TYPE Original Research PUBLISHED 06 January 2023 DOI 10.3389/fnagi.2022.1073258 Evaluation of AQP4 functional OPEN ACCESS variants and its association with EDITED BY Veronica Martinez Cerdeño, University of California, Davis, fragile X-associated United States REVIEWED BY tremor/ataxia syndrome Randi Jenssen Hagerman, University of California, Davis, United States Andrea Elias-Mas1,2,3 , Miriam Potrony4,5,6 , Jaume Bague5,7 , Jacob Raber, David J. Cutler8 , Maria Isabel Alvarez-Mora4,5,6 , Oregon Health & Science University, United States Teresa Torres5,6 , Tamara Barcos4,5 , *CORRESPONDENCE Joan Anton Puig-Butille5,6,9 , Marta Rubio2,10 , Laia Rodriguez-Revenga lbodi@clinic.cat Irene Madrigal4,5,6 , Susana Puig5,6,7 , Emily G. Allen8*† and Emily G. Allen emgrave@emory.edu Laia Rodriguez-Revenga4,5,6*† 1 † These authors have contributed Radiology Department, Hospital Universitari Mútua de Terrassa, Terrassa, Barcelona, Spain, 2 equally to this work and share senior Institute for Research and Innovation Parc Taulí (I3PT), Sabadell, Spain, 3 Genetics Doctorate authorship Program, Universitat de Barcelona (UB), Barcelona, Spain, 4 Biochemistry and Molecular Genetics Department, Hospital Clinic of Barcelona, Barcelona, Spain, 5 Institut d’Investigacions Biomèdiques SPECIALTY SECTION August Pi i Sunyer (IDIBAPS), Barcelona, Spain, 6 CIBER of Rare Diseases (CIBERER), Instituto de This article was submitted to Salud Carlos III, Barcelona, Spain, 7 Dermatology Department, Melanoma Unit, Hospital Clínic de Parkinson’s Disease and Aging-related Barcelona, Universitat de Barcelona, Barcelona, Spain, 8 Department of Human Genetics, Emory Movement Disorders, University School of Medicine, Atlanta, GA, United States, 9 Molecular Biology CORE, Hospital Clinic a section of the journal of Barcelona, Barcelona, Spain, 10 Department of Neurology, Parc Taulí Hospital Universitari, Frontiers in Aging Neuroscience Sabadell, Spain RECEIVED 24October 2022 ACCEPTED 19 December 2022 PUBLISHED 06 January 2023 CITATION Introduction: Fragile X-associated tremor/ataxia syndrome (FXTAS, OMIM# Elias-Mas A, Potrony M, Bague J, Cutler DJ, Alvarez-Mora MI, Torres T, 300623) is a late-onset neurodegenerative disorder with reduced penetrance Barcos T, Puig-Butille JA, Rubio M, that appears in adult FMR1 premutation carriers (55–200 CGGs). Clinical Madrigal I, Puig S, Allen EG and Rodriguez-Revenga L (2023) symptoms in FXTAS patients usually begin with an action tremor. After that, Evaluation of AQP4 functional different findings including ataxia, and more variably, loss of sensation in variants and its association with the distal lower extremities and autonomic dysfunction, may occur, and fragile X-associated tremor/ataxia syndrome. gradually progress. Cognitive deficits are also observed, and include memory Front. Aging Neurosci. 14:1073258. problems and executive function deficits, with a gradual progression to doi: 10.3389/fnagi.2022.1073258 dementia in some individuals. Aquaporin 4 (AQP4) is a commonly distributed COPYRIGHT © 2023 Elias-Mas, Potrony, Bague, water channel in astrocytes of the central nervous system. Changes in Cutler, Alvarez-Mora, Torres, Barcos, AQP4 activity and expression have been implicated in several central nervous Puig-Butille, Rubio, Madrigal, Puig, Allen and Rodriguez-Revenga. This is system disorders. Previous studies have suggested the associations of AQP4 an open-access article distributed single nucleotide polymorphisms (SNPs) with brain-water homeostasis, and under the terms of the Creative neurodegeneration disease. To date, this association has not been studied in Commons Attribution License (CC BY). The use, distribution or reproduction in FXTAS. other forums is permitted, provided the original author(s) and the copyright Methods: To investigate the association of AQP4 SNPs with the risk