Inflammation plays a large role in the etiology of the late onset, sporadic form of Alzheimer’s disease (AD), yet these critical factors are not adequately modeled in mice where inflammatory mechanisms often differ widely from primates. In contrast, aging rhesus macaques offer a powerful translational model for investigating how advancing age and inflammation initiate early-stage pathology in sporadic AD, and for evaluating preventive therapeutic strategies. Unlike rodents, macaques possess highly developed association cortices with magnified calcium signaling, human-like inflammatory responses, and are naturally homozygous for ApoE-ε4—factors that together contribute to the spontaneous emergence of tau and amyloid pathology alongside cognitive decline. Critically, macaques allow the detection of early, soluble forms of hyperphosphorylated tau (pTau), including pT217Tau, which rapidly dephosphorylates postmortem and is rarely observable in human brain tissue outside of biopsies. New findings reveal that soluble pTau is neurotoxic and capable of propagating pathology across cortical networks, with elevated pT217Tau in plasma. Growing evidence points to age-related inflammatory signaling as a key driver of calcium dysregulation, which in turn promotes tau hyperphosphorylation, amyloid-β (Aβ) accumulation, synapse loss and autophagic degeneration. Both GCPII (glutamate carboxypeptidase II) and kynurenic acid inflammatory signaling have expanded roles in the primate association cortices that contribute to cognitive deficits. Pharmacological interventions in aged macaques demonstrate that targeting inflammation and restoring calcium homeostasis can significantly reduce pTau pathology with minimal side effects—highlighting a promising path for early intervention in AD.

The common, late onset form of Alzheimer’s disease (AD) selectively impacts higher brain circuits, with tau pathology and neurodegeneration preferentially afflicting glutamatergic neurons in the limbic and association cortices. Understanding this selective vulnerability may help reveal the etiology of sporadic AD and therapeutic targets for prevention. The current review describes that these vulnerable circuits express magnified calcium signaling needed for higher cognition and memory, but that heightened calcium signaling becomes toxic when dysregulated by age and inflammation. Many of the earliest pathological events in AD are challenging to study in human brain, as proteins such as tau rapidly dephosphorylate postmortem. However, they can be studied in aging macaques, who are all APOE-ε4 homozygotes and naturally develop cognitive deficits, calcium dysregulation, synapse loss, tau and amyloid pathology and autophagic degeneration, including elevated plasma pT217Tau, a new blood biomarker of incipient AD. High resolution nanoscale imaging of aging macaque brains reveals the earliest stages of soluble tau pathology and its relationships with Aβ42 and calcium signaling. These data indicate that inflammation erodes regulation of calcium signaling leading to the activation of calpain-2, which drives tau hyperphosphorylation, APP cleavage to Aβ42 and autophagic degeneration. These in turn propel further calcium dysregulation to drive vicious cycles. Restoring calcium dysregulation, e.g., with calpain-2 inhibitors, thus may be a rational strategy for slowing or preventing AD pathology. Recent data show that an agent that reduces GCPII inflammation and restores mGluR3 regulation of calcium reduced tau pathology in aged macaques, encouraging this approach. Targeting inflammation and dysregulated calcium may be especially helpful for patients who are APOE-ε4 carriers and insufficiently aided by current anti-amyloid antibody treatments.

