Analysis of mRNA and Protein Levels of CAP2, DLG1 and ADAM10 Genes in Post-Mortem Brain of Schizophrenia, Parkinson’s and Alzheimer’s Disease Patients

Schizophrenia (SCZ) is a mental illness characterized by aberrant synaptic plasticity and connectivity. A large bulk of evidence suggests genetic and functional links between postsynaptic abnormalities and SCZ. Here, we performed quantitative PCR and Western blotting analysis in the dorsolateral prefrontal cortex (DLPFC) and hippocampus of SCZ patients to investigate the mRNA and protein expression of three key spine shapers: the actin-binding protein cyclase-associated protein 2 (CAP2), the sheddase a disintegrin and metalloproteinase 10 (ADAM10), and the synapse-associated protein 97 (SAP97). Our analysis of the SCZ post-mortem brain indicated increased DLG1 mRNA in DLPFC and decreased CAP2 mRNA in the hippocampus of SCZ patients, compared to non-psychiatric control subjects, while the ADAM10 transcript was unaffected. Conversely, no differences in CAP2, SAP97, and ADAM10 protein levels were detected between SCZ and control individuals in both brain regions. To assess whether DLG1 and CAP2 transcript alterations were selective for SCZ, we also measured their expression in the superior frontal gyrus of patients affected by neurodegenerative disorders, like Parkinson’s and Alzheimer’s disease. Interestingly, also in Parkinson’s disease patients, we found a selective reduction of CAP2 mRNA levels relative to controls but unaltered protein levels. Taken together, we reported for the first time altered CAP2 expression in the brain of patients with psychiatric and neurological disorders, thus suggesting that aberrant expression of this gene may contribute to synaptic dysfunction in these neuropathologies.


Introduction
Schizophrenia (SCZ) is a polygenic and multifactorial disorder with complex phenotypes encompassing multiple domains, such as delusions and hallucinations (positive symptoms), avolition, anhedonia, lack of social interaction (negative symptoms), and

Colocalization of CAP2, SAP97, and ADAM10 in Primary Hippocampal Neurons
The actin-binding proteins CAP2, ADAM10, and its binding partner SAP97 are synaptic proteins enriched in the postsynaptic compartment ( Figure 1a) and their synaptic localization is finely modulated by activity-dependent synaptic plasticity [23,26]. To assess their colocalization in dendritic spines, we performed an immunocytochemistry experiment in primary hippocampal neuronal cultures. As shown in Figure 1b, the staining revealed that the spine shapers are localized along dendrites and in dendritic spines, where they colocalize in the head of the spines.

Correlation Analysis of CAP2, DLG1, and ADAM10 mRNA Expression with Age and PMI in the Post-Mortem Dorsolateral Prefrontal Cortex and Hippocampus of Schizophrenia Patients
Here we investigated the correlations between CAP2, DLG1 (encoding SAP97 protein), and ADAM10 mRNA expression with demographic and post-mortem storage characteristics such as age and PMI, respectively (Table 1).

Analysis of mRNA Expression of CAP2, DLG1, and ADAM10 in the Post-Mortem Dorsolateral Prefrontal Cortex and Hippocampus of SCZ Patients
Here, we evaluated potential alterations in CAP2, DLG1, and ADAM10 mRNA expression in the DLPFC of SCZ patients compared to control subjects. Statistical analysis indicated no significant difference in CAP2 and ADAM10 transcript levels in SCZ patients, compared to non-psychiatric controls (CAP2: p = 0.0893, ADAM10: p = 0.4780; Mann-Whitney test) (Figure 3a,c). Conversely, we reported increased levels of DLG1 transcript in SCZ patients compared to control subjects (p = 0.0407; Mann-Whitney test) (Figure 3b). Next, we extended gene expression analysis of CAP2, DLG1, and ADAM10 to the postmortem hippocampus of SCZ patients and control individuals. Interestingly, RT-PCR experiments revealed decreased CAP2 mRNA levels in the SCZ group compared to control subjects (p = 0.0309; Mann-Whitney test) (Figure 3d). On the other hand, no significant difference in DLG1 and ADAM10 transcripts was observed between the two diagnostic groups (DLG1: p = 0.2485; ADAM10: p = 0.1682; Mann-Whitney test) (Figure 3e,f).

