Characterizing the Most Vulnerable Prefrontal Cortical Neurons in Schizophrenia
Publication: American Journal of Psychiatry
Understanding the neurobiological underpinnings of schizophrenia remains a daunting challenge—what is happening in our highest-order cortical circuits to lead to such distortions in abstract thinking? Symptoms of thought disorder are highly correlated with reduced functioning of the dorsolateral prefrontal cortex (DLPFC), a recently evolved brain region that subserves working memory, abstract reasoning, and executive functions and that exists only in primates (1). A major clue has been known for almost three decades: patients with schizophrenia have reduced numbers of spines and dendrites on the pyramidal cells in deep layer III of the DLPFC (2–4). Deep layer III expands greatly in primate evolution and contains the recurrent excitatory microcircuits that interconnect on dendritic spines to generate the mental representations that are foundational to working memory and abstract thought (1). These dendritic alterations in schizophrenia are highly specific to region, cell type, and lamina, as they are not seen in DLPFC layers V or VI, or even in superficial layer III, and they are not seen in the primary visual cortex (2–4). The neuropathological data are consonant with the more gross-level in vivo brain imaging of individuals at risk for schizophrenia, which shows a wave of gray matter loss during adolescence in subjects at risk as symptoms emerge (5) and a loss of synapses early in the course of disease (6). Altogether, these data have suggested excessive, erroneous pruning of neuronal connections, with early and pronounced loss in the DLPFC but progressing to envelop multiple association cortices (5). What are these cells that lose so many spines and connections?
In this issue, Arion et al. (7) try to capture the identity of these most vulnerable pyramidal cells by characterizing transcriptional changes between prepuberty, postpuberty, and young adulthood in macaque DLPFC. It had previously been shown that there are two largely nonoverlapping populations of pyramidal cells in layer III: those that project to the contralateral DLPFC (callosal projection cells, or CP cells) and those that project ipsilaterally to the parietal cortex (ipsilateral projection cells, or IP cells). Arion et al. have now performed a remarkably challenging study—combining tract tracing with laser capture microdissection and transcriptomics—and have discovered that the IP cells reach a mature state early in life, whereas the CP cells mature later, during the transition from adolescence to young adulthood, the age when schizophrenia symptoms typically begin. These data suggest that it is the CP cells that may be especially vulnerable to genetic or environmental events that could cause dendritic spine loss during adolescence.
What is different about the CP cells compared to the IP cells transcriptionally? Arion et al. found that the CP cells could be characterized by a number of differences from the IP cells; for example, they had low levels of SCN1B, which encodes a subunit of a voltage-gated sodium channel, whereas the IP cells had high levels of SCN1B. Most intriguingly, Arion et al. show that the CP cells express much higher levels of ADCYAP1, which encodes the “master stress peptide,” pituitary adenylate cyclase-activating polypeptide (PACAP). PACAP is highly expressed in circuits that orchestrate the stress response, and it has a critical role in initiating stress actions (8). As described below, chronic stress causes loss of spine and dendrites from prefrontal cortex (PFC) pyramidal cells (9), and thus an enriched “stress transcriptome” would be of particular interest when trying to understand why a specific layer of pyramidal cells is so vulnerable to spine loss in schizophrenia. Furthermore, PACAP and its receptors have genetic links to schizophrenia (10). Thus, it is especially interesting that CP pyramidal cells have uniquely high levels of this transcript in the primate DLPFC.
Stress, Schizophrenia, and the DLPFC
Stress exposure has long been known to be an aggravating factor in the descent into illness, including the alterations that underlie thought disorder in schizophrenia (1). In rodents, chronic stress causes loss of dendrites and spines from pyramidal cells in the medial PFC and is correlated with impaired cognition, and in humans is associated with reduced PFC gray matter (9). In patients, the wave of gray matter loss at the onset of schizophrenia is related to stress and inflammation (5), suggesting that these factors play a role in the structural changes that drive pathology. Thus, understanding how stress impairs PFC function and leads to dendritic spine loss may provide important clues about the etiology of the illness.
