Effect of Certain Genetic Variants in Nicotinic AChR on Substance Addictions: A Structural Equation Modeling Approach

doi:10.21203/rs.3.rs-3406839/v1

Download PDF

Research Article

Effect of Certain Genetic Variants in Nicotinic AChR on Substance Addictions: A Structural Equation Modeling Approach

https://doi.org/10.21203/rs.3.rs-3406839/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Nicotine addiction, like many other addictions, is a complex form of behavior with both genetic and environmental aspects. Although there is a need for studies investigating the interaction between environmental and genetic factors in the field of substance addiction, the number of studies that consider both aspects is limited.

This study explored how certain genetic variants in the nicotine acetylcholine receptor (AChR) genes interact with nicotine and other types of substance addiction and environment.

From the genetic data of 66936 variants, scores generated by Polygenic Risk Score (PRS) analysis for certain genetic variants found in nicotinic AChR genes. The relationship between this score and Wisconsin Index of Smoking Dependence Motives (WISDM), socio-economic status (SES), nicotine and other substance addictions was examined using structural equation modeling (SEM). The data set contains information of 2969 subjects (Female (1786) Male (1168)).

PRS had a significant effect on the FTND as expected. PRS also showed a significant effect on alcohol addiction and marijuana addiction with higher coefficients than FTND. SES significantly affects WISDM. WISDM showed a significant effect on alcohol, marijuana, cocaine and opiate addiction. Thus, WISDM was examined as mediator of the relationship between SES and substance addictions. Alcohol addiction significantly affects nicotine, cocaine and marijuana addiction.

These findings suggest that certain genetic variants in nicotinic AChR known to affect one type of addiction may also have an effect on other types of substance addictions. In addition, nicotinic AChR may be a clinically significant target for prevention of substance addictions other than nicotine.

nicotine dependence

substance dependence

socio-economic status

structural equation modelling

polygenic risk score

Today, 30% of the world's population uses tobacco. According to the current data of the World Health Organization (WHO), nearly 8 million people worldwide die each year due to diseases developed as a result of tobacco and tobacco use. For this reason, WHO defines tobacco use as one of the most common public health problems and preventable causes of early death (Vink et al. 2005). Smoking addiction, like many other addictions, is a complex form of behavior that has both genetic and environmental aspects. Genetic transmission in smoking addiction; researched through family studies, twin studies, and molecular genetic studies. In a study, it was stated that nicotine addiction was caused by 60% genetic, 40% environmental, and other factors (Mackillop et al. 2010). Studies have shown that environmental factors play a more prominent role in the beginning to smoking and genetic factors play a more prominent role in the transition from regular smoking to addiction (Bierut 2011, Akgün GA et al. 2011).

Smoking is the leading cause of preventable morbidity and mortality and become an important public health issue that motivates studies to understand the biological processes involved in smoking behaviour (Do 2017, López-Flores et al. 2017). In recent years, a tremendous amount of effort has been put into the study of the molecular mechanisms underlying tobacco use-related behaviors and diseases, and many susceptibility genes have been identified through many genomic approaches (Fang 2017).

Nicotine crosses the blood-brain barrier and stimulates nicotinic receptors. Nicotinic receptors activate the mesolimbic dopamine system responsible for the euphoric and energizing effect of nicotine. Nicotine taken with the use of tobacco products causes upregulation in the ventral tegmental area neuron receptors in the brain, which means an increase in the number and functional capacity (Biala and Weglinska 2004). There are 5 subunits of nicotinic acetylcholine receptors. Of these, the α3, α5, β4 subunit gene clusters are located on chromosome 15 (CHRNA5-CHRNA3-CHRNB4), while the α6 and β3 subunit gene clusters are located on chromosome 8 (CHRNA6-CHRNB3) (The Tobacco and Genetics Consortium 2010). Many genome association studies have shown that the diversity in the CHRNA5-CHRNA3-CHRNB4 nicotinic cholinergic receptor subunit gene clusters located at chromosome 15q24-25.1 is important in the development of nicotine addiction, smoking and lung cancer (Bierut et al. 2006). Genome-wide association studies have found that the diversity of CHRNA5-CHRNA3-CHRNB4 nicotinic acetylcholine receptor genes are effecting nicotine addiction and withdrawal symptoms (Changeux 2010). Another genome association study found correlations between nicotine tolerance, severity of withdrawal symptoms, ability to stop smoking, and CHRNA5-CHRNA3-CHRNB4 gene clusters (Wittenberg et al. 2020). It was stated that these CHRNA5 and CHRNB4 gene clusters were determinants of nicotine withdrawal severity in a 12-week follow-up study (Ray et al. 2010). In pharmacogenetic studies, single point polymorphisms on these genes were found to be determinative in response rates to nicotine replacement therapy (NRT) and varenicline (King et al. 2011).