Analysis of mRNA Expression of CAP2, DLG1, and ADAM10 in the Post-Mortem Dorsolateral Prefrontal Cortex and Hippocampus of SCZ Patients
Here, we evaluated potential alterations in CAP2, DLG1, and ADAM10 mRNA expression in the DLPFC of SCZ patients compared to control subjects. Statistical analysis indicated no significant difference in CAP2 and ADAM10 transcript levels in SCZ patients, compared to non-psychiatric controls (CAP2: p = 0.0893, ADAM10: p = 0.4780; Mann-Whitney test) (Figure 3a,c). Conversely, we reported increased levels of DLG1 transcript in SCZ patients compared to control subjects (p = 0.0407; Mann-Whitney test) ( Figure 3b). Next, we extended gene expression analysis of CAP2, DLG1, and ADAM10 to the postmortem hippocampus of SCZ patients and control individuals. Interestingly, RT-PCR experiments revealed decreased CAP2 mRNA levels in the SCZ group compared to control subjects (p = 0.0309; Mann-Whitney test) (Figure 3d). On the other hand, no significant difference in DLG1 and ADAM10 transcripts was observed between the two diagnostic groups (DLG1: p = 0.2485; ADAM10: p = 0.1682; Mann-Whitney test) ( Figure  3e,f).     . Correlation analysis between PMI and protein levels of CAP2, SAP97, and ADAM10 in the dorsolateral prefrontal cortex of (c,g,k) control subjects (n = 20) and (d,h,l) schizophrenia patients (n = 20). Analysis of correlation between age and protein levels of CAP2, SAP97, and ADAM10 in the hippocampus of (m,q,u) control subjects (n = 20) and (n,r,v) schizophrenia patients (n = 20). Correlation analysis between PMI and protein levels of CAP2, SAP97, and ADAM10 in the hippocampus of (o,s,w) control subjects (n = 20) and (p,t,x) schizophrenia patients (n = 20).

Correlation Analysis of CAP2, SAP97, and ADAM10 mRNA and Protein Expression with Age and PMI in the Post-Mortem Superior Frontal Gyrus of Alzheimer's and Parkinson's Disease Patients
Synaptic dysfunction has been considered a major determinant of many neurological diseases, including AD, PD, and Huntington's Disease [27,28]. Therefore, we extended our assessment of CAP2, DLG1, and ADAM10 gene and protein expression levels to the post-mortem superior frontal gyrus (SFG) of patients affected by neurodegenerative diseases such as PD and AD.
First, we examined the correlations of CAP2, DLG1, and ADAM10 mRNA with age and PMI in the SFG of AD and PD patients and the control group (Table 2). Synaptic dysfunction has been considered a major determinant of many neurological diseases, including AD, PD, and Huntington's Disease [27,28]. Therefore, we extended our assessment of CAP2, DLG1, and ADAM10 gene and protein expression levels to the postmortem superior frontal gyrus (SFG) of patients affected by neurodegenerative diseases such as PD and AD.
First, we examined the correlations of CAP2, DLG1, and ADAM10 mRNA with age and PMI in the SFG of AD and PD patients and the control group (Table 2)     Next, we investigated if the protein expression of these genes is correlated with both age and PMI in the same brain region. We found a significant negative correlation of CAP2 and  transcript in PD patients and a decreasing trend in AD patients, compared with nonneurological subjects (CTRL vs. PD p = 0.0111; CTRL vs. AD p = 0.0541; Mann-Whitney test) (Figure 7a). Conversely, we did not observe any alteration in DLG1 and ADAM10 mRNA expression levels between groups (DLG1: CTRL vs. PD p = 0.7430; CTRL vs. AD p = 0.8785; ADAM10: CTRL vs. PD p = 0.3823; CTRL vs. AD p = 0.1304; Mann-Whitney test) (Figure 7b,c).