Abstract Background The newly evolved dorsolateral prefrontal cortex (dlPFC) with pyramidal cell circuits in layer 3 are essential to higher cognition1,2. However, these pyramidal cells are uniquely susceptible in psychiatric disorders, where layer 3 is the focus of synapse dendritic atrophy in schizophrenia and degenerative disorders such as Alzheimer’ s disease (AD), correlating with the degree of cognitive deficits3-6. In particular, mutations in CACNA1C (L-type-calcium-channel Cav1.2) are consistently linked to increased risk of mental disorders but it is unknown why these channels are critical to cognition7-9, and whether they effect layer 3 pyramidal cells in dlPFC that are especially vulnerable in cognitive disorders. We examined the transcriptomic signatures of dlPFC layer 3 pyramidal cells, as well as additional studies of their connections and molecular regulation in macaque dlPFC, and found that they express magnified calcium signaling needed to sustain mental representations. However, higher levels, during chronic stress, caused a loss of firing and impaired cognition. Aims & Objectives The current study examined the molecular mechanisms expressed in layer III pyramidal cells in primate dlPFC. Methods The current study integrated single-nucleus RNA-sequencing in human and rhesus macaque dlPFC, multi-label immunofluorescence (MLIF), super-resolution immunoelectron microscopy (immunoEM), in vivo physiology coupled with iontophoresis of pharmacological compounds and behavior in rhesus macaque dlPFC to evaluate the calcium interactome (CACNA1C, GRIN2B, KCNN3). Results We showed that, in both human and macaque dlPFC, these cells have especially high expression of GRIN2B, CACNA1C and KCNN3, encoding the NMDAR-GluN2B and LTCC Cav1.2 channels that flux high levels of calcium into the neuron, and the SK3 potassium channel that whose open state is increased by calcium. Macaque studies showed that the CALB1-enriched pyramidal cells preferentially connect to the contralateral dlPFC, suggesting that increased calcium may be needed to maintain firing across the corpus callosum. ImmunoEM demonstrated that Cav1.2, SK3 channels and B1-AR are all concentrated on layer III dendritic spines, similar to NMDAR-GluN2B, with Cav1.2 on the plasma membrane near the calcium-storing spine apparatus, positioned to further increase calcium actions via internal release. Physiological recordings from cognitively engaged macaques showed that either inadequate or excessive LTCC actions, the latter driven by B1-AR stimulation, markedly reduced Delay cell firing needed for working memory via opening of SK potassium channels. Comparable effects were seen at the behavioral level, with stress-induced working memory impairment rescued by LTCC or B1-AR blockade. Discussion & Conclusion These data reveal a mechanism by which stress impairs dlPFC cognitive function, and suggests that either loss-of-function or gain-of-function mutations in CACNA1C would be harmful to dlPFC function and increase risk of neuropsychiatric disorders. As elevated calcium signaling is known to drive AD pathology, the data also explain why this subset of pyramidal cells with an enriched calcium interactome is especially susceptible to tau pathology and degeneration. The current findings are a rare example where transcriptomic and genomic data are related to the dysfunction of higher cortical circuits, illuminating how molecular insults give rise to symptoms of cognitive deficits. References 1Arnsten, A. F. T., Datta, D. &Wang, M. The genie in the bottle-magnified calcium signaling in dorsolateral prefrontal cortex. Mol Psychiatry 26, 3684-3700, doi:10.1038/s41380-020-00973-3 (2021). 2Goldman-Rakic, P. S. Cellular basis of working memory. Neuron 14, 477-485 (1995). 3Glausier, J. R. &Lewis, D. A. Dendritic spine pathology in schizophrenia. Neuroscience 251, 90-107, doi:10.1016/j.neuroscience.2012.04.044 (2013). 4Hoftman, G. D., Datta, D. &Lewis, D. A. Layer 3 Excitatory and Inhibitory Circuitry in the Prefrontal Cortex: Developmental Trajectories and Alterations in Schizophrenia. Biol Psychiatry 81, 862-873, doi:10.1016/j.biopsych.2016.05.022 (2017). 5Kahn, R. S. &Keefe, R. S. Schizophrenia is a cognitive illness: time for a change in focus. JAMA Psychiatry 70, 1107-1112, doi:10.1001/jamapsychiatry.2013.155 (2013). 6Perlstein, W. M., Carter, C. S., Noll, D. C. &Cohen, J. D. Relation of prefrontal cortex dysfunction to working memory and symptoms in schizophrenia. Am J Psychiatry 158, 1105-1113 (2001). 7Bigos, K. L. et al. Genetic variation in CACNA1C affects brain circuitries related to mental illness. Arch Gen Psychiatry 67, 939-945, doi:10.1001/archgenpsychiatry.2010.96 (2010). 8Sekar, A. et al. Schizophrenia risk from complex variation of complement component 4. Nature 530, 177-183, doi:10.1038/nature16549 (2016). 9Trubetskoy, V. et al. Mapping genomic loci implicates genes and synaptic biology in schizophrenia. Nature 604, 502-508, doi:10.1038/s41586-022-04434-5 (2022).

In-context learning (ICL) has revolutionized natural language processing by enabling models to adapt to diverse tasks with only a few illustrative examples. However, the exploration of ICL within the field of computer vision remains limited. Inspired by Chain-of-Thought (CoT) prompting in the language domain, we propose Chain-of-Focus (CoF) Prompting, which enhances vision models by enabling step-by-step visual comprehension. CoF Prompting addresses the challenges of absent logical structure in visual data by generating intermediate reasoning steps through visual saliency. Moreover, it provides a solution for creating tailored prompts from visual inputs by selecting contextually informative prompts based on query similarity and target richness. The significance of CoF prompting is demonstrated by the recent introduction of Large Autoregressive Vision Models (LAVMs), which predict downstream targets via in-context learning with pure visual inputs. By integrating intermediate reasoning steps into visual prompts and effectively selecting the informative ones, the LAVMs are capable of generating significantly better inferences. Extensive experiments on downstream visual understanding tasks validate the effectiveness of our proposed method for visual in-context learning.

NR2A and NR2B are the major GluR2 subunits of N-methyl-D-aspartate (NMDA) receptor. NR2B-containing NMDA receptor was found to be exclusively expressed in post-synapses in layer III pyramidal cells of the prefrontal cortex (PFC). Many studies have suggested the importance of PFC NR2B-containing NMDA receptor for working memory, especially for the persistent delay cell firing. However, direct evidence for the necessity of PFC NR2B-containing NMDA receptor on working memory is still absent, especially in non-human primates. Here, we directly evaluated the necessity of PFC synaptic NR2B for working memory in both rats and monkeys. We first examined the synaptosome expression ratio of NR2B/2A in the PFC, hippocampus and visual cortex, and confirmed a relatively higher expression ratio of NR2B/2A in the PFC than in the hippocampus and visual cortex in both species. We then investigated the effects of intra-PFC blockade of NR2B on the performance of spatial working memory and pattern discrimination, and found that the spatial working memory, but not pattern discrimination, was robustly impaired in a delay length-dependent way in both species. The present study provided behavioral and neuropharmacological evidence for the critical role of PFC NR2B-containing receptors in working memory performance in non-human primates.

