A CNS-targeted atovaquone nanotherapy mitigates complement activation and neuromuscular degeneration in ALS

Complement-mediated glial inflammation and synaptic injury are increasingly implicated in amyotrophic lateral sclerosis (ALS), but clinically translatable complement-modulating therapies remain limited. Atovaquone (ATQ) is an approved, poorly water-soluble small molecule with reported immunomodulatory activity. We investigated whether ATQ could modulate C1qb or C5aR1-associated pathology and whether lactoferrin (Lf)-derived nanomicelles could improve delivery to the spinal cord. ALS spinal-cord single-cell transcriptomic and pathway-enrichment analyses were combined with screening of an FDA-approved drug library, molecular docking and surface plasmon resonance (SPR). Lf-ATQ nanomicelles were characterised by dynamic light scattering, transmission electron microscopy and fluorescence distribution studies. Complement/inflammasome responses were examined in microglial and motor-neuron-microglia models. Therapeutic effects were evaluated in mice expressing the human superoxide dismutase 1 G93A variant (SOD1^G93A) by motor testing, compound muscle action potential recording, histology, immunostaining, western blotting and proteomics. Complement genes, particularly C1qb and C5aR1, were enriched in microglia, whereas glutamatergic-synapse and axon-guidance programmes were negatively enriched. In cellular models and SOD1^G93A spinal cord, Lf-ATQ reduced C1qb, NLRP3 and multiple inflammatory mediators, attenuated glial reactivity and altered autophagy-related proteins. Treatment preserved rotarod performance, grip strength, gait parameters, compound muscle action potential amplitude and gastrocnemius morphology. Proteomic analysis indicated remodelling of complement/coagulation, vesicular-transport and synaptic modules. These findings support Lf-ATQ as a central nervous system (CNS) -oriented, complement-regulating nanotherapy candidate for ALS. Direct functional target-inhibition assays, quantitative ATQ pharmacokinetics, long-term safety assessment and validation in non-SOD1 models remain necessary before translational conclusions can be drawn.