Understanding the heterogeneity of freezing responses in mice by addressing homeostatic mechanisms in brain defense circuit

Irina KovlyaginaPhD student in the lab pg Beat Lutz (A06)

Preclinical studies rely on functional patterns of defensive behaviours in rodents as basis for conceptualizing human fear/anxiety. However, very often animal models do not account for the role of inter-individual differences in stimulus perception and processing in shaping the behavioural response.

By applying cued fear conditioning paradigm with prolonged retrieval session, we observed that mice respond to fearful stimuli in an individual manner. Using an unsupervised clustering approach, we identified two distinct endophenotypes (phasic and sustained responders) in a population of male and female wild-type mice (C57BL/6J). The individual freezing behaviour during the retrievals correlates with the approach/avoidance behaviour in anxiety tests conducted before and after conditioning further proving intrinsic differences between animals. Therefore, we hypothesized that a wide range of homeostatic mechanisms acting at the molecular, cellular, and circuit levels underly distinct anxiety phenotypes.

Transcriptomic analysis of key brain regions in defense circuit of phasic and sustained female responders confirmed substantial differences in transcriptomes of anterior cingulate cortex (ACC), basolateral amygdala (BLA), central amygdala (CeA), and bed nucleus of stria terminalis (BNST). Functional annotation analysis revealed that DEGs from both regions are enriched for genes involved in an inflammatory response. Additionally, extracellular matrix organization pathways were significantly deregulated in CeA and ACC. Interestingly, among top DEGs in CeA, we found the transcription factors NeuroD2 and NeuroD6 that were previously reported to be involved in homeostatic scaling of post-synaptic AMPA receptors.

In an ongoing collaboration of the Lutz and von Engelhardt labs, we are investigating alterations in intrinsic neuronal excitability and synaptic transmission between phasic and sustained responders to further elucidate potential homeostatic mechanisms underlying inter-individual variability of fear expression in inbred mice.

Modelling spatial distribution of cytoplasmic and membrane-bound AMPA receptors

Surbhit Wagle, Tatjana Tchumatchenko group (C03)Maximilian Eggl, Tatjana Tchumatchenko group (C03)Maximilian Ken Kracht, Amparo Acker-Palmer group (B04)Markus Matthias Middeke, Amparo Acker-Palmer group (B04)

Neuronal networks need to dynamically adjust their synaptic strength to external stimuli for normal brain function. This adaptive ability, termed synaptic plasticity, is implemented by the interplay of several molecular mechanisms. In this process, a critical molecular event is the trafficking of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPARs) in glutamatergic neurons (Bissen et al., 2019; Bissen et al., 2021). AMPARs are ion channels that are ligand-gated and are the main mediators of fast excitatory neurotransmission in the brain (Shepherd and Huganir, 2007). AMPARs are tetrameric structures consisting of four closely related genes that express the four subunits of glutamatergic receptors (GluA)1-4 (Shepherd and Huganir, 2007). Receptors composed of GluA2/GluA3 serve to recycle continuously synaptic receptors independently of neuronal activity to replace pre-existing receptors (Shi et al., 2001), whereas heteromeric receptors containing GluA1 and GluA2 are delivered in an activity-dependent manner (Wenthold et al., 1996; Shi et al., 2001). Previous hypotheses suggested that AMPARs are exocytosed upon presenting a plasticity cue. This activity-driven exocytosis of receptors can be facilitated by a local, dendritic pool of readily available receptors to be incorporated into the membrane. Other studies found mRNA and ribosomes localized at dendrites, indicating that local synthesis is the crucial factor rather than exocytosis (Cajigas et al., 2012). We will take a multidisciplinary approach to understand the homeostatic set-point of synaptic AMPARs better. First, we will build a theoretical model of the processes which balance exocytosis and endocytosis of AMPARs to get a spatiotemporal profile of their copy numbers. Second, from our model, we will derive experimentally testable predictions, which we will compare with the experimental data provided by the Acker-Palmer group (B04). In addition, we will also develop a tool to perform several statistical analyses of the data provided by the Acker-Palmer group (B04).

The presynaptic roles of the Cav1 and PMCA for graded synaptic transmission

Lea Deneke, Carsten Duch group (B12)Burak Gür, Marion Silies group (C06)

We have recently discovered the co-localization of different types of voltage gated calcium channels, namely Cav1 and Cav2 with distinctly different functions at glutamatergic presynaptic terminals of Drosophila motoneurons. While Cav2 is mandatory for evoked release of synaptic vesicles (SVs), Cav1 modulates short-term synaptic plasticity and augments synaptic vesicle recycling. Functional separation in the minimal space of the presynaptic terminal is achieved by a strategic localization of the membrane bound calcium ATPase PMCA (Krick et al., 2021). In B12 we are currently investigating the function of this arrangement of two voltage gated calcium channels and a membrane anchored calcium extrusion pump for homeostatic synaptic plasticity at the Drosophila NMJ and in hippocampal synapses.

Here, we aim to first probe the functions of PMCA, Cav1, and Cav2 in different types of central synapses that are required for stable contrast processing in the Drosophila visual system. Project C06 investigates fast homeostatic mechanisms that ensure contrast constancy, i.e. visual responses are able to respond robustly to the same contrast even if background illumination is quickly changing. The peripheral visual circuitry implementing contrast constant responses have been recently identified (Ketkar et al. 2020, Ketkar et al. 2022, Gür et al., unpublished). The respective networks contain synapses with graded SV release, and knockdown of Cav2 has little effect on these synapses. We propose that the slower kinetics, higher single channel conductances, and the loose coupling of Cav1 channels are well suited for graded synaptic transmission and presynaptic homeostatic plasticity. Combining the knowledge, techniques, and tools of B12 with that of C06 will allow to test novel roles of different presynaptic voltage gated calcium channels, which we have discovered at the NMJ model synapse, for their function in information encoding in central circuitry. In particular, we propose prominent roles of presynaptic Cav1 in graded synaptic transmission and of presynaptic Cav2 in fast presynaptic homeostatic compensation that is required for contrast constancy.