Intercepting Alzheimer’s early: Preventing neuronal hyperactivity by ­targeting Aβ monomers

Alzheimer’s disease (AD) is one of the most pressing challenges in neuroscience. While its clinical symptoms – memory loss and cognitive decline – manifest late, the underlying pathology begins years or even decades earlier. Among the earliest detectable changes is neuronal hyperactivity, particularly in the hippocampus, a region essential for learning and memory. This hyperactivity precedes amyloid plaque deposition and brain atrophy, hence offering a promising target for preventive strategies.

According to the amyloid hypothesis [1], which has guided much of AD research, β-amyloid (Aβ) peptides and their aggregates are central to disease progression. While plaques are a hallmark of advanced AD, soluble aggregates – dimers and oligomers– appear earlier and are considered the most neurotoxic species (Fig. 1). They disrupt synaptic signaling, trigger inflammation, and promote tau pathology. Recently, the Aβ-binding antibodies Lecanemab and Donanemab have been found to clear amyloid plaques and modestly slow cognitive decline in AD patients [2], leading to their approval as treatments for AD. Still, the relatively high rate of neuropathological side-effects motivates the continued search for safe and effective AD treatments.

 

Aβ monomer scavenging restores neuronal activity levels

In our work [2], we explored a different approach: preventing toxicity at its source by scavenging Aβ monomers before they aggregate. We use an engineered protein called an anticalin [3], designed to bind Aβ monomers with high affinity. This anticalin is based on the human lipocalin 2 scaffold (an abundant plasma protein) and has been selected and optimized for stability and specificity. Unlike antibodies, anticalins are small, exhibit low immunogenicity, and can be produced efficiently in bacterial host organisms, making them attractive candidates for therapeutic development [4].

We tested the Aβ-anticalin in two transgenic mouse models of AD endogenously overproducing Aβ: APP23xPS45 mice, which develop amyloid pathology early, and APP23 mice, which do so later. We focused on young animals (two to four months old), before plaque formation (Fig. 2), to mimic early disease stages. To bypass delivery challenges, we applied the anticalin directly into the hippocampal CA1 region using a fine pipette. Neuronal activity was monitored at single-cell resolution using two-photon calcium imaging, a technique that enables the visualization of activity patterns within the brain. In parallel, we assessed synaptic function using glutamate imaging with fluorescent molecular sensors.

Under baseline conditions, as expected [5], AD mice exhibited pronounced hyperactivity: A subset of neurons fired excessively, with more than 20 calcium transients per minute. After local application of the Aβ-anticalin, however, this hyperactivity was rapidly suppressed. Activity levels returned to those seen in wild-type mice, and the effect was reversible after washout, confirming that the anticalin acted acutely and specifically by scavenging monomeric Aβ in the brain (Fig. 3), even at low concentrations. An important finding was that the anticalin had no effect in wild-type mice, demonstrating that it does not interfere with normal neuronal function.

At the synaptic level, AD mice showed excessive extracellular glutamate during stimulation – a sign of impaired synaptic clearance and excitotoxicity. Application of the anticalin normalized these glutamate transients. Control experiments with the wild type lipocalin (the parent scaffold, which does not bind Aβ) had no effect, confirming that the observed rescue was due to specific Aβ binding.

Mechanistic insights

Why does scavenging of Aβ monomers work? Our experiments revealed that freshly secreted Aβ monomers are not toxic on their own. When we applied synthetic Aβ monomers to the brains of wild-type mice, they did not induce hyperactivity, even at high concentrations. However, when monomers were allowed to “age” for 90-120 minutes – thus forming oligomers – they caused robust hyperactivity. This supports the idea that toxicity arises during early Aβ aggregation. 

We confirmed this biochemically. Fresh Aβ monomers rapidly began forming aggregates in vitro, but when the anticalin was added at 
the start, fibril formation was completely blocked. Adding the anticalin after aggregates had formed did not reverse the process, nor did it neutralize their toxicity. Thus, the anticalin acts by preventing aggregation, not by dismantling existing Aβ oligomers. This explains why its effect was strong in young but marginal in older, plaque-bearing animals. We also tested a γ-secretase inhibitor, which blocks the endogenous Aβ production entirely. This treatment produced similar benefits as the anticalin, reinforcing the idea that ongoing release of Aβ monomers drives early dysfunction. Together, these findings highlight that continuous aggregation of nascent Aβ monomers into oligomers constitutes a key pathogenic mechanism.

Implications

Our results demonstrate that an early intervention targeting Aβ monomers can restore normal neuronal and synaptic function in Alzheimer’s models. By intercepting monomers before they aggregate, we prevented the formation of toxic Aβ species and reversed neuronal hyperactivity—a hallmark of early AD. The beneficial biochemical properties of the anticalin make it a promising drug candidate for preventive treatment. If these findings translate to humans, they could shift the therapeutic paradigm from clearing plaques to blocking the earliest steps of toxicity.
 

Future directions

Based on these promising findings, we are now focusing on the development of a modified Aβ-anticalin with brain-targeting properties, which can be applied peripherally, for long-term in vivo studies of efficacy and toxicity. To this end, we have modified the anticalin in two ways: (1) To extend its plasma half-life, we have fused the protein to a conformationally disordered voluminous polypeptide chain; and (2) we have fused it with blood-brain-barrier shuttle. Experiments are ongoing, and we have obtained the first promising preliminary results.

Other activities of the Focus Group

Additionally, the focus group is investigating neuronal and glial dysfunctions in AD from various perspectives. I contributed to a recent study demonstrating that neuronal silencing, a hallmark of later AD stages, already starts at an earlier time point than expected and is caused by a synaptic uncoupling of individual neurons from the rest of the circuit [6]. Preliminary data from this and other work enabled me to apply for and win an ERC starting grant, which will allow me to continue this ambitious research program at TUM Hospital for at least the next five years.

Additionally, we have utilized a portion of the TUM-IAS Fellowship to investigate brain waste clearance, another significant contributor to neurodegenerative diseases. This allows me to bridge the gap between basic neuroscience and my clinical activities as a neuroradiologist. 

Selected publications

• B. Zott et al., “β-amyloid monomer scavenging by an anticalin protein prevents neuronal hyperactivity in mouse models of Alzheimer’s Disease,” Nature Communications, vol. 15, no. 1, p. 5819, 2024/07/10 2024, doi: 10.1038/s41467-024-50153-y.
   
• Y. Zhang et al., “Amyloid β–dependent neuronal silencing through synaptic decoupling,” Proceedings of the National Academy of Sciences, vol. 122, no. 35, p. e2515113122, 2025, doi:10.1073/pnas.2515113122.
   
• B. Zott and A. Konnerth, “Impairments of glutamatergic synaptic transmission in Alzheimer’s disease,” Seminars in Cell & Developmental
   
• Biology, vol. 139, pp. 24-34, 2023, doi: https://doi.org/10.1016/j.semcdb.2022.03.013.
   
• J. Zimmermann et al., “Total cerebral blood volume changes drive macroscopic cerebrospinal fluid flux in humans,” PLOS Biology, vol. 23, no. 4, p. e3003138, 2025, doi: 10.1371/journal.p bio.3003138.
   
• J. Zimmermann et al., “Impaired macroscopic CSF flow by sevoflurane in humans - both during and after anesthesia,” Anesthesiology, Jan 8 2025, doi: 10.1097/aln.0000000000005360.

Full list of publications