Importance:The risk of mental disorders is consistently associated with variants in CACNA1C (L-type calcium channel Cav1.2) but it is not known why these channels are critical to cognition, and whether they affect the layer III pyramidal cells in the dorsolateral prefrontal cortex that are especially vulnerable in cognitive disorders. Objective:To examine the molecular mechanisms expressed in layer III pyramidal cells in primate dorsolateral prefrontal cortices. Design, Setting, and Participants:The design included transcriptomic analyses from human and macaque dorsolateral prefrontal cortex, and connectivity, protein expression, physiology, and cognitive behavior in macaques. The research was performed in academic laboratories at Yale, Harvard, Princeton, and the University of Pittsburgh. As dorsolateral prefrontal cortex only exists in primates, the work evaluated humans and macaques. Main Outcomes and Measures:Outcome measures included transcriptomic signatures of human and macaque pyramidal cells, protein expression and interactions in layer III macaque pyramidal cells using light and electron microscopy, changes in neuronal firing during spatial working memory, and working memory performance following pharmacological treatments. Results:Layer III pyramidal cells in dorsolateral prefrontal cortex coexpress a constellation of calcium-related proteins, delineated by CALB1 (calbindin), and high levels of CACNA1C (Cav1.2), GRIN2B (NMDA receptor GluN2B), and KCNN3 (SK3 potassium channel), concentrated in dendritic spines near the calcium-storing smooth endoplasmic reticulum. L-type calcium channels influenced neuronal firing needed for working memory, where either blockade or increased drive by β1-adrenoceptors, reduced neuronal firing by a mean (SD) 37.3% (5.5%) or 40% (6.3%), respectively, the latter via SK potassium channel opening. An L-type calcium channel blocker or β1-adrenoceptor antagonist protected working memory from stress. Conclusions and Relevance:The layer III pyramidal cells in the dorsolateral prefrontal cortex especially vulnerable in cognitive disorders differentially express calbindin and a constellation of calcium-related proteins including L-type calcium channels Cav1.2 (CACNA1C), GluN2B-NMDA receptors (GRIN2B), and SK3 potassium channels (KCNN3), which influence memory-related neuronal firing. The finding that either inadequate or excessive L-type calcium channel activation reduced neuronal firing explains why either loss- or gain-of-function variants in CACNA1C were associated with increased risk of cognitive disorders. The selective expression of calbindin in these pyramidal cells highlights the importance of regulatory mechanisms in neurons with high calcium signaling, consistent with Alzheimer tau pathology emerging when calbindin is lost with age and/or inflammation.

Sustained cognitive deficits are a common and debilitating feature of “long COVID”, but currently there are no FDA-approved treatments. The cognitive functions of the dorsolateral prefrontal cortex (dlPFC) are the most consistently afflicted by long COVID, including deficits in working memory, motivation, and executive functioning. COVID-19 infection greatly increases kynurenic acid (KYNA) and glutamate carboxypeptidase II (GCPII) in brain, both of which can be particularly deleterious to PFC function. KYNA blocks both NMDA and nicotinic-alpha-7 receptors, the two receptors required for dlPFC neurotransmission, and GCPII reduces mGluR3 regulation of cAMP-calcium-potassium channel signaling, which weakens dlPFC network connectivity and reduces dlPFC neuronal firing. Two agents approved for other indications may be helpful in restoring dlPFC physiology: the antioxidant N-acetyl cysteine inhibits the production of KYNA, and the α2A-adrenoceptor agonist guanfacine regulates cAMP-calcium-potassium channel signaling in dlPFC and is also anti-inflammatory. Thus, these agents may be helpful in treating the cognitive symptoms of long COVID.

Genetic variations in L-type-calcium-channels (LTCC) Cav1.2 (CACNA1C) are frequently associated with stress-related cognitive disorders. Here we examined Cav1.2-LTCC actions in macaque dorsolateral prefrontal cortical (dlPFC) NMDAR-GluN2B circuits that mediate cognition. LTCCs were essential for working memory-related neuronal firing. However, high LTCC calcium influx, driven by β1-adrenoceptors (β1-ARs) -similar to the stress response in heart- reduced firing via SK channel opening. Likewise, stress-induced cognitive deficits were blocked by β1-AR or LTCC antagonists. A calcium-enriched transcriptome was also identified in subsets of human dlPFC pyramidal cells with high CACNA1C, co-expressing GRIN2B (NMDAR-GluN2B), KCNN3 (SK3-channels), ADRB1 (β1-AR), and uniquely co-expressing CALB1 (calbindin), a marker of future vulnerability in Alzheimer’s disease (AD) when calbindin is lost with age. This pattern was recapitulated in macaques, including protein expression in layer III spines, with CALB1-expressing cells projecting to contralateral dlPFC. Thus we’ve identified a subgroup of calcium-enriched neurons essential to cognition, but especially susceptible to disease.Funding Information: This research was primarily funded by NSF 2015276 to AFTA and SAM, with additional funding from 2R37DA023999-12 (PR).Declaration of Interests: The authors have nothing to declare.Ethics Approval Statement: All research was conducted according to NIH guidelines and approved by the Yale sC (macaque) and the Partners Human Research Committee (humans).

Glutamate carboxypeptidase-II (GCPII) expression in brain is increased by inflammation, e.g. by COVID19 infection, where it reduces NAAG stimulation of metabotropic glutamate receptor type 3 (mGluR3). GCPII-mGluR3 signaling is increasingly linked to higher cognition, as genetic alterations that weaken mGluR3 or increase GCPII signaling are associated with impaired cognition in humans. Recent evidence from macaque dorsolateral prefrontal cortex (dlPFC) shows that mGluR3 are expressed on dendritic spines, where they regulate cAMP-PKA opening of potassium (K+) channels to enhance neuronal firing during working memory. However, little is known about GCPII expression and function in the primate dlPFC, despite its relevance to inflammatory disorders. The present study used multiple label immunofluorescence and immunoelectron microscopy to localize GCPII in aging macaque dlPFC, and examined the effects of GCPII inhibition on dlPFC neuronal physiology and working memory function. GCPII was observed in astrocytes as expected, but also on neurons, including extensive expression in dendritic spines. Recordings in dlPFC from aged monkeys performing a working memory task found that iontophoresis of the GCPII inhibitors 2-MPPA or 2-PMPA markedly increased working memory-related neuronal firing and spatial tuning, enhancing neural representations. These beneficial effects were reversed by an mGluR2/3 antagonist, or by a cAMP-PKA activator, consistent with mGluR3 inhibition of cAMP-PKA-K+ channel signaling. Systemic administration of the brain penetrant inhibitor, 2-MPPA, significantly improved working memory performance without apparent side effects, with largest effects in the oldest monkeys. Taken together, these data endorse GCPII inhibition as a potential strategy for treating cognitive disorders associated with aging and/or neuroinflammation.

