Schizophrenia, a serious psychiatric disorder, is characterized by psychotic symptoms and abnormal behaviors. Approximately 1% of the world's population have this disease. The symptoms of schizophrenia are currently classified into three categories: positive symptoms (loss of contact with reality), negative symptoms (defects in states of basic emotional and behavioral processes), and cognitive impairments. Schizophrenia patients usually have persistent social disability. The social and economic burdens caused by schizophrenia are thought to be among the most substantial of all mental illnesses.[4–6]
Researchers have attempted to reveal the causes of schizophrenia from a psychological perspective since the early 1950s. According to earlier psychological theories, schizophrenia was caused by abnormal metabolisms in the brain. It was not until traditional sedatives (such as haloperidol and chlorpromazine) were found to improve schizophrenia symptoms that scientists started to explore antipsychotics and investigate the mechanisms of schizophrenia.
Currently, antipsychotic drugs are the most commonly used method to control the progression of schizophrenia. Antipsychotic drugs can be classified into “typical antipsychotics” and “atypical antipsychotics”. Typical antipsychotics, also called first-generation antipsychotics, can reduce positive symptoms in schizophrenia. Atypical antipsychotics, or second-generation antipsychotics, can be beneficial in reducing both positive and negative symptoms.[10,11]
However, the antipsychotic drugs currently used in clinical practice give unsatisfactory results, and still lack the ability to effectively reduce the most serious symptoms of schizophrenia.[12,13] Although antipsychotic drugs reduce delusions and hallucinations in schizophrenia patients, they are unable to solve patients’ disabilities or enhance their functional recovery (for example, by allowing employment). Although antipsychotic drugs can effectively reduce the positive symptoms of schizophrenia, they are not effective in decreasing the occurrence of multiple negative symptoms and cognitive impairments that are closely related to the disabilities of schizophrenia patients.
In addition to their unsatisfactory outcomes, current antipsychotic drugs frequently result in additional adverse drug reactions (ADRs) that may worsen patients’ conditions.[12,13] It has been noted that typical antipsychotics, although effective in lessening certain symptoms of schizophrenia, often cause problems in motor function. Atypical antipsychotic drugs lead to fewer problems in motor function, but can cause other metabolic abnormalities, such as weight gain. Significant efforts have been made to identify more sensitive and accurate signs of antipsychotic drug efficacy and ADRs, to improve the efficacy of antipsychotic drugs and minimize their ADRs. Novel markers of antipsychotic drug responses will help to optimize individualized treatments in the initial stage of schizophrenia, for example by avoiding the use of ineffective or toxic drugs, and will decrease the incidence of relevant ADRs.[14,15]
An incomplete understanding of current antischizophrenic drugs creates a substantial challenge for scientists designing new drugs. This lack of understanding is reflected in the confusion in drug classification, as well as in the lack of drug targets that are currently available. With regards to the accurate classification of antischizophrenic drugs, these drugs have frequently been classified into typical and atypical antipsychotic drugs based on their corresponding ADRs and pharmacodynamics. However, several independent studies have shown that there are no differences between typical and atypical antipsychotics. It has been suggested that inappropriate generalizations of the different types of antipsychotic drugs may have caused errors and confusion in earlier conclusions regarding the differences between typical and atypical antipsychotic drugs; that is, certain schizophrenic drugs with different pharmacology and side effects were mistakenly classified as one group, because antipsychotics contain a number of drugs with very different properties. Therefore, a greater understanding of the mechanisms of antischizophrenic drug treatments is needed to design a more appropriate classification system for these drugs.
In addition, the identification of new drug targets is also urgently needed to be able to design more effective antischizophrenic drugs. Another bottleneck in designing an advanced treatment for schizophrenia is the lack of knowledge around effective drug targets. Several studies have demonstrated that the emergence and progression of schizophrenia are related to the activities of dopamine, glutamate, and γ-aminobutyric acid. Currently, mainstream theories of the etiology of schizophrenia include the dopamine hypothesis, 5-hydroxytryptamine (5-HT) hypothesis, glutamate hypothesis, and oxidative stress hypothesis. The most common drug targets include dopamine D2 receptors, 5-HT 2 receptors, and glutamate. A certain proportion of “new” antipsychotic drugs still aim to block central dopamine D2 receptors. In addition, the targets of the three antipsychotic drugs (cariprazine, aripiprazole lauroxil, and brexpiprazole) that were approved by the US Food and Drug Administration in 2015 are dopamine and 5-HT receptors, and the majority of studies of antipsychotic drugs focus on the dopamine and 5-HT systems. However, the use of D2 antagonists over time may contribute to a decline in drug efficacy, worsened symptoms, and the emergence of ADRs. Identification of novel genetic markers and therapeutic agents with improved efficacy would therefore help scientists to design better treatments for the different types of symptoms and dysfunction that occur in schizophrenia patients.[12,13]
The development of multi-omic studies can improve the treatment of schizophrenia by helping to develop effective drugs and minimize ADRs. More advanced omics technologies are required to systematically discover markers of drug response and ADRs in schizophrenia therapies. The present review discusses the value of genomics, transcriptomics, epigenomics, proteomics, and metabolomics in studying drug targets in schizophrenia, and explains how they can be used to improve drug efficacy and minimize the occurrence of ADRs.
Database search strategy
We performed a series of electronic searches in PubMed database and Google Scholar (2014–2019) using the following key terms: schizophrenia, antipsychotic treatment, omics technologies, and multi-omics. Articles in reference lists were also included in this review. We reviewed all the English articles that focus on schizophrenia and antipsychotics. Articles were excluded if less than one omics technologies was applied. We first screened the title, abstract, and keywords to select for the articles related to our study. Then we collected the relevant information in the full text to summarize their methods and findings.
