Altered functional dynamics gradient in schizophrenia with cigarette smoking
Glasgow College, University of Electronic Science and Technology of China, No. 2006, Xiyuan Ave, West Hi-Tech Zone, Chengdu 611731, China
High-Field Magnetic Resonance Brain Imaging Key Laboratory of Sichuan Province, School of Life Science and Technology, University of Electronic Science and Technology of China, No. 2006, Xiyuan Ave, West Hi-Tech Zone, Chengdu 611731, China
Abstract
Schizophrenia is associated with a high prevalence of cigarette smoking. Neural dynamics are spatially structured and shaped by both microscale molecular and macroscale functional architectures, which are disturbed in the diseased brain. The neural mechanism underlying the schizophrenia-nicotine dependence comorbidity remains unknown. In this study, we aimed to test whether there is an interaction between schizophrenia and smoking in brain neural dynamics, and how the main effect of the 2 factors related to the molecular architecture. Functional magnetic resonance imaging data were obtained from 4 groups: schizophrenia and healthy controls with/without smoking. We identified 2 dynamics gradients combined with over 5,000 statistical features of the brain region's time series. The interaction effect was found in the high-order functional network, and the main effect of schizophrenia was in the bilateral orbitofrontal cortices. Moreover, the disease- and smoking-related alteration in brain pattern was associated with spatial distribution of serotonin, cannabinoid, and glutamate. Collectively, these findings supported the self-medication hypothesis in schizophrenia-nicotine dependence with a neural intrinsic dynamics perspective.
Untitled section
Keywords: cigarette smoking, gradient, intrinsic dynamics, schizophrenia, self-medication
Article notes
Untitled section
Received 2022 Dec 1; Revised 2023 Jan 18; Accepted 2023 Jan 19; Collection date 2023 Jun 1.
Introduction
Schizophrenia, characterized by recurrent psychotic symptoms, is one of the most severe psychiatric illnesses (Jauhar et al. 2022; Organization 2022). It displays a disability in cognitive domains and everyday functioning and thus affects approximately 1% of the world population (Fleischhacker et al. 2014). More than 80% of individuals with schizophrenia have been cigarette smokers (Coustals et al. 2020; Hunter et al. 2020); the prevalence of the disease is high in the general population (de Leon and Diaz 2005; Collaborators 2017), and it is even more than patients with other mental illnesses (Firth et al. 2019). This high schizophrenia–nicotine dependence comorbidity would be supported by 2 concurrent hypotheses: “self-medication” (Winterer 2010) and “addictive vulnerability” (Krystal et al. 2006). Although a significant but complex link between schizophrenia and nicotine dependence is clear (Isuru and Rajasuriya 2019), the neurophysiological mechanism attributed to it remains unclear.
Mounting evidence from human multimodal neuroimaging has investigated the neurobiological basis for the link between schizophrenia and nicotine dependence. The lateral ventricular enlargement and broad brain volume reductions have been substantially corroborated in schizophrenia (Liloia et al. 2021; Pico-Perez et al. 2022). Nonetheless, brain morphometric alterations have subtlety and even failure to find smoking-related disease effects in schizophrenia (Yokoyama et al. 2018; Ringin et al. 2021). The presence of brain functional brain inactions between schizophrenia and smoking is also manifested, initially reported at smoking-related cues or cravings (Potvin et al. 2016; Khan et al. 2018) task functional magnetic resonance imaging (fMRI). Moreover, in schizophrenia-nicotine dependence comorbidity recent task-free (resting-state) fMRI studies found that cigarette smoking restores intrinsic brain activity in the right striatal and prefrontal cortices (Liu et al. 2018), and a homotopic functional connectivity antagonistically affects the bilateral ventrolateral prefrontal cortices (Liao et al. 2019). But, abnormalities of cortical activity and connectivity did not take the form solely of local dysfunction. Our previous studies further suggest that cigarette smoking affects schizophrenia at the dynamic of spatial influences among the neuropsychiatric triple networks (Liao et al. 2018) and dynamic intrinsic activity of temporal evolution (Yang et al. 2019). Collectively, these findings are consistent with the self-medication hypothesis. However, a significant limitation is that these studies focused on specific, manually selected spontaneous fluctuations (e.g. fMRI time-series) features.
