Rhushikesh A. Phadke (Apellis Pharmaceuticals, Inc.) and Alberto Cruz-Martín (The CU Anschutz Medical Campus)

This blog is a 'Behind the Paper' post related to our recent paper by Phadke et al. (2024) published in Molecular Psychiatry, https://tinyurl.com/5dn2x9dr

Phadke, R.A., Brack, A., Fournier, L.A. et al. The schizophrenia risk gene C4 induces pathological synaptic loss by impairing AMPAR trafficking. Mol Psychiatry (2024). https://doi.org/10.1038/s41380-024-02701-7

Readers can also view an older version (Phadke et al. 2023)  of this story as a preprint on bioRxiv: https://www.biorxiv.org/content/10.1101/2023.09.09.556388v2

The complement pathway is a part of the immune system that enhances the ability of antibodies and phagocytic cells to clear microbes and damaged cells from an organism, promote inflammation, and attack the pathogen's cell membrane. It plays a crucial role in innate immunity and the body's defense against infections. For more than a decade, research has shown that complement members are expressed in various brain cell types, such as neurons and glia, where they play a role in healthy brain function and are dysregulated in disease^1–3. 

It is thought that microglia, the brain's resident macrophages, play a role in developmental synapse pruning through the engulfment of synapses. A process that, at least in the sensory system, is dependent on the C3 and C receptor 3 (CR3) signaling pathway¹. Microglia and macrophages are the only cells in the brain that express the CR3, which allows these cells to recognize and phagocytose material tagged by the complement pathway¹. A long-standing dogma is that glia can release complement proteins into the extracellular space, where they modify the connectivity of neurons through the activation of microglia complement receptors and recruitment of this immune brain cell type^1,2.

 In 2019, graduate students Rhushikesh A. Phadke and Ashley L. Comer in our lab used a gene transfer approach, in utero electroporation, and the CR3 KO mouse line to present intriguing results demonstrating that elevated levels of the schizophrenia risk gene C4 cause synaptic loss independently of the main phagocytic CR3. At the time, we were preparing our first preprint (Comer and Jinadasa et al., 2019; bioRxiv), which showed that both mouse and human homologs of this neuroimmune gene cause prefrontal cortex circuit deficits and prefrontal cortex-associated behavioral changes in mice. Our story was motivated by results from Sekar et al.⁵ and the fact that many human diseases are associated with the upregulation of the complement pathway. However, current neuroimmune studies were using KO/loss of function mice to understand the role of these proteins in the brain and their contribution to disease. In this context, our upcoming paper presented the first complement overexpression model in a brain circuit relevant to neuropsychiatric disorders like schizophrenia and neurodegenerative diseases (Comer and Jinadasa et al., 2020; Plos Biology)⁴.

 By early 2020, we had analyzed a complete data set, demonstrating that overexpression of C4 leads to a decrease in connectivity independent of the phagocytic receptor CR3. The results greatly surprised and puzzled my group as, for more than a decade, most studies have concentrated on complement-driven activation of microglia as the primary mechanism for synaptic loss. As we discussed the interpretation of these results, the pandemic arrived, and nonessential laboratories were shutting down across the world. During a challenging time, we held several video conference meetings over the summer, where we formulated a new hypothesis for a non-canonical pathway that could explain complement-driven synaptic loss.

 Rhush and lab members made two key observations over that summer. First, Comer et al.⁶ showed that increased levels of C4 caused specific deficits in filopodia function. Major contributors to synaptogenesis are dendritic filopodia, which are thin, specialized postsynaptic structures that orchestrate synapse formation by dynamically sampling potential presynaptic partners, thus optimizing circuit wiring. Separately, Druart et al. 2021⁷ performed in vivo 2P imaging to monitor spine dynamics, demonstrating that overexpression of C4 led to a decrease in the fraction of gained dendritic spines in the cortex. Both observations were consistent with increased levels of C4 causing synaptic loss through alterations in synaptic formation. As during synaptic development, accelerated synaptic formation is followed by synaptic elimination or pruning, we asked whether the effects classically assigned to complement-driven synaptic loss through microglia-dependent synaptic pruning could be a side-effect of early deficits in synaptic formation. We have also previously demonstrated that besides the changes in connectivity, increased levels of this neuroimmune gene did not affect the gross morphology or the excitability of transfected neurons, suggesting that, at least early in the pathology, C4 overexpression does not compromise overall neuronal health. Therefore, we also started thinking about whether our in utero electroporation approach could be used as a general model to study deficits in synaptic formation in neurodegenerative diseases associated with synaptic loss (see Restrepo et al., 2022)⁸. Additionally, could this model describe the early stages of neurodegenerative diseases, such as Alzheimer's disease, where there is still synaptic loss without cell death?