Analysis of Protein Expression of CAP2, SAP97, and ADAM10 in the Post-Mortem Superior Frontal Gyrus of Alzheimer's and Parkinson's Disease Patients
Finally, in the same brain samples, we performed Western blot analysis to investigate eventual variations in CAP2, SAP97, and ADAM10 protein content in individuals with PD and AD. Our analyses showed unaltered CAP2, SAP97, and ADAM10 protein levels in the SFG of PD and AD patients compared to control subjects (CAP2: CTRL vs.

Analysis of Protein Expression of CAP2, SAP97, and ADAM10 in the Post-Mortem Superior Frontal Gyrus of Alzheimer's and Parkinson's Disease Patients
Finally, in the same brain samples, we performed Western blot analysis to investigate eventual variations in CAP2, SAP97, and ADAM10 protein content in individuals with PD and AD. Our analyses showed unaltered CAP2, SAP97, and ADAM10 protein levels in the SFG of PD and AD patients compared to control subjects (CAP2: CTRL vs. PD neurological subjects (CTRL vs. PD p = 0.0111; CTRL vs. AD p = 0.0541; Mann-Whitney test) (Figure 7a). Conversely, we did not observe any alteration in DLG1 and ADAM10 mRNA expression levels between groups (DLG1: CTRL vs. PD p = 0.7430; CTRL vs. AD p = 0.8785; ADAM10: CTRL vs. PD p = 0.3823; CTRL vs. AD p = 0.1304; Mann-Whitney test) (Figure 7b,c).

Analysis of Protein Expression of CAP2, SAP97, and ADAM10 in the Post-Mortem Superior Frontal Gyrus of Alzheimer's and Parkinson's Disease Patients
Finally, in the same brain samples, we performed Western blot analysis to investigate eventual variations in CAP2, SAP97, and ADAM10 protein content in individuals with PD and AD. Our analyses showed unaltered CAP2, SAP97, and ADAM10 protein levels in the SFG of PD and AD patients compared to control subjects (CAP2: CTRL vs.