Imaging-based phenotypes, especially intrinsic neural dynamics, are more subtle than simple statistics or single features (Fulcher et al. 2013; Fulcher and Jones 2017). Specifically, the intrinsic neural dynamics give rise to spatially distributed functional networks that emerged from brain hierarchical organization that supported complex cognition (Cabral et al. 2014). Previous studies suggested that neural dynamics altered individuals with schizophrenia (Damaraju et al. 2014; Miller et al. 2016; Yang et al. 2019), and differentiated from patients with other mental disorders and healthy controls (Rashid et al. 2014; Kottaram et al. 2018). Moreover, the influence of nicotine on brain functions was also of interest (Lawrence et al. 2002; Cole et al. 2010; Tanabe et al. 2011). In addition, intrinsic neural dynamics at each brain region can also be used to collect large feature sets of time series properties (Fulcher and Jones 2017). In intact brain, these dynamics can be used to compare temporal profile similarity, and so-called gradient, revealing functionally dynamical hierarchies (Shafiei et al. 2020). Specifically, the principal functional dynamics gradient (FDG) follows a ventromedial-dorsolateral axis, in which features contribute mostly toward describing the autocorrelation of time series (Shafiei et al. 2020). Previous studies demonstrated that brain regions with greater autocorrelation show greater connectivity (Fallon et al. 2020). The second FDG follows a unimodal-to-transmodal gradient, in which contributed features describe the shape and distribution of the time series amplitude. The second gradient is related to the canonical functional gradient, intracortical myelin, and cortical thickness (Shafiei et al. 2020). Therefore, a comprehensive and scientific description of the FDG would help investigate the link between nicotine dependence and schizophrenia.
In this study, we examined brain functional dynamics in 4 groups: schizophrenia groups with/without smoking, and healthy controls with/without smoking. We hypothesized that schizophrenia and smoking factors would interact with brain functional dynamics. We then further investigate whether the interaction would support the self-medication or addiction vulnerability hypothesis. Both hypotheses might result from deficits in nicotinic neurotransmission or decreasing dopaminergic activity secondary to nicotinic cholinergic receptor desensitization in schizophrenia (Liu and Kenny 2017). Thus, we further speculated that the corresponding patterns of alterations relate to the serotonin and dopaminergic system. To assess the molecular underpinnings, we examined the main effect of schizophrenia and smoking in functional dynamics alterations, which were spatially correlated with patterned receptors/transporters.
Methods and materials
Subjects
This study included 4 age- and sex-matched groups: healthy controls without smoking (HCnon, n = 21), healthy controls with smoking (HCsm, n = 22), schizophrenia patients without smoking (SCHnon, n = 21), and schizophrenia patients with smoking (SCHsm, n = 22). Written informed consent was obtained from all participants. All examinations were carried out under the guidance of the Declaration of Helsinki 1975. This study was reviewed and approved by the Local Medical Ethics Committee of the First Affiliated Hospital of Chongqing Medical University.
According to the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders (SCID-I/P, Chinese version), patients with schizophrenia were diagnosed by experienced physicians or clinical psychologists. Patients were excluded if (i) they were under the age of 16; (ii) they had comorbidities with substance addiction other than cigarettes; or (iii) if they were comorbid with neurological or other psychiatric disorders. Totally, 43 patients were confirmed for further analysis. We asked them about the onset age, and duration of illness. A total of 5 patients were first-episode, and 38 were chronic patients; 24 patients who took antipsychotic medication were clinically medication stable (>3 months with no change in medicine). Their symptoms severity was assessed using the Positive and Negative Syndrome Scale (PANSS) (Kay et al. 1987).
Participants were defined as smokers if they smoked any number of cigarettes daily for more than 1 year. Participants were considered nonsmokers if they never smoked or did not use nicotine products at all. Nicotine addiction severity and dependence was assessed using the Fagerstrom Test for Nicotine Dependence (FTND) (Heatherton et al. 1991) and the Revised Tolerance Questionnaire (RTQ) (Tate and Schmitz 1993). Smokers are encouraged to smoke before the scan to avoid the withdrawal effect. But they are prohibited from smoking 30 min before the scan to avoid an immediate effect (Galvan et al. 2011; Addicott et al. 2015). In addition, we recorded data on the current use of tobacco products, the number of cigarettes smoked per day, and the onset age of smoking.
Imaging protocol
Functional data were acquired on a 3.0-Tesla MRI system (GE Medical Systems, Waukesha, WI, United States) at the First Affiliated Hospital of Chongqing Medical University. All participants were required to keep their eyes closed but not fall asleep. All participants were asked if they had fallen asleep during the scanning. These resting-state functional magnetic resonance imaging (rs-fMRI) were acquired using an echo-planar imaging sequence: repetition time (TR) = 2000 ms, echo time (TE) = 30 ms, flip angle = 90°, field of view = 220 × 240 mm2, matrix = 64 × 64, and slice thickness = 4 mm. A total of 33 transverse slices that aligned along the anterior commissure–posterior commissure line were acquired with a total scan time of 400 s.