We have shown that uncontrollable stress rapidly impairs DLPFC function, switching control of behavior to more primitive circuits—for example, the amygdala—via high levels of dopamine and norepinephrine release in the DLPFC (1). High levels of dopamine and norepinephrine stimulate dopamine D1 receptors (D1R), β1-adrenoceptors (β1-AR), and alpha-1 adrenergic receptors (α1-AR) on layer III dendritic spines that drive feedforward cAMP-calcium signaling, which in turn opens nearby potassium channels to rapidly weaken synaptic efficacy and reduce PFC neuronal firing (1). For example, recent research shows that high levels of norepinephrine activate β1-AR on dendritic spines, which drives PKA opening of nearby Cav1.2 L-type calcium channels, which in turn opens SK potassium channels to reduce firing and impair working memory (11). This process may be dysregulated by inflammation, prolonging detrimental actions (1).
In parallel studies, transcriptomic analyses of bulk-tissue macaque and human DLPFC found that superficial (CUX2-expressing) pyramidal cells express high levels of calcium-related signaling molecules that are also linked to genetic risk for schizophrenia and other mental illnesses (11). These pyramidal cells have high levels of CALB1, encoding the calcium binding protein, calbindin, indicative of high calcium usage; high levels of GRIN2B encoding the NMDAR-GluN2B subunits that flux high levels of calcium and are essential to DLPFC neurotransmission; and high levels of CACNA1C, encoding the L-type voltage-gated calcium channel Cav1.2 (11). Genetic alterations in CACNA1C are a common risk factor for mental illness, including gain-of-function mutations related to schizophrenia (12). Interestingly, these same superficial pyramidal cells also have exceptionally high expression of CHP1, which encodes an endogenous inhibitor of the phosphatase calcineurin, and calcineurin inhibitors can induce psychosis (13). However, it is not known which of these superficial pyramidal cells is the subgroup identified by Arion et al. as the CP cells that are malleable in adolescence. Although Arion et al. performed laser capture microdissection of layer III pyramidal cells of young animals (3.2–4.7 years), and our study employed more general single-nucleus RNA sequencing of older animals (6.6–8.2 years), might we try to bridge across studies by identifying a subgroup of macaque DLPFC CUX2 (superficial layer) pyramidal cells with exceptionally high ADCYAP1 (PACAP), and very low SCN1B?
“Molecular Fingerprint” of CP and CUX2B Pyramidal Cells
Our analyses identified three subgroups of CUX2 (superficial layer) pyramidal cells: CUX2A, CUX2B, and CUX2C (11). Of these, subgroup CUX2B expresses the highest ADCYAP1 (PACAP) and the lowest SCN1B, suggesting that at least some of the CUX2B subgroup may be some of the CP cells that project across the corpus callosum (Table 1). While a rigorous bridge across studies would require quantitative comparisons of a very large number of genes, a focus on a few transcripts related to DLPFC neurotransmission and stress indicates interesting characteristics of the CUX2B cells. Thus, in addition to their very high expression of ADCYAP1, they also have the highest levels of CALB1, indicative of high calcium use, and GRIN2B, which encodes NMDAR-GluN2B, needed for neurotransmission. This elevated level of calcium may be needed to sustain recurrent excitatory firing across the hemispheres. Perhaps most relevant to neuronal vulnerability, CUX2B pyramidal cells express the highest levels of KCNN3, which encode the SK3 potassium channels that weaken connectivity and reduce neuronal firing with high levels of calcium (Table 1).