Back to table of contents Previous article Next article EditorialsFull AccessThe Evolutionary Expansion of mGluR3-NAAG-GCPII Signaling: Relevance to Human Intelligence and Cognitive DisordersAmy F.T. Arnsten, Ph.D., Min Wang, Ph.D.Amy F.T. Arnsten, Ph.D., Min Wang, Ph.D.Published Online:1 Dec 2020https://doi.org/10.1176/appi.ajp.2020.20101458AboutSectionsView articleView PDFView EPUB ToolsAdd to favoritesDownload CitationsTrack Citations ShareShare onFacebookTwitterLinked InEmail View articleIn this issue of the Journal, Zink et al. (1) report remarkable new findings relating variance in the folate hydrolase 1 (FOLH1) gene encoding for glutamate carboxypeptidase II (GCPII) to human intelligence. Although there have been many genetic insults that have been linked to severe intellectual disability, it is rare to find a genetic variant that is related to IQ within the normal range. The authors found that a missense mutation in FOLH1 that leads to greater expression of GCPII in both healthy subjects and patients with schizophrenia is associated with reduced levels of its substrate, N-acetylaspartylglutamate (NAAG), and with lower IQ. Consistent with lower intelligence, the authors found inefficient processing by the dorsolateral prefrontal cortex (DLPFC), a newly evolved cortical region that subserves working memory, abstract reasoning, and executive functioning.NAAG is a highly prevalent (yet underappreciated) neurotransmitter that is coreleased with glutamate (Figure 1). Importantly, NAAG selectively stimulates metabotropic glutamate receptor type 3 (mGluR3, encoded by the GRM3 gene). Genetic alterations in GRM3 are consistently linked to increased risk of schizophrenia by genome-wide association studies (2), further emphasizing the importance of mGluR3-GCPII signaling to human cognition and cognitive disorders. However, the links to higher cognition have been perplexing to many scientists, as mGluR3s are classically viewed as glial receptors.FIGURE 1. Classic view of metabotropic glutamate receptor type 3 (mGluR3) signaling in neurons based on rodent studiesaPrevious research on mGluR3s has emphasized their localization presynaptically on glutamate axon terminals where they inhibit glutamate release. mGluR3s are stimulated by both glutamate and N-acetylaspartylglutamate (NAAG), which is coreleased with glutamate, but is selective for mGluR3. NAAG is catabolized by glutamate carboxypeptidase II (GCPII). mGluR3s are also localized on astrocytes, where they enhance glutamate uptake from the synapse through excitatory amino acid transporters. Thus, their classic role is to reduce glutamate signaling through both presynaptic and glial actions, which would result in reduced neuronal firing.mGluR3s are localized on astrocytes, where they enhance glutamate uptake through excitatory amino acid transporters (reviewed by Neale et al. [3]). Studies in rodents show that they can also play a related role in classic neuronal circuits, where they often reside on presynaptic terminals and reduce glutamate release (3) (Figure 1). Thus, mGluR3s traditionally have been seen as providing negative feedback on glutamate signaling and can be protective against excitotoxicity in rodent models (4). However, recent data indicate that their role in some neurons has changed and expanded with cortical evolution.Unique Role of NAAG-mGluR3 Signaling in the Newly Evolved Cortical Circuits Subserving Higher CognitionStudies of the rhesus monkey DLPFC have discovered that mGluR3 and GCPII have a novel role in higher cortical circuits in primates—strengthening the connectivity of layer 3 DLPFC circuits that mediate working memory—helping to explain their genetic links to cognition and cognitive disorders.The prefrontal cortex (PFC) expanded greatly over brain evolution, especially the DLPFC and frontal pole regions that do not exist in rodents, and greatly increases in size across nonhumans to human primates. Research by Goldman-Rakic and Gonzalez-Burgos has shown that the primate DLPFC contains extensive recurrent excitatory connections in deep layer 3 (Figure 2A), where pyramidal cells excite each other through N-methyl-d-aspartate receptor (NMDAR) synapses (Figure 2B) to keep information "in mind" (reviewed in [5]). The information is refined by lateral inhibition from GABA interneurons to generate a precise representation. In this way, layer 3 microcircuits are able to maintain representations of information across a delay period without the need for sensory stimulation, the foundation of abstract thought. These NMDAR synapses are powerfully and uniquely modulated by the arousal systems (e.g., where exposure to an uncontrollable stressor can weaken synaptic connectivity by driving high levels of calcium-cyclic AMP-dependent protein kinase [cAMP-PKA] signaling), which in turn open potassium (K+) channels near the synapse to decrease recurrent network excitation and reduce neuronal firing (Figure 2B) (reviewed by Arnsten et al. [5]). Recent data suggest that mGluR3 plays a major role in regulating these signaling events in primate DLPFC, inhibiting cAMP-PKA-K+ signaling to enhance synaptic strength and persistent neuronal firing.FIGURE 2. Recently discovered role of metabotropic glutamate receptor type 3 (mGluR3) in newly evolved microcircuits in primate layer 3 dorsolateral prefrontal cortex (DLPFC) that subserve working memoryaa Deep layer 3 pyramidal cell microcircuits in primate DLPFC have extensive recurrent excitation (panel A), where pyramidal cells excite each other through N-methyl-d-aspartate receptor (NMDAR) synapses to keep information "in mind." In this example of spatial working memory, a group of pyramidal cells representing 80° excite each other to maintain the memory of this spatial position during a spatial working memory task. In layer 3 of primate DLPFC, mGluR3s are localized on spines (panel B), positioned near the spine apparatus (pink; the extension of the endoplasmic reticulum into the spine) to inhibit cyclic-AMP (cAMP) calcium signaling. Layer 3 DLPFC spines contain the molecular machinery for feed-forward cAMP-calcium signaling, where cAMP drives internal calcium release from spine apparatus, which in turn drives more cAMP production. High levels of calcium-cAMP protein kinase (PKA) signaling open nearby K+ channels, which weaken synaptic connectivity and reduce neuronal firing. mGluR3s are positioned to regulate cAMP-calcium feed-forward signaling, reducing K+ channel openings, and thus strengthening synaptic connectivity and neuronal firing. mGluR3s are engaged by both glutamate and N-acetylaspartylglutamate (NAAG), which is selective for mGluR3. Native FOLH1 leads to normal levels of glutamate carboxypeptidase II (GCPII) expression, where there is sufficient NAAG stimulation of mGluR3 to regulate cAMP-K+ signaling and support higher cognition. The study by Zink et al. (1) shows that a missense mutation in FOLH1 leads to greater GCPII expression, reduced NAAG levels, and lower intelligence (panel C). This genetic insult may impair cognition by reducing NAAG stimulation of mGluR3 on spines, weakening synaptic connectivity and reducing the neuronal firing needed for mental representations in higher cortical circuits. The effects of GCPII on the firing of a DLPFC Delay cell in an aged monkey performing a spatial working memory task are shown in panel D. Delay cells maintain representations of spatial information over a delay period in a spatial working memory task but have lower levels of firing in aged monkeys. This Delay cell is from an aged monkey (20 years) with relatively low levels of task-related firing. Direct iontophoretic application of the GCPII inhibitor, ZJ43, onto the Delay cell enhances its firing in a dose-related manner. Similar enhancing effects were seen with iontophoresis of NAAG. The enhancing effects of the GCPII inhibitor were reversed by increasing cAMP signaling (not shown), consistent with the signaling events represented in panels B and C. Panel E shows the dose-response effects of the GCPII inhibitor, ZJ43, on the average firing rate of DLPFC Delay cells across the delay period in aged monkeys performing a working memory task, where the greatest GCPII inhibition causes the largest increase in neuronal firing. Panels D and E are adapted with permission from Jin et al. (6).Immunoelectron microscopy (immunoEM) can reveal the subcellular location of proteins in brain. ImmunoEM of the rhesus monkey layer 3 DLPFC showed that mGluR3s are concentrated postsynaptically on spines (Figure 2B), completely different from their classic location on presynaptic terminals in rodent circuits. mGluR3s are also localized on astrocytes in primate DLPFC, but the presynaptic receptors on glutamate axon terminals are exclusively mGluR2 rather than mGluR3 (6). mGluR3s on spines can be found within the postsynaptic density but are mostly localized on the spine membrane near the calcium-containing spine apparatus, the extension of the smooth endoplasmic reticulum into the spine, which stores and releases calcium. As illustrated in Figure 2B, cAMP-PKA signaling drives calcium release from the spine apparatus, which in turn stimulates more cAMP production, thus creating a rapid rise in calcium-cAMP-PKA signaling. PKA signaling can also increase calcium flux through NMDAR and through voltage-gated calcium channels, such as Cav1.2, magnifying cytosolic calcium (reviewed by Arnsten et al. [5]). As mentioned above, high levels of calcium-cAMP-PKA signaling open nearby K+ channels to weaken connectivity and reduce neuronal firing. Conversely, inhibition of cAMP signaling closes K+ channels, strengthens connectivity, and enhances the persistent neuronal firing needed for working memory (5, 7). These beneficial actions have been seen with NAAG-mGluR3 signaling in primate DLPFC, where iontophoretic application of either NAAG or a GCPII inhibitor (Figure 2D and E) directly onto DLPFC neurons greatly enhances working memory-related neuronal firing by inhibiting cAMP-PKA-K+ channel signaling (6). This can be seen as a dose-response, where increasing amounts of GCPII activity lead to decreasing levels of DLPFC neuronal firing. Thus, NAAG-mGluR3 signaling strengthens the connectivity of higher cortical glutamatergic circuits and increases DLPFC neuronal firing in primates, opposite to the reduction in glutamate release classically associated with mGluR3 presynaptic actions in rodents.The new study by Zink et al. highlights the consequences to cognition when a genetic variation increases the expression of GCPII and reduces NAAG levels in brain. Consistent with a reduction in beneficial NAAG-mGluR3 actions, the authors found a positive correlation between NAAG levels measured by magnetic resonance spectroscopy and cognitive performance. The study also found that carriers of the missense FOLH1 variant had inefficient activation of DLPFC during working memory and lower IQ scores. Coupled with the findings in monkeys, the data suggest that the human subjects with the FOLH1 variant would have less NAAG stimulation of mGluR3, greater opening of K+ channels, and weaker DLPFC network connectivity needed for working memory and abstract reasoning (Figure 2C) and thus would have to recruit greater amounts of DLPFC to perform a cognitive task (i.e., inefficient action of the DLPFC as seen in the Zink et al. study). Because GCPII expression is also increased by inflammation, we can speculate that similar insults to cortical connectivity and cognitive function may occur during an infection, perhaps helping to explain why we often have "brain fog" during illness.Relevance to Cognitive DisordersThe results from human and monkey studies suggest that either genetic or environmental insults that increase GCPII and/or reduce NAAG-mGluR3 signaling can weaken higher cortical connections and increase risk of cognitive disorders. For example, there is an increased risk of schizophrenia with mutations to GRM3 (2), or with perinatal inflammation, which in animal studies, increases GCPII expression in the brains of offspring (8). Other genetic insults that increase risk of schizophrenia may create a similar phenotype of increased cAMP-calcium signaling in the DLPFC (e.g., duplication of VIPR2 [VPAC2] or loss of function translocation in DISC1 anchoring of phosphodiesterase 4 [PDE4] has been shown to increase cAMP signaling, while alterations in CACNA1C [Cav1.2] increase calcium flux [reviewed by Arnsten et al. [5]). While these signaling events are needed for normal function, excessive cAMP-calcium signaling in layer 3 DLPFC circuits is detrimental, opening large numbers of K+ channels to reduce firing and likely contributing to dendritic atrophy when sustained by chronic conditions (5).The increase in GCPII expression with inflammation may also contribute to cognitive deficits in aging and Alzheimer's disease (9). Advancing age is associated with loss of PDE4 (10) and mGluR3 (11) regulation of cAMP-calcium signaling in the aged PFC. Studies of aging rhesus monkeys have shown that this dysregulated cAMP-PKA-K+ signaling in the DLPFC contributes to age-related loss of persistent neuronal firing needed for working memory (12) and to phosphorylation of tau, which leads to neurofibrillary tangles (10). Elevated GCPII signaling also appears to contribute to cognitive deficits in multiple sclerosis, where hippocampal NAAG levels correlate with cognitive abilities (13). It is possible that elevated GCPII due to infectious conditions, such as COVID-19, may also contribute to the delirium and cognitive deficits that can accompany this disease.Relevance to Potential TreatmentsGCPII inhibitors are currently under development for the treatment of inflammatory and cognitive disorders (14). For example, they have shown success in animal models of neuropathic pain, Alzheimer's disease, and multiple sclerosis (14–16), and studies of aging monkeys are in progress. Assays of GCPII inhibitors in human and animal studies to date have shown that these compounds appear to have excellent side-effect profiles, encouraging their use and potential utility for disease prevention. However, current compounds have limited brain penetrance, and this will need to be overcome to be of practical benefit in cognitive disorders.In closing, the findings by Zink et al., coupled with the links between GRM3 and schizophrenia, serve as genetic "flashing lights," alerting us to the importance of a previously underappreciated signaling pathway important to human cognition. These genetic associations are further illuminated by the recent data from macaque DLPFC showing that NAAG-mGluR3 signaling strengthens network connectivity in the circuits that generate the mental representations underlying working memory and abstract reasoning. Taken together, they provide a rational story, where the role of GCPII-NAAG-mGluR3 signaling expands with cortical evolution to regulate the circuits underlying human intelligence.Department of Neuroscience, Yale University School of Medicine, New Haven Conn. (Arnsten, Wang).Address correspondence to Dr. Arnsten ([email protected]).Dr. Arnsten and Yale University receive royalties from the U.S. sales of Intuniv from Shire/Takeda Pharmaceuticals. Dr. Wang reports no financial relationships with commercial interests.Supported by NeuroNex NSF (grant 2015276) and PHS (NIH grant R01 AG061190-01). The original mGluR3 physiology and immunoEM data were funded by NIH grants R01AG043430-01A1 and RO1 MH100064-01A1.References1 Zink CF, Barker PB, Sawa A, et al.: Association of missense mutation in FOLH1 with decreased NAAG levels and impaired working memory circuitry and cognition. Am J Psychiatry 2020; 177:1129–1139Abstract, Google Scholar2 Saini SM, Mancuso SG, Mostaid MS, et al.: Meta-analysis supports GWAS-implicated link between GRM3 and schizophrenia risk. Transl Psychiatry 2017; 7:e1196Crossref, Medline, Google Scholar3 Neale JH, Olszewski RT, Zuo D, et al.: Advances in understanding the peptide neurotransmitter NAAG and appearance of a new member of the NAAG neuropeptide family. J Neurochem 2011; 118:490–498Crossref, Medline, Google Scholar4 Cao Y, Gao Y, Xu S, et al.: Glutamate carboxypeptidase II gene knockout attenuates oxidative stress and cortical apoptosis after traumatic brain injury. BMC Neurosci 2016; 17:15Crossref, Medline, Google Scholar5 Arnsten AFT, Datta D, Wang M: The Genie in the bottle: magnified calcium signaling in dorsolateral prefrontal cortex. Mol Psychiatry 2020 (in press)Google Scholar6 Jin LE, Wang M, Galvin VC, et al.: mGluR2 vs. mGluR3 in primate prefrontal cortex: postsynaptic mGluR3 strengthen cognitive networks. Cereb Cortex 2018; 28:974–987Crossref, Medline, Google Scholar7 Wang M, Ramos BP, Paspalas CD, et al.: Alpha2A-adrenoceptors strengthen working memory networks by inhibiting cAMP-HCN channel signaling in prefrontal cortex. Cell 2007; 129:397–410Crossref, Medline, Google Scholar8 Zhang Z, Bassam B, Thomas AG, et al.: Maternal inflammation leads to impaired glutamate homeostasis and up-regulation of glutamate carboxypeptidase II in activated microglia in the fetal/newborn rabbit brain. Neurobiol Dis 2016; 94:116–128Crossref, Medline, Google Scholar9 Passani LA, Vonsattel JP, Carter RE, et al.: N-acetylaspartylglutamate, N-acetylaspartate, and N-acetylated alpha-linked acidic dipeptidase in human brain and their alterations in Huntington and Alzheimer's diseases. Mol Chem Neuropathol 1997; 31:97–118Crossref, Medline, Google Scholar10 Carlyle BC, Nairn AC, Wang M, et al.: cAMP-PKA phosphorylation of tau confers risk for degeneration in aging association cortex. Proc Natl Acad Sci USA 2014; 111:5036–5041Crossref, Medline, Google Scholar11 Hernandez CM, McQuail JA, Schwabe MR, et al.: Age-related declines in prefrontal cortical expression of metabotropic glutamate receptors that support working memory. eNeuro 2018; 5:ENEURO.0164-18.2018Crossref, Google Scholar12 Wang M, Gamo NJ, Yang Y, et al.: Neuronal basis of age-related working memory decline. Nature 2011; 476:210–213Crossref, Medline, Google Scholar13 Rahn KA, Watkins CC, Alt J, et al.: Inhibition of glutamate carboxypeptidase II (GCPII) activity as a treatment for cognitive impairment in multiple sclerosis. Proc Natl Acad Sci USA 2012; 109:20101–20106Crossref, Medline, Google Scholar14 Vornov JJ, Hollinger KR, Jackson PF, et al.: Still NAAG'ing after all these years: the continuing pursuit of GCPII inhibitors. Adv Pharmacol 2016; 76:215–255Crossref, Medline, Google Scholar15 Olszewski RT, Janczura KJ, Bzdega T, et al.: NAAG peptidase inhibitors act via mGluR3: animal models of memory, Alzheimer's, and ethanol intoxication. Neurochem Res 2017; 42:2646–2657Crossref, Medline, Google Scholar16 Neale JH, Olszewski R: A role for N-acetylaspartylglutamate (NAAG) and mGluR3 in cognition. Neurobiol Learn Mem 2019; 158:9–13Crossref, Medline, Google Scholar FiguresReferencesCited byDetailsCited ByMolecules and Circuits Implicated in Schizophrenia Provide Leads for Novel TreatmentsNed H. Kalin, M.D.1 December 2020 | American Journal of Psychiatry, Vol. 177, No. 12 Volume 177Issue 12 December 01, 2020Pages 1103-1106 Metrics KeywordsCognitive NeuroscienceNeurocircuitry History Accepted 9 October 2020 Published online 1 December 2020 Published in print 1 December 2020

