Linking Y‐chromosomal short tandem repeat loci to human male impulsive aggression

Abstract Introduction Men are more susceptible to impulsive behavior than women. Epidemiological studies revealed that the impulsive aggressive behavior is affected by genetic factors, and the male‐specific Y chromosome plays an important role in this behavior. In this study, we investigated the association between the impulsive aggressive behavior and Y‐chromosomal short tandem repeats (Y‐STRs) loci. Methods The collected biologic samples from 271 offenders with impulsive aggressive behavior and 492 healthy individuals without impulsive aggressive behavior were amplified by PowerPlexRY23 PCR System and the resultant products were separated by electrophoresis and further genotyped. Then, comparisons in allele and haplotype frequencies of the selected 22 Y‐STRs were made in the two groups. Results Our results showed that there were significant differences in allele frequencies at DYS448 and DYS456 between offenders and controls (p < .05). Univariate analysis further revealed significant frequency differences for alleles 18 and 22 at DYS448 (0.18 vs 0.27, compared to the controls, p = .003, OR=0.57,95% CI=0.39–0.82; 0.03 vs 0.01, compared to the controls, p = .003, OR=7.45, 95% CI=1.57–35.35, respectively) and for allele 17 at DYS456 (0.07 vs 0.14, compared to the controls, p = .006, OR=0.48, 95% CI =0.28–0.82) between two groups. Interestingly, the frequency of haploid haplotype 22‐15 on the DYS448‐DYS456 (DYS448‐DYS456‐22‐15) was significantly higher in offenders than in controls (0.033 vs 0.004, compared to the control, p = .001, OR = 8.42, 95%CI =1.81–39.24). Moreover, there were no significant differences in allele frequencies of other Y‐STRs loci between two groups. Furthermore, the unconditional logistic regression analysis confirmed that alleles 18 and 22 at DYS448 and allele 17 at DYS456 are associated with male impulsive aggression. However, the DYS448‐DYS456‐22‐15 is less related to impulsive aggression. Conclusion Our results suggest a link between Y‐chromosomal allele types and male impulsive aggression.


| INTRODUCTION
Human aggressive behavior is a behavior of physical, psychological, or verbal violence against other persons with hostile, threat, harmful, or destructive components. Impulsive aggressive behavior is a non-premeditated aggressive act that an individual engages in hastily without any consideration of its consequences (Coccaro, Lee, & McCloskey, 2014;Siever, 2008). It usually occurs as an exaggerated response to emotion-provoking events, leading to undesirable consequences. Individuals with impulsive aggressive behavior often have deficit in controlling their impulses. Most of them feel regret after their impulsive aggression. The behavior has complex biologic and social causes, and often results from the interactions of genetic and environmental factors (Chester et al., 2015). There are significant sex differences in response to stressful events. For example, men are more susceptible to impulsive behavior than women (Lovell-Badge, 2005).
Epidemiological studies revealed that the impulsive aggressive behavior is affected by genetic factors (Craig & Halton, 2009), and the male-specific Y chromosome plays an important role in this behavior (Shah, Ayub, Firasat, Kaiser, & Mehdi, 2009). Several studies (Carey, 1992;Hesselbrock, 1991) showed that aggressive/violent behavior is associated with Y chromosome. Shah et al. (2009) found that haploid R2 and R1a1 on the human Y chromosome may be linked to the aggressive behavior. Lee and Harley (2012) proposed that male-specific sex-determining region on the Y chromosome (SRY), as genetic basis for sex-related responses, has a direct relationship with men impulsive aggressive behavior and control this behavior by regulating the release of catecholamines hormones (such as noradrenaline), 5-HT and DA. Recently, Yang et al. (2013Yang et al. ( , 2015 showed that there is an association between euchromosome STR loci (TH1 and TPOX) on the non-Y-chromosome and male impulsive violent behavior. However, the association between Y-chromosomal STR loci and the male impulsive aggressive behavior has not been reported. In this study, we used

