Unveiling the brain’s symphony: exploring the necessity and sufficiency of neural networks in behavior control

Since the pioneering work by Broca and Wernicke in the 19^th century, who examined individuals with brain lesions to associate them with specific behaviors, it was evident that behaviors are complex and cannot be fully attributable to specific brain areas alone. Instead, they involve connectivity among brain areas, whether close or distant. At that time, this approach was considered the optimal way to dissect brain circuitry and function. These pioneering efforts opened the field to explore the necessity or sufficiency of brain areas in controlling behavior and hence dissecting brain function. However, the connectivity of the brain and the mechanisms through which various brain regions regulate specific behaviors, either individually or collaboratively, remain largely elusive. Utilizing animal models, researchers have endeavored to unravel the necessity or sufficiency of specific brain areas in influencing behavior; however, no clear associations have been firmly established.

Necessity or sufficiency is related terms that can answer different questions regarding brain function. While necessity asks whether a brain area is crucial for a behavior by observing what happens when that region is disrupted or removed, sufficiency asks whether a brain area alone can drive a behavior by artificially activating or enhancing its activity. With the emergence of new technologies, five main approaches (Figure 1) are now commonly used to study the necessity or sufficiency of a brain area in controlling behavior: (1) Noninvasive approaches facilitating the measurement of brain activity are commonly employed to investigate the relationship between function and active brain areas. For instance, researchers can map active brain areas during specific tasks by utilizing functional magnetic resonance imaging to measure changes in blood flow as an estimate of brain activity; magnetoencephalography for assessing endogenous magnetic fields produced by the oscillations in neuronal activity; magnetic resonance spectroscopy to analyze biochemical changes related to brain function; and positron emission tomography using radioactive compounds to generate 3D brain reconstructions and analyze metabolic activity, providing valuable insights into the neural basis of various behaviors and cognitive processes. The mandatory use of specialized equipment in a laboratory setting, however, limits their application. Nonetheless, the utilization of electroencephalogram to measure brain activity or functional ultrasound to monitor changes in cerebral blood volume has opened a new avenue for analyzing brain activity with significantly fewer equipment requirements, allowing for field-based studies. Furthermore, despite being more invasive than the aforementioned examples, the application of genetically encoded calcium indicators together with miniaturized microscopes, or in vivo electrophysiology permits the analysis of brain activity in freely moving animals. (2) Lesioning a particular brain region, involves using a physical or chemical method to directly lesion a brain area, enabling the determination of the necessity of a specific region in controlling behavior. Targeting a specific area is troublesome, specially using chemicals, since it is difficult to circumscribe the lesion, and spillover can damage other contiguous regions without apparent damage. Moreover, lesioning a deep brain area leaves a path of lesions that also needs to be considered when analyzing the results. (3) Using whole-animal knockouts or local knockouts through specific viral delivery using the Cre/Lox or the CRISPR/Cas9 systems, helps elucidate the role of specific genes in regulating behavior, either in the entire animal brain (systemic delivery) or by targeting a specific brain region to determine the necessity of the brain area in the study. The use of specific promoters, whether for expressing Cre or employing the CRISPR/Cas9 system to target particular cell types or brain regions, is a common and powerful approach. Moreover, the incorporation of inducible promoters offers precise control over the timing of mutations and can address challenges related to embryonic developmental lethality. Additionally, a myriad of spontaneous animal models carrying mutations akin to those causing pathology in humans provides a valuable resource for studying brain function. (4) The use of optogenetics, serving as a targeted stimulation or repression approach, enables the activation or repression of specific brain areas using optical fibers, surface activation, or the innovative implant-free deep brain optogenetics (Tye and Deisseroth, 2012). Precise targeting of the brain area of interest is crucial for both viral delivery and the placement of optical fibers for stimulation if needed. However, the use of transgenic animals expressing engineered opsins broadly or in specific cell populations, achieved through Cre/Lox animal crosses, allows for the expression of functional transgenes in the experimental animals. In addition to optogenetics for studying brain function and wiring, the use of designer receptors exclusively activated by designer drugs (DREADDs) as chemogenetic tools has emerged as a promising approach (Goutaudier et al., 2019). It enables the activation or repression of a targeted population through downstream signaling of G protein-coupled receptors with specifically designed drugs. Despite their usefulness, recent studies have shown that the drugs used for designer receptors exclusively activated by designer drugs activation may have off-target effects, leading to unintended consequences in cells. (5) A transcranial magnetic stimulation approach that employs magnets to activate the brain activity of a particular neuronal population. However, it comes with challenges, such as the difficulty in targeting specific regions and the lack of focal specificity. Moreover, the use of special equipment and a laboratory setting is essential for experimentation. All these approaches have been widely used to study brain function, wiring, and the necessity of brain areas for controlling particular behaviors, as well as the role of gene knockouts or mutations in many neurological pathologies, including autism spectrum disorders (ASD) where there is still much to be learned (Genon et al., 2018).