In this study, causal relationships between nicotine, alcohol, marijuana, opiate, cocaine addiction, SES and the PRS representing genetic aspect were modeled using SEM. As in modeling many complex cases, SEM method was also used in substance addiction modeling (Creswell et al. 2015, Voci 2014, Courtney et al. 2021). The difference of this study is that besides environmental variables, genetic data are also included in the modeling process with PRS analysis. Despite the need of studies explore the interaction between environmental and genetic factors in the field of substance addiction, the number of studies taking into account both these aspects is limited. This study was needed to make up for this deficiency.

We hypothesized that; PRS generated from certain genetic variants known to effect smoking addiction may also have an effect on other substance addictions. Another hypothesis of this study is that, SEM method may reveal some latent casual relationships between substance addictions, WISDM and SES.

2.1. Data Source

For this study, data was obtained from the dbGap database run by NCBI (National Center for Biotechnology Information), which contains genotype and phenotype data sets. The dataset titled “CIDR, NCI, NIDA Sequencing of Targeted Genomic Regions Associated with Smoking” includes examples from two projects: the Collaborative Genetic Study of Nicotine Addiction (COGEND) and the Genetic Study of Nicotine Addiction in African Americans (AAND). Cases are defined by the Nicotine Dependence (FTND) Fagerstrom Test with a valid score of 4 or more. Control status is defined as an individual who smoked at least 100 cigarettes during their lifetime but never became dependent (lifetime FTND < 2). In this dataset, there are 2969 samples containing 117 variables. Molecular data is presented in SNP Genotypes (Array) type and contains information on 66936 variants (NCBI dbGaP Study Accession: phs000813.v1.p1). It is stated in the study page that dataset includes no personal identifying information for research participants and Institutional Review Board (IRB) approval is not required to use it. Also permission to share data and samples was obtained from all subjects in the informed consent form. Informed consent was obtained from all subjects by trained research assistants. Prior to signing the consent form, a research assistant reviewed the form with the subject and answered any questions.

2.2. Measures

Environmental data includes scales measuring the symptoms of nicotine, alcohol, marijuana, cocaine, and opiate, the fagerstorm scale (FTND), the Wisconsin Index of Smoking Dependence Motives (WISDM) scale. Apart from these scales, features related to physical characteristics (age, height, weight) and SES (income, education level) are also available in the data set. PRS variable was created with PRS analysis from genetic data and the effect of this score on the substance addictions were evaluated as given in the hypothesis of this study.

2.3. Missing Data Analysis

Before analysis, data was checked if there are missing values using SPSS 21 software. Missing data analysis showed that, demographical variables had missing values ranging between 0.5% and 6%. Other variables had no missing data. Missing data in demographical variables was handled by using multiple imputation approach because maximum likelihood estimator used in SEM analysis cannot work with missing data.

2.4. PRS Analysis

Polygenic risk scores summarize genome-related data as a single variable that measures genetic susceptibility to a trait. PRS is calculated by summing the number of risk alleles carried by an individual and weighting by the effect size (Lewis and Vassos 2017). The effect size for this study is the odds ratio calculated by Eq. 1 for each SNP. Since the PRS analysis requires independent base and target data, we split the dataset randomly by keeping the case/control ratio constant and got two datasets. Base data which used to calculate odds ratios for selected SNPs consist of 1500 samples and the target data which used to calculate PRS cosist of 1469 samples. The target data also used to explore associations between PRS and phenotypic traits in SEM analysis.

