henhaobio
Sign in

bioRxivNeurosciencePreprint

Impaired astrocyte-to-neuron cholesterol trafficking drives synaptic dysfunction in Rett syndrome

Francesca M PostognaiD, Noemi Giancroce, Cecilia CabasinoiD, Fabio BiellaiD, Ottavia Maria RoggeroiD, Martina Breccia, Laura MorelliiD, Diego ColomboiD, Alessandro Arcari, Giulia LunghiiD, Manuela ValsecchiiD, Elena ChiricozziiD, Nicoletta LandsbergeriD, Marta ValenzaiD, Angelisa FrascaiD

PDFDOI
In-depth full-text digest available · Jump to digest ↓

Abstract

Rett syndrome (RTT) is a severe X-linked neurodevelopmental disorder caused by loss-of-function mutations in the MECP2 gene and characterized by profound impairments in neuronal maturation and synaptic connectivity. Increasing evidence indicates that astrocyte dysfunction contributes to RTT pathogenesis through non-cell-autonomous mechanisms, although the molecular pathways underlying defective astrocyte-neuron communication are only partially understood. Astrocytes are the primary source of cholesterol in the brain and support neuronal maturation and synaptic function by supplying cholesterol through ApoE-containing lipoproteins. Although alterations in brain cholesterol metabolism have been reported in RTT, the underlying cellular mechanisms and their functional consequences remain poorly investigated. Here, we studied cholesterol homeostasis in Mecp2 knock-out (KO) astrocytes and its impact on neuron-astrocyte communication. Mecp2 KO astrocytes exhibited reduced nuclear localization of the transcriptional regulator Srebp2, together with the downregulation of genes involved in cholesterol biosynthesis and transport. These molecular alterations were associated with intracellular cholesterol and desmosterol accumulation, reduced Abca1 expression and defective ApoE lipidation, despite preserved ApoE expression and cholesterol secretion. Importantly, similar alterations were detected in acutely isolated astrocytes and in the cerebral cortex of Mecp2 deficient mice, demonstrating that impaired cholesterol homeostasis extends beyond in vitro models. Functionally, cholesterol supplementation of astrocyte-conditioned medium rescued the synaptic defects induced in wild-type neurons by Mecp2 KO astrocytes. Moreover, cholesterol treatment restored pre- and post-synaptic density, as well as axon initial segment length in Mecp2 heterozygous (HET) neurons. Together, these findings identify defective astrocyte-to- neuron cholesterol trafficking as a key mechanism contributing to neuronal dysfunction in RTT and suggest that strategies aimed at restoring cholesterol functional availability might represent a promising therapeutic avenue for RTT.

In-depth summary

Claude Opus 5.5 · xhigh effort · 7 Nov 2026 · from the full text of version 1 (31 Jul 2026) · CC BY 4.0

Mecp2 knock-out mouse astrocytes accumulate cholesterol yet release poorly lipidated ApoE; cholesterol supplementation restores synaptic puncta density in neurons exposed to their conditioned medium and in Mecp2 heterozygous neurons.

Medium from Mecp2-deficient astrocytes impairs neuronal development and cholesterol metabolism is altered in Rett syndrome (RTT) mice and patients, but the authors note no prior test of whether astrocytic cholesterol defects affect astrocyte-neuron crosstalk or whether supplementation helps. This work places the defect in apolipoprotein E (ApoE) lipidation rather than in total or secreted cholesterol, and tests supplementation on RTT model neurons.

Key results

  1. ShownAdding cholesterol (0.1 μg/ml, complexed with methyl-β-cyclodextrin, 24 h) reverses pre- and post-synaptic puncta loss in wild-type (WT) neurons exposed to medium conditioned by Mecp2 knock-out (KO) astrocytes (KO-ACM) and in Mecp2 heterozygous (HET) neurons.