Neurons in the association cortices are particularly vulnerable in cognitive disorders such as schizophrenia and Alzheimer’s disease, while those in primary visual cortex remain relatively resilient. This review proposes that the special molecular mechanisms needed for higher cognitive operations confer vulnerability to dysfunction, atrophy, and neurodegeneration when regulation is lost due to genetic and/or environmental insults. Accumulating data suggest that higher cortical circuits rely on magnified levels of calcium (from NMDAR, calcium channels, and/or internal release from the smooth endoplasmic reticulum) near the postsynaptic density to promote the persistent firing needed to maintain, manipulate, and store information without “bottom-up” sensory stimulation. For example, dendritic spines in the primate dorsolateral prefrontal cortex (dlPFC) express the molecular machinery for feedforward, cAMP–PKA–calcium signaling. PKA can drive internal calcium release and promote calcium flow through NMDAR and calcium channels, while in turn, calcium activates adenylyl cyclases to produce more cAMP–PKA signaling. Excessive levels of cAMP–calcium signaling can have a number of detrimental effects: for example, opening nearby K + channels to weaken synaptic efficacy and reduce neuronal firing, and over a longer timeframe, driving calcium overload of mitochondria to induce inflammation and dendritic atrophy. Thus, calcium–cAMP signaling must be tightly regulated, e.g., by agents that catabolize cAMP or inhibit its production (PDE4, mGluR3), and by proteins that bind calcium in the cytosol (calbindin). Many genetic or inflammatory insults early in life weaken the regulation of calcium–cAMP signaling and are associated with increased risk of schizophrenia (e.g., GRM3 ). Age-related loss of regulatory proteins which result in elevated calcium–cAMP signaling over a long lifespan can additionally drive tau phosphorylation, amyloid pathology, and neurodegeneration, especially when protective calcium binding proteins are lost from the cytosol. Thus, the “genie” we need for our remarkable cognitive abilities may make us vulnerable to cognitive disorders when we lose essential regulation.

