henhao
Sign in
NeurosciencebioRxiv
Preprint

PTEN Subcellular Localization Dictates Function

Nicole M. DesmetiD, Cheyenne F. Griffin, H. Nephi SeoiD, Annaliese OuYang, Penelope Tir, Mackenzi L. PrinaiD, Wei WangiD, Meijie LiiD, Bryan W. LuikartiD

Abstract

Mutations in phosphatase and tensin homolog (PTEN) drive unregulated activation of the phosphatidylinositol-3-kinase (PI3K) pathway, resulting in neuronal hypertrophy, and are strongly associated with autism spectrum disorder (ASD). Several PTEN mutations alter subcellular localization, yet how localization governs PTEN function in developing neurons remains unclear. Although PTEN has been reported broadly distributed throughout neurons, here, live imaging of HaloTagged PTEN reveals dynamically regulated localization, suggesting spatial control of its signaling. We then used retroviral-mediated genetic manipulation to delete endogenous Pten in developing hippocampal neurons while simultaneously expressing PTEN fused to defined localization motifs, allowing us to directly test how subcellular targeting regulates neuronal morphology. Loss of Pten produces neurons characterized by enlarged somata, more elaborate dendritic arbors, and increased spine density, length, and head area. Nuclear-excluded PTEN fully rescued these phenotypes, whereas targeting PTEN to filopodia via fusion to the FBAR domain of srGAP3 or to the postsynaptic density via Homer1C corrected or corrected all morphological abnormalities in PTEN-deficient neurons and simplified dendritic arborization compared to wild-type. In contrast, nuclear-localized PTEN produced only partial rescue, normalizing soma size and spine head area but not dendritic complexity or spine density. These findings indicate that PTEN acts locally to restrain growth and structural connectivity, whereas regulation of spine head size can be mediated by PTEN both inside and outside the nucleus, potentially through transcriptional or splicing-dependent mechanisms. Together, our results identify subcellular localization as a critical determinant of PTEN function and reveal spatially distinct mechanisms through which PTEN sculpts neuronal development.

PDFDOIData
Full-text digest ↓

02 /Specification & Classification

Click any tag to browse matching papers
Citations
1
OpenAlex Topics

In-depth summary

✦ Multi-agent verified full-text synthesis

Claude Opus 5.5 · xhigh effort · 7 Nov 2026 · from the full text of version 2 (8 Oct 2026) · CC BY 4.0

Nuclear-excluded PTEN rescued all knockout overgrowth phenotypes in mouse dentate granule neurons; nuclear-restricted PTEN rescued only spine head area and partly soma size, so localization shapes function.

Several PTEN mutations alter its subcellular localization, but how localization governs PTEN function in developing neurons was unclear. Here endogenous Pten is deleted in single neurons and replaced with PTEN fused to compartment-targeting motifs, so each localization can be tested for rescue.

Key results

  1. ShownForcing PTEN into the nucleus only partly rescues the soma enlargement of Pten knockout granule neurons. Nuclear-excluded PTEN and the filopodia- and PSD-targeted fusions bring soma size back to near wildtype.

    Soma cross-sectional area: wildtype 111.4 µm^2, Pten KO 222.0, NES-PTEN 98.32, PTEN-FBAR 103.8, PTEN-HOMER 103.7, NLS-PTEN 135.8 (p<0.0001 vs both wildtype and KO). · Nuclear PTEN Partially Regulates Soma…; Figure legends

  2. ShownNuclear PTEN does not normalize dendritic architecture. NLS-PTEN neurons resemble knockouts in primary dendrite number and inner/middle molecular layer branching, while NES-PTEN matches wildtype.

    Primary dendrites: KO 2.429, NLS-PTEN 2.346, wildtype/PTEN/NES-PTEN 1.000. Total dendritic length: KO 1573 µm, wildtype 1218, NES-PTEN 1276, NLS-PTEN 1435 (a trend only). · Nuclear PTEN Does Not Regulate Dendritic…; Figure legends

  3. ShownTargeting PTEN to filopodia (FBAR) or the postsynaptic density (HOMER) reduces dendritic branching relative to wildtype in the inner and middle molecular layer. Total dendritic length only trends lower and is not significantly different from wildtype.