$$OR=\frac{{D}_{E}/{D}_{N}}{{H}_{E}/{H}_{N}} \left(1\right)$$

Here, ${D}_{E}$ represents the number of samples that carry the relevant SNP risk allele and are in the case group according to the measure in the data set, and ${D}_{N}$ represents the number of samples that do not carry the relevant SNP risk allele and are also in the case group. Likewise, ${H}_{E}$ refers to the number of samples that carry relevant SNP risk allele and in the control group, and ${H}_{N}$ refers to the number of samples that do not carry the relevant SNP risk allele and are in the control group. Calculated odds ratios were used to calculate PRS scores for each sample as given in Eq. 2.

$$PRS=\sum _{i=1\dots x}^{n}{\beta }_{i}.{G}_{i} \left(2\right)$$

$${\beta }_{i}=\text{log}\left(odds ratio\right)$$

$${G}_{i}=alel count$$

PRS analysis was performed for selected 10 SNPs on nicotinic acetylcholine receptor genes whose relationship with nicotine addiction has been revealed in the literature. The Variant Call Format (VCF) format file provided in the data set was used for PRS analysis. PRS analysis run in Anaconda environment with PYTHON 3.8. Firstly, the sample-based genotype data of 10 SNPs were filtered. Then, odds ratio was calculated according to Eq. 1 for each of these SNPs. Table 1 lists the 10 SNPs selected for PRS analysis and their effects on nicotine addiction as stated in the literature and odds ratios calculated for the selected SNPs.

Table 1

Odds ratios calculated for selected SNPs
SNP rs_id	Gene	Ref. Allele	Risk Allele	odds ratio	SE	P
rs16969968 (Pandey et al. 2017)	CHRNA5	G	A	1.7609	0.335	0.009
rs680244 (Chen et al. 2012)	CHRNA5	A	G	1.0985	0.188	0.019
rs3743074 (Robinson et al. 2013)	CHRNA3	A	G	1.1254	0.175	0.004
rs578776 (Robinson et al. 2013)	CHRNA3	C	T	0.7346	0.124	0.006
rs588765 (Wang et al. 2009)	CHRNA5	C	T	0.9511	0.089	0.001
rs684513 (Zhang et al. 2010)	CHRNA5	C	G	0.7404	0.117	0.023
rs1051730 (Zhang et al. 2010)	CHRNA3	C	T	1.3396	0.206	0.003
rs3743078 (Zhang et al. 2010)	CHRNA3	C	G	0.7612	0.161	0.008
rs17487223 (Stevens et al. 2008)	CHRNB4	C	T	1.2263	0.170	0.001
rs950776 (Stevens et al. 2008)	CHRNB4	C	T	0.9203	0.093	0.006

Finally, as given in Eq. 1, PRS score was calculated and recorded for each sample using odds ratio and genotype information. The PRS scores calculated for the samples were added to the measurement model obtained in the previous step as observed variable. The next step of the study is to establish the conceptual model and examine the causal relationships between concepts with SEM analysis.

2.5. Structural Equation Modeling

In this study, SEM was used to explore how certain genetic variants on nicotinic acetylcholine receptors genes effect types of substance addiction with genetic and environmental variables. While SEMs are theory-driven and can represent hypotheses about causal networks, they allow interpretation of the model even in the presence of multiple predictive and complex causal link (Deutsch et al. 2020). SEM is a statistical method that continues to evolve, allowing analysis with known statistical methods such as factor analysis and regression analysis. The most important feature of SEM is that it provides an easy and descriptive graphic environment by combining the complementary aspects of these statistical methods. SEM can also be described as a generalization, integration and extension of known combined statistical methods. SEM allows us to test the model or models created with the help of hypotheses that can explain the phenomenon that is the subject of the research. Direct observation and measurement of some variables is not always possible. These types of variables can only be derived directly and indirectly from observed variables. Consequently, observations will naturally contain measurement errors. SEM takes into account measurement errors and allows complex analysis by using hidden variables. SEM also provides the ability to analyze the direction and strength of causality between complex variables by testing hypothetically designed models through observational and latent variables (Hoyle 2014).

SEM was run with AMOS 24 software. The default estimator of SEM in AMOS was maximum likelihood (ML) estimator. Since the scales of our model have both ordinal and continuous items and ıt is stated that ML estimator is robust to non-normal categorical variables (Muthen and Kaplan 1992), we used ML estimator. First, a measurement model was created for the scales to be used in the model and the validity of the scales was examined. Then, a conceptual SEM model was created by adding connections between latent variables. The PRS was added to this model. The conceptual SEM model created is shown in Fig. 1.

