Knockdown of circHomer1 in the orbitofrontal cortex results in differential encoding of salient stimuli

doi:10.21203/rs.3.rs-1862282/v2

Download PDF

Article

Knockdown of circHomer1 in the orbitofrontal cortex results in differential encoding of salient stimuli

https://doi.org/10.21203/rs.3.rs-1862282/v2

This work is licensed under a CC BY 4.0 License

Version 2

posted

You are reading this older preprint version

Read the latest preprint version →

Background

CircHomer1 is an activity-dependent circular RNA (circRNA) isoform produced from back-splicing of the Homer1 transcript. Homer1 isoforms are well-known regulators of homeostatic synaptic plasticity through post-synaptic density scaffold regulation. Homer1 polymorphisms have been associated with psychiatric diseases including schizophrenia (SCZ) and bipolar disorder (BD). Postmortem tissue from patients with SCZ and BD showed reduced circHomer1 levels within the orbitofrontal cortex (OFC), a region associated with behavioral flexibility. While dysregulation of circHomer1 expression has recently been identified across multiple psychiatric and neurodegenerative disorders and is associated with impaired behavioral flexibility in mice, it is unknown whether circHomer1 can induce electrophysiological signatures relevant to cognitive dysfunction in these disorders.

Methods

CircHomer1 was knocked down in bilateral orbitofrontal cortex of C57BL/6J male mice and in vivo microarray recordings were captured throughout a touchscreen reversal learning task to identify electrophysiological changes associated with reduction of circHomer1. Following task completion, qRT-PCR was used to quantify transcriptional changes following circHomer1 knockdown.

Results

Knockdown of circHomer1 within the OFC induced robust changes in multiunit firing rate and local field potential coordination and power to salient stimuli during reversal learning. Further, these electrophysiological changes were associated with transcriptional downregulation of glutamatergic signaling effectors and behavioral alterations leading to impaired cognitive flexibility.

Conclusions

CircHomer1 is a stable biomolecule, whose knockdown in rodent OFC produces lasting electrophysiological and transcriptional changes important for efficient reversal learning. This is, to our knowledge, the first demonstration of a psychiatric-associated circRNA contributing to electrophysiological, transcriptional, and behavioral alterations relevant to psychiatric phenotypes.

circular RNA

orbitofrontal cortex

reversal learning

schizophrenia

bipolar disorder

electrophysiology

Schizophrenia (SCZ) and Bipolar Disorder (BD) are heterogeneous disorders that combined affect approximately 2-3.5 percent of the population[1, 2]. While considered distinct disorders, there is a high degree of overlap in symptoms and genetic factors that predispose an individual to SCZ and BD[3-7]. Both disorders are highly polygenic with an enrichment of genes involved in synaptic plasticity implicating synaptic dysfunction as an underlying mechanism[3-10].

Homeostatic synaptic plasticity is an intricate mechanism whereby synapses balance their respective weights to evenly tune a circuit and prevent hyperexcitability within the complex system[11-14]. HOMER1 is a homeostatically regulated synaptic scaffold protein[15] that has been previously linked to psychiatric disease and synaptic signaling[9, 16-20]. Additionally, polymorphisms in this gene have been associated with psychiatric phenotypes at the genome wide significance level[6, 20, 21]. The Homer1 gene undergoes alternative splicing to encode for multiple postsynaptic density (PSD) proteins[22, 23] that interact through a scaffolding structure with surface NMDA (N-methyl-D-aspartate receptor) receptors [24], type 1 metabotropic glutamate receptors (mGluR1/5)[25-27] and intracellular proteins including SHANK [24] and IP₃Rs (inositol 1,4,5-triphosphate receptor) to control intracellular Ca²⁺ signaling[28] (see review[18]). Long isoforms, HOMER1B/C, are constitutively expressed and contain an EVH1 domain to interact with other PSD proteins and a coiled-coil domain that allows for the formation of tetrameric complexes important for protein scaffolding[18, 24]. The short, activity-dependent isoform, Homer1a, is transcribed during heightened network excitability[23, 29] and acts as a dominant negative to displace the long isoforms by interacting through its EVH1 domain[29]. However, HOMER1A lacks the coiled-coil domain and cannot dimerize, effectively breaking down PSD scaffolds and reducing postsynaptic signaling[29] through a homeostatic downscaling mechanism involving AMPAR (α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor) internalization[15, 30]. Recently, circHomer1, another activity dependent Homer1 splice isoform was described[31-33]. CircHomer1 is a circular isoform created through back-splicing in which exons 2 and 5 covalently join to form a closed-loop structure[34, 35] that has not been shown to be translated into protein, but is able to bind other RNA species and proteins[31, 32, 36, 37] suggesting a more complex regulatory role. CircHomer1 is an inducible isoform[33] with RNA expression similar to that of Homer1a[31]. Reduction of circHomer1 has been found in numerous psychiatric and neurological disorders including schizophrenia (SCZ) and bipolar disorder (BD)[31, 32], Alzheimer’s Disease [38-40] and frontotemporal lobe degeneration[40], Mesial temporal lobe epilepsy[41], and Parkinson’s Disease [42]. Additionally, loss of circHomer1 in a mouse model resulted in differential gene expression within the orbitofrontal cortex and impaired cognitive function on a touchscreen executive function task[32].

The orbitofrontal cortex (OFC) has been identified as an important frontocortical region for outcome-monitoring involved in goal-directed action selection during decision-making[43-45]. Through its reciprocal connectivity with the striatum, the OFC tracks when reward contingencies change and encodes the value of the reward[45-48] in its particular circumstance prompting either a value-guided or habitual response through its interactions with dorsal medial or dorsal lateral striatum, respectively[44, 48, 49]. Dysfunction of this region is commonly found in individuals with psychiatric phenotypes including BD and SCZ and is believed to underlie some of the detrimental behaviors associated with poor decision-making in these populations[50, 51].

Recent work has uncovered electrophysiological signatures that correlate with neural dysfunction in patients with SCZ and BD[52, 53]. Importantly, similar signatures have been found in animal modeling of traits seen in these disorders including psychosis and cognitive dysfunction[54, 55]. Cognitive-associated electrophysiological signatures are conserved across species[53, 54] and provide a more reliable approach for identifying the underlying cognitive dysfunction compared to behavior alone. Previous work[32] predicted a disruption in synaptic activity following the knockdown (KD) of circHomer1 within the OFC using RNA sequencing and pathway analysis; yet, to our knowledge, no studies have determined whether a disease-relevant circular RNA influences electrophysiological function relevant to cognition. We paired targeted knockdown of circHomer1 within mouse OFC with in vivo electrophysiological recordings during a touchscreen visual reversal task to determine the role of circHomer1 on neuronal signaling within the OFC during learning.

For more detailed methodology, see Supplemental Experimental Procedures.

Animals

The Institutional Care and Use Committee (IACUC) at the University of New Mexico Health Sciences Center approved all experimental procedures. WT mice used in our study were adult male C57BL/6J mice (The Jackson Laboratory, Bar Harbor, ME). Investigators were blinded where possible to animal treatment groups.