Discussion
Several psychiatric and neurological disorders are associated with alterations in spine number and dendritic arborization [29,30]. The spine shapers CAP2, SAP97, and ADAM10 are localized in the postsynaptic compartment and their localization is under the control of activity-dependent synaptic plasticity [23,26]. CAP2 translocation into spines is required for spine enlargement upon long-term potentiation induction [23]. SAP97 controls the abundance of ADAM10 at the synapse and their association is essential for the long-term depression-triggered spine shrinkage [26]. Based on the well-recognized synaptic defects in the SCZ brain [12], here we investigated the mRNA and protein levels of CAP2, SAP97, and ADAM10 in the post-mortem brain of SCZ patients. In particular, we performed the analysis in total homogenates of DLPFC and hippocampus, two brain regions involved in deregulated cortical-subcortical network typical of this psychiatric disorder [31][32][33][34].
Our data indicated a significant increase of DLG1 mRNA levels within DLPFC, but not in the hippocampus of SCZ patients, when compared to controls. Conversely, no changes were observed in the expression levels of the protein encoded by the DLG1 gene, SAP97, in both brain regions.
Taken together, these results are puzzling since it has been demonstrated a pivotal role of this synaptic element in regulating the trafficking of the glutamate AMPA receptors, whose expression, and function have been reported to be altered in SCZ [35]. In this regard, compelling evidence, including genetic-imaging studies, suggests that genetic variants of the DLG1 gene are functionally linked to abnormal cognitive patterns in SCZ patients [36][37][38][39][40]. In particular, Xu and colleagues reported that SAP97 rs3915512 polymorphism in patients with first-episode schizophrenia was associated with low structural and functional connectivity within the orbitofrontal-striatal-thalamic circuitry [41]. Despite this finding, protein expression analysis of SAP97 in DLPFC reported conflicting data, showing either a reduction [42] or increased protein amount in SCZ brains, compared to healthy controls [43]. In addition, no difference was detected in DLG1 gene expression in both DLPFC and occipital cortex of post-mortem old adults with SCZ [44].
Notably, we observed a decrease of CAP2 transcript level, selectively in the hippocampus, but not in the DLPFC of SCZ patients, while no changes in CAP2 protein levels were detected in both brain regions. To the best of our knowledge, this is the first work reporting CAP2 mRNA reduction in the post-mortem brain of SCZ patients. In humans, the CAP2 gene is located on chromosome 6p22.3. Interestingly, patients with an interstitial 6p22−24 deletion syndrome show a complex and variable phenotype including a developmental delay and cognitive disorders [45,46]. It is worth mentioning that the expression of CAP1, another member of the CAP family, is increased in the mediodorsal thalamus of SCZ patients compared to control subjects [47] and is altered in the cortex of an animal model of SCZ [48]. Despite CAP2 transcript alteration in the hippocampus of SCZ patients, we found that the protein levels were unaffected in this brain area thus making it difficult to argue a direct consequence of mRNA reduction in modulating excitatory synaptic transmission. Further investigations are warranted to understand the relevance of CAP2 at the glutamatergic synapse of the SCZ brain.
Overall, our data indicate alterations in both DLG1 (increase) and CAP2 (decrease) gene expression in SCZ brains but not in their corresponding protein levels. It should be remarked that our analysis has been carried out using total homogenate extracts. Therefore, future studies are needed to assess whether differences in the expression levels of these proteins occur in selective neuronal compartments, like the postsynaptic fraction where they exert their specific function.
Our data also indicated an unaltered mRNA and protein expression of ADAM10 in the post-mortem brain of SCZ patients, compared to normal controls in both DLPFC and hippocampus. These results are in line with a recent report indicating no changes in transcript and protein expression of ADAM10 in post-mortem PFC of SCZ patients [49]. However, a very recent study showed a significant reduction of ADAM10 mRNA in several brain regions of SCZ patients [50]. The apparent discrepancy between our results on CAP2, SAP97, and ADAM10 mRNA and protein expression and those obtained from other cohorts of post-mortem SCZ brains should take into account diverse confounding variables, including antipsychotics, age, disease duration, and gender. In particular, pharmacological therapy could have played a role in regulating the expression of these synaptic proteins at multiple levels, such as type of antipsychotic, dosage, and duration of treatment. Consistent with this assumption, evidence suggests that the expression of key scaffold proteins can be significantly affected in animal models exposed to antipsychotic drugs [51][52][53]. Moreover, we found that the age of SCZ patients was significantly lower than control individuals. This is in line with literature reporting a reduced life expectancy in patients with SCZ compared with the general population [54,55]. On the other hand, we did not find any effect of gender, so that, therefore, should not have affected our results.
In the present work, as a reference of synaptopathy-related diseases, we analyzed CAP2, SAP97, and ADAM10 expression in post-mortem brains of patients affected by PD and AD. Indeed, aberrant synapse functioning is a common trait of several brain disorders, including the aforementioned neurodegenerative diseases [27]. Moreover, the relevance of postsynaptic proteins in spine pathology is also supported by evidence showing alterations of spine shapers in AD [23,56]. Consistent with this, CAP2 protein levels were found to be reduced in the hippocampus of patients and animal models of AD [23]. It is noteworthy that even though our results show a remarkable decreasing tendency of CAP2 mRNA levels in the SFG of AD individuals, CAP2 protein levels were not affected, confirming the results obtained in previous studies [23]. CAP2 has never been analyzed in PD, while ADAM10 and SAP97 gene variants have already been associated with PD [57,58]. In the present work, we show a significant reduction of CAP2 mRNA levels in the SFG of the PD group, while no alterations in SAP97 and ADAM10 expression were detected in patients, compared to controls. In contrast to our observations, earlier immunohistochemical studies carried out in the human post-mortem hippocampus of PD patients showed a significant increase of SAP97 [59]. This synapse-associated protein was also altered in the striatum of animal models of PD and Levodopa-induced dyskinesia [60]. We can hypothesize that the alterations in SAP97 expression in PD can depend on different factors, including the brain area analyzed, the disease stage, and drug administration. In conclusion, our data suggest that aberrant gene expression of spine shapers may contribute significantly to synaptic dysfunction-related neuropsychiatric disorders. The involvement of key molecules of the synapse may highlight the need for a more vigorous translational approach towards the search of novel molecular targets paving the way for innovative treatments beyond or in augmentation with the ones presently available.