    Single primary dendrite in both groups; Sholl p<0.0001 vs wildtype. Total length 1079 µm (FBAR) and 1082 µm (HOMER) vs 1218 µm wildtype (p=0.9760 and p=0.9980). · Filopodial and PSD-localized PTEN Decrease…; Spatial enrichment of PTEN restricts dendritic…

  4. ShownExtra-nuclear PTEN restores spine density, length and head area. NLS-PTEN restores spine head area but not spine length, and spine density only trends higher. A second, independent two-photon thresholding analysis also found head area rescued by NLS-PTEN.

    Spine length: wildtype 0.9386 µm, KO 1.1790, NLS-PTEN 1.1180 (p<0.0001 vs wildtype). Head area: NLS-PTEN 0.2726 µm^2 vs wildtype 0.2840 (p=0.9860). Density: KO 3.102/µm, NLS-PTEN 2.558. · Extra-Nuclear PTEN Rescues Dendritic Spine…; Compartment-specific regulation of dendritic…

  5. SuggestedLive imaging of PTEN-Halo9 shows non-uniform localization, including growth cones and filopodia. A pulse-chase assay gives a PTEN lifetime of about 2.5 days in dentate granule neurons.

    Lifetime 2.50±0.14 days, averaged over 24h-96h chases with 2 mice per time point. The PTEN-Halo9 soma area of 91.42 µm^2 was not significantly different from wildtype (p=0.0580). · PTEN Localization is Regulated in Developing…; PTEN Lifetime Assessed Via HaloTag Pulse-Chase; PTEN lifetime and localization

Methods, data and reproducibility

Retroviral Cre delivery at P7 into the dentate gyrus of Pten flox/flox x Ai14 mice deleted endogenous Pten and expressed PTEN fused to NLS, NES, srGAP3 FBAR or Homer1C. Morphology was quantified at 21 days post-injection by soma area, Sholl analysis and spine measures; HaloTag live imaging and pulse-chase tracked PTEN.

  • Dentate granule neurons infected with retroviral constructs; soma area, Sholl/Neurolucida reconstructions, spine density, length and head area · mouse (Pten flox/flox x Ai14 on C57BL/6J background) · n = n=animals unless noted; per-group n in Table 1; PTEN OE in wildtype n=17
  • Fusion-protein localization by PTEN immunohistochemistry (cytoplasm/nucleus and spine/dendrite intensity ratios) and in HEK293ft cells · mouse (neurons); human cell line (HEK293ft)
  • HaloTag pulse-chase for PTEN lifetime, chase intervals 1.5h to 96h · mouse (C57BL/6J) · n = 2 mice per chase time point
  • Live two-photon imaging of PTEN-Halo9 in acute slices (two supplemental videos) · mouse · n = 2 example movies
  • Independent two-photon spine head area dataset from a second set of animals · mouse
Methods & tools
HaloTag (Halo9), DELTA pulse-chase, AlphaFold, Neurolucida 360, NeuronStudio, ImageJ/Fiji, GraphPad Prism, Stata mixed-effects models, pRubi retrovirus, Ai14 tdTomato reporter
Data & accessions
https://neuromorpho.org/
Reproducibility gaps
No original code is reported. Neuron reconstructions are on Neuromorpho.org; image stacks, ROI sets and spreadsheets are available only on request from the lead contact. Viral constructs are being submitted to Addgene.
Limitations (5)
  • Fusion motifs bias PTEN toward compartments rather than confine it; PTEN trafficks dynamically and NLS/NES tags likely allow some nucleocytoplasmic shuttling. · stated by the authors
  • Only structural outcomes were measured; electrophysiology is needed to establish effects on synaptic function and excitability. · stated by the authors
  • Optical measurements overestimate absolute submicron spine sizes and should be read as relative differences between groups. The growth-cone observation in the live imaging needs quantitation. · stated by the authors
  • The FBAR and HOMER fusions shifted spine/dendrite ratios only modestly (0.883 for PTEN-Halo9 vs 0.985 and 1.020). · noted in this summary
  • Wildtype and Pten KO control data partly reuse previously published data, and the lifetime estimate rests on 2 mice per time point. · noted in this summary

Who should read:Read this if you study PTEN, PI3K pathway or autism-associated mutations and want in vivo evidence on how subcellular localization affects neuronal soma, dendrite and spine phenotypes.

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
THISLedderose '22Bracha '26Pigoni '22Abdulkareem '26Dhaliwal '26Yonan '24
71% semantic matchNeuroscience
Julia M T Ledderose, Jorge A. Benítez, Amanda J. Roberts+6 more·Brain·May 23, 2022· 7 citations