of owner(s) are credited and that the presenting FXTAS, a total of seven common AQP4 SNPs were selected and original publication in this journal is cited, in accordance with accepted genotyped in 95 FMR1 premutation carriers with FXTAS and in 65 FMR1 academic practice. No use, distribution premutation carriers without FXTAS. or reproduction is permitted which does not comply with these terms. Results: The frequency of AQP4-haplotype was compared between groups, denoting 26 heterozygous individuals and 5 homozygotes as carriers of the Frontiers in Aging Neuroscience 01 frontiersin.org Elias-Mas et al. 10.3389/fnagi.2022.1073258 minor allele in the FXTAS group and 25 heterozygous and 2 homozygotes in the no-FXTAS group. Statistical analyses showed no significant associations between AQP4 SNPs/haplotypes and development of FXTAS. Discussion: Although AQP4 has been implicated in a wide range of brain disorders, its involvement in FXTAS remains unclear. The identification of novel genetic markers predisposing to FXTAS or modulating disease progression is critical for future research involving predictors and treatments. KEYWORDS FXTAS, AQP4, FMR1 premutation, genetic variation, glymphatic system Introduction Despite the relationship between glymphatic dysfunction and neurodegenerative diseases, dysfunction of glymphatic system The brain is a high-energy consuming organ with a high has not yet been studied in Fragile X-associated tremor/ataxia metabolic activity, producing a substantial amount of interstitial syndrome (FXTAS) and its association with AQP4 genetic waste products. Efficient clearance of the brain’s metabolic waste variants is unknown. FXTAS is a neurodegenerative disorder is needed in order to avoid their accumulation, causing several linked to FMR1 gene premutation carriers (55–200 CGG neurological diseases (Kaur et al., 2021). Since there is a lack of repeats) that is associated with cognitive loss that can evolve conventional lymphoid circulation in the brain, the glymphatic into dementia. Intranuclear inclusions and increased β amyloid system has been postulated as an alternative clearance for the load has been discovered in brains of patients with FXTAS brain waste product (Iliff et al., 2015), though evidence is still (Cabal-Herrera et al., 2020; Salcedo-Arellano et al., 2021a). On incomplete (Hladky and Barrand, 2022). the basis of these observations, we analyzed AQP4 functional In the glymphatic system, cerebrospinal fluid (CSF) flows variants with the aim to investigate whether AQP4 could be a into the brain parenchyma within the periarterial spaces that new genetic risk factor for FXTAS. surround the penetrating cerebral arteries, also called the perivascular spaces. Facilitated by aquaporin 4 (AQP4), CSF flows from the periarterial space into the brain interstitium and mixes with interstitial fluid, which, along with interstitial Materials and methods solutes, travels into the perivenous spaces, draining the fluid and its contents into the deep veins and into the basal meningeal Study population and cervical lymphatic vessels (Hablitz and Nedergaard, 2021). AQP4 is the most abundant water channel in the brain, and, The present study was conducted on genotype data from since it has a role regulating fluid exchange between perivascular a total of 160 unrelated FMR1 premutation carriers (95 spaces and the rest of the glymphatic system, it is considered presenting FXTAS symptoms and 65 without FXTAS clinical the most important element in it (Nagelhus and Ottersen, 2013; symptoms). Participants were identified through previous Papadopoulos and Verkman, 2013; Szczygielski et al., 2021). fragile X syndrome research projects at Emory University Alterations of glymphatic fluid circulation through (Atlanta, GA, USA), through recruitment efforts at scientific AQPs variations are