3.1. Measurement Model

The created measurement model was tested with confirmatory factor analysis and the validity of items in the scales confirmed. Table 2 shows the standardized regression coefficients and types of items on the scales. All relationships were found to be significant.

Table 2

Items in the scales of measurement model
Scale	Items	Type	Categories	Std. Estimate
SES	ed_grade	encoded integer	17	,957
SES	income	encoded integer	10	,973
Alcohol Dependence	alc_sx1	encoded integer	3	,509
	alc_sx2	encoded integer	3	,593
	alc_sx3	encoded integer	3	,581
	alc_sx4	encoded integer	3	,573
	alc_sx5	encoded integer	3	,540
	alc_sx6	encoded integer	3	,551
	alc_sx7	encoded integer	3	,575
Marihuana Dependence	MJ_sx1	encoded integer	3	,683
	MJ_sx2	encoded integer	3	,647
	MJ_sx3	encoded integer	3	,713
	MJ_sx4	encoded integer	3	,727
	MJ_sx5	encoded integer	3	,715
	MJ_sx6	encoded integer	3	,575
	MJ_sx7	encoded integer	3	,681
FTND	FTND_1	encoded integer	5	,902
	FTND_2	encoded integer	3	,555
	FTND_3	encoded integer	3	,661
	FTND_4	encoded integer	5	,703
	FTND_5	encoded integer	3	,569
	FTND_6	encoded integer	3	,632
Nicotine Dependence	nic_sx1	encoded integer	3	,531
	nic_sx2	encoded integer	3	,619
	nic_sx3	encoded integer	3	,521
	nic_sx4	encoded integer	3	,537
	nic_sx5	encoded integer	3	,601
	nic_sx6	encoded integer	3	,513
	nic_sx7	encoded integer	3	,578
Cocaine Dependence	COC_sx1	encoded integer	3	,878
	COC_sx2	encoded integer	3	,890
	COC_sx3	encoded integer	3	,866
	COC_sx4	encoded integer	3	,878
	COC_sx5	encoded integer	3	,877
	COC_sx6	encoded integer	3	,860
	COC_sx7	encoded integer	3	,859
Opiate Dependence	op_sx1	encoded integer	3	,917
	op_sx2	encoded integer	3	,954
	op_sx3	encoded integer	3	,939
	op_sx4	encoded integer	3	,900
	op_sx5	encoded integer	3	,928
	op_sx6	encoded integer	3	,891
	op_sx7	encoded integer	3	,815
WISDM	WISDM_Attach	decimal		,606
	WISDM_Auto	decimal		,625
	WISDM_Control	decimal		,631
	WISDM_Behav	decimal		,687
	WISDM_Cognitive	decimal		,584
	WISDM_Craving	decimal		,564
	WISDM_Cue	decimal		,554
	WISDM_Negative	decimal		,506
	WISDM_Positive	decimal		,541
	WISDM_Goads	decimal		,521
	WISDM_Senses	decimal		,511
	WISDM_Tolerance	decimal		,687
	WISDM_WtCntrl	decimal		,608

Measurement model showed good fit to the data (CMIN/DF = 4,713; CFI = ,966; RMSEA = ,038).

3.2. Structural Model

The conceptual model was examined by SEM analysis. Table 4 shows the coefficients and significance levels obtained as a result of SEM analysis.