Figure 1.

Schematic representation of the approaches used to study the necessity or sufficiency of brain areas.

Traditionally, five methods have been employed to investigate brain function in wild-type animals, involving the measurement, activation, or inactivation of specific brain areas. These approaches include MRI, lesions, local or whole KO, optogenetics, or transcranial stimulations. We propose utilizing the Flailer animal model, characterized by a genetic mutation leading to a defective entire brain, to re-activate specific brain areas and evaluate their sufficiency in behavior. The Flailer mutation arises from a spontaneous recombination and duplication event, forming a fusion of exons 1–2 of gnb5 with exons 26–40 of myo5a, connected by a mixed intron. As the mutation involves duplication, the endogenous copies of gnb5 and myo5a remain intact. Removing the Flailer duplication presented a challenge, as using CRISPR/Cas9 to target Flailer would also affect the endogenous copies (indicated by red arrows). To address this, we employed DN-CRISPRs to specifically remove the Flailer mutation without altering the endogenous copies of the genes. With this approach, only a nick is produced in gnb5 and myo5a, and the Flailer gene is edited, producing its knockout. Consequently, we were able to recover the entire brain or target specific brain areas. We propose that this system can be employed to dissect brain regions and their association with behaviors. Additionally, we suggest investigating the developmental component of behaviors when reactivating brain areas at distinct developmental stages. Created with Adobe Illustrator. CRISPR: Clustered regularly interspaced short palindromic repeats; DREADDs: designer receptors exclusively activated by designer drugs; EEG: electroencephalogram; fMRI: functional magnetic resonance imaging; fUS: functional ultrasound; KO: knockout; MEG: magnetoencephalography; MRS: magnetic resonance spectroscopy; PET: positron emission tomography; WT: wild type.