Operant behavioral testing and stereotaxic surgery

Twenty male C57BL/6J mice (n = 10 per group) were acquired from The Jackson Laboratories and used for in vivo electrophysiological testing. Behavioral acclimation and operant testing were performed as described previously[31, 32]. Following pre-training, animals underwent stereotaxic surgery, which involved delivery of either a scrambled sh-control lentiviral vector or an sh-circHomer1 lentiviral vector targeted to the unique backsplice junction of exons 2 and 5 of the mature circHomer1 RNA product[31, 32].

In vivo electrophysiological recordings

Neuronal activity was recorded using a custom 16-channel (2x2x4) fixed microarray with OFC-specific separation (2.75mm) (Innovative Neurophysiology, Inc.) and collected using a digital multichannel acquisition processor (OmniPlexD, Plexon, Dallas, TX, USA), as previously described[48, 55, 56] during seven sessions throughout discrimination and reversal on the following days: day two of discrimination, discrimination criterion (85% correct), first day of reversal (R1), three consecutive days at chance reversal (CR1-3 - starting on the day after the animal performed at least 10/30 correct trials), and reversal criterion (R85 - the day following the animal achieving 85% correct).

Tissue harvesting and RNA extraction/circRNA quantification

Following completion of behavioral testing, OFC micropunches were obtained and processed as described previously[31, 32] for mRNA and circRNA quantification by qRt-PCR.

KD of circHomer1 in the OFC of mice during a visual discrimination and reversal paradigm

Harnessing a translational touchscreen approach paired with in vivo electrophysiological recordings, we captured real-time activity from excitatory neurons within the OFC while an animal was performing an executive function task. Using lentiviral-mediated infection (Fig. 1A-B), we employed shRNA knockdown of circHomer1 within the bilateral OFC (Fig. 1A), which was previously shown to impair mid-stage (chance) reversal learning and alter gene isoform expression favoring a reduction in synaptic activity[32]. Using an asymmetrically designed shRNA targeting the unique circHomer1 backsplice junction[32] of exons two and five (Fig. 1B), we obtained an approximately 40 percent reduction in the mature circHomer1 transcript within the OFC (Fig. 1E). Viral delivery and electrode implantation occurred after initial training to ensure all animals appropriately learn to perform the touchscreen task (Fig. 1d). Bilateral microarrays were implanted into the OFC (AP + 2.6, ML, +/- 1.35, DV -2.6) with eight separate wires per hemisphere spaced 150µm apart to increase the likelihood of capturing distinct neuronal units[57] (Fig. 1A). Following two weeks of recovery allowing for stable viral expression and acclimation to the electrode implant, mice were trained to the initial discrimination, in which they had to correctly choose one shape (Fig. 1C) 85 percent of the time. Once criterion was achieved, they were switched to reversal and had to re-attain 85 percent correct responding on the previously unrewarded shape (Fig. 1C). Animals were recorded seven times across the discrimination and reversal paradigm (Fig. 1D).

Behavior in recorded animals recapitulates previously reported chance reversal impairment following KD of circHomer1 in the OFC

Behavior for recorded animals with confirmed KD was assessed and recapitulated the phenotype previously shown following circHomer1 KD in the OFC[32] (Supplementary Fig. 1). Correct trials were rewarded with strawberry Nesquik (Supplementary Fig. 1A), and incorrect trials were punished with a ten-second timeout and illumination of a houselight (Supplementary Fig. 1A). No significant difference was found in the number of sessions to obtain the initial discrimination between circHomer1 KD animals and controls (Supplementary Fig. 1B). However, in those animals with confirmed KD of circHomer1, significantly (p<0.05) more correct (Supplementary Fig. 1C) and incorrect (Supplementary Fig. 1D) trials were required to move through chance reversal (33-66% correct) compared to controls, while early and late reversal stages were not significantly impacted. The number of correction trials (a measure of perseveration) did not significantly differ between groups during chance reversal, but the general trend was apparent (Supplementary Fig. 1E). Additional analysis of specific trial types (Supplementary Fig. 1F) during chance reversal confirmed previous work[32] indicating reduction of circHomer1 in the OFC significantly (p<0.0001) impairs chance reversal measured by an increase in the number of all trial types during chance reversal (Supplementary Fig.1G).

Knockdown of circHomer1 within the OFC dynamically regulates choice-responsive firing rates.

To determine neuronal responsiveness to stimuli that may contribute to this learning impairment, we analyzed neuronal firing rates across a four second (1 second pre- to 3 second post-choice) epoch for correct and incorrect choices (Fig. 2A-B). For correct choices, a robust increase in event-responsive firing was observed in the one-second post-correct choice epoch, in which a secondary reinforcing tone is played to signal a correct choice (Figure 2C-G). Firing rates from both circHomer1 KD animals and controls increased relative to the one second pre-correct choice baseline during the conditioned stimulus (CS) tone period across all reversal recording days (Fig. 2C-G). Alternatively, for incorrect choices, neurons from both groups showed a slow ramping in activity following the incorrect choice (Fig. 2H-I). On the first day of reversal (R1), the average OFC neuronal firing rate from circHomer1 KD animals was increased significantly (p<0.0001) over controls in the one-second post-correct choice window when the CS is presented before significantly dropping below control animals in the two- (p<0.05) and three- (p<0.0001) second post-choice epochs, when the reward is consumed. (Fig. 1C). Similarly, for mid-chance reversal (CR1-3) and criterion reversal (R85) sessions, firing rates remained elevated in the 2-3 second post-correct choice epoch for OFC neurons from control animals compared to those from circHomer1 KD animals (p<0.0001) (Fig. 2D-G). Only on the second day of chance reversal (CR2) did correct choice-responsive firing rate from OFC neurons in control animals significantly (p<0.001) increase over circHomer1 KD during the CS period (Fig. 2E). These findings demonstrate a shift in correct-choice responsive neural firing from reward-responding to cue-responding in circHomer1 KD animals.

For incorrect choices, average firing rate of OFC neurons from circHomer1 KD animals significantly (p<0.05) increased over controls on the first day of reversal (Fig. 2H) for the entire three-second post-choice epoch. However, firing rate of OFC neurons from circHomer1 KD animals significantly dropped below controls on the first (p<0.001) (Fig. 2I) and second day of chance reversal (Fig. 2J) during the post-incorrect choice epoch before increasing again over controls on the last day of reversal (p<0.05) (Fig 2L). A modest reduction in firing rate (p<0.05 for the end of the post-choice window) was seen on chance reversal day 2 (Fig. 2J), while no differences were seen on chance reversal day 3 (Fig. 2K). Together, these results confirm previous work demonstrating firing rates of OFC neurons are dynamic across learning stages[48, 55] and provide novel evidence that they are differentially modulated following a choice when circHomer1 expression is reduced.

