Testa F, Villarreal L, Kurtzman GJ, Iwaki Y, Colosi P. Adeno- Surace EM, Rossi S, Lyubarsky A, Arruda VR, et al. associated virus vectors can be efficiently produced Safety and efficacy of gene transfer for Leber’s without helper virus. Gene Ther. 1998; 5:938–45. congenital amaurosis. N Engl J Med. 2008; 358:2240– https://doi.org/10.1038/sj.gt.3300680 48. https://doi.org/10.1056/NEJMoa0802315 PMID:9813665 PMID:18441370 117. Ayuso E, Mingozzi F, Montane J, Leon X, Anguela XM, 108. Naldini L. Gene therapy returns to centre stage. Haurigot V, Edmonson SA, Africa L, Zhou S, High KA, Nature. 2015; 526:351–60. Bosch F, Wright JF. High AAV vector purity results in https://doi.org/10.1038/nature15818 serotype- and tissue-independent enhancement of PMID:26469046 transduction efficiency. Gene Ther. 2010; 17:503–10. 109. Thomas CE, Ehrhardt A, Kay MA. Progress and https://doi.org/10.1038/gt.2009.157 problems with the use of viral vectors for gene PMID:19956269 therapy. Nat Rev Genet. 2003; 4:346–58. 118. Zijlmans JM, Martens UM, Poon SS, Raap AK, Tanke https://doi.org/10.1038/nrg1066 PMID:12728277 HJ, Ward RK, Lansdorp PM. Telomeres in the mouse 110. Muñoz-Lorente MA, Martínez P, Tejera Á, have large inter-chromosomal variations in the Whittemore K, Moisés-Silva AC, Bosch F, Blasco MA. number of T2AG3 repeats. Proc Natl Acad Sci USA. AAV9-mediated telomerase activation does not 1997; 94:7423–28. accelerate tumorigenesis in the context of oncogenic https://doi.org/10.1073/pnas.94.14.7423 K-Ras-induced lung cancer. PLoS Genet. 2018; PMID:9207107 14:e1007562. 119. González-Suárez E, Samper E, Flores JM, Blasco MA. https://doi.org/10.1371/journal.pgen.1007562 Telomerase-deficient mice with short telomeres are PMID:30114189 resistant to skin tumorigenesis. Nat Genet. 2000; 111. Harley CB. Telomerase is not an oncogene. 26:114–17. https://doi.org/10.1038/79089 Oncogene. 2002; 21:494–502. PMID:10973262 https://doi.org/10.1038/sj.onc.1205076 120. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, PMID:11850774 Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld 112. Conboy IM, Conboy MJ, Wagers AJ, Girma ER, S, Schmid B, Tinevez JY, White DJ, Hartenstein V, et Weissman IL, Rando TA. Rejuvenation of aged al. Fiji: an open-source platform for biological-image progenitor cells by exposure to a young systemic analysis. Nat Methods. 2012; 9:676–82. environment. Nature. 2005; 433:760–64. https://doi.org/10.1038/nmeth.2019 https://doi.org/10.1038/nature03260 PMID:22743772 PMID:15716955 121. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to 113. Conboy IM, Rando TA. Heterochronic parabiosis for ImageJ: 25 years of image analysis. Nat Methods. the study of the effects of aging on stem cells and 2012; 9:671–75. www.aging-us.com 2943 AGING https://doi.org/10.1038/nmeth.2089 memory deficits at 4-months of age in the triple- PMID:22930834 transgenic mouse model of Alzheimer’s disease. PLoS One. 2013; 8:e80355. 122. Mulero F, Donate LE, Serrano M. Imaging cancer in https://doi.org/10.1371/journal.pone.0080355 mice by PET, CT, and combined PET-CT. Current PMID:24236177 Protocols in Mouse Biology. Hoboken, NJ, USA: John Wiley & Sons, Inc.; 2011. p. 85–103. 132. Mouzon B, Chaytow H, Crynen G, Bachmeier C, https://doi.org/10.1002/9780470942390.mo100137 Stewart J, Mullan M, Stewart W, Crawford F. PMID:26068988 Repetitive mild traumatic brain injury in a mouse model produces learning and memory deficits 123. Hamm RJ, Pike BR, O’Dell DM, Lyeth BG, Jenkins LW. accompanied by histological changes. J Neurotrauma. The rotarod test: an evaluation of its effectiveness in assessing motor deficits following traumatic brain 2012; 29:2761–73. https://doi.org/10.1089/neu.2012.2498 injury. J Neurotrauma. 1994; 11:187–96. PMID:22900595 https://doi.org/10.1089/neu.1994.11.187 PMID:7932797 133. Meijering E, Dzyubachyk O, Smal I. Methods for cell and particle tracking. Methods Enzymol. 2012; 124. Ingram DK, Reynolds MA. Assessing the predictive 504:183–200. https://doi.org/10.1016/B978-0-12- validity of psychomotor tests as measures of 391857-4.00009-4 PMID:22264535 biological age in mice. Exp Aging Res. 1986; 12:155– 62. https://doi.org/10.1080/03610738608259454 PMID:3830235 125. Matheu A, Maraver A, Klatt P, Flores I, Garcia-Cao I, Borras C, Flores JM, Viña J, Blasco MA, Serrano M. Delayed ageing through damage protection by the Arf/p53 pathway. Nature. 