Thanks to the development of sequencing technologies, DNA sequencing has become less expensive and faster, leading to the rapid accumulation of genomic data, such as the completion of the human genome project. These features have attracted growing interest from researchers wanting to study the functions of genetic polymorphisms (including phenotypic differences, susceptibility to diseases, and individual drug efficacy) using these big datasets. As a result, pharmacogenomics and pharmacogenetics are becoming increasingly popular for studying drugs of interest and novel drug targets at a genetic level. These investigations attempt to explain individual variations in drug reactions using an individual's unique genomic information, and to improve current antipsychotic treatments using genetic signatures of drug targets, drug efficacy, and ADRs.
Genome-wide association studies (GWAS) are a commonly used method to analyze genetic data, and enable the successful detection of relationships between genetic variations and complex diseases. However, most GWAS studies have been performed in European populations, and it is thus necessary to carry out GWAS studies in different populations.
Understanding ADRs at a genetic level can help to detect markers of ADRs. For example, a recent study revealed that the side effects of antipsychotic medications could be decreased using CYP2C19 genetic information when treating patients. Another study reported the importance of the link between two loci (on HLA-DQB1 and HLA-B, respectively) and clozapine-induced agranulocytosis/granulocytopenia in predicting ADRs. Carriers of these 2 alleles have an increased risk (0.3–0.4%) of some ADRs of interest. In particular, 13 single nucleotide polymorphisms (SNPs) in nine genes, including the rs3813929 SNP in HTR2C, were significantly associated with antipsychotic-induced weight gain.[28,29] These ADR marker candidates, as discussed above, may be used to predict and prevent ADRs in the future.
In addition to ADRs, genomic studies have successfully revealed multiple markers of drug efficacy. In many studies, drug of interest are transported and metabolized in vivo, and this is thought to be regulated by specific proteins (eg, drug-metabolizing enzymes and transporters). In this process, drug efficacy is suspected to be influenced by certain gene mutations that affect the coding of key proteins. For example, variations in CYP2D6 have been found to affect activities of the polymorphic CYP2D6 enzyme, which is related to metabolism of the antipsychotic drugs aripiprazole and risperidone. This study also showed that the CYP2D6 genotype contributed to poorer metabolism of active moiety exposure compared with normal metabolizers (∼1.6 times), and suggested that poor and intermediate metabolizers should take lower daily doses of risperidone and aripiprazole to obtain better drug effects. Additionally, plasma and tissue drug concentrations have also been found to affect drug efficacy. For example, genetic polymorphisms in D2 receptors are significantly associated with different drug efficacy levels of clozapine, aripiprazole, and risperidone.DRD4 (120-bp duplication), another newly discovered marker of drug response, is associated with improved drug efficacy of clozapine.
Genetic markers of drug efficacy will help with the discovery of new drug targets in future personalized medicine. The rapid progress of genomic studies and the use of genome-wide screening methods have also provided insights into novel candidate drug targets. Certain genetic information can improve the success rate of drugs up to twofold compared with the use of drugs without such genetic support.DRD2, a star candidate drug target in schizophrenia, is one of the main targets of typical antipsychotics.[21,33] Recently, Yu et al reported that the SNPs rs72790443 in MEGF10, rs1471786 in SLC1A1, rs9291547 in PCDH7, rs12711680 in CNTNAP5, and rs6444970 in TNIK were significantly associated with the response to antipsychotic treatments. In addition, the Psychiatric GWAS Consortium discovered 108 loci associated with schizophrenia using GWAS, while Lam et al identified 19 schizophrenia-associated loci in patients with East Asian ancestry. Furthermore, Xiao et al demonstrated a significant association between the rs5177 SNP in LRP8 and schizophrenia using a meta-analysis of GWAS data sets and the Genotype-Tissue Expression project. The aforementioned genes may be associated with the effects of schizophrenia treatment, and may serve as effective targets for future antipsychotic drugs.
In summary, genomic studies and an increasing accumulation of genomic data have enabled scientists to identify effective drug targets as well as genetic markers of drug efficacy and ADRs (Table 1). Effective markers of drug efficacy and ADRs may be used to improve the safety and effectiveness of antipsychotic drugs in the future. Meanwhile, it should be noted that genetic factors are not the only factors in the development of new drugs with optimized drug efficacy and minimized ADRs, because there are also complex interactions between genes and the environment. Genomics studies can therefore be regarded as a starting point for multi-omics studies, as discussed in the following sections.
Transcriptomic analysis is an important component of individualized genomic analysis. Transcriptomic studies on RNA in cells may reflect specific gene expression and can be used to infer mRNA variations of interest. In particular, non-coding RNAs regulate protein expression through post-transcriptional regulation and the inhibition of translation. Non-coding RNAs, such as transfer RNA and ribosomal RNA (protein synthesis), small nucleolar RNA (RNA modification), and microRNA and small interfering RNA (post-transcriptional silencing), play important roles in various biological processes, including drug metabolism.[37,38] A “transcriptomic fingerprint” has been reported to reflect the unique pharmacological characteristics of drug treatments.[38,39] Thus, RNA transcripts can be used to classify drugs, predict drug efficacy and ADRs, and discover new drug targets.