Because baseline firing rate can be indicative of synaptic properties and intrinsic excitability[58], we compared baseline firing frequency from the one-second pre-choice epoch across reversal sessions and found that it, too, was dynamically regulated throughout reversal (Fig. 2M). Baseline firing rates differed significantly between groups on chance reversal days 1 (p<0.0001, correct choice; P<0.001, incorrect choice) and 2 (p<0.01 correct choice; p<0.05 incorrect choices) and criterion reversal (p<0.001 correct choice; p<0.0001 incorrect choices) for the one-second epoch preceding choices (Fig. 2M) demonstrating differential rate modulation of neuronal firing frequency across learning stages that is impacted by reduction of circHomer1 within OFC neurons. Firing rate modulation in the immediate window preceding a choice is associated with attention allocated to the choice and anticipation of a predicted outcome[45]. For this reason, we analyzed firing rate in the 100 ms immediately before a choice when the animal would be oriented to the screen and found similar rate modulation to the entire baseline window across reversal. However, the firing rate within this 100 ms window was higher than the baseline firing rate across the entire one second baseline window (Fig. 2N). Anticipatory firing rate was significantly higher in OFC neurons from circHomer1 KD animals on the first and last day of reversal for correct choices (p<0.001) and incorrect choices (p<0.0001) (Fig. 2N) when the outcome-associations are less ambiguous. These results suggest circHomer1 is important for maintaining baseline firing characteristics of OFC neurons.

While the firing rate of neuronal populations is likely to change in response to a salient event, the proportion of OFC neurons recruited to respond to a specific event has also been shown to be dynamically regulated throughout learning[48]. To determine whether neuronal units were differentially recruited in KD and control animals across reversal stages, we analyzed the percentage of neuronal units that significantly (p<0.05) increased their firing rate over the pre-choice baseline across reversal learning stages. Significant differences between groups arose on the second day of chance reversal, whereby control animals had a significantly (p<0.0001) higher proportion of correct choice-responsive units compared to circHomer1 KD animals, and this difference was maintained throughout the remaining reversal learning sessions (Fig. 2O). Oppositely, the percentage of incorrect choice-responsive units was significantly higher in circHomer1 KD animals compared to controls on chance reversal days 2 (p<0.05) and 3 (p<0.0001), while the proportion of incorrect-responsive units was significantly reduced in circHomer1 KD animals compared to controls on the final session of reversal learning (p<0.0001) (Fig. 2P). These results indicate that reduction of circHomer1 within the OFC alters neuronal recruitment to a choice through enhanced error-responsive recruitment and reduction of correct-responsive recruitment during chance reversal learning.

To further assess baseline firing properties at the cellular level following the loss of circHomer1, we knocked down circHomer1 in mouse hippocampal cultures using the same lentiviral vector and shRNA used in vivo. No significant changes were found in inward sodium (Supplementary Fig. 2A and B) or outward potassium current (Supplementary Fig. 2A and C). We confirmed a significant reduction in spontaneous excitatory postsynaptic current (sEPSC) frequency (p<0.05) (Supplementary Fig. 2D and E) and a non-significant (p=0.06) reduction in burst frequency (Supplementary Fig. 2G and H) with no change in amplitude (Supplementary Fig. 2F) demonstrating loss of circHomer1 in vitro recapitulates the reduction in baseline firing frequency found in vivo. Altogether, these data implicate a role for circHomer1 in maintenance of both baseline neuronal characteristics at the cellular and network level as well as in an evoked response to salient stimuli.

ITPC is aberrantly increased across reversal learning following circHomer1 KD in the OFC

Local field potential (LFP) oscillatory phase alignment was assessed across trials to determine event-locked shifts in coordinated oscillatory activity in response to correct and incorrect choices. Inter-trial phase consistency (ITPC) is a measure of the coordinated activity of a neuronal population and is thought to be important in the processing of salient events[59-61]. An ITPC value of one represents the phase of oscillations aligning perfectly at a given time, while a value of 0 represents no phase alignment. ITPC was aligned to correct and incorrect choices and analyzed within the same one second pre-choice and three second post-choice window as used for spike-firing analysis. In accordance with previous work[48, 55], frequency-specific region of interest (ROI) analyses were performed across the following frequency ranges: low delta (1-2 Hz), theta (4-8 Hz), alpha (9-13 Hz), beta (14-30 Hz), and gamma (30-50 Hz) during the 0-250ms post-choice epoch (Fig. 3B, black boxes) to examine direct choice-responsive phase alignment. As expected, correct-choice associated ITPC peaked early in chance reversal (Fig. 3 A-E, Supplementary Figure 3A-E) and dissipated for the remaining reversal sessions for all tf-ROIs.

No group differences were observed for correct choice-responsive ITPC on the first (Fig. 3A and F) and final (Fig. 3E and H) session of reversal across tf-ROIs. A significant increase in ITPC was observed across tf-ROIs in the OFC of circHomer1 KD animals coincident with the behavioral deficit across chance reversal sessions (Fig. 3B-D and G).

Incorrect choice-associated ITPC similarly peaked on the first day of chance reversal (Fig. 3I, Supplementary Fig. 3F). A modest (p<0.1) increase in incorrect choice-associated ITPC was observed in circHomer1 KD animals on the first day of reversal (Fig. 3I and N). A more robust (p<0.0001) increase in incorrect choice-associated ITPC was found during chance reversal sessions in circHomer1 KD animals, with particularly strong effects in low frequency delta and theta tf-ROIs (Fig. 3J-L and O). By the final reversal session, ITPC associated with incorrect choices was no longer significantly increased above control levels (Fig. 3M and P). These data point to aberrantly sustained low frequency ITPC following both correct and incorrect choices across critical learning stages. The dynamic modulation of OFC ITPC across reversal stages recapitulates previous findings[55] and suggests circHomer1 is important for the proper timed coordination of oscillatory activity during learning.

Reward-related alterations in time-frequency power are present in the OFC of circHomer1 KD animals

Time-frequency power is a measure of the amplitude of a signal and has been used as an electrophysiological marker to assess cognitive control[53, 62, 63]. The same frequency range definitions from our ITPC analysis were used to assess power (Fig. 4A-E and I-M); however, two separate time-windows were analyzed for their relevance to reward- (250-550ms) (Fig. 4B black boxes) and punishment-related (500-800ms) (Fig. 4J black boxes) signaling as described previously[64]. No significant group differences were observed across time-frequency ROIs (tf-ROIs) on the first day of reversal (Fig. 4A and F) following correct choices. Interestingly, reward-related power within the tf-ROIs was significantly (p<0.05) diminished during chance reversal (Fig. 4B-D and G) in circHomer1 KD animals and remained lower (p<0.05) than controls at reversal criterion (Fig. 4E and H). Notably, data are presented as a change from the one-second pre-choice baseline (dB), with control animals exhibiting a general increase in reward-related power relative to baseline during chance and criterion reversal, while circHomer1 KD animals exhibited either decreased or unchanged reward-related power during these reversal stages (Fig. 4G-H, Supplementary Fig. 4A-E).

Reduction of tf-ROI power relative to baseline was observed following incorrect choices consistent with previous work measuring punishment-associated power changes in rodents and humans[64] (Fig. 4I-P). Again, no significant group differences were present on the first day of reversal across the tf-ROIs measured (Fig. 4I and N). Strikingly, however, a robust difference was observed between circHomer1 KD animals and controls during chance reversal, particularly for mid-range frequencies (theta-beta) (Fig. 4J-L and O). During chance reversal, incorrect choice-responsive tf-ROI power was significantly reduced relative to baseline in controls when compared to circHomer1 KD animals (Fig. 4O). No group differences in incorrect choice-responsive tf-ROI power were measured at reversal criterion (Fig. 4M and P). In contrast to correct choice-responsive power, control animals tended to show a reduction in incorrect-choice associated power, while circHomer1 KD animals showed a markedly diminished change from their baseline power (Supplementary Fig 4F-J). These tf-ROI power analyses are consistent with previously reported event-responsive power changes[64]. Importantly, they indicate reduction of circHomer1 within the OFC induces reward- and error-related neural alterations during chance reversal, the stage where behavior is significantly impaired in circHomer1 KD animals.