Table 3

Parameter estimates of first SEM model (*** p < 0.001)
			Estimate	S.E.	C.R.	P
SES	<---	PRS	-1,351	0,103	-13,157	***
WISDM	<---	SES	-0,65	0,044	-14,919	***
Alcohol_Dep	<---	SES	,048	,026	1,890	,059
Marihuana_Dep	<---	SES	-,067	,021	-3,078	,062
Cocaine_Dep	<---	SES	-0,63	0,020	-3,684	,129
FTND	<---	SES	-,030	,015	-2,076	,103
Nicotine_Dep	<---	SES	,020	,016	1,246	,213
Opiate_Dep	<---	SES	-,006	,010	-,568	,570
WISDM	<---	PRS	-0,469	0,109	-4,291	***
Alcohol_Dep	<---	WISDM	0,118	0,015	8,027	***
Alcohol_Dep	<---	PRS	-0,342	0,068	-5,05	***
Marihuana_Dep	<---	WISDM	0,079	0,012	6,508	***
Marihuana_Dep	<---	PRS	0,267	0,056	4,799	***
Marihuana_Dep	<---	Alcohol_Dep	0,338	0,027	12,498	***
FTND	<---	WISDM	0,575	0,01	55,22	***
Cocaine_Dep	<---	WISDM	0,045	0,01	4,382	***
Cocaine_Dep	<---	Marihuana_Dep	0,151	0,021	7,027	***
Cocaine_Dep	<---	Alcohol_Dep	0,14	0,022	6,234	***
FTND	<---	PRS	-0,101	0,038	-2,619	0,009
Cocaine_Dep	<---	PRS	0,053	0,047	1,127	0,26
Nicotine_Dep	<---	FTND	0,548	0,026	20,781	***
Opiate_Dep	<---	WISDM	0,033	0,006	5,321	***
Nicotine_Dep	<---	Alcohol_Dep	0,153	0,02	7,676	***
Opiate_Dep	<---	Cocaine_Dep	0,095	0,012	7,684	***
Nicotine_Dep	<---	PRS	0,077	0,045	1,721	0,085
Opiate_Dep	<---	PRS	-0,023	0,028	-0,801	0,423

According to the p value, insignificant connections were removed from the model and the final model given in Fig. 4 was obtained. All relationships were found significant in this last model. The model has been found to have a good fit in terms of all goodness of fit statistics (CMIN/DF = 3,885; CFI = ,965; RMSEA = ,033).

It was observed that the PRS had the highest level of significant relationship with SES (b = 0.301, p < 0.05). The relationship between the PRS and the FTND was also found to be significant (b = 0.034, p < 0.05). The PRS also showed a significant relationship with alcohol and marijuana addiction with higher coefficients than FTND (b = 0.116, p < 0.05; b = 0.097, p < 0.05). As expected, FTND has a very high effect on nicotine addiction (b = 0.781, p < 0.05). WISDM has the strongest relationship in the model with FTND (b = 0.925, p < 0.05). WISDM is significantly affected by SES (b = 0.489, p < 0.05). WISDM also shows significant effects on alcohol, marijuana, cocaine, and opiate addiction (b = 0.2, p < 0.05; b = 0.094, p < 0.05; b = 0.142, p < 0.05; b = 0.107, p < 0.05). Alcohol addiction significantly affects nicotine, cocaine and marijuana addiction (b = 0.2, p < 0.05; b = 0.163, p < 0.05; b = 0.359, p < 0.05). Marijuana addiction positively affects cocaine addiction (b = 0.175, p < 0.05). Finally, cocaine addiction positively affects opiate addiction (b = 0.156, p < 0.05).

In SEM analysis, PRS was taken to the center. PRS was expected to be effective on nicotine addiction and FTND. However, the effect of PRS on nicotine addiction was not found to be significant. However, its effect on FTND was significant, even if it was weak. Whether the PRS has an effect on SES, WISDM and other types of addiction was also examined. PRS, which has no significant effect on WISDM, cocaine and opiate, exerts a stronger effect on alcohol and marijuana addiction than it does on FTND. This result supports some studies in the literature (Berrettini and Doyle 2011, Hartz and Bierut 2010). In addition, a strong effect of the PRS on the SES was observed. Although this is a statistically significant result, it is insignificant in terms of field knowledge.

The strong effect of the SES on WISDM and the significant effect of the WISDM variable on all addiction types are one of the key results of this study. These effects show that SES affects nicotine addiction motivation and that smoking addiction motivation affects not only smoking addiction but also other substance addictions. Although WISDM has been associated with FTND and alcohol addiction in the literature (Cook et al. 2012), there is no information about its association with marijuana, opiate and cocaine addiction. Here, WISDM is the mediator between SES and addiction to smoking, alcohol, marijuana, opiate and cocaine. Another key result of this study is the effect of alcohol addiction on nicotine, marijuana, and cocaine addiction. The interaction between alcohol addiction and nicotine addiction has been demonstrated in many studies (Falk et al. 2006, Tarren and Bartlett 2016). While the results obtained support these studies, they also show that alcohol addiction is a trigger for nicotine, marijuana and cocaine addiction.