2007; 448:375–79. https://doi.org/10.1038/nature05949 PMID:17637672 126. Bevins RA, Besheer J. Object recognition in rats and mice: a one-trial non-matching-to-sample learning task to study ‘recognition memory’. Nat Protoc. 2006; 1:1306–11. https://doi.org/10.1038/nprot.2006.205 PMID:17406415 127. Obrenovic Z, Gasevic D. Open source software: all you do is put it together. IEEE Softw. 2007; 24:86– 95. https://doi.org/10.1109/MS.2007.141 128. Brooks SP, Trueman RC, Dunnett SB. Assessment of Motor Coordination and Balance in Mice Using the Rotarod, Elevated Bridge, and Footprint Tests. Curr Protoc Mouse Biol. 2012; 2:37–53. https://doi.org/10.1002/9780470942390.mo110165 PMID:26069004 129. Brooks SP, Dunnett SB. Tests to assess motor phenotype in mice: a user’s guide. Nat Rev Neurosci. 2009; 10:519–29. https://doi.org/10.1038/nrn2652 PMID:19513088 130. Barnes CA. Memory deficits associated with senescence: a neurophysiological and behavioral study in the rat. J Comp Physiol Psychol. 1979; 93:74–104. https://doi.org/10.1037/h0077579 PMID:221551 131. Attar A, Liu T, Chan WT, Hayes J, Nejad M, Lei K, Bitan G. A shortened Barnes maze protocol reveals www.aging-us.com 2944 AGING SUPPLEMENTARY MATERIAL Figure S1. Younger mice and wild-type mice have more doublecortin in the cerebral cortex than older and G3 Tert-/- mice and more neurogenic niches in the parietal subventricular zone. (A) Percent doublecortin in parietotemporal region of cerebral cortex. (B) Percent doublecortin in occipital region of cerebral cortex. (C) The level of neurogenic niches in the parietal subventricular zone quantified from hematoxylin and eosin stained paraffin sections. www.aging-us.com 2945 AGING Figure S2. Mice deficient for telomerase are more susceptible to Parkinson’s-like symptoms induced by MPTP neurotoxin. (A) Scheme of the MPTP neurotoxin experiment. Wild-type or G3 Tert-/- mice, were injected with different doses of MPTP or saline solution. Behavior tests were performed, and the mice were sacrificed 7 days after injection. The brain was preserved in formalin. (B) Hind paw stride length from footprint test. (C) The immobility time in the tail suspension test. (D) Immunofluorescence results for tyrosine hydroxylase in the substantia nigra for mice treated with different doses of MPTP neurotoxin. (E) Immuno- fluorescence results for tyrosine hydroxylase in the substantia nigra for young and old wild-type and G3 Tert-/- mice. Data represent the mean ±SE of analyzed mice within each group. For the histopathology results, the number of mice analyzed per group is indicated, as well as the number of fields of view, and the number of positive cells. The t-test was used for statistical analysis. *p<0.05; **p<0.01; ***p<0.001. www.aging-us.com 2946 AGING Figure S3. Mice deficient for telomerase perform more poorly in the tightrope test, rotarod test, tail suspension test, and object recognition test, and they also have less glucose uptake in the brain. (A) Percent success in the tightrope test for wild-type and Tert-/- mice at 70-90 weeks of age. (B) Time in rotarod test for wild-type, G2 Tert-/-, and G4 Tert-/- mice at 70-100 weeks of age. (C) Immobility time in tail suspension test for old wild-type, G2 Tert-/-, G3 Tert-/-, and G4 Tert-/- mice. (D) Ratio of time spent investigating a novel object (Time B) to the time spent investigating both objects (Time A+B) in the object recognition test. (E) Standard uptake value (SUV) of 18F-FDG (18F-fluorodeoxyglucose) into the brain in a positron emission tomography (PET) test. Representative heatmaps displaying the quantity of glucose uptake are also shown. Data represent the mean ±SE of analyzed mice within each group. The number of mice analyzed per group is indicated. The t-test was used for statistical analysis. *p<0.05; **p<0.01. www.aging-us.com 2947 AGING Figure S4. Body weights of the mice treated with AAV9-Tert or AAV9-null. These weights were determined at the humane endpoint. The age of death, the weeks of treatment, and the generation for each group is presented. www.aging-us.com 2948 AGING