| Gene | Protein | CUX2A | CUX2B | CUX2C |
|---|---|---|---|---|
| ADCYAP1 | PACAP | 0.998 | 1.374 | 0.637 |
| SCN1B | Na+ voltage-gated channel | 1.151 | 0.267 | 1.565 |
| CALB1 | Calbindin | 0.976 | 1.468 | 0.676 |
| GRIN2B | NMDAR-GluN2B | 72.854 | 79.838 | 63.193 |
| KCNN3 | SK3 potassium channel | 6.626 | 9.5044 | 4.173 |
| ADRB1 | Norepinephrine β1 adrenoceptor | 1.666 | 1.466 | 1.347 |
| ADCY1 | Adenylyl cyclase 1 | 19.416 | 27.758 | 18.192 |
| PRKACB | PKA catalytic subunit β | 27.785 | 33.832 | 29.140 |
| AKAP5 | A-kinase anchoring protein | 0.720 | 2.128 | 0.222 |
| PDE4A | Phosphodiesterase 4A | 4.663 | 4.749 | 8.266 |
| PDE4B | Phosphodiesterase 4B | 56.163 | 28.124 | 38.150 |
| PDE4D | Phosphodiesterase 4D | 193.844 | 166.759 | 152.161 |
| DISC1 | DISC1 (anchoring protein) | 1.809 | 1.379 | 1.775 |
| DRD1 | Dopamine D1 receptor | 0.441 | 0.924 | 0.030 |
| TIAM1 | TIAM Rac1 associated GEF1 | 2.320 | 3.695 | 1.805 |
| RAC1 | Rac family small GTPase 1 | 0.519 | 0.612 | 0.492 |
a
Expression values are log10 normalized counts of the number of transcripts per 100,000 transcripts in each cell type.
The CUX2B subgroup also has higher levels of stress-related signaling molecules (Table 1), including ADRB1, which encodes the β1-adrenoceptor (β1-AR) that drives cAMP-PKA and Cav1.2 signaling to reduce firing during stress (11); ADCY1, which encodes adenyl cyclase 1 (AC1), the major synthetic enzyme for cAMP; PRKACB, which encodes a PKA catalytic subunit; and much higher levels of AKAP5, which anchors PKA, β-AR, and Cav1.2 together in the plasma membrane to drive calcium currents (14). Conversely, CUX2B pyramidal cells have lower levels of the phosphodiesterases that catabolize cAMP (Table 1; lower PDE4A, PDE4B, and DISC1, which anchors PDE4s to the calcium-storing smooth endoplasmic reticulum). An earlier study by Arion et al. (15), of just the 3-year-old animals, found similar results, with CP cells enriched in cAMP-calcium signaling, including the PKA subunit PRKACB, and the PKA anchoring protein AKAP5, suggesting that at least a substantial subset of CUX2B cells are the CP neurons. Interestingly, the CUX2B cell group also expresses the highest levels of DRD1, which encodes the dopamine D1 receptor known to mediate the stress response (16). It is noteworthy that the dopamine innervation of macaque DLPFC layer III is increased in adolescence (17), which may increase the stress response in these D1R-enriched pyramidal cells. Patients in the early stages of schizophrenia have increased expression of D1R in the DLPFC (18), suggesting that an even larger stress response may occur in patients at this early stage.
The very high expression of potassium channels in layer III spines, and especially the very high expression of SK3 channels in the CUX2B (putative CP) pyramidal cells, may be a fundamental factor in why these cells may “overprune” dendritic spines. While very high levels of cAMP-calcium signaling can strengthen connections in many neurons—for example, in the hippocampus—the very high expression of KCNN3 (SK potassium channels) in the CUX2B subgroup switches cAMP-calcium signals from strengthening to weakening connectivity, which may drive spine loss when sustained for long periods of time.
Arion et al. speculate that PACAP may be involved in driving spine loss, as it can increase RAC1 and TIAM1 signaling to alter the actin skeleton and cause spine collapse, signaling pathways that are altered in layer III DLPFC pyramidal cells in patients with schizophrenia (19). The CUX2B cells have higher levels of TIAM1 and RAC1 compared to other CUX2 (superficial) pyramidal cells (Table 1), consistent with this hypothesis. Rac1 alteration of the actin cytoskeleton depends on activating PKCα (20), and in rodents PKC inhibition prevents stress-induced spine loss in medial PFC and working memory impairment (21). In this regard, it is noteworthy that all atypical antipsychotic and antimanic medications inhibit PKC (22), for example, by blocking 5-HT2A and α1-A receptors, or through more direct effects on IP3R-related calcium signaling. Thus, these agents may have some of their therapeutic actions by protecting PFC dendritic spines from toxic stress signaling.