Neuronal Cultures Preparation and Immunocytochemistry
Hippocampal neuronal primary cultures were prepared from embryonic day 19 (E19) rat hippocampi as previously described [61]. For colocalization studies, day in vitro (DIV) 14 hippocampal neurons were fixed with 4% Paraformaldehyde (PFA)-4% sucrose in PBS solution for 5 min at 4 • C and washed several times with PBS. Cells were permeabilized with 0.1% Triton X-100 in PBS for 15 min at room temperature (RT) and then blocked with 5% bovine serum albumin (BSA) in PBS for 1 h at RT. Cells were then labeled with primary antibodies at 4 • C overnight. The following antibodies were used: anti-CAP2 (anta Cruz Biotechnology, cod. SC-167378), anti-SAP97 (Enzo, cod. ADI-VAM-PS005), and anti-emphADAM10 (Abcam, cod. ab39153). Cells were washed and then incubated with secondary antibodies for 1 h at RT. The Alexa Fluor dye secondary antibodies used (donkey anti-rabbit-Alexa488, donkey anti-mouse-Alexa555, donkey anti-goat-Alexa647) were purchased from Thermo Fisher Scientific. After, the cells were washed in PBS and mounted on glass slides with Fluoromount mounting medium (Sigma Aldrich 20149,Milano, Italy). Images for the analysis of neuronal spine morphology were acquired with an Airyscan (resolution 100-120 nm) microscopy using Zeiss LSM 900, 63× oil objective, PLAN Apochromat, NA 1.42.

Human Post-Mortem Tissue Collection
Human tissue collection DLPFC and hippocampus samples from post-mortem brains of non-psychiatric controls and SCZ patients (n = 20/brain region/clinical condition) were obtained from The Human Brain and Spinal Fluid Resource Center (Los Angeles Healthcare Center, Los Angeles, CA, USA). Clinical diagnosis of SCZ was performed according to DSMIII-R criteria. Demographic characteristics of control and SCZ subjects are described in Table 1 and Table S1. We obtained human SFG samples of normal controls, PD, and AD patients from The Netherlands Brain Bank (Netherlands Institute for Neuroscience, Amsterdam, open access: www.brainbank.nl, accessed on 27 January 2022). We selected cases with a clinical diagnosis of AD (n = 10) and neuropathological staging of Braak ≥ 5. PD patients (n = 10) were characterized by Braak LB stage ≥ 4. Controls (n = 10) were adults without cognitive decline and Braak ≤ 3 in accordance with the Braak and Braak criteria [62]. The control subjects had no known clinical history of neurological or psychiatric disorders and were also fully neuropathologically evaluated to confirm that they were free of neurodegenerative pathologies. AD patients had a clinical diagnosis of dementia or probable AD, according to the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders Association (NINCDS-ADRDA) criteria [63]. Clinical diagnosis of PD was based on diagnostic procedure according to the UK Brain Bank criteria for PD [64] and confirmed by neuropathological findings [65]. Frozen tissues were pulverized in liquid nitrogen and stored at −80 • C for subsequent processing.

