Intracerebroventricular diphtheria toxin causes off-target toxicity in CD11b-DTR and wild-type mice, revealing...

Diphtheria toxin receptor (DTR)-based depletion models are widely used to study microglial and macrophage function, yet the extent to which diphtheria toxin (DT) produces off-target effects remains incompletely defined. Here, we examined tolerability, behavioural outcomes, and cellular responses following intracerebroventricular (i.c.v.) DT administration in wild-type (WT) and CD11b-DTR mice. Mice received bilateral i.c.v. infusions of DT or vehicle over a 10-day period and were assessed for survival, motor and cognitive behaviour, myeloid cell changes, and neuropathology. Unexpectedly, DT induced dose-dependent mortality in WT mice, demonstrating that toxicity can occur independently of DTR expression. CD11b-DTR mice exhibited greater susceptibility, with reduced survival and the emergence of illness at lower DT doses. Behavioural testing revealed significant dose-dependent impairments in rotarod performance and Y-maze spontaneous alternation in both genotypes, while open-field mobility was largely preserved among animals. Region-specific analysis of myeloid cells in CD11b-DTR mice showed robust depletion in the midbrain at higher DT doses, whereas hippocampal cell numbers remained unchanged with marked morphological signs of activation. These findings indicate that DT-mediated myeloid cell responses vary across brain regions, potentially reflecting differential toxin exposure following ventricular delivery. Consistent with this, focal abnormalities in the brain-including ventriculitis, meningoencephalitis and spongiotic changes-were observed in a subset of clinically affected DT-treated animals, whereas peripheral organs were largely unremarkable and haematological changes were infrequent. Together, these data demonstrate that i.c.v. DT administration can induce mortality, behavioural dysfunction, and focal CNS pathology in both WT and CD11b-DTR mice, with transgene expression amplifying susceptibility. Our findings highlight the need for careful dose optimisation, appropriate DT-treated controls, and cautious interpretation of behavioural phenotypes when employing this model.