Data Preprocessing
Functional image preprocessing was performed using the DPABI (Yan et al. 2016) (DPABI2.3, http://rfmri.org/dpabi), which is based on statistical parametric mapping (SPM12, http://www.fil.ion.ucl.ac.uk/spm). Specifically, the first 10 images were discarded, and the remaining images were corrected for temporal differences and head motion. Then the images were spatially normalized to Montreal Neurologic Institute (MNI) space with 3 × 3 × 3 voxel resolution. We next regressed the nuisance variables, including white matter signals, mean ventricular signals, and motion parameters (Friston 24-parameter model) (Friston et al. 1996). We calculated the frame-wise displacement (FD) for each time point (Power et al. 2012). Participants were excluded if (a) mean FD exceeded 0.5 mm, or (b) head motion exceeded 3 mm or 3°. To further correct motion effects, the scrubbing parameters were regressed out. The scrubbing points were identified with FD exceeding 0.5 mm as well as 1forward and 2 back neighbors. After that, images were spatially smoothed using a Gaussian kernel of full width at a half-maximum of 8 mm. Finally, images were detrended and temporal band-pass filtered (0.01–0.08 Hz).
Functional dynamics gradient
To comprehensively characterize the rich temporal pattern of neural activity, we described the massive features of spontaneous blood oxygenation level-dependent signals. To this end, we first used the Schaefer200 parcellation to extract the mean time series for each brain region of the preprocessed functional images. We then used the highly comparative time-series analysis (hctsa) toolbox to obtain the massive temporal features for each region (Fulcher and Jones 2017). This toolbox computed over 7,000 regional time-series features based on a wide range of operations. These measures comprehensively described the distribution of the time series, autocorrelation structure, stationary properties, information theoretic measures of entropy and temporal predictability, the linear and nonlinear model fits, etc. (Fulcher et al. 2013). After the feature extraction, the error operations were removed, and the remaining features normalized across regions using outlier-robust sigmoidal transformation. For the following comparisons between groups, we analyzed the shared features in all participants, resulting in a 200 × 5,253 temporal features matrix for each participant. This matrix described the regional functional dynamics, which reflected the local rhythms shaped by the micro- and macro-architectural properties.
To investigate the topological organization of brain functional dynamics, we mapped the gradient pattern of the time-series features. We performed principal component analysis (PCA) on the temporal features matrix, yielding a set of components that explained the variance of the matrix. We referred to the components as FDGs. In the following exploratory, we mainly focused on the first 2 components: FDG1 and FDG2.
Statistical analysis
Demographic and clinical data were analyzed using GraphPad Prism 8 (GraphPad Software, Inc.). The Chi-square test (χ2) was used for sex and handedness. One-way analysis of variance (ANOVA) or the Kruskal–Wallis test (the data were not normally distributed) was used for the other demographic characteristics and/or clinical scores. A significant threshold was set at P < 0.05. A 2 × 2 ANOVA with disease and addiction factors was employed for FDG comparison, regressing out the age, sex, and education years. The FDG parcels were corrected with false-positive adjustment, 1/n = 1/200 (Fornito et al. 2011; Ji et al. 2017). The averaged FDG value was extracted from each participant in each significant brain region. Post hoc analyses were performed to determine whether interactions were synergistic or antagonistic.
Associations analyses
Following the identification of the interaction effect in brain FDG, we carried out the correlation analyses between FTND, RTQ, and PANSS scores with significant brain parcels, with Bonferroni adjustment for multiple comparisons. In addition to interaction analysis, we also investigated the associations between spatial patterns of the main effects of disease and addiction with neuro-transmitter maps derived by positron emission tomography (PET). To this end, we obtained PET maps from the JuSpace toolbox, covering various neurotransmitter systems, including dopaminergic, serotonergic, noradrenergic, and gamma-aminobutric acid (GABA)ergic neurotransmission. We computed Spearman's correlation between the t-maps (unthresholded) of the main effect of disease and addiction and PET maps. We performed 5,000 spin test permutations to control the spatial autocorrelation (Alexander-Bloch et al. 2013).
Validation analyses
In this section, we performed additional analyses to validate the interaction effect of schizophrenia and smoking. First, we consider that the 4 groups are close to differ in gender, which may be impacting the results. To clarify this, we repeated the analysis on male participants and then examined the interaction effect. We further assessed the association between the disease- and smoking-related effect on the functional dynamics with the PET-derived neurotransmitters distribution. Second, the number of parcels in cortical parcellation may impact the results. To validate our findings, this work also repeated the analyses using the Schaefer100 and Scharfer400 parcellations (Schaefer et al. 2018).