now emerging as central elements conferences, and through collaborations with other research in the pathophysiology of different brain conditions. In groups. All participants were enrolled from families with fact, dysfunction of AQP4 have been implicated in the members known to be affected with fragile X-associated pathogenesis of many degenerative disorders, including conditions and molecularly diagnosed. Table 1 summarizes Alzheimer’s disease (AD), vascular cognitive impairment, general demographics of the participants. FXTAS subjects were idiopathic normal-pressure hydrocephalus, Parkinson’s disease screened for eligibility as described in Kong et al. (2022). Briefly, dementia, frontotemporal dementia and Creutzfeldt-Jakob case subjects were male or female premutation carriers with disease (Zeppenfeld et al., 2017; Nedergaard and Goldman, symptoms of tremor or ataxia before age 65, as reviewed by 2020; Silva et al., 2021; Wang et al., 2022). Furthermore, a neurologist. Control individuals, named as the no-FXTAS evidence indicates that genetic variation in AQP4 modulates group, were male premutation carriers that reached age 68 sleep quality and architecture, amyloid-β burden and rate without significant tremor or ataxia symptoms, as reviewed by and progression of cognitive decline in AD patients (Burfeind a neurologist. The protocols and consent forms were approved et al., 2017; Rainey-Smith et al., 2018; Ulv Larsen et al., 2020). by the Institutional Review Board at Emory University, and Frontiers in Aging Neuroscience 02 frontiersin.org Elias-Mas et al. 10.3389/fnagi.2022.1073258 written informed consent was obtained from all subjects Results (IRB00074941). Genotype data from 95 FXTAS and 65 no-FXTAS individuals were analyzed. Table 1 shows the demographic data SNPs of the AQP4 gene and haplotype of FXTAS and no-FXTAS group. Significant differences were analysis found for the age and the CGG repeat size when comparing both groups (p < 0.0001 and p < 0.0001, respectively). Age difference Seven tag single nucleotide polymorphisms (SNPs) across can be attributed to a bias in the recruitment of no-FXTAS AQP4 gene (NM_001650) were selected according to their individuals. In order to make sure that FMR1 permutations location and known functions, based on earlier reports on in the no-FXTAS group did not have neurologic symptoms their associations with clinical phenotypes (Ulv Larsen et al., older men were included in this cohort. As for the CGG repeat 2020). The SNPs were considered for those above 15% in Utah Residents with Northern and Western European Ancestry TABLE 1 Description of the individuals recruited in this study. (CEU) population according to minor allele frequency (MAF: FXTAS cases No-FXTAS 0.15∼0.26). These SNPs included rs162007 (Chr18:26865883, (n = 95) cases (n = 65) Upstream, MAF 0.16), rs162008 (Chr18:26865728, 0 Males/female [no (%)] 80 (84%)/15 (16%) 65 (100%)/0 5 UTR, MAF 0.20), rs63514 (Chr18:26863457, intron, Age (mean ± SD, Y)* 67.70 ± 10.81 77.01 ± 6.38 MAF 0.17), rs335931 (Chr18:26859108, intron, MAF 0.15), rs335930 (Chr18:26856961, intron, MAF 0.23), Age min, max (Y) 27–94 66–98 rs335929 (Chr18:26855623, 30 UTR, MAF 0.14), rs16942851 CGG repeat size (mean ± SD)* 93.15 ± 20.33 75.00 ± 15.14 (Chr18:26851725, downstream, MAF 0.14). The chromosome CGG repeat size min, max 55–150 56–150 positions are based on hg38. *p < 0.0001 using Student’s t-test for comparison of means. TABLE 2 Genotype frequency for each SNP according to presence of FXTAS. Genotyping of AQP4 SNPs Genotype FXTAS No-FXTAS Whole genome sequencing was performed on samples using (n = 95) (n = 65) Illumina platforms at Hudson Alpha or Novogene as described rs162007 GG 64 (67.4%) 37 (56.9%) in Kong et al. (2022). All samples were mapped using PEMapper GA 26 (27.3%) 25 (38.5%) and called using PECaller (Johnston et al., 2017). Genomic data AA 5 (5.3%) 3 (4.6%) have been uploaded to the National Institute of Mental Health rs162008 CC 64 (67.4%) 36 (55.4%) (NIMH) Data Archive.1 CT 26 (27.3%) 26 (40%) TT 5 (5.3%) 3 (4.6%) Statistical analysis rs63514 CC 63 (66.3%) 37 (56.9%) CT 27 (28.4%) 25 (38.5%) To test the population homogeneity of the study subjects, TT 5 (5.3%) 3 (4.6%) the allele frequencies were tested against Hardy-Weinberg rs335931 AA 63 (66.3%) 38 (58.5%) equilibrium (HWE) by the χ2-test. The plink v1.07 toolset was AG 27 (28.4%) 25 (38.5%) used to perform SNP association and haplotype2 (Purcell et al., GG 5 (5.3%) 2 (3%) 2007). The power analysis was performed using the “Quanto” tool.3 Statistical analyses were performed using commercially rs335930 AA 57 (60%) 36 (55.4%) available software (SPSS SmartViewer, version 18.0; SPSS, AC 32 (33.7%) 26 (40%) Chicago, IL, USA). P-values < 0.05 were considered statistically CC 6 (6.3) 3 (4.6%) significant. Association tests were corrected using the Benjamini rs335929 AA 63 (66.3%) 38 (58.5%) and Hochberg step-up False Discovery Rate (FDR) for multiple AC 27 (28.4%) 24 (36.9%) comparisons. CC 5 (5.3%) 3 (4.6%) rs16942851 TT 64 (67.4%) 38 (58.5%) 1 https://nda.nih.gov/edit_collection.html?id=2380 TG 26 (27.3%) 24 (36.9%) 2 https://zzz.bwh.harvard.edu/plink GG 5 (5.3%) 3 (4.6%) 3 https://bio.tools/QUANTO Frontiers in Aging Neuroscience 03 frontiersin.org Elias-Mas et al. 10.3389/fnagi.2022.1073258 TABLE 3 Single nucleotide polymorphism (SNP) allele association analysis. SNP FXTAS MAF No-FXTAS MAF OR (95% CI) P-value Adj P-value rs16942851 0.19 0.22 0.81 (0.47–1.41) 0.463 0.628 rs335929 0.19 0.22 0.84 (0.49–1.46) 0.538 0.628 rs335930 0.23 0.24 0.96 (0.57–1.63) 0.887 0.887 rs335931 0.19 0.22 0.84 (0.49–1.46) 0.538 0.628 rs63514 0.19 0.24 0.77 (0.45–1.33) 0.348 0.628 rs162008 0.19 0.25 0.72 (0.42–1.23) 0.224 0.628 rs162007 0.19 0.24 0.75 (0.43–1.28) 0.290 0.628 MAF, minor allele frequency; OR, odd-ratio. Adj p-value, adjusted p-value using Benjamini and Hochberg step-up false discovery rate, for multiple comparisons. size, the difference found might account for the CGG-repeat their association with clinical traits. In the present study we dependence described in clinical and neuropathologic features selected this method to investigate the role of AQP4 gene of FXTAS. variants in FXTAS susceptibility. We hypothesized that AQP4 In agreement with HWE, there was no deviation detected polymorphisms could play a role as risk factors for FXTAS. in any of the analyzed SNPs (p > 0.3). All SNPs studied However, we did not find any significant difference in the were in linkage disequilibrium (LD), and the pairwise LD distributions of alleles, genotypes, and haplotypes between coefficient (r2 ) ranged between 0.8 and 1. Table 2 shows the FXTAS and no-FXTAS individuals, after correction for multiple genotype frequency for each SNP according to the presence testing. of FXTAS symptoms. After correction of p-values for multiple A myriad of different studies point out AQP4 gene as a comparisons, there was no significant difference in frequencies novel candidate gene for brain plasticity and associated with of any of the analyzed SNPs between FXTAS and no-FXTAS neuropsychiatric and neurodegenerative disorders. According group (Table 3). Adjustment for sex did not change these to the human postmortem brain microarray data from the Allen results (data not shown). Table 4 shows the results of the Brain Atlas resources,4 AQP4 is most highly expressed in fronto- genotype association analysis between AQP4 polymorphisms limbic and temporal cortical regions. Both neuroanatomical and risk of FXTAS, according to different genetic inheritance areas are linked to cognitive and executive processes, and its models. disturbance leads to the neuropsychological changes described