These new data from macaque DLPFC suggest that the CP pyramidal cells, which develop later than their neighbors, may have a “stress-enriched” molecular fingerprint that makes them most vulnerable to atrophy, including the higher expression of ADCYAP1 (PACAP). It is noteworthy that PACAP acts at three receptors that increase cAMP signaling: PAC1 (ADCYAP1R1), VPAC1 (VIPR1), and VPAC2 (VIPR2) receptors. VPAC2 is of special interest, as duplication of its gene, VIPR2, is a rare risk factor of very large effect for schizophrenia (23), suggesting that elevated stress signaling in DLPFC may be particularly impactful for mental disorders.
Transcriptomic signatures provide a bridge across studies and species, where additional tract tracing experiments in macaque begin to illuminate the molecular and connectional properties of the cells most vulnerable in schizophrenia, and how they may vary across developmental stages and may be differentially affected by stress, inflammation, and genetic alterations over the lifespan. The discovery by Arion et al. that CP cells are the latest to mature, during a developmental period when schizophrenia onset is most common, suggests that these cells may be a particular target of pathology, with a molecular fingerprint that renders them especially vulnerable to stress.
Acknowledgments
The authors appreciate the invaluable help of Dr. Fenna Krienen.
References
1.
Arnsten AFT, Woo E, Yang S, et al: Unusual molecular regulation of dorsolateral prefrontal cortex layer III synapses increases vulnerability to genetic and environmental insults in schizophrenia. Biol Psychiatry 2022; 92:480–490
2.
Selemon LD, Goldman-Rakic PS: The reduced neuropil hypothesis: a circuit based model of schizophrenia. Biol Psychiatry 1999; 45:17–25
3.
Glantz LA, Lewis DA: Decreased dendritic spine density on prefrontal cortical pyramidal neurons in schizophrenia. Arch Gen Psychiatry 2000; 57:65–73
4.
Kolluri N, Sun Z, Sampson AR, et al: Lamina-specific reductions in dendritic spine density in the prefrontal cortex of subjects with schizophrenia. Am J Psychiatry 2005; 162:1200–1202
5.
Cannon TD, Chung Y, He G, et al: Progressive reduction in cortical thickness as psychosis develops: a multisite longitudinal neuroimaging study of youth at elevated clinical risk. Biol Psychiatry 2015; 77:147–157
6.
Yoon JH, Zhang Z, Mormino E, et al: Reductions in synaptic marker SV2A in early-course schizophrenia. J Psychiatr Res 2023; 161:213–217
7.
Arion D, Enwright JF III, Gonzalez-Burgos G, et al: Cell type–specific profiles and developmental trajectories of transcriptomes in primate prefrontal layer 3 pyramidal neurons: implications for schizophrenia. Am J Psychiatry 2024; 181:920–934
8.
Mustafa T: Pituitary adenylate cyclase-activating polypeptide (PACAP): a master regulator in central and peripheral stress responses. Adv Pharmacol 2013; 68:445–457
9.
Woo E, Sansing LH, Arnsten AFT, et al: Chronic stress weakens connectivity in the prefrontal cortex: architectural and molecular changes. Chronic Stress (Thousand Oaks) 2021; 5:24705470211029254
10.
Hashimoto R, Hashimoto H, Shintani N, et al: Pituitary adenylate cyclase–activating polypeptide is associated with schizophrenia. Mol Psychiatry 2007; 12:1026–1032
11.
Datta D, Yang S, Joyce MKP, et al: Key roles of CACNA1C/Cav1.2 and CALB1/calbindin in prefrontal neurons altered in cognitive disorders. JAMA Psychiatry (Online ahead of print, May 22, 2024)
12.