Rationale Metabotropic glutamate type 5 receptor (mGluR5) antagonists are under development for treating cognitive disorders such as Fragile X syndrome and Alzheimer’s disease, largely based on success in mouse models, where post-synaptic mGluR5 stimulation weakens synaptic functions in hippocampus. However, human trials of mGluR5 antagonists have yet to be successful. This may be due in part to the differing effects of mGluR5 in hippocampus vs. prefrontal cortex, as mGluR5 are primarily post-synaptic in rodent hippocampus, but are both pre- and post-synaptic in the dorsolateral prefrontal cortical (dlPFC) circuits known to subserve working memory. Objectives and methods The current study examined the effects of the selective mGluR5 negative allosteric modulator, MTEP (3-((2-Methyl-1,3-thiazol-4-yl)ethynyl)pyridine hydrochloride), on neuronal firing and working memory performance in aging rhesus monkeys with naturally occurring impairments in neuronal firing and cognitive performance. Results We found that iontophoresis of MTEP directly onto dlPFC “Delay cells” had an inverted U dose-response, where low doses tended to enhance task-related firing, but higher doses suppressed neuronal firing. Similar effects were seen on cognitive performance following systemic MTEP administration (0.0001–0.1 mg/kg), with MTEP producing erratic dose-response curves. In the subset of monkeys (50%) that showed replicable improvement with MTEP, co-administration with the mGluR5 PAM, CDPPB (3-Cyano- N -(1,3-diphenyl-1 H -pyrazol-5-yl)benzamide), blocked MTEP beneficial effects, consistent with mGluR5 actions. Conclusions The mixed effects of MTEP on cognitive performance may arise from opposing actions at pre- vs. post-synaptic mGluR5 in dlPFC. These data from monkeys suggest that future clinical trials should include low doses, and identification of potential subgroup responders.