in many different movement disorders (Robertson et al., 2016). Although indirectly, genome-wide linkage studies have repeatedly pointed out the role of AQP4 in the development Discussion of brain disorders (Dadgostar et al., 2021). Genetic variations, abnormal distribution and quantitative abnormalities of AQP4 Fragile X-associated tremor/ataxia syndrome is a late- have also been associated with several neurodegenerative onset neurodegenerative disorder with reduced penetrance, disorders, such as AD, Parkinson’s disease and amyotrophic meaning that not all FMR1 premutation carriers will develop lateral sclerosis (reviewed in Mader and Brimberg, 2019). it (Hagerman et al., 2001). Among FMR1 premutation carriers Recently the rs162008, the most prevalent genetic variant older than 50 years, it has been estimated that 40% of men of AQP4, has been associated with a ∼15% change in and 16% of women will develop FXTAS symptoms, although AQP4 expression. In AD, genetic variations in AQP4 were there is significant variability in the progression of neurological shown to be associated with changes in sleep pattern and dysfunction (Coffey et al., 2008; Rodriguez-Revenga et al., increased β-amyloid (Rainey-Smith et al., 2018), as well as 2009). The description and characterization of FXTAS is of to β-amyloid accumulation and disease stage progression great interest, because the prevalence of FMR1 premutation (Burfeind et al., 2017; Chandra et al., 2021). Taking everything in the general population is relatively high. It has been into account, it is implied that AQP4 distribution and estimated that FMR1 premutation affects ∼1 out of 400 males regulation might have crucial role in neuronal activity and and 1 out of 200 females (Tassone et al., 2012), leading to function. symptoms of FXTAS in up to 1 in 3000 men older than Apart from intention tremor and cerebellar ataxia, core 50 years. Even though the FMR1 premutation is the major clinical features of FXTAS include executive dysfunction which risk factor for FXTAS, there are still some unknown genetic, may progress to dementia in some cases (Hagerman et al., epigenetic or environmental factors that might be affecting gene penetrance. Candidate gene SNP association analysis is a commonly used approach to identify risk alleles and 4 http://www.brain-map.org Frontiers in Aging Neuroscience 04 frontiersin.org Elias-Mas et al. 10.3389/fnagi.2022.1073258 TABLE 4 Genotype association using different genetic models. SNP Allele Test OR (95% CI) P-value Adj P-value rs16942851 G ADD 1.22 (0.52–2.83) 0.647 0.978 DOM 0.68 (0.35–1.31) 0.251 0.365 REC 1.75 (0.33–9.31) 0.512 0.854 rs335929 C ADD 1.23 (0.53–2.86) 0.634 0.978 DOM 0.71 (0.37–1.37) 0.313 0.365 REC 1.75 (0.33–9.31) 0.512 0.854 rs335930 C ADD 1.38 (0.60–3.15) 0.449 0.978 DOM 0.83 (0.43–1.57) 0.561 0.561 REC 2.12 (0.41–10.87) 0.366 0.854 rs335931 G ADD 1.23 (0.53–2.86) 0.634 0.978 DOM 0.71 (0.37–1.37) 0.313 0.365 REC 1.75 (0.33–9.31) 0.512 0.854 rs63514 T ADD 0.99 (0.47–2.08) 0.978 0.978 DOM 0.67 (0.35–1.29) 0.229 0.365 REC 1.15 (0.26–4.98) 0.854 0.854 rs162008 T ADD 0.97 (0.46–2.04) 0.932 0.978 DOM 0.60 (0.31–1.15) 0.125 0.365 REC 1.15 (0.26–4.98) 0.854 0.854 rs162007 A ADD 0.98 (0.47–2.06) 0.961 0.978 DOM 0.64 (0.33–1.23) 0.180 0.365 REC 1.15 (0.26–4.98) 0.854 0.854 ADD, additive model; DOM, dominant model; REC, recessive model. Adj p-value, adjusted p-value using Benjamini and Hochberg step-up false discovery rate, for multiple comparisons. 