KD of circHomer1 alters transcriptional landscape of post-synaptic signaling targets

Both our in vitro and in vivo recording data implicated a reduction in synaptic firing potential. Our previous work also demonstrated that KD of circHomer1 within the OFC alters specific isoforms important for synaptic signaling including Grin2b and Fmr1[32]. We quantified mRNA expression for known regulators of excitatory glutamatergic neurotransmission (Fig. 5A) including immediate early gene Homer1a and PSD scaffold component Homer1b; genes encoding NMDAR subunits, Grin2a and b; genes coding for AMPAR subunits, Gria1 and 2, and Fmr1, which encodes the RNA binding protein FMRP, to identify a potential mechanism by which KD of circHomer1 is able to influence synaptic signaling. A main effect of group was observed, whereby KD of circHomer1 resulted in a general reduction of all synaptic transcripts tested (p= 0.001); however, no individual transcripts were found to be significantly reduced following Bonferroni’s post-hoc correction (Fig. 5B). Because these transcripts are generally regulated in unison due to their role in excitatory neurotransmission, we expected the correlation among transcripts to be robust. As expected, we found that when all samples were combined (Fig. 5C), circHomer1 was strongly correlated with Homer1a (r = 0.90, p = 0.0004) and Homer1a and b were correlated (r = 0.71, p = 0.02). Additionally, post-synaptic targets were primarily positively correlated with each other. Fmr1 was positively correlated with Gria1 (r = 0.7, p = 0.024), Gria2 (r = 0.92, p<0.0001), Grin2b (r = 0.87, p = 0.001), and Grin2a (r = 0.71, p = 0.02). Gria1 and Gria2 were positively correlated, (r = 0.85, p = 0.002) as were Grin2b and Gria1 (r = 0.73, p = 0.02).

We then assessed correlation between these target genes separately for control (Fig. 5D) and circHomer1 KD animals (Fig. 5E). While the sample size is limiting, the relationships among transcripts are altered between these groups. Specifically, circHomer1 remained highly correlated with Homer1a in controls (r = 0.93, p = 0.007), but this correlation was completely lost in OFC tissue from KD animals (r = -0.009, p = 0.99), while the correlation between circHomer1 and Homer1b became negative (r = -0.35, p = 0.65). Interestingly, the direction of correlation was changed between several of the post-synaptic receptor genes in OFC tissue from circHomer1 KD animals. Grin2b became more strongly negatively correlated with Homer1a (r = -0.62, p = 0.38) and Homer1b (r = -0.85, p = 0.15) in KD animals. Moreover, Grin2a was positively correlated with Homer1 isoforms in KD animals (r = 0.74, p = 0.26 for circHomer1; r = 0.39, p = 0.61 for Homer1a; and r = 0.18, p = 0.82 for Homer1b), but negatively correlated in control animals (r = -0.71, p = 0.11 for circHomer1; r = -0.69, p = 0.13 for Homer1a; and r = -0.24, p = 0.65 for Homer1b). These correlations were not statistically significant, likely due to the small numbers of animals in each group but may indicate a compensatory change in glutamatergic receptor composition and coordination with PSD scaffolds following KD of circHomer1.

Low expression of circHomer1 in rodent OFC induces psychiatric disease-relevant phenotypes

The goal of the current study was to determine whether reducing expression of circHomer1 within the orbitofrontal cortex, as has been reported in psychiatric[31, 32] and neurodegenerative disorders[38-40, 42], could produce electrophysiological signatures similar to those observed in these patient populations[65-67]. It has been reported that individuals with psychiatric disorders have behavioral manifestations of inappropriate salience signaling[68-70], whereby they assign incentive salience to cues rather than outcomes. Mistimed neuronal firing can influence inappropriate salience assignment and motivation that may drive behavioral inflexibility[55, 71] and misguide reward-seeking behavior. This is a key behavioral manifestation associated with substance abuse and poor decision-making[68, 70]. We demonstrate that knockdown of circHomer1 within the OFC, a frontocortical region associated with outcome-monitoring, shifts the neuronal signaling to cue-responding (i.e. increased firing rate to the tone) rather than reward-responding (increased firing rate to the reward) as observed in control animals. This heightened response to the cue appears early in reversal learning, while a loss of reward responding occurs during chance reversal, when it is important for the OFC to monitor rewarded outcomes to update behavior[48]. The reduced firing rate of circHomer1 KD animals during the reward window coincides with behavioral impairment in these animals[32], marked by an increase in the number of trials to complete chance reversal[32], suggesting neuronal response to reward in this timeframe may be key to update value-guided behavior. This behavioral phenotype has been observed in patients with SCZ[72] and in those with reduced OFC-striatal activity[73]. We also found the proportion of correct choice-responsive neurons decreases throughout chance reversal, while the proportion of incorrect choice-responsive neurons increases in circHomer1 KD animals, indicating additional disruption in OFC neural recruitment tied to inappropriate salience assignment during a critical learning window.

KD of circHomer1 was previously predicted to result in reduced synaptic activity through gene expression analysis[32]; however, the molecular mechanism by which this occurs is unclear. Homer1 is homeostatically regulated through alternative splicing[23], and knockdown of the activity-dependent isoform, circHomer1, has been shown to increase both constitutive, Homer1b, and activity-dependent, Homer1a, mRNA localization into the synapse[31, 32]. An increase in synaptic Homer1b has been associated with neuronal hyperexcitability in disease[74], while increased synaptic Homer1a is often associated with a homeostatic renormalization[15, 23, 30] and reduced neuronal excitability[75, 76]. Over-expression of Homer1a has also been shown to attenuate intracellular calcium release through group 1 mGluR activation[28]. We found that KD of circHomer1 resulted in a global down-regulation of glutamatergic receptor gene expression (Gria1 and 2 and Grin2a and b) as well as Fmr1, which encodes the RNA binding protein FMRP responsible for regulating synaptic protein translation important for mGluR-mediated LTD[77-79]. These results confirm the previous finding that KD of circHomer1 in the OFC results in transcriptional downregulation of key synaptic mediators[32]. The altered correlations of glutamatergic receptors with Homer1 isoforms in KD animals might additionally suggest a compensatory change in receptor composition to homeostatically regulate the synapse in an attempt to prevent prolonged hyperexcitability. However, we did not measure gene expression specifically in synaptosomes in this study and synaptic transcript localization may differ from total tissue expression as we have observed previously[31]. It is probable that the activity-dependent transcriptional regulation of these isoforms may result in differential signaling depending on the timing and network activity demands.