To specifically investigate the sufficiency of a particular brain area, we propose a novel approach that could be integrated into this matrix. Rather than intervening in a specific brain region in a completely healthy animal to assess its necessity, we suggest employing an animal model with an entire defective mutant brain and subsequently recovering only a precise region of interest (Bustos et al., 2023). This approach enables the assessment of the sufficiency of this brain area, as the entire brain is malfunctioning, and only the targeted region is restored. For this purpose, we have recently validated an ASD/anxiety animal model named Flailer, which expresses an extra protein acting as a dominant negative for MyosinVa, which serves as a plus-end motor traveling along actin filaments in cells. Flailer causes severe disruption of synaptic component transport and synaptic function, leading among others, to a lack of long-term depression (Pandian et al., 2020). Moreover, Flailer animals display several abnormal behaviors resembling anxiety and ASD, including seizures, memory deficits, asocial behaviors, anxiety, and repetitive behaviors (Pandian et al., 2020). As part of a myosin family, MyosinVa is responsible for transporting various components of the synaptic complex, including scaffolding proteins, receptors, and mRNAs to dendritic spines (Hammer and Sellers, 2012). Therefore, its function is crucial for the correct operation of synapses and synaptic plasticity. The Flailer animal was identified by Jones et al. (2000), who were able to characterize its genetic mutation describing it as a spontaneous non-homologous recombination event produced by exon shuffling between gnb5 and moy5a, giving rise to the flailer gene. The flailer genomic mutation is peculiar since endogenous copies of gnb5 and myo5a are normal and the extra flailer gene is also expressed. All three of them are localized to chromosome 9. In particular, the flailer gene is composed of exons 1–2 of gnb5 fused in frame by a mixed intronic sequence to exons 26–40 of myo5a (Figure 1). The expression of the flailer gene is regulated by the highly and broadly expressed promoter of gnb5 in the central nervous system. This genetic rearrangement results in the expression of the new Flailer protein. Consequently, the resulting protein binds to cargo but lacks the actin-binding domain of MyosinVa, necessary for binding and walking to the plus end of actin filaments. When the Flailer protein is present in a 1:1 ratio with wild-type MyosinVa, it functions as a dominant negative, leading to reduced transport of synaptic components to the synapse. Importantly, removing at least one copy of the flailer gene can recover synaptic function and behavior in the mouse model (Bustos et al., 2023).

Using the Flailer animal model, we can investigate the brain regions responsible for behaviors associated with anxiety and ASD, determining their sufficiency for the studied behavior. Since Flailer’s brain is entirely perturbed, recovering particular brain areas with the help of gene editing technologies lets us evaluate the sufficiency of a particular brain area controlling behavior. To achieve this, we designed a gene-editing approach using CRISPR/Cas9 to modify the flailer gene (Cong et al., 2013). Importantly, since the two endogenous copies of gnb5 and myo5a are present, we devised a strategy to target the flailer duplication coding sequence without altering the endogenous copies of gnb5 and myo5a. We employed double-nicking CRISPRs (DN-CRISPRs), where guide RNAs are designed on opposite strands (Ran et al., 2013). It has been described that when two of them are in close proximity (< 300 bp offset), CRISPRs are capable of removing the targeted region. Previous studies have shown that when sgRNAs are separated by more than 100–300 bp, excision efficiency diminishes significantly. However, in our study to target Flailer, we specifically aimed to target the coding region of gnb5 with one guide RNA and part of the intron of myo5a with another guide RNA. We had to push the limits of DN-CRISPRs to address a ~700 bp offset between them. This demonstrates the feasibility of using DN-CRISPR to target and remove a larger fragment (> 700 bp) of DNA as part of a partial genomic duplication, indicating that using DN-CRISPR with sgRNAs featuring large offsets is indeed feasible and can be equally or even more efficient than using sgRNAs with small offsets. This insight is important to consider when evaluating the off-target effects of DN-CRISPRs and opens possibilities to target other gene regions specifically. After identifying the optimal combination of guide sgRNAs, we successfully edited the flailer gene, resulting in the knockout of its expression (Figure 1). With this tool in hand, we analyzed the recovery of synaptic protein transport and synaptic function, observing that the gene editing of Flailer is capable of restoring the phenotypes observed in Flailer neurons, including the induction of long-term depression (Bustos et al., 2023).