2001; Hall et al., 2016). In addition, several conditions affecting the SNPs analyzed and the risk of developing FXTAS (Table 4). sleep quality have been frequently described among FXATS Furthermore, no association was detected when comparing patients (Hamlin et al., 2011; Summers et al., 2014). FXTAS frequency distribution of the two major AQP4-haplotypes. In can coexist with other neurodegenerative disorders, such as fact, the frequency detected did not differ from the one described Parkinson’s disease and AD (Aydin et al., 2020; Salcedo-Arellano among CEU population (Table 5). Although no relationship et al., 2021b), suggesting a synergistic effect on the progression between genetic variants in AQP4 gene and FXTAS was found, of disease symptoms. On the basis of these observations and no association with changes in the development of the disease the evidence of the consequences of AQP4 dysfunction in has been assessed due to lack of clinical information. Similarly neurological conditions, we analyzed genetic variation of AQP4 with the AD, AQP4 SNPs have been associated with some aspects gene among FXTAS patients. We compared frequency of alleles, of the clinical course, such as faster cognitive decline, rather genotypes, and haplotypes of AQP4 between FXTAs and no- than the presence or absence of the disease (Burfeind et al., FXTAS cases. Results did not find any association between any of 2017). As previously described (Ulv Larsen et al., 2020), examination of the SNPs revealed two conserved haplotypes: TABLE 5 Major allele and minor allele haplotype (HTMa and HTMi) HtMa (haplotype for the major allele) and HtMi (haplotype for frequencies of AQP4 functional SNPs across the CEU (Utah residents with ancestry from Northern and Western Europe) population and the the minor allele). Haplotype frequency comparison by means FXTAS and no-FXTAS groups. of dominant analysis between FXTAS and no-FXTAS group did not show significant differences (p > 0.05). Moreover, both Haplotype FXTAS No-FXTAS frequency CEU (n = 190) (n = 130) groups showed similar haplotype frequency compared to the CEU population (Table 5). HTMa 0.767 0.76842105 0.73846154 Given the sample size this study had limited power. Post HTMi 0.1966 0.18947368 0.22307692 hoc power analyses showed that the power to detect the Frontiers in Aging Neuroscience 05 frontiersin.org Elias-Mas et al. 10.3389/fnagi.2022.1073258 observed odds ratios for FXTAS cases vs. no-FXTAS ranged Author contributions from 0.05 to 0.19. This study has two main limitations that might be explained All authors listed have made a substantial, direct, and because of the rarity of the disorder. First the relatively small intellectual contribution to the work, and approved it sample size. Power ranged only from 0.05 to 0.19. Second, for publication. the age differences between groups and the fact that those individuals considered as no-FXTAS may develop clinical symptoms later in life, masking differences among groups. Nonetheless, to our knowledge we are reporting for the Funding first time, that the AQP4 SNPs (rs162007, rs162008, rs63514, This work was supported by the Instituto de Salud Carlos rs335931, rs335930, rs335929, and rs16942851) and haplotypes III (ISCIII), (through the projects PI17/01067 and PI21/01085), were not associated with susceptibility of FXTAS in Caucasian co-funded by the European Union, and AGAUR from the population. Despite this lack of association, further studies are Autonomous Catalan Government (2017SGR1134). The CIBER necessary to fully discard the role of AQP4 and glymphatic de Enfermedades Raras is an initiative of the Instituto de system in the pathology of FXTAS. There is a need to describe Salud Carlos III. new evidence into how the glymphatic system functions, and how AQP4 dysfunction might take part into FXTAS disease progression. Conflict of interest Data availability statement The authors declare that the research was conducted in the absence of any commercial or financial relationships that could The original contributions presented in this study are be construed as a potential conflict of interest. included in the article, further inquiries can be directed to the corresponding