Regarding neuronal firing properties at baseline, we found that, although baseline (pre-choice) firing rate is dynamic across reversal learning, the profound reduction in baseline firing rate of circHomer1 KD neurons during chance reversal may be detrimental to learning at this stage and represent an altered homeostatic response to network activity[80]. Homer1 isoforms are well-known for their activity-regulated homeostatic reorganization of the postsynaptic neuron[18] , and this work reinforces the idea that circHomer1 may be influencing linear Homer1 isoform localization at the synapse[31] in response to increased network demands required for efficient learning.

CircHomer1 maintains cellular firing potential

Previous experiments have shown that heightened network activity (induced by bicuculline) results in a robust increase in circHomer1 expression and localization to the dendrites[33]. In this way, circHomer1 appears to act as an immediate early gene similar to its linear counterpart Homer1a, which scales down synaptic weights through scaffold disruption and AMPAR internalization[18, 30]. We have previously demonstrated that circHomer1 KD increases both Homer1b and Homer1a expression within the synapse[31], suggesting a potential compensatory mechanism triggered to maintain balance of postsynaptic firing potential. Previous work indicated circHomer1 expression is inversely correlated with duration of illness[32], which could indicate disease progression reflects downregulation of important synaptic effectors including circHomer1. This is corroborated by our finding that KD of circHomer1 produces long-lasting changes in transcriptional regulation of post-synaptic regulators of glutamatergic neurotransmission. Chronic reduction of circHomer1, as seen in psychiatric and neurodegenerative disease, could potentially disrupt the activity-dependent regulation resulting in sustained downscaling as a homeostatic mechanism to prevent repeated hyperexcitation. If this were the case, we would expect to see the chronic loss of circHomer1 to result in diminished cellular firing potential. Our in vitro experiments confirmed reduced baseline firing frequency following circHomer1 KD suggesting that reduction of circHomer1 is sufficient to impair spontaneous neuronal activity and disturb cellular firing potential. The reduction in sEPSC frequency in circHomer1 KD neurons may be a result of reduced presynaptic activity through a feedback mechanism or diminished AMPA receptor unsilencing in surrounding postsynaptic terminals resulting from a disruption of Homer1 scaffold integrity. Given the trend towards reduced burst firing in circHomer1 KD neurons, we hypothesize circHomer1 plays a role in maintaining excitatory synaptic weights[81] and its loss may reduce the likelihood of synaptic potentiation upon stimulation. This may, in part, explain how the reduced baseline firing rate observed in vivo impairs appropriate signaling during chance reversal. Homer1 family members differentially modulate group 1 mGluRs (mGluR1/5), whose activation is important in the regulation of glutamatergic signaling and promotes burst firing[82] pointing to a potential mechanism whereby circHomer1 either directly or indirectly, via upregulation of synaptic Homer1a localization²⁹ and downregulation of Fmr1[79], influences mGluR signaling resulting in attenuated burst firing[83]. Reduction of both EPSC frequency and burst firing, as seen in circHomer1 KD animals, has been observed in other genetic mouse models of schizophrenia[84], further underlining the significance of circHomer1 expression to this disorder.

CircHomer1 influences OFC coordination to salient events

Timing of population activity is thought to underlie signal propagation to downstream regions[61, 85]. The primary projections from the lateral OFC are into the dorsal striatum, a region important for producing goal-directed movement in response to outcomes[86]. Aberrant or mistimed signaling between the OFC and striatum has been shown to disrupt behavioral flexibility[55]. We found that circHomer1 KD in the OFC resulted in aberrantly increased and sustained coordinated activity as measured by ITPC. In particular, theta ITPC, which has been shown to be especially important for long-range signal propagation,[87-90] was increased in circHomer1 KD animals throughout late-stage reversal to both correct and incorrect choices. High ITPC within the OFC is associated with salient, changing event contingencies and typically peaks during early-to-mid chance reversal when the new association is being formed[48]. ITPC should, however, decrease as the outcome association is solidified indicating the outcome is no longer unexpected. Aberrantly increased theta ITPC has been found in individuals with psychiatric conditions[91] and in animal models of these disorders[55, 92]. Given that pyramidal neurons of the OFC project onto dopaminergic neurons within the dorsal striatum, loss of circHomer1 in the OFC may alter downstream dopaminergic signaling linked to incentive salience and value-guided behavior[51, 68, 93]. Moreover, this increase in coordination of oscillatory activity may represent a compensatory mechanism whereby intact synapses can homeostatically maintain their signaling when excessive downscaling occurs[14, 94]. Assessment of striatal activity following loss of circHomer1 will be necessary to reveal this mechanism.

Endogenous circHomer1 and Homer1b expression is temporally regulated across reversal learning[31]. Specifically, circHomer1 expression transiently drops in wild-type mice during chance reversal and Homer1b expression is increased at this stage, while both remain unchanged from a “baseline” state during all other reversal stages. The dynamic regulation of these isoforms is known to be critical for homeostatic synaptic plasticity[15, 18, 23, 30]; therefore, we hypothesize that the transient dip in circHomer1 expression during chance reversal is likely prompting scaffold rearrangement[18] through Homer1b shuttling within the synapse[31]. However, the inability for circHomer1 expression to return to normal, as in circHomer1 KD animals, would potentially inhibit the synaptic renormalization critical for synapse stabilization, important for learning. The flexibility of synaptic tuning via Homer1 isoform regulation thus appears to be critical during early-mid reversal learning, yet, detrimental during mid-to-late stage reversal. This is similar to the dynamic nature of NMDAR subunits across reversal learning; whereby, the GluN2B subunit, associated with flexibility, is critical during early-to-mid reversal[56] and the GluN2A subunit, associated with stability, dominates during mid-to-late stage reversal[95]. circHomer1 KD animals exhibit a learning deficit during mid-stage (chance) reversal, indicating circHomer1 is likely acting in the transition from plastic to stable synapses. We found a reduction in Grin2a and b isoforms following KD of circHomer1 and differential correlation of these isoforms between circHomer1 KD and control animals, which likely contributes to the behavioral and electrophysiological phenotypes observed.

CircHomer1 expression in the OFC is a cognitive-related biomarker

CircHomer1 expression is reduced in multiple patient populations with psychiatric[31, 32] and neurodegenerative disease[38-42]. Its expression is correlated with duration of illness in human patients[32], cognitive performance in rodents during a translational touchscreen task[31], and alterations in neuronal signaling. Importantly, circHomer1 is more stable than linear RNA due to its lack of free ends[37, 96] and its expression in peripheral blood cells correlates with disease[31, 42] and is amenable to antipsychotics[32] indicating its potential role as a predictive biomarker[96] of cognitive dysfunction. While there is a degree of overlap in circHomer1 expression in patients and controls [31, 32] as well as in our animal model[32] (and current work), its expression may be more predictive of a particular behavioral phenotype such as behavioral inflexibility that spans multiple disorders rather than any specific disease as a whole. We present findings that suggest circHomer1 influences neuronal signaling and coordination within the OFC during specific windows critical for efficient learning. We hypothesize based on this work and previous studies that circHomer1 does this by balancing linear Homer1 isoform expression in an activity-dependent manner, and loss of circHomer1 results in unchecked linear isoform regulation required for homeostatic maintenance. Future work will be needed to determine associations with circHomer1 expression and human cognitive flexibility, but our results provide a promising rationale for this work. Because circHomer1 expression is associated with multiple neurological and psychiatric disorders such as Alzheimer’s disease[38-40], Parkinson’s disease[42], mesial temporal lobe epilepsy[41] and SCZ and BD[31, 32], there is likely a convergent mechanism by which synaptic dysfunction results in the ultimate reduction of circHomer1 that promotes brain region-specific phenotypes.

