Methamphetamine (METH), a well-known powerful central nervous psychostimulant, is currently the most commonly used illicit drug in China (1). METH can cause damage to multiple organs, such as the heart, gut, and brain (2). Meanwhile, individuals with a history of chronic METH use are likely to develop METH-related psychosis, including auditory hallucinations and paranoid thinking (3). METH dependence is a huge financial burden on people with addiction and their families. In addition, it can provoke a series of violent events, leading to several social problems. Although the precise mechanism underlying METH dependence remains unknown, METH-induced changes in gene expression were shown to be closely related to severe dysregulation of normal neurophysiological brain activity. More recently, genome-wide association studies have been extensively employed to detect correlations between genetic variants and complex diseases, such as schizophrenia, coronary heart disease, and height in samples from various populations (4, 5). Numerous loci that influence complex traits, such as alcohol and other substance abuse, have been identified, facilitating our understanding of underlying molecular mechanisms (6–8). However, data on METH dependence remains scarce despite multiple reports of candidate genes (9).

Epigenetic alterations can lead to persistent structural chromatin adaptations, indicating that epigenetic modifications may play a critical role in the METH-induced gene expression changes (10, 11). Furthermore, abnormal microRNA expression is known to mediate METH dependence (12). Histone deacetylases (HDACs) are proteins involved in histone acetylation and can be classified into four classes according to the homologous similarity of their sequences. Class I, including HDAC1, HDAC2, HDAC3, and HDAC8, is considered critical for transcriptional repression and epigenetic modulation (13). As HDAC3 is highly expressed in the adult brain, its effects on transcriptional regulation related to learning and memory have attracted considerable attention (14). In the METH-induced conditioned place preference (CPP) model, an increased histone 3 acetylation was identified in the limbic forebrain of mice and found in specific gene-promoter regions related to synaptic plasticity, such as Nrxn, Gria1, Grin2a, and Grin2b, thus indicating its essential role in METH dependence (15). An increased expression of HDAC3 mRNA has been noted in Fos-positive neurons in the dorsomedial striatum after withdrawal following METH self-administration (16). Studies have shown that sodium butyrate, a non-selective inhibitor of class I/II HDACs, can help overcome a previously established CPP and suppress the reinstatement of METH-induced CPP (17). These findings suggest that HDAC3 may be a key molecule for regulating METH-associated gene expression and clarifying the underlying biological mechanism. However, the relationship between the HDAC3 gene and METH dependence remains elusive. In this study, we aimed to systematically explore the risk susceptibility of HDAC3 to METH dependence among Han Chinese population.

A variety of neurotransmitter and receptor system-related genes, including BDNF, DRD2, and GABRB2, were found to contain genetic variants that contribute to the vulnerability of METH use disorder (9). Although multiple lines of evidence have associated epigenetic alterations with METH dependence (25, 26), no relevant population-based studies focusing on epigenetic-related genes have been conducted. To the best of our knowledge, this study is the first to link DNA variants of HDAC3 and vulnerability to METH dependence in a population-based study. Recently, Rudzinskas et al. have indicated that METH alters HDAC and DNA methyltransferase (DNMT) activity in the posterior dorsal medial amygdala of rats (26); therefore, our findings align with these results. Genetic markers of specific genes may modify the effects of METH on human epigenetic patterns.

Significant associations between haplotypes and vulnerability to METH dependence have been previously reported. Both SNPs rs56221992 and rs11741808 showed moderate levels of LD with rs14251 (r² = 0.3 for rs56221992 and r² = 0.3 for rs11741808, respectively). Although we cannot exclude the possibility that this haplotypic association signal occurred by chance or originated from other underlying independent association signals, it could probably arise from SNP rs14251.