Drug-induced gene expression is useful in studying the mechanisms of ADRs. It has been reported that a number of genetic variants in non-coding regions regulate biological processes by varying gene expression. The differentially expressed genes related to ADRs may be used to predict ADR occurrence in the future. Five obesity-related genes (GPER, LTF, MMP8, OLR1, and OLFM4) and four diabetes-related genes (ALPL, LTF, MMP8, and OLR1) have been found to be differentially expressed in patients who received atypical antipsychotics treatment. This result suggests that altered gene expression caused by atypical antipsychotics may lead to obesity and diabetes in these patients.
Transcriptomics is frequently used to understand antischizophrenic drugs because the drug-response transcripts are similar when treated with the same drugs. For example, haloperidol (a typical antipsychotic) induces Fos expression in the striatum, while clozapine (an atypical antipsychotic) induces more Fos expression in the cortex. Sakuma et al have therefore proposed a new classification method for antipsychotic drugs. In their study, haloperidol and aripiprazole were found to have similar transcriptomic fingerprints, while olanzapine had different transcriptional mechanisms. Similarly, Readhead et al were able to accurately identify 18 drugs in the same group using 52 different drug-induced transcripts in human induced pluripotent stem cell-derived neural progenitor cells.
The drug efficacy of antipsychotics in schizophrenia can be predicted using relevant RNA information. Some typical and atypical antipsychotics change the regulation of excitability and the maintenance of neuronal electrical balance by affecting Homer1a gene expression. Crespo-Facorro et al found that six genes with high expressions associated with positive symptoms (ADAMTS2, CD177, CNTNAP3, ENTPD2, RFX2, and UNC45B) returned to normal levels after atypical antipsychotic drug treatment. In addition, Zhang et al estimated that the coefficient of variation (the ratio of the standard deviation to the mean) of RNA and microRNA was increased by over 20% in schizophrenic patients. This study also demonstrated that the coefficient of variation of gene expression decreased after patients received oral second-generation antipsychotic treatment.
Transcriptomic studies have been used to discover novel schizophrenia-related genes in a large number of investigations. Recently, Enwright Iii et al identified more than 800 differentially expressed transcripts in parvalbumin neurons in the dorsolateral prefrontal cortex, most of which had never been previously reported. These transcripts were related to mitochondrial function and signal transduction pathways. In addition, Klarer et al demonstrated that afferent nerve conduction in the subdiaphragmatic vagus nerve in male rats may lead to changes in transcription in the functional network, which has been annotated in two brain regions, the nucleus accumbens and prefrontal cortex, associated with schizophrenia. The differentially expressed genes in the nucleus accumbens were revealed to be mainly schizophrenia-related (Sema3a, Sst, Calb1, Reln, Grm3, and Grik5), while differentially expressed genes in the prefrontal cortex were associated with neurodegeneration (Bcl2l1, Mag, Klk6, Gjb1, and Gjc2). These differently expressed genes in schizophrenia may serve as new drug targets in the future.
Transcriptomics is a bridge that connects individual information from the genome and the environment. Using transcriptomics, researchers are able to explore the molecular basis of diseases as well as the mechanisms of antipsychotic drugs by examining variations in gene expression (Table 2). Moreover, it has been reported that gene expression levels in peripheral blood lymphocytes are similar to those in the nervous system; therefore, gene expression in peripheral blood transcriptomes may be used to characterize drug targets, drug efficacy, and ADRs of interest. However, the traditional approach to predicting drug efficacy using transcription patterns still needs more evidence. It appears that the novel classification of drugs based on gene expression profiles could accelerate progress and improve accuracy in the screening of new psychoactive compounds and the development of new drugs.
Proteomics and metabolomics
Proteomic biomarkers are useful to reflect the real-time physiological state. In recent years, proteomics have been frequently used in the discovery of biomarkers for early diagnosis, the detection of pharmacodynamics and ADRs, and the design of new therapeutic methods. Metabolomics has become a key method for identifying novel biomarker candidates that respond to variations in the genome and environment, based on changes in the metabolic state.
Mass spectrometry is a tool that can be used for the qualitative and quantitative identification of biomarkers and proteins (sequence changes, post-translational modifications, protein interactions, and bioactive peptides) in cell models, animal models, and human tissue samples.[48,49] Mass spectrometry has been widely used to study the effects of antipsychotics and of psychiatric disorders such as schizophrenia.[48,49]
Proteomic and metabolomic studies have revealed additional details regarding the molecular mechanisms underlying ADR occurrence, and the results have greatly facilitated the design of new therapies with reduced side effects for clinical practice. Antipsychotic drugs are known to mainly affect the production of energy, proteins in metabolic pathways, and certain lipids. Scientists have proposed that metabolic pathway disruption may be a major contributor to metabolic syndrome risk. A recent study successfully identified a number of effective biomarkers of energy metabolism by analyzing 122 schizophrenic patients who received antipsychotic drugs. Another study showed that risperidone treatment varied the expression levels of 17 proteins relating to metabolic pathways. The aforementioned studies have significantly contributed to our understanding of the detailed mechanisms of ADRs relevant to antipsychotic drug treatment.
As well as ADRs, proteomic studies have also revealed multiple biomarkers that are significantly associated with drug efficacy. For example, Steiner et al reported that the expression levels of serum prolactin in schizophrenic patients were proportional to the effectiveness of antipsychotic treatments. In addition, expression of the heart form of fatty acid binding protein was associated with the efficacy of olanzapine. Cassoli et al demonstrated that haloperidol repaired damage in calcium ion and G protein signal transduction in schizophrenia patients. Therefore, the drug efficacy of haloperidol may be optimal in schizophrenia patients with symptoms of inflammation, because haloperidol activates oxidative stress and apoptotic pathways. In addition, risperidone may have a better therapeutic effect in schizophrenia patients with energy metabolism disorders. These results may provide important clues to accurately predict the efficacy of drugs of interest in the future.