Limitations

Due to the highly specific design needed for an shRNA to target a circRNA backsplice junction[37], we are only able to use a single shRNA. We have previously demonstrated the specificity of this shRNA[31, 32]. Only male mice were used in this study to be consistent with previously published work characterizing circHomer1 KD animals[32]; however, recent work has suggested sex differences may be present regarding loss of circHomer1[39]. We recognize that the sample size is limited in the behavioral and transcriptional analysis due to experimental design trade-offs whereby we selected a random subset of animals to be used for PCR quantification and chose to use the same animals for behavioral analysis as this was a replication study following our previous work[32]. The molecular and behavioral characteristics of this subset are, however, in accordance with previously published phenotypes[31, 32].

Acknowledgements and financial disclosures

This work was supported by a NARSAD Young Investigator Grant (FP00000839—N.M.) by the Brain & Behavior Research Foundation, an NIH P20 exploratory grant (P20GM121176-N.M.),

a high priority short-term R56 award from the NIMH (1R56MH119150-01-N.M. and J.W.), R01s from NIAAA (1R01AA025652-J.B.) and NINDS (1R01NS116051-J.W.), and an NRSA research fellowship (F31 MH122121-01A1-A.Z.). Authors would like to thank David Gregg, Madison Otero, Dr. Kristin Marquardt, Dr. James Cavanagh and Dr. David Linsenbardt for technical and analytical assistance. A previous version of the manuscript was uploaded as a preprint to ResearchSquare: https://doi.org/10.21203/rs.3.rs-1862282/v1.

Author conflict of interest:

N.M. has a financial interest as co-founder of Circular Genomics Inc. and as inventor of patents related to the use of circRNAs for brain disease diagnostics. N.M. was not involved in data acquisition for this manuscript. A.J.Z., J.P.W., G.P., and J.L.B. have no conflicts to disclose.

Code availability

MATLAB code for analysis of electrophysiological data is available from the corresponding author upon request.

1. McGrath, J., et al., Schizophrenia: a concise overview of incidence, prevalence, and mortality. Epidemiol Rev, 2008. 30: p. 67-76.

2. Merikangas, K.R., et al., Lifetime prevalence of mental disorders in U.S. adolescents: results from the National Comorbidity Survey Replication--Adolescent Supplement (NCS-A). J Am Acad Child Adolesc Psychiatry, 2010. 49(10): p. 980-9.

3. Bigdeli, T.B., et al., Genome-Wide Association Studies of Schizophrenia and Bipolar Disorder in a Diverse Cohort of US Veterans. Schizophr Bull, 2021. 47(2): p. 517-529.

4. Bipolar, D., et al., Genomic Dissection of Bipolar Disorder and Schizophrenia, Including 28 Subphenotypes. Cell, 2018. 173(7): p. 1705-1715 e16.

5. Craddock, N., M.C. O'Donovan, and M.J. Owen, The genetics of schizophrenia and bipolar disorder: dissecting psychosis. J Med Genet, 2005. 42(3): p. 193-204.

6. Mullins, N., et al., Genome-wide association study of more than 40,000 bipolar disorder cases provides new insights into the underlying biology. Nat Genet, 2021. 53(6): p. 817-829.

7. Stahl, E.A., et al., Genome-wide association study identifies 30 loci associated with bipolar disorder. Nat Genet, 2019. 51(5): p. 793-803.

8. Barry, G., et al., The long non-coding RNA Gomafu is acutely regulated in response to neuronal activation and involved in schizophrenia-associated alternative splicing. Mol Psychiatry, 2014. 19(4): p. 486-94.

9. Akula, N., et al., An Integrative Genomic Study Implicates the Postsynaptic Density in the Pathogenesis of Bipolar Disorder. Neuropsychopharmacology, 2016. 41(3): p. 886-95.

10. Schizophrenia Working Group of the Psychiatric Genomics, C., Biological insights from 108 schizophrenia-associated genetic loci. Nature, 2014. 511(7510): p. 421-7.

11. Surmeier, D.J. and R. Foehring, A mechanism for homeostatic plasticity. Nat Neurosci, 2004. 7(7): p. 691-2.

12. Turrigiano, G.G. and S.B. Nelson, Homeostatic plasticity in the developing nervous system. Nat Rev Neurosci, 2004. 5(2): p. 97-107.

13. Turrigiano, G.G. and S.B. Nelson, Hebb and homeostasis in neuronal plasticity. Curr Opin Neurobiol, 2000. 10(3): p. 358-64.

14. Fernandes, D. and A.L. Carvalho, Mechanisms of homeostatic plasticity in the excitatory synapse. J Neurochem, 2016. 139(6): p. 973-996.

15. Hu, J.H., et al., Homeostatic scaling requires group I mGluR activation mediated by Homer1a. Neuron, 2010. 68(6): p. 1128-42.

16. Norton, N., et al., Mutation screening of the Homer gene family and association analysis in schizophrenia. Am J Med Genet B Neuropsychiatr Genet, 2003. 120B(1): p. 18-21.

17. Szumlinski, K.K., P.W. Kalivas, and P.F. Worley, Homer proteins: implications for neuropsychiatric disorders. Curr Opin Neurobiol, 2006. 16(3): p. 251-7.

18. de Bartolomeis, A., et al., The Homer1 family of proteins at the crossroad of dopamine-glutamate signaling: An emerging molecular "Lego" in the pathophysiology of psychiatric disorders. A systematic review and translational insight. Neurosci Biobehav Rev, 2022. 136: p. 104596.

19. Szumlinski, K.K., et al., Behavioral and neurochemical phenotyping of Homer1 mutant mice: possible relevance to schizophrenia. Genes Brain Behav, 2005. 4(5): p. 273-88.

20. Rietschel, M., et al., Genome-wide association-, replication-, and neuroimaging study implicates HOMER1 in the etiology of major depression. Biol Psychiatry, 2010. 68(6): p. 578-85.

21. Peyrot, W.J. and A.L. Price, Identifying loci with different allele frequencies among cases of eight psychiatric disorders using CC-GWAS. Nat Genet, 2021. 53(4): p. 445-454.

22. Soloviev, M.M., et al., Molecular characterisation of two structurally distinct groups of human homers, generated by extensive alternative splicing. J Mol Biol, 2000. 295(5): p. 1185-200.

23. Thalhammer, A., F. Jaudon, and L.A. Cingolani, Emerging Roles of Activity-Dependent Alternative Splicing in Homeostatic Plasticity. Front Cell Neurosci, 2020. 14: p. 104.

24. Hayashi, M.K., et al., The postsynaptic density proteins Homer and Shank form a polymeric network structure. Cell, 2009. 137(1): p. 159-71.

25. Ciruela, F., M.M. Soloviev, and R.A. McIlhinney, Co-expression of metabotropic glutamate receptor type 1alpha with homer-1a/Vesl-1S increases the cell surface expression of the receptor. Biochem J, 1999. 341 ( Pt 3): p. 795-803.