 As the summer ended, we added two more pieces to the puzzle. First, my lab discussed a paper from Claudia Kemper's lab demonstrating that in the immune system, intracellular complement activation sustains T cell homeostasis and mediates effector differentiation (Liszewski et al., 2013)⁹. Although this paper was beyond our scope of study, it broadened our horizon as they stated that "Importantly, intracellular C3a was observed in all examined cell populations, suggesting that intracellular complement activation might be of broad physiological significance." We also discussed a paper by Huttlin et al. (Bioplex Interactome)¹⁰, a comprehensive resource that maps protein interactions in human cells that identified the endosomal protein sorting nexin 27 (SNX27) as a potential intracellular interacting partner of C4. In the lab, we subsequently validated these findings for mouse homologs of the proteins by performing a co-immunoprecipitation assay in cell lines transfected with C4 and FLAG-tagged SNX27. These results started shaping our hypothesis that complement dysregulation could be altering neuronal intracellular pathways linked to synaptic plasticity and long-term potentiation, which are required for learning and memory. 

 Over the next two years, we performed loss and gain of function experiments in acute brain slices (with Alison Brack and Luke Fournier), demonstrating that knockdown of SNX27 mimicked the synaptic loss observed with increased levels of C4 and that co-overexpression of SNX27 with C4 rescued the synaptic phenotype in layer 1 circuit of the prefrontal cortex (presented at the Keystone Symposia: Neuroimmune interactions, Phadke et al. 2022). Interestingly, axonal projections to superficial L1 are part of a circuit that controls consciousness, attention, and learning states (Schuman et al., 2021)¹¹.  In the summer of 2022, Alberto also presented a talk at Aarhus University, where he had the opportunity to meet with Gregers Rom Andersen, who studies the human immune system using crystal structures and biochemical assays. Despite his skepticism of our hypothesis regarding the intracellular function of complement proteins in neurons, Gregers asked insightful questions that helped us pinpoint potential C4 mutants lacking complement activity. We discussed the possibility of overexpressing these mutants in neurons to see if we would still observe synaptic deficits. If this were the case, it would support our hypothesis that the synaptic alterations occur through a non-canonical intracellular pathway independent of CR3. 

 A year later, we had a more complete story, obtaining results that challenged established views on the role of the complement pathway in synaptic plasticity. Our conclusion was now supported by decreased connectivity observed after overexpressing C4 mutants lacking C3 convertase activity, co-IP of C4 with SNX27, the colocalization of C4 and SNX27 in dendrites and spines, and the observation that Snx27 KD mimicked the effects induced by C4 overexpression. The results described a new intracellular neuronal pathway through which C4 can modify synaptic physiology.

 Lastly, graduate students Ezra Kruzich, Alison Brack, and Rhushikesh Phadke used STED super-resolution imaging in L1 prefrontal cortex circuits and biochemical assays in cell lines to demonstrate that increased levels of C4 led to alterations in GluR1 trafficking, as observed in the mislocalization of the postsynaptic receptor in the recycling and endo-lysosomal compartments and elevated GluR1 protein levels (what we call “C4-driven GluR limbo") deficits that are consistent with perturbations of the endo-lysosomal pathway and dysfunction of the sorting nexin retromer complex. These findings are also important because, to our knowledge, besides complement-mediated microglia engulfment of excitatory synapses and its molecular components, no other neuroimmune mechanisms linking complement overactivation to glutamatergic dysfunction, which is associated with schizophrenia pathology have been studied or identified. Our preprint describing an intracellular complement pathway was finally uploaded in 2023¹².

 One controversy that remains is how we reconcile our new hypothesis with previous findings supporting microglia as major players in complement-dependent synaptic loss in the brain¹. Microglia may play a role in synaptic phagocytosis with complement dysregulation. However, this could be achieved by eliminating already weakened connections or the molecular remnants of the connections (see Weinhard et al., 2018)¹³. In our model, and because of the function of SNX27, microglia-dependent synaptic phagocytosis is triggered by reduced synapse formation or an inability to stabilize the synapse. Alternatively, synaptic material colocalizing in microglia could result from various biological processes, such as neuronal apoptosis and subsequent phagocytosis of cell bodies and debris or exosome communication between neurons and microglia, rather than just phagocytosis of synaptic products. 