Proteomic and metabolomic data may also be used to study and design novel therapies for schizophrenia. Studies of stress-induced biomarkers have revealed details of the whole metabolic state of organisms of interest and have provided a unique perspective of local metabolism. For example, action pathways of drugs have been speculated and combination therapies have been proposed based on abnormal metabolisms of patients. Using mass spectrometry, Farrelly et al revealed that prenatal iron deficiency affects metabolic processes (such as the tricarboxylic acid cycle) and increases the risk of schizophrenia by detecting 100 differentially expressed proteins in the frontal cortex of a rat model of prenatal iron deficiency. Another study revealed that better therapeutic effects and a reduction in the stress response may be achieved by supplementing ATP fuel, antioxidants, and polyunsaturated fatty acids when patients received corresponding antipsychotic drug treatments. In addition, increased levels of interleukin-1 beta in the cerebrospinal fluid of schizophrenic patients, and inferred inflammation, have been found to actively participate in the regulation of the cerebral nervous system. Thus, several researchers have begun to explore the potential of anti-inflammatory drugs, such as aspirin, combined with traditional antipsychotics; these have been shown to alleviate the symptoms of schizophrenia.
Proteomic and metabolomic studies are effective in identifying biomarkers of interest. These findings of genetic signatures have deepened our understanding of schizophrenia and helped to improve the diagnosis and treatment of schizophrenia in clinical practice. Previously identified proteomic biomarkers have been used to adjust pharmacodynamics and reduce ADRs in a number of small-scale studies. However, we still lack solid evidence and further validation of these biomarkers from studies with larger sample sizes. Therefore, large-scale proteomic studies are needed before these novel biomarkers can be applied to predict drug effects and drug side effects in schizophrenia in real clinical settings.
Epigenetic mechanisms, including histone modification and DNA methylation, regulate gene expression and generate heritable mutations without altering DNA sequences. Histone acetylation promotes gene transcription by opening chromosome structure. DNA methylation is known to directly modify and regulate gene expression, and is catalyzed by a series of methyltransferases. Histone modification and DNA methylation are dynamic processes that may partially explain the complex interactions between genes and the environment. However, it is still a challenge to understand the inheritance model of epigenetic changes. Epigenome-wide association studies are the main approach used in epigenomic research, and these use array and sequencing techniques. The results of these studies may help us to better understand the process of epigenetic regulation in patients receiving antipsychotic drug treatment. Effective markers of clinical responses would also help with the design of novel, advanced drugs and treatments for individual patients.
Epigenomics have been used to investigate variations in epigenetic modifications induced by the drug of interest. The results of these analyses may help us to understand the mechanisms of action of antipsychotic drugs. For example, quetiapine has been shown to reduce methylation of the SLC6A4 promoter and normalize 5-HT levels in the synaptic cleft. Olanzapine has been demonstrated to induce hypermethylation of 15 specific genes in the hippocampus, and then downregulate the dopamine pathway. Clozapine and sulpiride have been reported to downregulate Reln and Gad1 in the γ-aminobutyric acidergic system. Furthermore, risperidone decreases histone H3 phosphorylation in the medial prefrontal cortex in a rat model of schizophrenia. Additionally, clozapine and sulpiride were reported to promote demethylation in the prefrontal cortex and striatum of rats in a recent investigation.
Certain markers related to drug efficacy have been detected using epigenomic information. Drug efficacy of antipsychotics has been noted to vary across different individuals; this may be caused by the unique DNA methylation and histone modification profiles of each person. The DNA methylation/demethylation and histone acetylation status may reflect drug efficacy. For example, increased expression of mGlu2 has been shown to reduce resistance to antipsychotics in schizophrenia. Miura et al revealed that aripiprazole has a more important role in patients with hypermethylation of CpG387 in ankyrin repeat and kinase domain containing 1 compared with patients without this hypermethylation. In a recent twin experiment, levels of DNA methylation in peripheral blood were lower in schizophrenia patients treated with drugs, and Santoro et al reported that the promoter regions of Galr3 and Clra1 were hypermethylated in rats treated with risperidone or haloperidol. It has also been reported that regulation of histone acetylation may be critical in the treatment of schizophrenia.
The therapeutic effects of antipsychotics are reportedly related to epigenetic modifications, which may offer a possible new field of drug targets. MS-275 (Entinostat) was developed based on the knowledge that benzamide can cross the blood–brain barrier to promote histone acetylation. Currently, proteins with epigenetic modifications that have been used to develop antischizophrenic drugs include histone deacetylase inhibitors and valproate (a weak inhibitor of histone acetylation). Both histone deacetylases and valproate have regulatory potential and may be used to treat multiple psychiatric diseases, including schizophrenia. Memory loss has also been successfully ameliorated by increasing the histone acetylation of H4K12 and H3K9 using histone deacetylase 2 inhibitors in mouse models of neurodegeneration. This discovery may facilitate the advancement of schizophrenia treatments and allow the prevention of cognitive decline in schizophrenic patients.