AbstractBackgroundEarly onset forms of Alzheimer’s disease (AD) are driven by rare dominantly inherited genes; however, the predominant form of the disease is sporadic and occurs later in life through an unknown etiology. Sporadic AD emerges after age 60‐65 and preferentially afflicts the association cortex. Thus, a better understanding of how age‐related changes in vulnerable brain regions relate to AD pathology is crucial to developing novel treatments for the disease. AD is characterized by two main pathological hallmarks, tau neurofibrillary tangles and amyloid‐beta plaques. Tau pathology correlates with cognitive decline while amyloid pathology does not. Thus, we focused on the role of age‐related signaling changes on tau pathology. Dysregulated calcium signaling has long been hypothesized to play a major role in age‐related vulnerability to AD1. Thus, we investigated the impact of aging on calcium signaling in the vulnerable dorsolateral prefrontal cortex (dlPFC) of rhesus monkeys, focusing on regulation of and by cAMP‐PKA signaling in relationship to tau pathology. Monkeys provide a unique opportunity to study higher order cognitive circuits unique to primates while still being able to probe molecular networks to a degree that is not possible in human post‐mortem samples. Overall, we found a significant age‐related dysregulation of calcium‐cAMP‐PKA signaling that strongly associated with early stage tau pathology in rhesus monkey dlPFC. (1) Khachaturian, Z. S. C Alzheimer’s & Dementia 13, 178‐182.e17 (2017)MethodPatterns of molecular signaling were assessed via western blot and immuno‐electron microscopy of monkey dlPFC in animals ranging from 8 to 28 years of age. Mechanistic studies of the relationship between intracellular calcium and tau phosphorylation were performed in rat primary cortical cultures.ResultWe observed a consistent pattern of calcium‐cAMP‐PKA dysregulation in the vulnerable dlPFC region of rhesus monkeys. We found a highly significant association between leaky ryanodine receptor calcium channels and early stage tau phosphorylation (pS214 and pS356). The link between these phosphorylation sites and intracellular calcium was validated through in vitro mechanistic studies.ConclusionAge‐related dysregulation of calcium‐cAMP in the dlPFC is associated with early stage tau pathology and may explain the vulnerability of these circuits to degeneration.