 The Cruz-Martín lab is very excited to continue this work on the effects of intracellular complement activity and how its dysregulation contributes to brain pathology. Our story has generated more questions than answers; in particular, the fundamental questions that we are currently pursuing is whether the intracellular complement regulates the physiology of other brain cell types like microglia and astrocytes in the healthy brain and diseases such as Alzheimer's and schizophrenia, and what are the identity of the molecular pathways and downstream targets of intracellular complement proteins. Lastly, we are investigating whether this mechanism is specific to the trafficking of GluR1s or if it also alters the trafficking of other AMPA and NMDAR subunits (see the NMDA hypothesis of schizophrenia).

 In summary, our study uncovers a novel mechanism through which complement dysregulation mediates synapse loss. Our results link excessive intracellular complement activity to an endo-lysosomal trafficking pathway altering synapse formation. This finding is crucial for understanding mechanisms of learning and memory and diseases associated with complement-driven synaptic pathology. Given that therapies targeting the complement pathway or microglia-mediated pruning may adversely affect both the nervous and immune systems, our discovery offers a new approach (for example, targeting the endo-lysosomal pathway) to mitigating the harmful effects of pathological complement activity on the physiology of neurons.

Main paper:

Phadke, R.A., Brack, A., Fournier, L.A. et al. The schizophrenia risk gene C4 induces pathological synaptic loss by impairing AMPAR trafficking. Mol Psychiatry (2024). https://doi.org/10.1038/s41380-024-02701-7

References

1. Hammond, T. R., Robinton, D. & Stevens, B. Microglia and the Brain: Complementary Partners in Development and Disease. Annual Review of Cell and Developmental Biology 34, 523–544 (2018).
2. Druart, M. & Le Magueresse, C. Emerging Roles of Complement in Psychiatric Disorders. Frontiers in Psychiatry 10, (2019).
3. Comer, A. L., Carrier, M., Tremblay, M.-È. & Cruz-Martín, A. The Inflamed Brain in Schizophrenia: The Convergence of Genetic and Environmental Risk Factors That Lead to Uncontrolled Neuroinflammation. Front Cell Neurosci 14, 274 (2020).
4. Comer, A. L. et al. Increased expression of schizophrenia-associated gene C4 leads to hypoconnectivity of prefrontal cortex and reduced social interaction. PLoS Biol 18, e3000604 (2020).
5. Sekar, A. et al. Schizophrenia risk from complex variation of complement component 4. Nature 530, 177–183 (2016).
6. Comer, A. L. et al. Increased expression of schizophrenia-associated gene C4 leads to hypoconnectivity of prefrontal cortex and reduced social interaction. PLoS Biol 18, e3000604 (2020).
7. Druart, M. et al. Elevated expression of complement C4 in the mouse prefrontal cortex causes schizophrenia-associated phenotypes. Mol Psychiatry 26, 3489–3501 (2021).
8. Restrepo, L. J. et al. γ-secretase promotes Drosophila postsynaptic development through the cleavage of a Wnt receptor. Dev Cell 57, 1643-1660.e7 (2022).
9. Liszewski, M. K. et al. Intracellular complement activation sustains T cell homeostasis and mediates effector differentiation. Immunity 39, 1143–1157 (2013).
10. Huttlin, E. L. et al. The BioPlex Network: A Systematic Exploration of the Human Interactome. Cell 162, 425–440 (2015).
11. Schuman, B., Dellal, S., Prönneke, A., Machold, R. & Rudy, B. Neocortical Layer 1: An Elegant Solution to Top-Down and Bottom-Up Integration. Annu Rev Neurosci 44, 221–252 (2021).
12. Phadke, R. A. et al. C4 induces pathological synaptic loss by impairing AMPAR trafficking. bioRxiv2023.09.09.556388 (2023) doi:10.1101/2023.09.09.556388.
13. Weinhard, L. et al. Microglia remodel synapses by presynaptic trogocytosis and spine head filopodia induction. Nat Commun 9, 1228 (2018).