authors. Publisher’s note Ethics statement All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated The studies involving human participants were reviewed organizations, or those of the publisher, the editors and the and approved by the Institutional Review Board at Emory reviewers. Any product that may be evaluated in this article, or University (IRB00074941). The patients/participants provided claim that may be made by its manufacturer, is not guaranteed their written informed consent to participate in this study. or endorsed by the publisher. References Aydin, E., Schneider, A., Protic, D., Wang, J., Martínez-Cerdeño, V., Tassone, Dadgostar, E., Tajiknia, V., Shamsaki, N., Naderi-Taheri, M., Aschner, F., et al. (2020). Rapidly progressing neurocognitive disorder in a male with M., Mirzaei, H., et al. (2021). Aquaporin 4 and brain-related disorders: FXTAS and Alzheimer’s Disease. Clin. Interv. Aging 15, 285–292. doi: 10.2147/CIA. Insights into its apoptosis roles. EXCLI J. 20, 983–994. doi: 10.17179/excli202 S240314 1-3735 Burfeind, K., Murchison, C., Westaway, S., Simon, M., Erten-Lyons, D., Hablitz, L., and Nedergaard, M. (2021). The glymphatic system: A novel Kaye, J., et al. (2017). The effects of noncoding aquaporin-4 single-nucleotide component of fundamental neurobiology. J. Neurosci. 41, 7698–7711. doi: 10. polymorphisms on cognition and functional progression of Alzheimer’s disease. 1523/JNEUROSCI.0619-21.2021 Alzheimers Dement (N Y) 3, 348–359. doi: 10.1016/j.trci.2017.05.001 Hagerman, R., Leehey, M., Heinrichs, W., Tassone, F., Wilson, R., Hills, J., Cabal-Herrera, A., Tassanakijpanich, N., Salcedo-Arellano, M., and et al. (2001). Intention tremor, parkinsonism, and generalized brain atrophy Hagerman, R. (2020). Fragile X-associated tremor/ataxia syndrome in male carriers of fragile X. Neurology 57, 127–130. doi: 10.1212/wnl. (FXTAS): Pathophysiology and clinical implications. Int. J. Mol. Sci. 21:4391. 57.1.12 doi: 10.3390/ijms21124391 Hall, D., Robertson, E., Shelton, A., Losh, M., Mila, M., Moreno, E., et al. Chandra, A., Farrell, C., Wilson, H., Dervenoulas, G., De Natale, E., and (2016). Update on the clinical, radiographic, and neurobehavioral manifestations Politis, M. (2021). Alzheimer’s Disease Neuroimaging Initiative. Aquaporin-4 in FXTAS and FMR1 premutation carriers. Cerebellum 15, 578–586. doi: 10.1007/ polymorphisms predict amyloid burden and clinical outcome in the Alzheimer’s s12311-016-0799-4 disease spectrum. Neurobiol. Aging 97, 1–9. Hamlin, A., Liu, Y., Nguyen, D., Tassone, F., Zhang, L., and Hagerman, R. Coffey, S., Cook, K., Tartaglia, N., Tassone, F., Nguyen, D., Pan, R., et al. (2008). (2011). Sleep apnea in fragile X premutation carriers with and without FXTAS. Expanded clinical phenotype of women with the FMR1 premutation. Am. J. Med. Am. J. Med. Genet. B Neuropsychiatr. Genet. 156B, 923–928. doi: 10.1002/ajmg.b. Genet. A 146A, 1009–1016. doi: 10.1002/ajmg.a.32060 31237 Frontiers in Aging Neuroscience 06 frontiersin.org Elias-Mas et al. 10.3389/fnagi.2022.1073258 Hladky, S., and Barrand, M. (2022). The glymphatic hypothesis: The theory and Rodriguez-Revenga, L., Madrigal, I., Pagonabarraga, J., Xunclà, M., Badenas, the evidence. Fluids Barriers CNS 19:9. doi: 10.1186/s12987-021-00282-z C., Kulisevsky, J., et al. (2009). Penetrance of FMR1 premutation associated pathologies in fragile X syndrome families. Eur. J. Hum. Genet. 17, 1359–1362. Iliff, J., Goldman, S., and Nedergaard, M. (2015). Implications of the discovery doi: 10.1038/ejhg.2009.51 of brain lymphatic pathways. Lancet Neurol. 14, 977–979. doi: 10.1016/S1474- 4422(15)00221-5 Salcedo-Arellano, M., Wang, J., McLennan, Y., Doan, M., Cabal-Herrera, A., Jimenez, S., et al. (2021a). Cerebral microbleeds in fragile X-associated Johnston, H., Chopra, P., Wingo, T., Patel, V., Epstein, M., Mulle, J., et al. tremor/ataxia syndrome. Mov. Disord. 