    Synapsin1/2, Shank2 and colocalized puncta, reduced by KO-ACM, fully restored with cholesterol (2-way ANOVA, n=40/50); HET deficits recovered (n=57/69). Axon initial segment (AIS, Ankyrin-G): mean length not different from WT with or without treatment; about 40% of HET neurons had AIS under 12 µm vs about 20% in treated groups (n=70/group); treatment shifted HET AIS distally (p<0.05). · Cholesterol treatment on neurons; Cholesterol supplementation prevents synaptic…; Figure legends

  2. ShownCultured Mecp2 KO cortical astrocytes show less nuclear sterol regulatory element-binding protein 2 (Srebp2) and lower expression of cholesterol synthesis and export genes than WT cells, with ApoE mRNA unaffected.

    Nuclear Srebp2 immunofluorescence reduced (p<0.01; n=25/30 fields from 6/7 biological replicates); total Srebp2 protein and mRNA unchanged. qRT-PCR in P1-P3 cultures: Nsdhl, Hmgcr, Mvk, Sqle, Abca1, Abcg1, Lcat reduced, most confirmed in bead-sorted P7 astrocytes (n=6/10). Western blot: Nsdhl and Abca1 protein reduced, Lcat only a trend (n=6/group). · Cholesterol metabolic pathway is impaired in…; Figure legends

  3. ShownMecp2 KO astrocytes accumulate intracellular sterols, yet the amount of cholesterol and ApoE they release into conditioned medium is unchanged.

    Gas chromatography-mass spectrometry (GC-MS): desmosterol increased (p<0.01, Mann-Whitney, n=6-8/group), cholesterol only a trend. Filipin (free cholesterol) density per astrocyte area increased (p<0.0001, n=80/100). Medium cholesterol unchanged by GC-MS and Amplex Red (n=6/group), and also in medium of acutely isolated astrocytes; medium ApoE unchanged by western blot (n=12/group). · Cholesterol quantitative analysis; ApoE lipidation is defective in cortical Mecp2…; Figure legends

  4. SuggestedLess of the ApoE released by Mecp2 KO astrocytes is in lipidated complexes; the authors propose the reduced Abca1, which transfers free cholesterol to ApoE, as the cause.

    Native (non-denaturing) gel of 10-fold concentrated conditioned medium probed for ApoE showed visibly fewer lipidated ApoE complexes in KO than WT medium. The assessment is described as qualitative; no densitometry or replicate number is reported for this blot. · Astrocyte-Conditioned Medium (ACM) preparation; Non-denaturing gel electrophoresis; ApoE lipidation is defective in cortical Mecp2…; Figure legends

  5. ShownMecp2-deficient mouse brain shows matching changes: higher cortical cholesterol by high-performance thin-layer chromatography (HPTLC), lower cholesterol-gene expression, reduced Nsdhl protein across regions, ages and a second Mecp2 model, and lower cortical Abca1 with ApoE and Lcat unchanged.

    HPTLC: cortical cholesterol up in P40 KO (p<0.05, n=8/group). GSEA of P60 cortex RNA-seq: cholesterol homeostasis NES -1.57, FDR q=0.044. P40 cortex qRT-PCR: Nsdhl, Hmgcr, Sqle, Cyp51 down. Nsdhl protein down in cortex, hippocampus, cerebellum at P20 and P40, about 50% lower in Y120D knock-in brain, cortex-only in P200 HET females. Cortical Abca1 down; ApoE, Lcat protein unchanged. · Molecular evidence of dysfunctional cholesterol…; Figure legends

Methods, data and reproducibility

Cholesterol-pathway gene and protein expression, Srebp2 localization, sterol levels (GC-MS, filipin, Amplex Red, HPTLC) and ApoE lipidation (native gel) were compared between Mecp2 KO and WT mouse astrocytes (cultured and bead-sorted) and brain tissue. WT and Mecp2 HET cortical neurons were then treated with conditioned medium and/or cholesterol and immunostained to score synaptic puncta and AIS length.