When the chi-square values of the models obtained in the confirmatory factor analysis and SEM analysis are examined, it is seen that the significance level is below 0.05. This situation is interpreted as the tested model is significantly different from the data. This value is required to be above 0.05 in order to accept the model. However, it is stated in the literature that since this statistic is a sensitive statistic that is strongly affected by the sample size, it should not be taken into account in cases where the sample size is over 200, and the CMIN / DF value obtained by dividing the chi-square value by the degree of freedom should be used instead of this statistic (Marsh and Hocevar 1985). For this reason, the chi-square significance statistic was ignored considering that the model showed perfect fit according to the known threshold values of all goodness of fit statistics (Browne and Cudeck 1992, Joreskog and Sorbom 1996, Bentler 1990).

Although nicotine addiction was the focus of this study, other substance addictions were also examined and interesting results were obtained. Especially the strong effect of the SES on WISDM and the effect of the WISDM on other substance addictions besides nicotine addiction are very important in terms of showing that SES is a phenomenon that indirectly affects all substance addictions. In addition, the effect of alcohol addiction on nicotine, marijuana, and cocaine addiction has revealed the triggering effect among substance addictions. The results of this study show that certain genetic variants known to affect one type of addiction may also have an effect on other substance addictions. In addition, the motivation created by the SES for one substance addiction may also play a role in the development of other substance addictions.

This study has some limitations. First, odds ratios and the scores calculated in the PRS analysis are specific to this study and cannot be generalized. It is possible to obtain different ratios and scores with another data set. Likewise, the model and coefficients obtained by SEM analysis may not be valid in a different data set. The results obtained from this study need to be verified with different data sets in order to increase their generalizability. In future studies, it is possible to increase the number of SNPs included in the PRS analysis, to perform PRS analysis for SNPs on different genes and include them in the model to examine the effects on nicotine addiction as well as other types of addictions. In addition, by including the PRS scores calculated for different SNP groups together into the SEM model, both their interactions with each other and with the addiction types can be examined.