Single nucleotide polymorphism rs14251 is located in the 3′-untranslated region of HDAC3. It is also located in the exonic region of RELL2 as a non-synonymous change. This double identity increases the complexity of predicting its functional consequences. Both SIFT and PolyPhen-2 predicted that this SNP exerts mild functional consequences on the protein structure of RELL2. Nevertheless, our in silico analyses revealed that SNP rs14251 could affect the expression levels of RELL2, FCHSD1, and PCDHGB6 (but not that of HDAC3) in various kinds of human tissues. A recent study has linked HDAC3 expression to schizophrenia (27). However, neither RELL2 nor FCHSD1 appears to be associated with psychiatric or brain-related traits. RELL2 reportedly encodes receptors expressed in lymphoid tissues such as 2 protein, which participates in the positive regulation of the p38MAPK cascade and is speculated to be a human tumor necrosis factor (28). FCHSD1 encodes F-BAR protein and double SH3 domains protein 1. A recent study by Kawasaki et al. has indicated that the loss of FCHSD1 could ameliorate chronic obstructive pulmonary disease (29). In addition, the third gene, PCDHGB6, encodes a calcium-dependent cell-adhesion protein, potentially related to the activity of specific neuronal connections in the human brain (30). However, this study results are insufficient to functionally map SNP rs14251 to any of the four genes. Therefore, it might be more appropriate to map this SNP to the 5q31.3 genomic region instead of a specific gene.

Although eQTL data in the GTEx database indicated that SNP rs14251 could be mapped to the three surrounding genes, the results of in silico analyses should be cautiously interpreted. First, SNP rs14251 is related to gene expressions in various kinds of human tissues; however, 13 types of brain tissues (e.g., cortex, cerebellar, and others) are not listed. In other words, METH-dependence targeted tissues are not included. In addition, the METH addiction status for patients whose data were obtained from the GTEx database remains largely unknown. A gene expression pattern has specific spatial and temporal features, and it may considerably differ between patients and healthy participants. Accordingly, animal-based functional studies remain crucial to properly map SNP rs14251 to its functionally relevant gene and unravel potential consequences.

The PPI network constructed in this study indicates that HDAC3 interacts with several other genes for implementing its biochemical and biological functions. Among the genes included in this network, PPARG is of particular interest. PPARG is known to encode peroxisome proliferator-activated receptor (PPAR) gamma, which is a member of the PPARs, a subfamily of nuclear receptors (31). A recent genome-association study has linked PPARG genetic polymorphisms to human cognitive measurements (32). As METH-use disorders are widely associated with cognitive functions, this study highlights the complexity of pathogenic mechanisms underlying METH dependence.

In addition, the limitations of this study need to be addressed. Although we have selected SNPs located at the gene region of HDAC3 with MAF >0.05, the genetic information coverage may be insufficient. Recent studies on several psychiatric disorders, including bipolar disorder and nicotine dependence, have indicated that genetic variants with low frequency might substantially contribute to the susceptibility of relevant disorders (33, 34). In the future, low-frequency and rare DNA variants should be explored to exhaustively examine the genetic contribution of HDAC3 to the vulnerability to METH dependence. Replication studies are also needed to rule out the “winner’s curse.” In addition, the controls recruited in this study might not be comparable with METH-dependent individuals. To examine the vulnerability to METH dependence, ideal controls would be those individuals who use METH but are not addicted; this, in turn, might affect the association measures.

In summary, we demonstrated that genetic polymorphism rs14251 located at 5q31.3 was significantly associated with the vulnerability to METH dependence in a Han Chinese population.

According to national legislation/guidelines, specifically the Administrative Regulations of the People’s Republic of China on Human Genetic Resources (http://www.gov.cn/zhengce/content/2019-06/10/content_5398829.htm, http://english.www.gov.cn/policies/latest_releases/2019/06/10/content_281476708945462. htm), no additional raw data is available at this time. Data of this project can be accessed after an approval application to the China National Genbank (CNGB, https://db.cngb.org/cnsa/). Please refer to https://db.cngb.org/, or email: CNGBdb@cngb.org for detailed application guidance. The accession code CNP0002872 should be included in the application.

The studies involving human participants were reviewed and approved by the Medical Ethics Committee of Xi’an Jiaotong University Health Science Center. The patients/participants provided their written informed consent to participate in this study.

FG, TZ, and HLin conceived and designed the study. JX performed the computational analyses and a candidate SNP selection and wrote the first draft of the manuscript. YM and XW performed the statistical analyses of genotype and haplotype data. CW, ML, WH, and HLiu conducted the study subjects’ phenotypic screening. HLiu, HWe, WZ, HWa, and LZ contributed to the collections and preparations of control subjects’ DNA samples. All authors revised the current manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.