Although mouse models of Alzheimer's disease (AD) have provided tremendous breakthroughs, the etiology of later onset AD remains unknown. In particular, tau pathology in the association cortex is poorly replicated in mouse models. Aging rhesus monkeys naturally develop cognitive deficits, amyloid plaques, and the same qualitative pattern and sequence of tau pathology as humans, with tangles in the oldest animals. Thus, aging rhesus monkeys can play a key role in AD research. For example, aging monkeys can help reveal how synapses in the prefrontal association cortex are uniquely regulated compared to the primary sensory cortex in ways that render them vulnerable to calcium dysregulation and tau phosphorylation, resulting in the selective localization of tau pathology observed in AD. The ability to assay early tau phosphorylation states and perform high-quality immunoelectron microscopy in monkeys is a great advantage, as one can capture early-stage degeneration as it naturally occurs in situ. Our immunoelectron microscopy studies show that phosphorylated tau can induce an "endosomal traffic jam" that drives amyloid precursor protein cleavage to amyloid-β in endosomes. As amyloid-β increases tau phosphorylation, this creates a vicious cycle where varied precipitating factors all lead to a similar phenotype. These data may help explain why circuits with aggressive tau pathology (e.g., entorhinal cortex) may degenerate prior to producing significant amyloid pathology. Aging monkeys therefore can play an important role in identifying and testing potential therapeutics to protect the association cortex, including preventive therapies that are challenging to test in humans.

Persistent activity generated in the PFC during the delay period of working memory tasks represents information about stimuli held in memory and determines working memory performance. Alternative models of working memory, depending on the rhythmicity of discharges or exclusively on short-term synaptic plasticity, are inconsistent with the neurophysiological data.Dual Perspectives Companion Paper:Working Memory: Delay Activity, Yes! Persistent Activity? Maybe Not, by Mikael Lundqvist, Pawel Herman, and Earl K. Miller.

The newly evolved circuits in layer III of primate dorsolateral prefrontal cortex (dlPFC) generate the neural representations that subserve working memory. These circuits are weakened by increased cAMP-K+ channel signaling, and are a focus of pathology in schizophrenia, aging, and Alzheimer's disease. Cognitive deficits in these disorders are increasingly associated with insults to mGluR3 metabotropic glutamate receptors, while reductions in mGluR2 appear protective. This has been perplexing, as mGluR3 has been considered glial receptors, and mGluR2 and mGluR3 have been thought to have similar functions, reducing glutamate transmission. We have discovered that, in addition to their astrocytic expression, mGluR3 is concentrated postsynaptically in spine synapses of layer III dlPFC, positioned to strengthen connectivity by inhibiting postsynaptic cAMP-K+ channel actions. In contrast, mGluR2 is principally presynaptic as expected, with only a minor postsynaptic component. Functionally, increase in the endogenous mGluR3 agonist, N-acetylaspartylglutamate, markedly enhanced dlPFC Delay cell firing during a working memory task via inhibition of cAMP signaling, while the mGluR2 positive allosteric modulator, BINA, produced an inverted-U dose-response on dlPFC Delay cell firing and working memory performance. These data illuminate why insults to mGluR3 would erode cognitive abilities, and support mGluR3 as a novel therapeutic target for higher cognitive disorders.