Network Homeostasis in Schizophrenia: Why Finding Reduction of a CSF Molecule May Advance Treatment

Membrane-bound ion channels are essential for neuronal functioning, but how might they play a role as regulators of network dynamics like excitation and inhibition?¹

Investigators identified a soluble form of a voltage-gated calcium channel subunit, α2δ-1, in cerebrospinal fluid (CSF), which is involved in network homeostasis.² For individuals with schizophrenia, the presence of soluble α2δ-1 in CSF is notably reduced. With identification of soluble α2δ-1 as playing a role in the excitatory/inhibitory network, investigators demonstrated that adding a synthetic analog of the molecule (synthetic ectodomain of Alpha2Delta-1 (SEAD1)) into a mouse model of schizophrenia improved synaptic and behavioral deficits.

Lead investigator Peter Penzes, PhD, shared more clinical context around this discovery and what it could mean for modeling and treating schizophrenia.

Psychiatric Times: How does framing schizophrenia as a problem of network homeostasis change the way clinicians should think about the disorder?

Peter Penzes, PhD: Current models of schizophrenia have been productive—they have explained some features of the illness and led to treatments that help with positive symptoms like hallucinations and delusions. But cognitive and negative symptoms remain largely untreatable, which tells us we need a new framework; network homeostasis offers that.

Network homeostasis refers to the brain's ability to maintain a functional balance between excitation and inhibition within neural circuits. A simple analogy: imagine a digital circuit where every switch is set to "on"—there is no way to encode a meaningful signal. The brain faces the same problem. When too many excitatory neurons are firing, a feedback signal is sent to dial activity back down and restore balance. That feedback process is what we mean by network homeostasis.

Our model incorporates elements that have already been well replicated in schizophrenia research—the role of parvalbumin interneurons, dysfunction at excitatory synapses—but adds an important new concept: a soluble protein that acts as a molecular messenger regulating this homeostatic balance. That is novel, and it matters because it can potentially be used both as a blood-based biomarker and as a therapeutic target.

PT: Can you explain what soluble α2δ-1 does in a healthy brain? Why does a reduction in soluble α2δ-1 in patients with schizophrenia matter?

Penzes: Soluble α2δ-1 is a fragment that is cleaved from α2δ-1, a calcium channel auxiliary subunit expressed on certain neurons. Once released, this soluble fragment diffuses through the extracellular space and acts as a signaling molecule that helps regulate the balance between excitatory and inhibitory drive in neural networks. In a healthy brain, when excitatory activity becomes too high, soluble α2δ-1 is produced and signals back to excitatory neurons to reduce their output, restoring normal network balance.

In patients with schizophrenia, this mechanism is impaired—levels of soluble α2δ-1 are reduced—which means the brain loses a critical feedback brake on excitation. That disruption in excitatory-inhibitory homeostasis is what we believe contributes to the cognitive and negative symptoms that current medications fail to address.

PT: α2δ-1 is the target of gabapentinoids like gabapentin and pregabalin. Do your findings change how clinicians should think about these drugs in patients with psychosis or schizophrenia?

Penzes: This is an important distinction. Gabapentinoids bind to α2δ-1 when it is embedded in the calcium channel complex—they work by interfering with the interaction between α2δ-1 and the channel itself. The soluble form of α2δ-1 that we study is not associated with calcium channels and does not affect their function. The binding pocket that gabapentinoids target—the interface between α2δ-1 and the calcium channel—simply does not exist on the soluble fragment. So our findings do not directly affect how clinicians should think about prescribing gabapentinoids.

As for a therapeutic based on soluble α2δ-1, that is not yet ready for clinical use. When it is, it will have gone through rigorous FDA approval, including comprehensive safety testing, before it reaches patients.

PT: Do you have hope this mechanism will be relevant in patients, and what are the biggest translational hurdles?

Penzes: I am cautiously optimistic, but it is impossible to predict with certainty—as with any new drug based on a novel mechanism, there are many steps between a laboratory discovery and a treatment that helps patients.

That said, we have designed a deliberate, stepwise derisking strategy. First, this is a peptide that the body naturally produces, which reduces the risk of adverse effects. Second, peptide therapeutics have been extraordinarily successful in recent years—glucaon-like peptide 1 receptor agonists are the clearest example—which demonstrates that this is a powerful and practical drug modality. A weekly injection is also far easier to track for adherence than multiple daily pills, which is particularly relevant in schizophrenia.

The biggest translational challenge is clinical trial design. Schizophrenia trials are costly and prolonged because of the heterogeneity of the patient population. Our approach is to start with a rare disease—16p11.2 duplication syndrome—which increases the risk of schizophrenia roughly 16-fold. Because these patients share the same genetic lesion, the signal-to-noise ratio in treatment response should be much better. This population would qualify for FDA orphan drug designation and accelerated trial pathways, allowing us to run small, focused trials with 10 to 20 patients. In a second phase, we plan to use reduced soluble α2δ-1 levels in blood as a biomarker to stratify patients with idiopathic schizophrenia and identify those most likely to respond. This biomarker-guided expansion strategy should significantly improve the probability of success when we move beyond the 16p11.2 population into broader schizophrenia trials.

PT: What is the most important thing a psychiatrist reading this study should take away today?

Penzes: The key takeaway is that cognitive and negative symptoms—the ones that most impair patients' daily functioning and that current antipsychotics do not adequately treat—may arise from a disruption in network homeostasis that we can now measure and potentially correct. For the first time, we have a molecular mechanism that links excitatory-inhibitory imbalance to a specific, quantifiable protein in the blood.

That matters for 2 reasons. First, it reframes how we think about treatment-resistant features of schizophrenia: not as an absence of treatment options, but as a problem we have not had a biological handle on until now. Second, it opens the door to a precision-medicine approach in which a blood-based biomarker could eventually help identify which patients are most likely to benefit from a targeted therapy—much the way oncology uses molecular markers to guide treatment selection. We are not there yet clinically, but the biological framework for stratified treatment in schizophrenia is taking shape, and psychiatrists should be aware of it.

Dr Penzes is the Ruth and Evelyn Dunbar professor of neuroscience and psychiatry at Northwestern University Feinberg School of Medicine and Director of the Northwestern Center for Autism and Neurodevelopment. His laboratory studies synaptic mechanisms underlying neurodevelopmental and psychiatric disorders, with a focus on translating molecular discoveries into novel therapeutics.

References

1. Alberts B, Johnson A, Lewis J, et al. Ion channels and the electrical properties of membranes. Molecular Biology of the Cell. 4th edition. Garland Science; 2002.

2. Dos Santos M, Forrest MP, Bomba-Warczak E, et al. Soluble α2δ-1, altered in disease CSF, modules network homeostasis and rescues deficits in a neuropsychiatric mouse model. Neuron. 2026.