| Subjects
The criminal records of all male offenders arrested from January 2003 to December 2014 were obtained and viewed from local police stations in Changzhou, Jiangsu Province. The repeatedly aggressive behaviors (such as fighting, killing, robbing, injuring, and assaulting) of these offenders resulted in either wounding (i.e., worse than minor injury) or death to the victims. By reviewing the criminal documents, we initially screened offenders with a response to spontaneous angerprovoking stimuli but no premeditation (regarded as impulsive violent behavior in contrast to premeditated aggression). The classification was further confirmed by the local Bureau of Public Security and welltrained psychiatrists from our hospital, a psychiatric center of Chinese People's Liberation Army (PLA).
The inclusion criteria for this study included (i) age ranged 10-80 years; (ii) a document of impulsive violent behavior. Offenders who had any of the following conditions were also excluded from the recruitment: (i) a serious mental disorder based on the psychiatrist's diagnosis or based on a history of a mental disorder that had previously been diagnosed by a psychiatrist; (ii) a history of premeditated aggressive behavior; (iii) a history of a head injury; (iv) an obvious physical illness; or (v) a history of substance abuse problems. A total of 814 male offenders with impulsive violent behavior were initially screened by telephone. Among them, 312 did not respond, and 183 refused to participate in this study. Subsequently, the voluntary participants were interviewed by two well-trained psychiatrists. In all, 48 offenders suffered from mental illness (16 with schizophrenia, 20 with bipolar disorders, and 12 with depression) were consistently diagnosed, and excluded in this study. Following this screening procedure, 271 male individuals with impulsive violent behavior were enrolled and signed informed consent in this study. The tools which were used for impulsive aggressive behavior involved the followings: sharp tools (216 males), blunt tools (37 males), hands (15 males), mouth bite (2 males), and 1 with firearm tool (1 male). Meanwhile, aggressive behaviors of these offenders were assessed with modified overt aggressive scale (MOAS) as described below by an experienced psychiatrist. The scale for aggressive behavior MOAS is widely used to assess impulsive aggressive behavior (Kay, Wolkenfeld, & Murrill, 1988

| Sample collection
5 ml peripheral venous blood from each participant was collected in ethylene diamine tetraacetic acid (EDTA) tubes, stored at −80°C until use.

| DNA preparation
Genomic DNA was extracted from white blood cell fractions of all samples using the chelex-100 protocol (Walsh, Metzger, & Higushi, 2013). Specifically, added 3 μl blood sample into 0.6 ml centrifuge tube, and then added 0.5 ml sterile deionized water (dH 2 O) at room temperature for 30 min; centrifuged at 12,000r/min for 5 min, then discarded the supernatant. Repeated the above steps one time, added 150 μl 5% Chlex-100 suspension; incubated 56°C for 30 min; vortexed and centrifuged at 12,000r/min for 5 min. Finally, the genomic DNA in Y23 System provides all materials necessary to amplify Y-STR regions of human genomic DNA, including a hot-start DNA polymerase (Thompson et al., 2013). The haploid composed by 22 Y-STR loci can be conservatively transmitted in the same paternal (with an exception of mutation), and the linked coding region information also will be transmitted. The related information of Y-STR loci is seen in Table 1.

| PCR amplification
The PowerPlex ® Y23 System (Promega, Wisconsin, USA), a multiplex amplification system with fluorescent detection, was used in this study.
The amplification was performed following the manufacturer's instructions. Extracted DNA was amplified in a total reaction volume of 10 μl,

| Statistical analysis
SPSS software version 19.0 (SPSS Inc., Chicago, IL, USA) was used to perform statistical comparisons between two groups, and the

| Allele frequencies at 22 Y-STR loci in offenders and controls
Characteristics of offenders and controls are summarized in Table 2.
No significant differences were observed in mean age (p = .156), marital status (p = .447), educational level (p = .124), or geographical region (p = .28) between the two groups. For MOAS test, all offenders showed physical aggression against other people (MOAS score >0), with or without verbal aggression and physical aggression against objects. In addition, all recruited offenders lacked expression of physical aggression against self. The total MOAS mean score of the offenders was 21.2. In contrast, because individuals with any aggressive behavior were excluded in the control group, the total MOAS score of the controls was 0.
We first compared allele frequencies of 22 STR loci between offenders and controls by chi-square test. As shown in Tables 3 and 4 and

| Haploid haplotype frequencies at DYS448-DYS456 in offenders and controls
As seen in Table 5, there was a significant difference in the frequency of haploid haplotype DYS448-DYS456 between the two groups (p < .05). Specifically, the frequency of DYS448-DYS456-22-15 in offenders and controls was 3.32% and 0.41%, respectively, showing a significant difference (p < .05/18). In contrast, there were no significant differences in the frequencies of other haploid haplotype DYS448-DYS456 between two groups.

| Multivariate logistic regression analysis on influential factors of impulsive aggressive behavior
To adjust the effect of confounding factor, the unconditional logis- Unfortunately, the DYS448-DYS456 haplotypes appear to be less related to the impulsive aggression (data not shown).

