Taconic at SfN 2026

Superimmunodeficient NOG mice support engraftment of human cells and tissues, but complex allele engineering in this background has been limited by the lack of germline-competent embryonic stem cells (ESCs). We conducted an exploratory study to derive, characterize, and genetically modify ESCs from NOG (NOD.Cg-Prkdc^scidIl2rg^tm1Sug/JicTac) mice and to establish feasibility for generating neuroscience-relevant humanized knock-in alleles. Blastocysts were isolated at 3.5 days post coitum and inner cell mass outgrowths were cultured on mouse embryonic fibroblast feeders in LIF/2i conditions. ESC lines were expanded, cryopreserved, and assessed for ploidy, sex, and pluripotency marker expression. Germline competence was tested by blastocyst injection. To demonstrate targeted engineering in this background, the endogenous mouse Tfrc locus was modified by humanized knock-in replacement of the extracellular domain with the orthologous human TFRC sequence, while retaining mouse transmembrane and intracellular domains. Targeted clones were evaluated by Southern blotting and ploidy analysis before blastocyst injection. Thirteen ESC lines were derived from 38 blastocysts; nine were validated for ploidy, sex, and pluripotency markers. Two of six tested ESC lines transmitted through the germline. Tfrc targeting in one ESC line yielded validated humanized TFRC knock-in clones; three targeted clones were injected, and one transmitted the humanized allele to the G1 generation. These data establish germline-competent ESCs from the NOG background and demonstrate that large, locus-specific humanized alleles can be generated directly in this immunodeficient strain. Humanized TFRC NOG mice may provide a platform for evaluating receptor-mediated blood–brain barrier drug delivery in an immunodeficient background, including studies that combine BBB-targeted therapeutics with human cell or tissue engraftment. This model context may be particularly relevant for neuroscience programs that require assessment of human-targeted BBB transport mechanisms alongside xenogeneic cell therapy approaches, human immune-cell engraftment, tumor xenografts, or other humanized tissue systems.

Efficient, tissue-specific delivery of biologics remains a major hurdle in therapeutic development. This challenge is especially acute in the central nervous system (CNS), where the blood–brain barrier (BBB) restricts drug access. TfR1-mediated transcytosis has emerged as a strategy to enhance CNS delivery, yet its brain-wide vascular and parenchymal biodistribution is not well quantified at cellular resolution. Moreover, widespread TfR1 expression in peripheral tissues raises concerns about off-target exposure that remain poorly characterized at the whole-organ level. 

Here, we address these challenges by using a humanized transferrin receptor 1 (hTfR1) mouse model (Taconic Biosciences, strain 23236; hTFRC). Using this model, we combined whole-organ 3D light-sheet fluorescence microscopy (LSFM) with AI-based image analysis to map the biodistribution of four clinically relevant biologics: Trontinemab and Tividenofusp alfa (TfR1-targeted), and Gantenerumab and Efavaleukin alfa (non–TfR-binding controls). 

Whole brains and hindlimbs were immunolabeled alongside CD31 to visualize vasculature. Deep learning–based segmentation enabled 3D vascular reconstruction and voxel-level classification of signal in vascular versus parenchymal compartments. Brain-wide antibody quantification was performed using the Allen Mouse Brain Atlas, while hindlimb analysis assessed peripheral exposure in the bone marrow and skeletal muscle, which are TfR1-expressing tissues.  

Our results indicated modality-dependent differences in antibody distribution. Specifically, TfR1-targeted biologics had enhanced parenchymal brain penetration compared to non-targeted control biologics, which remained largely vascular- or barrier-restricted. Regional analysis revealed heterogeneous distribution across brain and peripheral tissues, highlighting the dual impact of hTfR1 on CNS delivery and systemic exposure in hTFRC mice. 

This study establishes 1) the hTFRC mouse model as an adequate mouse model system to test biologics and 2) an LSFM and AI-driven platform for unbiased whole-organ biodistribution analysis. Together, this approach allows for simultaneous assessment of BBB transport efficiency, regional brain exposure, and peripheral liabilities to support the design of next-generation BBB-penetrant biologics.

Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by cognitive decline, amyloid-β (Aβ) pathology, and a range of neuropsychiatric and sleep-related disturbances. The ARTE10 transgenic mouse model, which co-expresses human APP and PS1 mutations associated with familial AD, develops early and robust cerebral Aβ deposition, providing a valuable platform for translational AD research.

In the present study, we performed a longitudinal characterization of ARTE10 mice to evaluate the development of electrophysiological, and sleep-related phenotypes across disease progression. N=32, n=16 male WT and 16 male homozygous ARTE10 mice were surgically prepared for use with the non-tethered TaiNi EEG system. All mice underwent monthly assessments from 5 to 15 months of age using quantitative electroencephalography (qEEG) and sleep architecture analysis.

Electrophysiological recordings revealed a pronounced hyperexcitability phenotype in ARTE10 mice. Compared with wild-type controls, transgenic animals exhibited significant alterations in sleep architecture, including reduced non-rapid eye movement (NREM) sleep at later ages [insert statistics]. qEEG analyses demonstrated persistently elevated spectral power across multiple frequency bands, including increased theta-band activity [insert statistics]. Furthermore, ARTE10 mice displayed a marked reduction in normal sleep spindle activity and an increased incidence of spike-and-wave discharges (SWDs), predominantly during wakefulness, indicative of aberrant network excitability and seizure-like activity.

Collectively, these findings demonstrate that the ARTE10 mouse develops robust and progressive phenotypes that recapitulate key pathological and neurophysiological features observed in AD patients, including sleep disruption and cortical hyperexcitability. The model exhibits sustained survival to 15 months of age while providing an early and measurable disease phenotype, supporting its utility for evaluating both disease-modifying and symptomatic therapeutic interventions targeting AD and its associated comorbidities.

Alzheimer's disease (AD) is characterized by a prolonged preclinical phase during which progressive brain dysfunction precedes overt cognitive decline. Identifying translational functional biomarkers and preclinical models that reliably capture early network alterations remains critical to improve disease characterization and therapeutic development. Electroencephalography (EEG), a highly translational technique, offers a unique opportunity to investigate disease-related brain dysfunction in patients and preclinical models.

The ARTE10 transgenic mouse model co-expresses mutant human amyloid precursor protein (APP) and presenilin-1 (PS1) and develops early, robust amyloid pathology. Despite its behavioral characterization, its neurophysiological phenotype remains poorly defined. Here, we performed longitudinal EEG and sleep phenotyping of ARTE10 mice and WT littermates to characterize spontaneous cortical activity, sleep-wake architecture, and auditory evoked responses.

Twenty-four-hour polysomnographic EEG recordings were conducted at 5 and 12 months, corresponding to early amyloid deposition and advanced disease, respectively. Auditory steady-state responses (ASSRs) and auditory event-related potentials (AERPs) were recorded longitudinally at 5, 7, 9, and 12 months to probe cortical synchronization and sensory processing.

ARTE10 mice exhibited age-dependent alterations in sleep-wake organization, with increased nocturnal wakefulness, reduced Non-Rapid Eye Movement (NREM) sleep, and altered REM dynamics, resembling disturbances reported in AD patients. EEG spectral analysis revealed an early-onset phenotype at 5 months characterized by cortical hyperexcitability, increased high-frequency activity, and epileptiform discharges. With aging, this phenotype progressively evolved toward cortical slowing and pronounced reduction in interhemispheric functional connectivity, reflecting large-scale network disorganization. Auditory paradigms uncovered early and persistent impairments in cortical processing, including reduced 40-Hz ASSR power and synchronization, atypical AERP waveforms, and impaired sensory gating. These abnormalities preceded overt cognitive impairment.

Together, these findings demonstrate that ARTE10 mice exhibit early, persistent, and progressive EEG abnormalities aligned with electrophysiological signatures of human AD. Longitudinal EEG, sleep-wake metrics, and auditory evoked measures provide robust translational biomarkers, supporting the ARTE10 model for investigating AD from prodromal to symptomatic stages and for evaluating therapeutic strategies.