Harrison PJ, Husain SM, Lee H, et al: CACNA1C (CaV1.2) and other L-type calcium channels in the pathophysiology and treatment of psychiatric disorders: advances from functional genomics and pharmacoepidemiology. Neuropharmacology 2022; 220:109262
13.
Gunther M, Jiang S, Banga A, et al: Delayed-onset psychosis secondary to tacrolimus neurotoxicity after lung transplant: a case report and systematic review. J Acad Consult Liaison Psychiatry 2023; 64:550–561
14.
Hall DD, Davare MA, Shi M, et al: Critical role of cAMP-dependent protein kinase anchoring to the L-type calcium channel Cav1.2 via A-kinase anchor protein 150 in neurons. Biochemistry 2007; 46:1635–1646
15.
Arion D, Enwright JF, Gonzalez-Burgos G, et al: Differential gene expression between callosal and ipsilateral projection neurons in the monkey dorsolateral prefrontal and posterior parietal cortices. Cereb Cortex 2023; 33:1581–1594
16.
Gamo NJ, Lur G, Higley MJ, et al: Stress impairs prefrontal cortical function via D1 dopamine receptor interactions with hyperpolarization-activated cyclic nucleotide-gated channels. Biol Psychiatry 2015; 78:860–870
17.
Rosenberg DR, Lewis DA: Postnatal maturation of the dopaminergic innervation of monkey prefrontal and motor cortices: a tyrosine hydroxylase immunohistochemical analysis. J Comp Neurol 1995; 358:383–400
18.
Abi-Dargham A, Xu X, Thompson JL, et al: Increased prefrontal cortical D1 receptors in drug naive patients with schizophrenia: a PET study with [11C]NNC112. J Psychopharmacol 2012; 26:794–805
19.
Datta D, Arion D, Corradi JP, et al: Altered expression of CDC42 signaling pathway components in cortical layer 3 pyramidal cells in schizophrenia. Biol Psychiatry 2015; 78:775–785
20.
Tu X, Yasuda R, Colgan LA: Rac1 is a downstream effector of PKCα in structural synaptic plasticity. Sci Rep 2020; 10:1777
21.
Hains AB, Vu MA, Maciejewski PK, et al: Inhibition of protein kinase C signaling protects prefrontal cortex dendritic spines and cognition from the effects of chronic stress. Proc Natl Acad Sci U S A 2009; 106:17957–17962
22.
Ahmad A, Sheikh S, Khan MA, et al: Endoxifen: a new protein kinase C inhibitor to treat acute and mixed mania associated with bipolar I disorder. Bipolar Disord 2021; 23:595–603
23.
Vacic V, McCarthy S, Malhotra D, et al: Duplications of the neuropeptide receptor gene VIPR2 confer significant risk for schizophrenia. Nature 2011; 471:499–503
Information & Authors
Information
Published In
History
Accepted: 12 August 2024
Published online: 1 October 2024
Published in print: October 01, 2024
Keywords
Authors
Funding Information
Dr. Arnsten is supported by NIMH grant R01 MH130538-01A1. Dr. Datta is supported by Career Development KL2 Scholar Award KL2TR001862.The authors report no financial relationships with commercial interests.
Metrics & Citations
Metrics
Citations
Export Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
For more information or tips please see 'Downloading to a citation manager' in the Help menu.
View Options
View options
PDF/EPUB
View PDF/EPUBLogin options
Already a subscriber? Access your subscription through your login credentials or your institution for full access to this article.
Personal login Institutional Login Open Athens loginNot a subscriber?
PsychiatryOnline subscription options offer access to the DSM-5-TR® library, books, journals, CME, and patient resources. This all-in-one virtual library provides psychiatrists and mental health professionals with key resources for diagnosis, treatment, research, and professional development.
Need more help? PsychiatryOnline Customer Service may be reached by emailing [email protected] or by calling 800-368-5777 (in the U.S.) or 703-907-7322 (outside the U.S.).