Epigenetic disorders are one of the major causes of schizophrenia. It has become important to search for molecular markers of schizophrenia and predict antipsychotic drug responses. Epigenomics have revealed that the production and maintenance of tissue-specific expression play an important role in the development and synaptogenesis of the brain. Finally, promoter methylation status and gene expression variation in a number of schizophrenia susceptibility genes (for example, RELN, SOX10, GRM2, GRM5, and BDNF) have been identified in multiple studies.[57,64]
Epigenomics is an important field for individualized medicine because its active dynamics respond to the environment of patients. Environmental factors (such as social stress, drug therapy, and nutrition) are significantly associated with differential epigenetic modifications and may also serve as potential drug targets (Table 3). The modifications that are associated with drug and tissue specificity may provide additional clues for early diagnosis, drug screening, and intervention. A better understanding of the relevant epigenetic mechanisms will be of great help in the development and screening of effective drugs to treat schizophrenia.
Omics data have accumulated at an unprecedented speed in recent years. A single level of omics data alone cannot provide enough information to completely understand biological processes in complex diseases. For example, drug metabolic processes involve multiple types of biomolecules, such as DNA, RNA, proteins, and other metabolites. It is now believed that a comprehensive analysis of different layers of omics data will provide more accurate conclusions and insights into mechanisms when studying biological processes. Therefore, integrative analytical methods of multi-omics have been designed to facilitate such joint studies. These studies usually take genomic, transcriptomic, epigenomic, proteomic, and metabolomic information into account to study underlying mechanisms and identify biomarkers of interest. Currently, there are a number of commonly used methods to perform the joint analysis of multi-omics. As discussed in the following paragraphs, one method is built based on the central dogma, while the other focuses on the change in environment from multiple aspects.
The first method combines genomic, transcriptomic, and proteomic data. For example, Starokadomskyy et al found an intron mutation of POLA1 (SNP rs24744696 A to G) in X-linked reticulate pigmentary disorder patients using genome-wide association analysis. Transcriptomic analysis then revealed that the expression of POLA1 in X-linked reticulate pigmentary disorder cells is downregulated, suggesting that the mutation causes this gene downregulation. Gandal et al successfully identified 64 risk genes of schizophrenia using a comprehensive analysis of genomic data and a transcriptome-wide association study. Additionally, Amiri et al identified 228 previously reported schizophrenia-related SNPs in 158 enhancers using RNA-sequencing, snRNA-sequencing, ChIP-sequencing, and chromatin conformation data. Meng et al reported that schizophrenia-related copy number variations and long noncoding RNAs in the protein-coding gene DGCR5 affected schizophrenia risk. In addition, a recent study demonstrated that macrophage migration inhibitory factor was significantly associated with fibrogenesis and fibrosis, which has been regarded as a potential new drug target. In this study, the transcriptional levels of 121 candidate genes related to hepatic fibrosis were analyzed along with their neighboring genes; results revealed biological differences across individuals with hepatic fibrosis.
Another method takes the influence of environmental and metabolic processes into consideration by analyzing both metabolomic and epigenomic data. For example, 21 common metabolic variations and 12 functional rare metabolic variations were found in Middle Eastern populations using whole exome sequencing and high-resolution metabolomics analysis (Metabolon's Discovery HD4 platform). Additionally, multiple studies have attempted to combine transcriptomic data with epigenomic data to study the relationship between methylation sites and gene expression. In these studies, a number of methylation sites (with varied gene expression levels) were identified as associated with Alzheimer's disease. Another study found that arhinia was related to SMCHD1 mutations using a new approach that combined whole exome sequencing, whole genome sequencing, DNA methylation, and transcriptomics.
Multi-omics has helped to broaden our understanding of genes, protein activities, and protein functions, and may contribute to relevant studies of pathology and pharmacodynamics. It is important that information obtained via multi-omics studies is complementary rather than antagonistic. Some multi-omics data may not be able to be directly explained. For instance, researchers are unable to observe RNA-level changes caused by gene mutations. Our continuously improving understanding of biological mechanisms (such as the degradation process) may serve as a key tool to link these different levels of biological data. For example, a recent study of genome data revealed that a variant of CHADL (SNP rs532464664) was associated with hip osteoarthritis. In contrast, this variant was only associated with the expression of short non-coding RNA in adipose tissue, and no short non-coding RNA was detected in joint tissue in another study. These inconsistent conclusions from adipose tissue and joint tissue may be explained by the idea that this rs532464664 SNP induced the abnormal degradation of nonsense-mediated decay in the joint tissue. Therefore, understanding the details of relevant biological mechanisms is crucial to combine and jointly analyze multi-omic data.
Appropriate integration of multi-omic data is a substantial challenge. As discussed previously, our knowledge of the underlying biological mechanisms provides critical clues to analyze and explain multi-omics data. Therefore, it is necessary to continue gaining a better understanding of biological mechanisms and other relevant properties, such as the physiochemical structures of the biological molecules of interest. In addition, the corresponding mathematical models and calculation methods customized for multi-omics data are crucial to obtain accurate conclusions. A number of computational tools (for example, Multiple Dataset Integration, Similarity Network Fusion, and jActiveModules) have been developed to facilitate such integrative analysis of multiple levels of biological data. However, more efficient tools still need to be developed to appropriately analyze multi-omic data in various forms, to solve specific computational challenges in studies with certain goals. For example, multi-omic datasets of interest may be obtained from different tissues. Therefore, more advanced statistical approaches are needed to adjust for such problems according to the subject of the study.
Integrative analyses of multi-omics have become increasingly popular to study complete biological processes and search for biomarkers. At the same time, they pose many challenges in the integration of these biological data at different levels.[67,76] A significant integrative effort is therefore needed to leverage all the different data, to better understand biological mechanisms using a multi-omics approach.