  • Primary cortical astrocyte cultures from P1-P3 WT and Mecp2 KO CD1 mice, and their conditioned medium · mouse · n = 6/10 (qRT-PCR); 6/exp.group (western blot); 6-8/exp.group (GC-MS)
  • Cortical astrocytes acutely isolated with anti-ACSA-2 microbeads from P7 and P40 WT and Mecp2 KO mice · mouse · n = 6/10 (P7 qRT-PCR)
  • Cerebral cortex, hippocampus and cerebellum of Mecp2 KO mice and WT littermates at P20, P40 and P60 · mouse · n = 8/group (HPTLC, qRT-PCR); 5-10 (Nsdhl blots); 8-13 (Abca1/Lcat/ApoE)
  • Previously published cortical RNA-seq from WT and Mecp2 KO mice at P60, reanalysed by GSEA · mouse
  • Brain tissue from Mecp2^(Y120D/y) knock-in mice (phospho-mimetic missense mutation) · mouse
  • Brain tissue from symptomatic Mecp2 HET female mice at P200 · mouse
  • Primary cortical neurons from E15.5 WT and Mecp2 HET embryos, treated with conditioned medium and/or cholesterol · mouse · n = 40/50 (KO-ACM rescue); 57/69 (HET puncta); 70/exp.group (AIS)
Methods & tools
GraphPad Prism 9, Fiji / ImageJ 2.14.0, GC-MS (Trace GC-Ultra 60 with Trace DSQ, Thermo Scientific), Amplex Red Cholesterol Assay kit, HPTLC, filipin staining, anti-ACSA-2 MicroBead Kit (Miltenyi Biotec), qRT-PCR, western blot and native gel electrophoresis, GSEA, immunofluorescence (Leica Stellaris confocal, Nikon Eclipse Ti2-E), cholesterol-methyl-β-cyclodextrin (Merck C4951)
Reagents & environment
GraphPad Prism 9 for statistics; ImageJ 2.14.0 (Java 1.8.0_322) for HPTLC band quantification; Fiji for puncta, Srebp2 and filipin image analysis.
Reproducibility gaps
No code or data availability statement; the reanalysed P60 cortical RNA-seq is cited as previously published with no accession given; primer sequences are in Tables 1-2, whose contents do not appear in the text.
Limitations (6)
  • The study does not establish a direct causal link between reduced ApoE lipidation and the synaptic defects; the authors present this as a hypothesis their data support. · stated by the authors
  • Brain tissue analyses were limited to cholesterol quantification and gene and protein expression; a fuller lipidomic profile of Mecp2-deficient brains is left for future work. · stated by the authors
  • Whether Nsdhl is a direct MeCP2 target, and whether its reduced expression alone disturbs astrocyte cholesterol homeostasis, remains to be tested. · stated by the authors
  • Lipid-raft restoration is proposed as one mechanism for the rescue, but the authors note that direct evidence of lipid raft changes in RTT is lacking. · stated by the authors
  • ApoE lipidation, the core of the trafficking claim, rests on a qualitative native-gel blot with no reported quantification or replicate number, and the rescue uses exogenous cyclodextrin-complexed cholesterol, which bypasses the ApoE route rather than testing it. · noted in this summary
  • Functional readouts are in vitro morphology only (puncta counts, Ankyrin-G AIS, a single 0.1 μg/ml 24 h dose); mean AIS length did not differ between HET and WT neurons, so the claimed AIS rescue rests on a frequency-distribution reading, and no electrophysiology or in vivo treatment is reported. · noted in this summary

Read this if you work on astrocyte-to-neuron lipid supply, ApoE lipoprotein biology or Rett syndrome mechanisms and want mouse culture and tissue data linking Mecp2 loss to cholesterol handling, plus an in vitro synaptic rescue.

Citations
1

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISAlbizzati '23Tomasello '24Ojha '26Korsakova '21Sinha '26Vartiainen '25
82% semantic matchNeuroscience
E. Albizzati, Elena Maria Florio, Massimo Breccia+8 more·bioRxiv·Feb 8, 2023· 1 citations