RNA Extraction and Quantitative RT-PCR Analysis
Total RNA was extracted from post-mortem tissues using RNeasy ® mini kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions (Querques et al. 2015). Total RNA was purified to eliminate potentially contaminating genomic DNA using recombinant DNase (Qiagen, Hilden, Germany). RNA integrity number (RIN) of samples was assessed using Agilent 2100 Bioanalyzer Expert (Santa Clara, CA, USA) and BioRad Experion Automated electrophoresis Station (Hercules, CA, USA) prior to cDNA synthesis using Transcriptor First Strand cDNA Synthesis kit (Roche Diagnostics, Mannheim, Germany). A total of 1 µg of total RNA of each sample was reverse transcribed with QuantiTect Reverse Transcription (Qiagen, Hilden, Germany) using oligo-dT and random primers according to the manufacturer's instructions. Quantitative RT-PCR with Real Time ready catalogue Assays (Roche Diagnostics) and LightCycler ® 480 Probe Master (Roche Diagnostics) was performed on a Light Cycler 480 Real Time PCR thermocycler with 96-well format (Roche Diagnostics). All measurements from each subject were performed in duplicate. CAP2, DLG1, and ADAM10 mRNA expression levels were normalized to the mean of two housekeeping genes: β-actin (ACTB) and cyclophilin (PPIA). The following protocol was used: 10 s for initial denaturation at 95 • C followed by 40 cycles consisting of 10 s at 94 • C for denaturation, 10 s at 60 • C for annealing, and 6 s for elongation at 72 • C temperature. The following primers were used for CAP2, DLG1, and ADAM10 cDNA amplification: CAP2 forward, 5 -GCC GCC TGG AGT CGC TGT C-3 and CAP2 reverse, 5 -AAA ACT CGG CCA CCA TAC TGT CCA-3 ; DLG1 forward, 5 -GAG ATG ACT CAA GTA TTT TCA TTA CCA-3 and DLG1 reverse, 5 -CAC GAA CAT CTA CTT CAT TTA CTC G-3 ; ADAM10 forward, 5 -CTGCCCAGCATCTGACCCTAA-3 , and ADAM10 reverse, 5 -TTG CCA TCA GAA CTG GCA CAC-3 . mRNA expression was calculated using the geometric mean of the two reference genes selected and the relative quantification method (2 −∆∆Ct ).

Western Blotting
Frozen, powdered samples from post-mortem DLPFC and hippocampus and from SFG of respective brain banks were sonicated in 1% SDS and boiled for 10 min. Aliquots (2 µL) of the homogenate were used for protein determination using a BioRad Protein Assay kit. Equal amounts of total proteins (30 µg) for each sample were loaded on precast 4-20% gradient gels (BioRad Laboratories, Hercules, CA, USA). Proteins were separated by SDS-PAGE and transferred to PVDF membranes (GE Healthcare, Chicago, IL, USA) via the Trans Blot Turbo System (BioRad Laboratories, Hercules, CA, USA). To investigate the targets of interest the blots were incubated with antibodies against CAP2 (1:1000; 15865-1-AP, Proteintech), SAP97 (1:1000; ADI-VAM-PS005, Enzo Life Sciences), and ADAM10 (1:4000; AbCaM Ab 39153). GAPDH (1:1000; sc-32233, Santa Cruz Biotechnology) was used to normalize the levels of analyzed proteins for variations in loading and transfer. All blots were incubated in horseradish peroxidase-conjugated secondary antibodies and target proteins visualized by ECL detection (Pierce, Rockford, IL, USA), followed by quantification through the "Quantity One" software (BioRad Laboratories, Hercules, CA, USA). Normalized values were then averaged and used for statistical comparisons. All representative blots shown in the figures arise from cut-out and pasted bands for reassembling the image. Of note, for each graph, the representative bands come from the same films.

Statistical Analysis
Normal distribution assumption for continuous variables was checked by Shapiro-Wilks and Kolmogorov-Smirnov tests. We observed a non-normal distribution of our data; therefore, a nonparametric approach was used for all statistical analyses. Data are reported as medians, along with interquartile range (first-third quartiles-IQR). Statistical analysis of qPCR and western blot experiments was performed in GraphPad Prism 7 by Mann-Whitney test. Statistical significance was also corrected for multiple comparisons using the Bonferroni-Dunn method (see Table S2). Spearman's nonparametric correlation was used to test possible associations between nonparametric variables. Asterisks denote statistical significance as calculated by the specific statistical tests (*, p < 0.05).

Conclusions
In conclusion, we reported increased expression of DLG1 transcript in DLPFC and a reduction in CAP2 mRNA expression in the hippocampus of post-mortem SCZ brains, thus suggesting an overall altered expression of the genes encoding these dendritic spine proteins in SCZ.  Informed Consent Statement: All material has been collected from donors for or from whom a written informed consent has been obtained.

Data Availability Statement:
The data that support the findings of this study are available from the corresponding author upon reasonable request.