This review discusses the roles of current genomic, transcriptomic, epigenomic, proteomic, and metabolomic investigations in improving the diagnosis and treatment of schizophrenia. Discoveries of novel genetic markers from multi-omic studies can be used to predict drug targets, reclassify drugs, predict drug efficacy, and prevent relevant ADRs. The development of multi-omics has enabled the prediction of pathogenic processes and the evaluation of biomarkers for therapeutic interventions. In the future, scientists may be able to develop personalized clinical tests for schizophrenia based on multi-omic data. Although genomics has successfully shifted the method for discovering drug targets from experience-based or accidental to systematic scientific research, genomics has limitations and cannot fully explain pathogenesis and drug metabolism processes. This is because RNA transcription and protein expression are regulated at multiple levels, including splicing, post-transcriptional regulation, and post-translational modification, as well as by the environment. Therefore, multi-omics are needed to gain a more comprehensive understanding of schizophrenia and the metabolism of antipsychotics, so that new drug targets can be discovered.
Currently, studies of genomics, transcriptomics, epigenomics, proteomics, or metabolomics alone cannot fully explain individual differences in treatment responses in complex diseases such as schizophrenia. Studies with multi-levels of omics data are helpful to accelerate the discovery of novel biomarkers of antipsychotic drug effects and relevant ADRs. Integrative multi-omic studies provide an effective approach to identify drug targets, improve drug efficacy, and predict drug response in schizophrenia. They will also facilitate the advancement of future personalized medicine design.
SQ, LH, and HH conceived the review. YS wrote the manuscript. WZ, LC, and CH revised the manuscript. All authors approved the final version of the paper.
This work was supported by grants from the 863 Program (No. 2012AA02A515, 2012AA021802), the National Natural Science Foundation of China (No. 81773818, 81273596, 30900799, 81671326), the National Key Research and Development Program of China (No. 2017YFC0909303, 2016YFC0905000, 2016YFC0905002, 2016YFC1200200, 2016YFC0906400), the 4th Three-year Action Plan for Public Health of Shanghai, China (No. 15GWZK0101), Shanghai Pujiang Program, China (No. 17PJD020), Shanghai Key Laboratory of Psychotic Disorders, China (No. 13dz2260500).
Conflicts of interest
The authors declare that they have no conflicts of interest.
. Carpenter WT Jr, Buchanan RW. Schizophrenia
. N Engl J Med 1994;330:681–690.
. Insel TR. Rethinking schizophrenia
. Nature 2010;468:187–193.
. Wiersma D, Wanderling J, Dragomirecka E, et al Social disability in schizophrenia
: its development and prediction over 15 years in incidence cohorts in six European centres. Psychol Med 2000;30:1155–1167.
. van Os J, Kapur S. Schizophrenia
. Lancet 2009;374:635–645.
. Wu EQ, Birnbaum HG, Shi L, et al The economic burden of schizophrenia
in the United States in 2002. J Clin Psychiatry 2005;66:1122–1129.
. Combating schizophrenia
. Nature 2010;468:133.
. Kapur S, Mamo D. Half a century of antipsychotics
and still a central role for dopamine D2 receptors. Prog Neuropsychopharmacol Biol Psychiatry 2003;27:1081–1090.
. Mueser KT, McGurk SR. Schizophrenia
. Lancet 2004;363:2063–2072.
. Kahn RS, Sommer IE, Murray RM, et al Schizophrenia
. Nat Rev Dis Primers 2015;1:15067.
. Zhang JP, Gallego JA, Robinson DG, et al Efficacy and safety of individual second-generation vs first-generation antipsychotics
in first-episode psychosis: a systematic review and meta-analysis. Int J Neuropsychopharmacol 2013;16:1205–1218.
. He X, Wu J, Jiang Y, et al Health care resource utilization and direct medical costs for patients with schizophrenia
initiating treatment with atypical versus typical antipsychotics
in Tianjin, China. BMC Health Serv Res 2015;15:149.
. Ruderfer DM, Charney AW, Readhead B, et al Polygenic overlap between schizophrenia
risk and antipsychotic response: a genomic medicine approach. Lancet Psychiatry 2016;3:350–357.
. Köster LS, Carbon M, Correll CU. Emerging drugs for schizophrenia
: an update. Expert Opin Emerg Drugs 2014;19:511–531.
. Jiang ML, Wang LT, Chen SA, et al Association between HLA-A*02:07 allele associates with clarithromycin-induced cutaneous adverse drug reactions in Chinese patients. Basic Clin Pharmacol Toxicol 2018;1:73–78.
. Yip VL, Hawcutt DB, Pirmohamed M. Pharmacogenetic markers of drug efficacy
and toxicity. Clin Pharmacol Ther 2015;98:61–70.
. Leucht S, Corves C, Arbter D, et al Second-generation versus first-generation antipsychotic drugs for schizophrenia
: a meta-analysis. Lancet 2009;373:31–41.
. Snyder SH. The dopamine hypothesis of schizophrenia
: focus on the dopamine receptor. Am J Psychiatry 1976;133:197–202.
. Shaw E, Woolley DW. Some serotoninlike activities of lysergic acid diethylamide. Science 1956;124:121–122.
. Tsai G, Coyle JT. Glutamatergic mechanisms in schizophrenia
. Annu Rev Pharmacol Toxicol 2002;42:165–179.
. Bitanihirwe BK, Woo TU. Oxidative stress in schizophrenia
: an integrated approach. Neurosci Biobehav Rev 2011;35:878–893.