Acknowledgment and/or disclaimers, if any

The authors acknowledge the contribution of data accessed through dbGaP from the project “CIDR, NCI, NIDA, Sequencing of Targeted Genetic Regions Associated with Smoking.” Funding support for collection of datasets and samples was provided by The Collaborative Genetic Study of Nicotine Dependence (COGEND; P01 CA089392) and Genetic Study of Nicotine Dependence in African Americans (R01 DA025888). Sequencing services were provided by the Center for Inherited Disease Research (CIDR). CIDR is funded through a federal contract from the National Institutes of Health to The Johns Hopkins University, contract numbers HHSN268200782096C, HHSN268201100011I and HHSN268201200008I.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Akgün GA, Tufan AE, Yurteri N, Erdoğan A (2011). Genetic Basis of Attention Deficit Hyperactivity Disorder. Current Approaches in Psychiatry 3:15-48.
Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107(2), 238–246.
Berrettini, W. H., Doyle, G. A. (2011). The CHRNA5–A3–B4 gene cluster in nicotine addiction. Molecular Psychiatry, 17(9), 856–866.
Biala, G., Weglinska, B. (2004). Calcium channel antagonists attenuate cross-sensitization to the rewarding and/or locomotor effects of nicotine, morphine and MK-801. Journal of Pharmacy and Pharmacology, 56(8), 1021–1028.
Bierut, L. J., Madden, P. A. F., Breslau, N., Johnson, E. O., Hatsukami, D., Pomerleau, O. F., Swan, G. E., Rutter, J., Bertelsen, S., Fox, L., Fugman, D., Goate, A. M., Hinrichs, A. L., Konvicka, K., Martin, N. G., Montgomery, G. W., Saccone, N. L., Saccone, S. F., Wang, J. C., Ballinger, D. G. (2006). Novel genes identified in a high-density genome wide association study for nicotine dependence. Human Molecular Genetics, 16(1), 24–35.
Bierut, L. J. (2011). Genetic Vulnerability and Susceptibility to Substance Dependence. Neuron, 69(4), 618–627.
Browne, M. W., Cudeck, R. (1992). Alternative Ways of Assessing Model Fit. Sociological Methods & Research, 21(2), 230–258.
Changeux, JP. Nicotine addiction and nicotinic receptors: lessons from genetically modified mice. Nat Rev Neurosci 11, 389–401 (2010).
Chen, L. S., Baker, T. B., Piper, M. E., Breslau, N., Cannon, D. S., Doheny, K. F., Gogarten, S. M., Johnson, E. O., Saccone, N. L., Wang, J. C., Weiss, R. B., Goate, A. M., Bierut, L. J. (2012). Interplay of genetic risk factors (CHRNA5-CHRNA3-CHRNB4) and cessation treatments in smoking cessation success. The American journal of psychiatry, 169(7), 735–742.
Cook, J. W., Fucito, L. M., Piasecki, T. M., Piper, M. E., Schlam, T. R., Berg, K. M., & Baker, T. B. (2012). Relations of alcohol consumption with smoking cessation milestones and tobacco dependence. Journal of consulting and clinical psychology, 80(6), 1075–1085.
Courtney R. Doxbeck, Joseph A. Jaeger, Jacob M. Bleasedale, 2021. Understanding pathways to e-cigarette use across sexual identity: A multi-group structural equation model, Addictive Behaviors, Volume 114.
Creswell, K. G., Chung, T., Wright, A. G. C., Clark, D. B., Black, J. J., Martin, C. S. (2015). Personality, negative affect coping, and drinking alone: a structural equation modeling approach to examine correlates of adolescent solitary drinking. Addiction, 110(5), 775–783.
Deutsch, E. S., Alameddine, I., Qian, S. S. (2020). Using structural equation modeling to better understand microcystis biovolume dynamics in a mediterranean hypereutrophic reservoir. Ecological Modelling, 435, 109282.
Do, Elizabeth K. (2017). A Multi-Method Exploration of the Genetic and Environmental Risks Contributing to Tobacco Use Behaviors in Young Adulthood. PHD Thesis, Virginia Commonwealth University.
Falk, D. E., Yi, H. Y., Hiller-Sturmhöfel, S. (2006). An epidemiologic analysis of co-occurring alcohol and tobacco use and disorders: findings from the National Epidemiologic Survey on Alcohol and Related Conditions. Alcohol research & health : the journal of the National Institute on Alcohol Abuse and Alcoholism, 29(3), 162–171.
Fang, Z., Yang, Y., Hu, Y., Li, M. D., Wang, J. (2017). GRONS: a comprehensive genetic resource of nicotine and smoking. Database : the journal of biological databases and curation, 2017, bax097.
Hartz, S. M., & Bierut, L. J. (2010). Genetics of addictions. The Psychiatric clinics of North America, 33(1), 107–124.
Hoyle, R. H. (2014). Handbook of Structural Equation Modeling (Illustrated ed.). The Guilford Press.
Joreskog, K. G., Sorbom, D. (1996). LISREL8: User’s reference guide. Mooresville: Scientific Software.