The combination of this tool and the Flailer animal presents a unique opportunity to dissect the brain, as its whole brain is affected, and we can stereotaxically target a virus coding for DN-CRISPRs to remove the Flailer gene and recover the brain area. To test this, we first aimed to recover a specific region, choosing the ventral hippocampus as it has been associated with memory formation and anxiety behaviors. Importantly, using single-nuclei PCR, we were able to show that approximately 70% of the targeted cells were edited in at least one copy; thus, in those cells, normal function was recovered. The experiments targeting the ventral hippocampus showed that memory formation recovered while the restoration of anxiety behavior remained only partial. This implies that anxiety behaviors are influenced to some extent by the ventral hippocampus, yet editing of this region alone is not sufficient to fully recover these behaviors through gene editing of the Flailer mutation. This implies that the ventral hippocampus is sufficient for memory formation but not necessary for the full recovery of anxiety behaviors. Therefore, the study provides insights into the sufficiency and necessity of specific brain regions in governing behaviors, demonstrating that a region may be sufficient for one behavior but not necessary for another. In addition, we utilized a novel adeno-associated virus capsid, PHP.eB, capable of infecting the entire brain when injected into the cerebral ventricles at postnatal day 0 (Brauer et al., 2023). With this approach, our aim was to demonstrate the effectiveness of our method in recovering the phenotype of the Flailer mouse. Through this, we achieved approximately 70% editing of at least one copy of the gene, as confirmed by single-cell nuclei PCR, and a significant reduction in its expression. One of the main phenotypes observed in Flailer mice is seizures, which were completely recovered in treated animals. Moreover, all behaviors tested in the animals were restored to levels comparable to those observed in wild-type animals. This demonstrates that the editing of the Flailer gene in the entire brain is sufficient to restore synaptic function and behavior in the Flailer mouse model.

The use of the Flailer animal model as a tool for studying brain functions introduces novel possibilities. As previously mentioned, employing local delivery of viruses enables the targeting of specific brain areas. Additionally, the incorporation of a specific promoter with the capability to target distinct cell types, such as excitatory or inhibitory neurons, astrocytes, microglia, provides a valuable avenue to delve into the roles of specific cell populations within the brain. This multifaceted approach not only enhances the precision of experimentation but also facilitates a detailed understanding of the intricate interactions within specialized cellular niches.

Through this comprehensive strategy, we can dissect the contributions of individual cell types to behavior, shedding light on the complex interplay between different neural components in the orchestration of cognitive and emotional processes. In addition, the versatility of gene editing that can be performed at different developmental stages opens avenues to explore the dependency of functional recovery in Flailer on developmental processes. This approach allows for a refined investigation into the therapeutic windows associated with each behavior and brain area, providing insights into the optimal timing for gene editing interventions and the reversal of phenotypic outcomes. By systematically manipulating the timing of gene editing, we could delineate critical periods during which interventions are most effective in restoring synaptic function and behavior. This dynamic perspective on developmental stages adds a temporal dimension to our understanding. As discussed earlier for other methods, local delivery of adeno-associated viruses into the brain, as in the case of Flailer, presents a challenge to stay confined to the target area without unintended spread to neighboring tissues. Using specific promoters and refining injection techniques are areas needing improvement for better precision and control.

While our study successfully demonstrates the recovery of synaptic function and behavior in the Flailer animal model by targeting specific brain regions, it is essential to recognize potential limitations. Even after the suppression of Flailer in a particular brain region, the presence of Flailer in the input and output areas of that region raises important considerations. It is plausible that disturbances in the input received by the rescued region may occur, potentially leading to compensatory effects in other brain regions. Moreover, the translation of the output from the rescued region into behavior may face disruptions. These caveats warrant further investigation to have a more comprehensive understanding of the implications of targeted interventions on the overall brain function.

Our work advances our understanding of brain wiring and function, offering insights into the complexities of behavior. It also lays the foundation for innovative therapies, particularly in addressing neurological conditions like ASD. Moving forward, utilizing our approach or similar methods will continue to unveil the sufficiency of specific brain regions in governing behavior within the intricate landscape of brain circuitry, holding promise for transformative interventions in neurological disorders.

This work was supported by ANID Fondecyt Iniciacion 11180540 (to FJB), ANID PAI 77180077 (to FJB), UNAB DI-02-22/REG (to FJB), Exploración-ANID 13220203 (to FJB), ANID-MILENIO (NCN2023_23, to FJB).

Additional file: Open peer review report 1 ^{(90.1KB, pdf)} .