. Amato D, Vernon AC, Papaleo F. Dopamine, the antipsychotic molecule: A perspective on mechanisms underlying antipsychotic response variability. Neurosci Biobehav Rev 2018;85:146–159.
. Nelson MR, Tipney H, Painter JL, et al The support of human genetic evidence for approved drug indications. Nat Genet 2015;47:856–860.
. Evans WE, Relling MV. Pharmacogenomics: translating functional genomics into rational therapeutics. Science 1999;286:487–491.
. Corvin A, Craddock N, Sullivan PF. Genome-wide association studies: a primer. Psychol Med 2010;40:1063–1077.
. Lam M, Chen CY, Li Z, et al Comparative genetic architectures of schizophrenia
in East Asian and European populations. BioRxiv 2018;445874.
. Walden LM, Brandl EJ, Tiwari AK, et al Genetic testing for CYP2D6 and CYP2C19 suggests improved outcome for antidepressant and antipsychotic medication. Psychiatry Res 2019;279:111–115.
. Goldstein JI, Jarskog LF, Hilliard C, et al Clozapine-induced agranulocytosis is associated with rare HLA-DQB1 and HLA-B alleles. Nat Commun 2014;5:4757.
. Luo C, Liu J, Wang X, et al Pharmacogenetic correlates of antipsychotic-induced weight gain in the Chinese population. Neurosci Bull 2019;35:561–580.
. Zhang JP, Lencz T, Zhang RX, et al Pharmacogenetic associations of antipsychotic drug-related weight gain: a systematic review and meta-analysis. Schizophr Bull 2016;42:1418–1437.
. Xu Q, Wu X, Xiong Y, et al Pharmacogenomics can improve antipsychotic treatment in schizophrenia
. Front Med 2013;7:180–190.
. Jukic MM, Smith RL, Haslemo T, et al Effect of CYP2D6 genotype on exposure and efficacy of risperidone and aripiprazole: a retrospective, cohort study. Lancet Psychiatry 2019;6:418–426.
. Rajagopal VM, Rajkumar AP, Jacob KS, et al Gene-gene interaction between DRD4 and COMT modulates clinical response to clozapine in treatment-resistant schizophrenia
. Pharmacogenet Genomics 2018;28:31–35.
. Cho M, Contreras A, Garza A, et al The impact of drug and gene interaction on the antipsychotic medication for schizophrenia
. Bipolar Disord 2017;3:1.
. Yu H, Yan H, Wang L, et al Five novel loci associated with antipsychotic treatment response in patients with schizophrenia
: a genome-wide association study. Lancet Psychiatry 2018;5:327–338.
Working Group of the Psychiatric Genomics Consortium. Biological insights from 108 schizophrenia
-associated genetic loci. Nature 2014;511:421–427.
. Xiao X, Yu H, Li J, et al Further evidence for the association between LRP8 and schizophrenia
. Schizophr Res 2017;doi: 10. 1016/j. schres. 2017. 05. 002.
. Mastriani E, Zhai R, Zhu S. Microarray-based microRNA expression data analysis with bioconductor. Methods Mol Biol 2018;1751:127–138.
. Zygmunt M, Piechota M, Rodriguez Parkitna J, et al Decoding the transcriptional programs activated by psychotropic drugs in the brain. Genes Brain Behav 2019;18:e12511.
. Sakuma K, Komatsu H, Maruyama M, et al Temporal and spatial transcriptional fingerprints by antipsychotic or propsychotic drugs in mouse brain. PLoS One 2015;10:e0118510.
. GTEx Consortium. Human genomics. The Genotype-Tissue Expression (GTEx) pilot analysis: multitissue gene regulation in humans. Science 2015;348:648–660.
. Crespo-Facorro B, Prieto C, Sainz J. Schizophrenia
gene expression profile reverted to normal levels by antipsychotics
. Int J Neuropsychopharmacol 2014;18:yu066.
. Readhead B, Hartley BJ, Eastwood BJ, et al Expression-based drug screening of neural progenitor cells from individuals with schizophrenia
. Nat Commun 2018;9:4412.
. Zhang F, Yao Shugart Y, Yue W, et al Increased variability of genomic transcription in schizophrenia
. Sci Rep 2015;5:17995.
. Enwright Iii JF, Huo Z, Arion D, et al Transcriptome alterations of prefrontal cortical parvalbumin neurons in schizophrenia
. Mol Psychiatry 2018;23:1606–1613.
. Klarer M, Krieger JP, Richetto J, et al Abdominal vagal afferents modulate the brain transcriptome and behaviors relevant to schizophrenia
. J Neurosci 2018;38:1634–1647.
. Steiner J, Guest PC, Martins-de-Souza D. Application of proteomic techniques for improved stratification and treatment of schizophrenia
patients. Adv Exp Med Biol 2017;974:3–19.
. Kageyama Y, Kasahara T, Morishita H, et al Search for plasma biomarkers in drug-free patients with bipolar disorder and schizophrenia
using metabolome analysis. Psychiatry Clin Neurosci 2017;71:115–123.
. Café-Mendes CC, Ferro ES, Torrão AS, et al Peptidomic analysis of the anterior temporal lobe and corpus callosum from schizophrenia
patients. J Proteomics 2017;151:97–105.
. Cassoli JS, Guest PC, Santana AG, et al Employing proteomics to unravel the molecular effects of antipsychotics
and their role in schizophrenia
. Proteomics Clin Appl 2016;10:442–455.
. Paredes RM, Quinones M, Marballi K, et al Metabolomic profiling of schizophrenia
patients at risk for metabolic syndrome. Int J Neuropsychopharmacol 2014;17:1139–1148.