King, D. P., Paciga, S., Pickering, E., Benowitz, N. L., Bierut, L. J., Conti, D. V., Kaprio, J., Lerman, C., Park, P. W. (2011). Smoking Cessation Pharmacogenetics: Analysis of Varenicline and Bupropion in Placebo-Controlled Clinical Trials. Neuropsychopharmacology, 37(3), 641–650.
Lewis, C. M., Vassos, E. (2017). Prospects for using risk scores in polygenic medicine. Genome medicine, 9(1), 96.
López-Flores, L. A., Pérez-Rubio, G., Falfán-Valencia, R. (2017). Distribution of polymorphic variants of CYP2A6 and their involvement in nicotine addiction. EXCLI journal, 16, 174–196.
Mackillop, J., Obasi, E., Amlung, M. T., McGeary, J. E., Knopik, V. S. (2010). The Role of Genetics in Nicotine Dependence: Mapping the Pathways from Genome to Syndrome. Current cardiovascular risk reports, 4(6), 446–453.
Marsh, H. W., Hocevar, D. (1985). Application of confirmatory factor analysis to the study of self-concept: First- and higher order factor models and their invariance across groups. Psychological Bulletin, 97(3), 562–582.
Muthen, B. and Kaplan, D. (1992), A comparison of some methodologies for the factor analysis of non‐normal Likert variables: A note on the size of the model. British Journal of Mathematical and Statistical Psychology, 45: 19-30.
NCBI's Database of Genotypes and Phenotypes: dbGaP, dbGaP Study Accession: phs000813.v1.p1
Pandey, N., Pal, S., Sharma, L. K., Guleria, R., Mohan, A., Srivastava, T. (2017). SNP rs16969968 as a Strong Predictor of Nicotine Dependence and Lung Cancer Risk in a North Indian Population. Asian Pacific journal of cancer prevention : APJCP, 18(11), 3073–3079.
Ray, R., Mitra, N., Baldwin, D., Guo, M., Patterson, F., Heitjan, D. F., Jepson, C., Wileyto, E. P., Wei, J., Payne, T., Ma, J. Z., Li, M. D., Lerman, C. (2010). Convergent Evidence that Choline Acetyltransferase Gene Variation is Associated with Prospective Smoking Cessation and Nicotine Dependence. Neuropsychopharmacology, 35(6), 1374–1382.
Robinson, J. D., Versace, F., Lam, C. Y., Minnix, J. A., Engelmann, J. M., Cui, Y., Karam-Hage, M., Shete, S. S., Tomlinson, G. E., Chen, T. T., Wetter, D. W., Green, C. E., & Cinciripini, P. M. (2013). The CHRNA3 rs578776 Variant is Associated with an Intrinsic Reward Sensitivity Deficit in Smokers. Frontiers in psychiatry, 4, 114.
Stevens VL, Bierut LJ, Talbot JT, Wang JC, Sun J, Hinrichs AL, Thun MJ, Goate A, Calle EE. Nicotinic receptor gene variants influence susceptibility to heavy smoking. Cancer Epidemiol Biomarkers Prev. 2008 Dec;17(12):3517-25.
Tarren, J. R., Bartlett, S. E. (2016). Alcohol and nicotine interactions: pre-clinical models of dependence. The American Journal of Drug and Alcohol Abuse, 43(2), 146–154.
The Tobacco and Genetics Consortium, (2010). Genome-wide meta-analyses identify multiple loci associated with smoking behavior. Nature Genetics, 42(5), 441–447.
Vink, J. M., Willemsen, G., Boomsma, D. I. (2005). Heritability of Smoking Initiation and Nicotine Dependence. Behavior Genetics, 35(4), 397–406.
Voci, Sabrina C., (2014). An examination of the relationships between causal attributions for smoking and smokers' treatment seeking and quit intentions: A structural equation modeling approach, Electronic Theses and Dissertations. Paper 5057.
Wang, J. C., Cruchaga, C., Saccone, N. L., Bertelsen, S., Liu, P., Budde, J. P., Duan, W., Fox, L., Grucza, R. A., Kern, J., Mayo, K., Reyes, O., Rice, J., Saccone, S. F., Spiegel, N., Steinbach, J. H., Stitzel, J. A., Anderson, M. W., You, M., Stevens, V. L. (2009). Risk for nicotine dependence and lung cancer is conferred by mRNA expression levels and amino acid change in CHRNA5. Human molecular genetics, 18(16), 3125–3135.
Wittenberg, R. E., Wolfman, S. L., De Biasi, M., Dani, J. A. (2020). Nicotinic acetylcholine receptors and nicotine addiction: A brief introduction. Neuropharmacology, 177, 108256.
Zhang, H., Kranzler, H., Poling, J., Variation in the Nicotinic Acetylcholine Receptor Gene Cluster CHRNA5–CHRNA3–CHRNB4 and Its Interaction with Recent Tobacco Use Influence Cognitive Flexibility. Neuropsychopharmacol 35, 2211–2224 (2010).

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Effect of Certain Genetic Variants in Nicotinic AChR on Substance Addictions: A Structural Equation Modeling Approach

Status:

Version 1

Abstract

Figures

1. Introduction

2. Materials and methods

2.1. Data Source

2.2. Measures

2.3. Missing Data Analysis

2.4. PRS Analysis

2.5. Structural Equation Modeling

3. Results

3.1. Measurement Model

3.2. Structural Model

4. Discussion

Declarations

References

Additional Declarations

Status:

Version 1