26. Ciruela, F., et al., Homer-1c/Vesl-1L modulates the cell surface targeting of metabotropic glutamate receptor type 1alpha: evidence for an anchoring function. Mol Cell Neurosci, 2000. 15(1): p. 36-50.

27. Brakeman, P.R., et al., Homer: a protein that selectively binds metabotropic glutamate receptors. Nature, 1997. 386(6622): p. 284-8.

28. Tu, J.C., et al., Homer binds a novel proline-rich motif and links group 1 metabotropic glutamate receptors with IP3 receptors. Neuron, 1998. 21(4): p. 717-26.

29. Sala, C., et al., Inhibition of dendritic spine morphogenesis and synaptic transmission by activity-inducible protein Homer1a. J Neurosci, 2003. 23(15): p. 6327-37.

30. Diering, G.H., et al., Homer1a drives homeostatic scaling-down of excitatory synapses during sleep. Science, 2017. 355(6324): p. 511-515.

31. Hafez, A.K., et al., A bidirectional competitive interaction between circHomer1 and Homer1b within the orbitofrontal cortex regulates reversal learning. Cell Rep, 2022. 38(3): p. 110282.

32. Zimmerman, A.J., et al., A psychiatric disease-related circular RNA controls synaptic gene expression and cognition. Mol Psychiatry, 2020. 25(11): p. 2712-2727.

33. You, X., et al., Neural circular RNAs are derived from synaptic genes and regulated by development and plasticity. Nat Neurosci, 2015. 18(4): p. 603-610.

34. Dai, X., et al., RNA-binding protein trinucleotide repeat-containing 6A regulates the formation of circular RNA circ0006916, with important functions in lung cancer cells. Carcinogenesis, 2018. 39(8): p. 981-992.

35. Glazar, P., P. Papavasileiou, and N. Rajewsky, circBase: a database for circular RNAs. RNA, 2014. 20(11): p. 1666-70.

36. Dell'Orco, M., R.J. Oliver, and N. Perrone-Bizzozero, HuD Binds to and Regulates Circular RNAs Derived From Neuronal Development- and Synaptic Plasticity-Associated Genes. Front Genet, 2020. 11: p. 790.

37. Cortes-Lopez, M. and P. Miura, Emerging Functions of Circular RNAs. Yale J Biol Med, 2016. 89(4): p. 527-537.

38. Dube, U., et al., An atlas of cortical circular RNA expression in Alzheimer disease brains demonstrates clinical and pathological associations. Nat Neurosci, 2019. 22(11): p. 1903-1912.

39. Urdanoz-Casado, A., et al., Gender-Dependent Deregulation of Linear and Circular RNA Variants of HOMER1 in the Entorhinal Cortex of Alzheimer's Disease. Int J Mol Sci, 2021. 22(17).

40. Cervera-Carles, L., et al., Assessing circular RNAs in Alzheimer's disease and frontotemporal lobar degeneration. Neurobiol Aging, 2020. 92: p. 7-11.

41. Gray, L.G., et al., Identification of Specific Circular RNA Expression Patterns and MicroRNA Interaction Networks in Mesial Temporal Lobe Epilepsy. Front Genet, 2020. 11: p. 564301.

42. Ravanidis, S., et al., Differentially Expressed Circular RNAs in Peripheral Blood Mononuclear Cells of Patients with Parkinson's Disease. Mov Disord, 2021. 36(5): p. 1170-1179.

43. Young, J.J. and M.L. Shapiro, The orbitofrontal cortex and response selection. Ann N Y Acad Sci, 2011. 1239: p. 25-32.

44. Gremel, C.M. and R.M. Costa, Orbitofrontal and striatal circuits dynamically encode the shift between goal-directed and habitual actions. Nat Commun, 2013. 4: p. 2264.

45. Schoenbaum, G., et al., A new perspective on the role of the orbitofrontal cortex in adaptive behaviour. Nat Rev Neurosci, 2009. 10(12): p. 885-92.

46. Wallis, J.D., Neuronal mechanisms in prefrontal cortex underlying adaptive choice behavior. Ann N Y Acad Sci, 2007. 1121: p. 447-60.

47. Knudsen, E.B. and J.D. Wallis, Closed-Loop Theta Stimulation in the Orbitofrontal Cortex Prevents Reward-Based Learning. Neuron, 2020. 106(3): p. 537-547 e4.

48. Marquardt, K., R. Sigdel, and J.L. Brigman, Corrigendum to "Touch-screen visual reversal learning is mediated by value encoding and signal propagation in the orbitofrontal cortex" [Neurobiol. Learn. Mem. 139 (2017) 179-188]. Neurobiol Learn Mem, 2021. 180: p. 107380.

49. Ragozzino, M.E., The contribution of the medial prefrontal cortex, orbitofrontal cortex, and dorsomedial striatum to behavioral flexibility. Ann N Y Acad Sci, 2007. 1121: p. 355-75.

50. Jackowski, A.P., et al., The involvement of the orbitofrontal cortex in psychiatric disorders: an update of neuroimaging findings. Braz J Psychiatry, 2012. 34(2): p. 207-12.

51. Fettes, P., L. Schulze, and J. Downar, Cortico-Striatal-Thalamic Loop Circuits of the Orbitofrontal Cortex: Promising Therapeutic Targets in Psychiatric Illness. Front Syst Neurosci, 2017. 11: p. 25.

52. Lee, H.S. and J.S. Kim, Implication of Electrophysiological Biomarkers in Psychosis: Focusing on Diagnosis and Treatment Response. J Pers Med, 2022. 12(1).

53. Javitt, D.C., et al., A roadmap for development of neuro-oscillations as translational biomarkers for treatment development in neuropsychopharmacology. Neuropsychopharmacology, 2020. 45(9): p. 1411-1422.

54. Cavanagh, J.F., et al., Electrophysiological biomarkers of behavioral dimensions from cross-species paradigms. Transl Psychiatry, 2021. 11(1): p. 482.

55. Marquardt, K., et al., Impaired cognitive flexibility following NMDAR-GluN2B deletion is associated with altered orbitofrontal-striatal function. Neuroscience, 2021. 475: p. 230-245.

56. Brigman, J.L., et al., GluN2B in corticostriatal circuits governs choice learning and choice shifting. Nat Neurosci, 2013. 16(8): p. 1101-10.

57. Buzsaki, G., Large-scale recording of neuronal ensembles. Nat Neurosci, 2004. 7(5): p. 446-51.

58. Turrigiano, G., Too many cooks? Intrinsic and synaptic homeostatic mechanisms in cortical circuit refinement. Annu Rev Neurosci, 2011. 34: p. 89-103.

59. Schroeder, C.E. and P. Lakatos, Low-frequency neuronal oscillations as instruments of sensory selection. Trends Neurosci, 2009. 32(1): p. 9-18.

60. Lakatos, P., et al., Entrainment of neuronal oscillations as a mechanism of attentional selection. Science, 2008. 320(5872): p. 110-3.

61. Varela, F., et al., The brainweb: phase synchronization and large-scale integration. Nat Rev Neurosci, 2001. 2(4): p. 229-39.