. Carboni L, Domenici E. Proteome effects of antipsychotic drugs: learning from preclinical models. Proteomics Clin Appl 2016;10:430–441.
. Cao B, Jin M, Brietzke E, et al Serum metabolic profiling using small molecular water-soluble metabolites in individuals with schizophrenia
: A longitudinal study using a pre-post-treatment design. Psychiatry Clin Neurosci 2019;73:100–108.
. Farrelly L, Rosato-Siri MV, Föcking M, et al The effects of prenatal iron deficiency and risperidone treatment on the rat frontal cortex: a proteomic analysis. Proteomics 2017;17:1600407.
. Cai HL, Jiang P, Tan QY, et al Therapeutic efficacy of atypical antipsychotic drugs by targeting multiple stress-related metabolic pathways. Transl Psychiatry 2017;7:e1130.
. Akbarian S. Epigenetic mechanisms in schizophrenia
. Dialogues Clin Neurosci 2014;16:405–417.
. Swathy B, Banerjee M. Understanding epigenetics of schizophrenia
in the backdrop of its antipsychotic drug therapy. Epigenomics 2017;9:721–736.
. Ovenden ES, McGregor NW, Emsley RA, et al DNA methylation and antipsychotic treatment mechanisms in schizophrenia
: Progress and future directions. Prog Neuropsychopharmacol Biol Psychiatry 2018;81:38–49.
. Bosia M, Pigoni A, Cavallaro R. Genomics and epigenomics in novel schizophrenia
drug discovery: translating animal models to clinical research and back. Expert Opin Drug Discov 2015;10:125–139.
. Karsli-Ceppioglu S. Epigenetic mechanisms in psychiatric diseases and epigenetic therapy. Drug Dev Res 2016;77:407–413.
. Miura I, Kunii Y, Hino M, et al DNA methylation of ANKK1 and response to aripiprazole in patients with acute schizophrenia
: A preliminary study. J Psychiatr Res 2018;100:84–87.
. Guidotti A, Grayson DR. DNA methylation and demethylation as targets for antipsychotic therapy. Dialogues Clin Neurosci 2014;16:419–429.
. Santoro ML, Ota VK, Stilhano RS, et al Effect of antipsychotic drugs on gene expression in the prefrontal cortex and nucleus accumbens in the spontaneously hypertensive rat (SHR). Schizophr Res 2014;157:163–168.
. Peedicayil J. Preclinical epigenetic models for screening epigenetic drugs for schizophrenia
. J Pharmacol Toxicol Methods 2016;77:1–5.
. Abdolmaleky HM, Pajouhanfar S, Faghankhani M, et al Antipsychotic drugs attenuate aberrant DNA methylation of DTNBP1 (dysbindin) promoter in saliva and post-mortem brain of patients with schizophrenia
and Psychotic bipolar disorder. Am J Med Genet B Neuropsychiatr Genet 2015;168:687–696.
. Macaulay IC, Ponting CP, Voet T. Single-cell multiomics: multiple measurements from single cells. Trends Genet 2017;33:155–168.
. Sun YV, Hu YJ. Integrative analysis of multi-omics
data for discovery and functional studies of complex human diseases. Adv Genet 2016;93:147–190.
. Meng C, Zeleznik OA, Thallinger GG, et al Dimension reduction techniques for the integrative analysis of multi-omics
data. Brief Bioinform 2016;17:628–641.
. Starokadomskyy P, Gemelli T, Rios JJ, et al DNA polymerase-( regulates the activation of type I interferons through cytosolic RNA:DNA synthesis. Nat Immunol 2016;17:495–504.
. Gandal MJ, Zhang P, Hadjimichael E, et al Transcriptome-wide isoform-level dysregulation in ASD, schizophrenia
, and bipolar disorder. Science 2018;362:eaat8127.
. Amiri A, Coppola G, Scuderi S, et al Transcriptome and epigenome landscape of human cortical development modeled in organoids. Science 2018;362:eaat6720.
. Meng Q, Wang K, Brunetti T, et al The DGCR5 long noncoding RNA may regulate expression of several schizophrenia
-related genes. Sci Transl Med 2018;10:eaat6912.
. Liu Z, Chalasani N, Lin J, et al Integrative omics
analysis identifies macrophage migration inhibitory factor signaling pathways underlying human hepatic fibrogenesis and fibrosis. J Bio-X Res 2019;2:16–24.
. Yousri NA, Fakhro KA, Robay A, et al Whole-exome sequencing identifies common and rare variant metabolic QTLs in a Middle Eastern population. Nat Commun 2018;9:333.
. Shaw ND, Brand H, Kupchinsky ZA, et al SMCHD1 mutations associated with a rare muscular dystrophy can also cause isolated arhinia and Bosma arhinia microphthalmia syndrome. Nat Genet 2017;49:238–248.
. Lozupone M, Seripa D, Stella E, et al Innovative biomarkers in psychiatric disorders: a major clinical challenge in psychiatry. Expert Rev Proteomics 2017;14:809–824.
. Styrkarsdottir U, Helgason H, Sigurdsson A, et al Whole-genome sequencing identifies rare genotypes in COMP and CHADL associated with high risk of hip osteoarthritis. Nat Genet 2017;49:801–805.
. Huang S, Chaudhary K, Garmire LX. More is better: recent progress in multi-omics
data integration methods. Front Genet 2017;8:84.
. Ramos M, Schiffer L, Re A, et al Software for the integration of multiomics experiments in bioconductor. Cancer Res 2017;77:e39–e42.