| DISCUSSION
This is the first study to investigate the association between Y-STR loci and human male impulsive aggression. By comparing allele and haplotype frequencies of the selected 22 Y-STRs between male offenders with impulsive aggression and controls, and further association analysis, we found that alleles 18 and 22 at DYS448 and allele 17 at DYS456 are associated with impulsive aggression.
Impulsive aggressive behavior often leads to serious consequences such as crime due to its uncontrollability and lack of rational judgment. Men tend to be more aggressive than women, since androgen is thought to play a key role in the impulsive aggressive behavior (de Almeida, Cabral, & Narvaes, 2015). Aluja, Garcia, Blanch, and Fibla (2011) found that men with less CAG repeats and more GGN repeats in androgen receptor gene was prone to exhibit aggressive behavior, suggesting that there is a correlation between polymorphisms of androgen receptor gene and the aggressive behavior. There is evidence (Carey, 1992;Hesselbrock, 1991)  gene in the Xp11-23~11-4 region (Manuck, Flory, Ferrell, Mann, & Muldoon, 2000), and estrogen receptor alpha gene (Westberg et al., 2003). Taken together, these findings indicate a significant genetic basis. Thus, it is necessary to expand the research scope of candidate genes for aggression.
Y chromosome is the smallest one in the human genome, accounting for 2%-3% of the entire haploid genome. 95% of Y chromosome The number in parentheses indicates frequency (%); allele frequencies with lower than 1% in both groups were removed; the significant test level is set at 0.05/6 ≈ 0.0083. The number in parentheses indicates frequency (%); allele frequencies with lower than 1% in both groups were removed; the significant test level was set at 0.05/6 ≈ 0.0083.
T A B L E 4 Comparisons of allele frequencies of DYS456 locus between offenders and controls genomes are male-specific, containing 23 million base pairs of malespecific euchromosome (Skaletsky et al., 2003).Y chromosome genetic analysis plays an important role in the research of human male evolution, complementary medicine, and forensic genetics (Jobling & Tyler-Smith, 2003). STR generally located within the non-coding regions of DNA throughout the human genome are repeating DNA sequences that contain 2-6 base pairs (Edwards, Civitello, Hammond, & Caskey, 1991 T A B L E 6 Significant association between allele types and impulsive aggression by unconditional logistic regression analysis Although the frequency of haploid haplotype 22-15 on the DYS448-DYS456 (DYS448-DYS456-22-15) was significantly higher in offenders than in controls, the multivariate regression analysis failed to show that the DYS448-DYS456-22-15 is associated with the aggression.
Based on these results, we concluded that DYS448-22 is susceptible to male impulsive aggressive behavior, whereas DYS456-17 and DYS448-18 are resistant to this behavior. Kittles et al. (1999) conducted one study on the association between alcoholism antisocial aggressive behavior and 7 Y-STR loci(DYS388, DYS389, DYS390, DYS391, DYS392, DYS393, and DYS394)in Finland, but failed to find any association. In their study, less Y-STR loci were used, without DYS448 and DYS456 leading to our positive results. Lee and Harley (2012) proposed the following possible mechanisms for male impulsive aggressive behaviors: SRY promotes the release of catecholamine hormones into the blood and further into the peripheral organs and muscles, inhibits monoamine oxidase A in the prefrontal brain, enhances physical activity by boosting the concentration of TH and DA in the substantia nigra, as well as improves the secretion of norepinephrine in the brain locus. Additionally, there is evidence that SRY downregulates the expression of the estrogen receptor, and in turn attenuates the inhibitory effects of estrogen on male impulsive behavior (Tao et al., 2012). DYS448 and DYS456 loci on the Y chromosome may be linked to SRY (Beltramo, Pena, & Lojo, 2015;Gopinath et al., 2013;Liu, Qiao, Li, Yan, & Chen, 2013).
However, whether the two loci affects male impulsive aggressive behavior by above mechanisms needs to be further investigated.
In summary, our results suggest that DYS448 and DYS456 loci are associated with male impulsive aggressive behavior. Specifically, DYS448-22 may be susceptible factors of male impulsive aggressive behavior, whereas DYS448-18 and DYS456-17 may be protective factors. Due to the limited number of samples and rare frequency of candidate alleles and haplotypes, we should be careful with our conclusions. Therefore, our findings need to be further confirmed in a larger-scale study with thousands or ten thousands of samples. Again, it is unclear how these genetic markers affect this behavior. Further research will require a great increase in the sample number, search specific candidate genes on these genetic regions, and further address the potential mechanisms.