62. Cavanagh, J.F. and M.J. Frank, Frontal theta as a mechanism for cognitive control. Trends Cogn Sci, 2014. 18(8): p. 414-21.

63. Cohen, M.X., A neural microcircuit for cognitive conflict detection and signaling. Trends Neurosci, 2014. 37(9): p. 480-90.

64. Cavanagh, J.F., et al., Amphetamine alters an EEG marker of reward processing in humans and mice. Psychopharmacology (Berl), 2022. 239(3): p. 923-933.

65. Angelopoulos, E., Brain functional connectivity and the pathophysiology of schizophrenia. Psychiatriki, 2014. 25(2): p. 91-4.

66. Cho, R.Y., R.O. Konecky, and C.S. Carter, Impairments in frontal cortical gamma synchrony and cognitive control in schizophrenia. Proc Natl Acad Sci U S A, 2006. 103(52): p. 19878-83.

67. Subramaniam, K., et al., Neural signal during immediate reward anticipation in schizophrenia: Relationship to real-world motivation and function. Neuroimage Clin, 2015. 9: p. 153-63.

68. Howes, O.D. and M.M. Nour, Dopamine and the aberrant salience hypothesis of schizophrenia. World Psychiatry, 2016. 15(1): p. 3-4.

69. Cicero, D.C., et al., Aberrant salience, self-concept clarity, and interview-rated psychotic-like experiences. J Pers Disord, 2015. 29(1): p. 79-99.

70. Berridge, K.C. and T.E. Robinson, Liking, wanting, and the incentive-sensitization theory of addiction. Am Psychol, 2016. 71(8): p. 670-679.

71. Ferguson, L.M., et al., Neurons of the Ventral Tegmental Area Encode Individual Differences in Motivational "Wanting" for Reward Cues. J Neurosci, 2020. 40(46): p. 8951-8963.

72. Leeson, V.C., et al., Discrimination learning, reversal, and set-shifting in first-episode schizophrenia: stability over six years and specific associations with medication type and disorganization syndrome. Biol Psychiatry, 2009. 66(6): p. 586-93.

73. Remijnse, P.L., et al., Reduced orbitofrontal-striatal activity on a reversal learning task in obsessive-compulsive disorder. Arch Gen Psychiatry, 2006. 63(11): p. 1225-36.

74. Cao, L., et al., Down-regulation of Homer1b/c protects against chemically induced seizures through inhibition of mTOR signaling. Cell Physiol Biochem, 2015. 35(4): p. 1633-42.

75. Wang, Y., et al., Scaffolding protein Homer1a protects against NMDA-induced neuronal injury. Cell Death Dis, 2015. 6: p. e1843.

76. Bockaert, J., J. Perroy, and F. Ango, The Complex Formed by Group I Metabotropic Glutamate Receptor (mGluR) and Homer1a Plays a Central Role in Metaplasticity and Homeostatic Synaptic Scaling. J Neurosci, 2021. 41(26): p. 5567-5578.

77. Clifton, N.E., et al., Regulation and Function of Activity-Dependent Homer in Synaptic Plasticity. Mol Neuropsychiatry, 2019. 5(3): p. 147-161.

78. Weiler, I.J., et al., Fragile X mental retardation protein is necessary for neurotransmitter-activated protein translation at synapses. Proc Natl Acad Sci U S A, 2004. 101(50): p. 17504-9.

79. Aloisi, E., et al., Altered surface mGluR5 dynamics provoke synaptic NMDAR dysfunction and cognitive defects in Fmr1 knockout mice. Nat Commun, 2017. 8(1): p. 1103.

80. Hengen, K.B., et al., Firing rate homeostasis in visual cortex of freely behaving rodents. Neuron, 2013. 80(2): p. 335-42.

81. Birtoli, B. and D. Ulrich, Firing mode-dependent synaptic plasticity in rat neocortical pyramidal neurons. J Neurosci, 2004. 24(21): p. 4935-40.

82. Collin, T., et al., Activation of metabotropic glutamate receptors induces periodic burst firing and concomitant cytosolic Ca2+ oscillations in cerebellar interneurons. J Neurosci, 2009. 29(29): p. 9281-91.

83. Li, Y., et al., Epileptiform stimulus increases Homer 1a expression to modulate synapse number and activity in hippocampal cultures. J Neurophysiol, 2013. 109(6): p. 1494-504.

84. Benamer, N., et al., GluD1, linked to schizophrenia, controls the burst firing of dopamine neurons. Mol Psychiatry, 2018. 23(3): p. 691-700.

85. Buzsaki, G. and A. Draguhn, Neuronal oscillations in cortical networks. Science, 2004. 304(5679): p. 1926-9.

86. Balleine, B.W., M.R. Delgado, and O. Hikosaka, The role of the dorsal striatum in reward and decision-making. J Neurosci, 2007. 27(31): p. 8161-5.

87. Tsujimoto, T., H. Shimazu, and Y. Isomura, Direct recording of theta oscillations in primate prefrontal and anterior cingulate cortices. J Neurophysiol, 2006. 95(5): p. 2987-3000.

88. von Stein, A., C. Chiang, and P. Konig, Top-down processing mediated by interareal synchronization. Proc Natl Acad Sci U S A, 2000. 97(26): p. 14748-53.

89. Huerta, P.T. and J.E. Lisman, Heightened synaptic plasticity of hippocampal CA1 neurons during a cholinergically induced rhythmic state. Nature, 1993. 364(6439): p. 723-5.

90. Buzsaki, G., Theta oscillations in the hippocampus. Neuron, 2002. 33(3): p. 325-40.

91. Roach, B.J., et al., Theta Phase Synchrony Is Sensitive to Corollary Discharge Abnormalities in Early Illness Schizophrenia but Not in the Psychosis Risk Syndrome. Schizophr Bull, 2021. 47(2): p. 415-423.

92. Caixeta, F.V., et al., Ketamine alters oscillatory coupling in the hippocampus. Sci Rep, 2013. 3: p. 2348.

93. Takahashi, Y.K., et al., Expectancy-related changes in firing of dopamine neurons depend on orbitofrontal cortex. Nat Neurosci, 2011. 14(12): p. 1590-7.

94. Bachmann, C., et al., Firing rate homeostasis counteracts changes in stability of recurrent neural networks caused by synapse loss in Alzheimer's disease. PLoS Comput Biol, 2020. 16(8): p. e1007790.

95. Brigman, J.L., et al., Impaired discrimination learning in mice lacking the NMDA receptor NR2A subunit. Learn Mem, 2008. 15(2): p. 50-4.

96. Liu, C.X. and L.L. Chen, Circular RNAs: Characterization, cellular roles, and applications. Cell, 2022. 185(12): p. 2016-2034.

Yes there is potential Competing Interest. N.M. has a financial interest as co-founder of Circular Genomics Inc. and as inventor of patents related to the use of circRNAs for brain disease diagnostics. N.M. was not involved in data acquisition for this manuscript.

CircHomerFINALExtendedDataFigs.pdf

Download PDF

Version 2

posted

You are reading this older preprint version

Read the latest preprint version →

Knockdown of circHomer1 in the orbitofrontal cortex results in differential encoding of salient stimuli

Status:

Version 2

Abstract

Figures

Introduction

Methods and Materials

Results

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 2