36, 1935–1943. doi: 10.1002/mds. (2017). PEMapper and PECaller provide a simplified approach to whole-genome 28559 sequencing. Proc. Natl. Acad. Sci. U.S.A. 114, E1923–E1932. doi: 10.1073/pnas. 1618065114 Salcedo-Arellano, M., Sanchez, D., Wang, J., McLennan, Y., Clark, C., Juarez, P., et al. (2021b). Case report: Coexistence of alzheimer-type neuropathology in Kaur, J., Fahmy, L., Davoodi-Bojd, E., Zhang, L., Ding, G., Hu, J., et al. (2021). fragile X-Associated tremor ataxia syndrome. Front. Neurosci. 15:720253. doi: Waste clearance in the brain. Front. Neuroanat. 15:665803. doi: 10.3389/fnana. 10.3389/fnins.2021.720253 2021.665803 Silva, I., Silva, J., Ferreira, R., and Trigo, D. (2021). Glymphatic system, AQP4, Kong, H., Lim, J., Linsalata, A., Kang, Y., Malik, I., Allen, E., et al. (2022). and their implications in Alzheimer’s disease. Neurol. Res. Pract. 3:5. doi: 10.1186/ Identification of PSMB5 as a genetic modifier of fragile X-associated tremor/ataxia s42466-021-00102-7 syndrome. Proc. Natl. Acad. Sci. U.S.A. 119:e2118124119. doi: 10.1073/pnas. 2118124119 Summers, S., Cogswell, J., Goodrich, J., Mu, Y., Nguyen, D., Brass, S., et al. (2014). Prevalence of restless legs syndrome and sleep quality in carriers Mader, S., and Brimberg, L. (2019). Aquaporin-4 water channel in the brain and of the fragile X premutation. Clin. Genet. 86, 181–184. doi: 10.1111/cge. its implication for health and disease. Cells 8:90. doi: 10.3390/cells8020090 12249 Nagelhus, E., and Ottersen, O. (2013). Physiological roles of aquaporin-4 in Szczygielski, J., Kopańska, M., Wysocka, A., and Oertel, J. (2021). Cerebral brain. Physiol. Rev. 93, 1543–1562. doi: 10.1152/physrev.00011.2013 microcirculation, perivascular unit, and glymphatic system: Role of Aquaporin- Nedergaard, M., and Goldman, S. (2020). Glymphatic failure as a final common 4 as the gatekeeper for water homeostasis. Front. Neurol. 12:767470. doi: 10.3389/ pathway to dementia. Science 370, 50–56. doi: 10.1126/science.abb8739 fneur.2021.767470 Papadopoulos, M., and Verkman, A. (2013). Aquaporin water channels in the Tassone, F., Iong, K., Tong, T., Lo, J., Gane, L., Berry-Kravis, E., et al. (2012). nervous system. Nat. Rev. Neurosci. 14, 265–277. doi: 10.1038/nrn3468 FMR1 CGG allele size and prevalence ascertained through newborn screening in the United States. Genome Med. 4:100. doi: 10.1186/gm401 Purcell, S., Neale, B., Todd-Brown, K., Thomas, L., Ferreira, M., Bender, D., et al. (2007). PLINK: A tool set for whole-genome association and population-based Ulv Larsen, S., Landolt, H., Berger, W., Nedergaard, M., Knudsen, G., and linkage analyses. Am. J. Hum. Genet. 81, 559–575. doi: 10.1086/519795 Holst, S. (2020). Haplotype of the astrocytic water channel AQP4 is associated with slow wave energy regulation in human NREM sleep. PLoS Biol. 18:e3000623. Rainey-Smith, S., Mazzucchelli, G., Villemagne, V., Brown, B., Porter, T., doi: 10.1371/journal.pbio.3000623 Weinborn, M., et al. (2018). Genetic variation in Aquaporin-4 moderates the relationship between sleep and brain Aβ-amyloid burden. Transl. Psychiatry 8:47. Wang, Y., Huang, C., Guo, Q., and Chu, H. (2022). Aquaporin-4 and cognitive doi: 10.1038/s41398-018-0094-x disorders. Aging Dis. 13, 61–72. doi: 10.14336/AD.2021.0731 Robertson, E., Hall, D., McAsey, A., and O’Keefe, J. (2016). Fragile Zeppenfeld, D., Simon, M., Haswell, J., D’Abreo, D., Murchison, C., Quinn, X-associated tremor/ataxia syndrome: Phenotypic comparisons with other J., et al. (2017). Association of Perivascular localization of aquaporin-4 With movement disorders. Clin. Neuropsychol. 30, 849–900. doi: 10.1080/13854046. cognition and alzheimer disease in aging brains. JAMA Neurol. 74, 91–99. doi: 2016.1202239 10.1001/jamaneurol.2016.4370 Frontiers in Aging Neuroscience 07 frontiersin.org