Results
Demographics and clinical characteristics
Demographic and clinical characteristics are shown in Table 1. Patients with schizophrenia did not differ from healthy controls in age, sex, and handedness. Schizophrenia smokers did not differ from schizophrenia nonsmokers in the onset of disease, type of first-episode or chronic patients, duration, anti-psychotic medication, and PANSS scores. Healthy smokers did not differ from healthy smokers in the onset of smoking, cigarettes per day, lifetime smoking (pack-years), FTND scores, and RTQ scores.
| Demographics | Healthy controls | Schizophrenia | Comparison | ||
|---|---|---|---|---|---|
| HCnon (n = 21) | HCsm (n = 22) | SCHnon (n = 21) | SCHsm (n = 22) | ||
| Gender (male/female) | 14/7 | 19/3 | 12/9 | 19/3 | χ2 = 7.253 (P = 0.064) |
| Handedness (left/right) | 0/21 | 0/22 | 0/21 | 0/22 | — |
| Age (years) | 31.43 ± 1.94 | 34.55 ± 2.14 | 30.05 ± 2.05 | 29.45 ± 2.12 | KW = 3.902 P = 0.27 |
| At onset of schizophrenia | — | — | 25.33 ± 8.25 | 23.36 ± 7.28 | U = 198.5 P = 0.436 |
| At onset of smoking | — | 19.64 ± 4.593 | — | 17.68 ± 2.70 | t = 1.72 P = 0.093 |
| Education (years) | 12.71 ± 3.48 | 14.59 ± 2.94 | 11.67 ± 3.022 | 11.05 ± 2.54 | KW = 17.379 P = 0.001 |
| Type (first-episode/chronic) | — | — | 1/20 | 4/18 | χ2 = 1.88 P = 0.17 |
| Duration of illness (years) | — | — | 4.74 ± 2.54 | 7.23 ± 8.57 | U = 221 0.5 P = 0.823 |
| Anti-psychotic medication (yes/no) | — | — | 4/17 | 5/17 | χ2 = 0.088 P = 0.767 |
| Cigarettes per day | — | 17.18 ± 1.86 | — | 23.09 ± 2.61 | U = 174.5 P = 0.106 |
| Lifetime cigarette use (pack years) | — | 12.06 ± 3.3 | — | 20.01 ± 4.3 | U = 174 P = 0.114 |
| FTND | — | 5.41 ± 0.55 | — | 5.86 ± 0.61 | t = −0.55 P = 0.583 |
| RTQ | — | 28.32 ± 1.51 | — | 29.55 ± 1.55 | t = −0.57 P = 0.57 |
| PANSS scores | — | — | 59.00 ± 18.19 | 67.14 ± 25.93 | U = 191.5 P = 0.344 |
| PANSS Positive | — | — | 11.67 ± 6.27 | 13.77 ± 7.47 | U = 208.5 P = 0.592 |
| PANSS Negative | — | — | 21.10 ± 8.76 | 19.32 ± 7.85 | U = 187 P = 0.283 |
The pattern of FDG in each group
To evaluate the topological organization of brain functional dynamics described by massive time-series features, we applied PCA to capture the first maximal variance axes, FDG1 and FDG2. The top 2 components accounted for more than 30% of the variance in time-series features. The FDG1 mainly reflects a ventromedial-dorsolateral transition, and the FDG2 captures a sensorimotor-association cortices transition (Fig. 1), similar to the previous findings (Shafiei et al. 2020).
The interaction effect and main effect
For FDG1, the interaction effect of schizophrenia and smoking was found in the bilateral posterior cingulate cortex (PCC), the left inferior frontal gyrus, and the left temporal parietal junction (Fig. 2A), mainly involving the default mode network and salience attention network. Planned post hoc analyses of the altered FDG1 are shown in Supplementary Fig. S1. The main effect of schizophrenia was found in the bilateral orbitofrontal cortices and the right posterior dorsal attention network, with patients exhibiting higher FDG1 than healthy controls. We did not find any parcels with a significant main effect of smoking. The interaction effect of FDG2 was found in the bilateral motor cortex and middle temporal gyrus (Supplementary Fig. S2). We did not find any parcels with a significant main effect of schizophrenia or smoking.
The strength (component scores) of FDG1 of the left PCC was negatively correlated with the PANSS scores (r = −0.476, P = 0.025) and negatively correlated with RTQ scores (r = −0.520, P = 0.013) in schizophrenia smokers (Fig. 3). The result suggested that increased FDG1 in schizophrenia smokers was associated with lower RTQ scores and low PANSS scores. We have further determined how the positive and negative PANSS scores related to the strength of the FDG1, and found that the positive PANSS was negatively correlated with the FDG1 (r = −0.426, P = 0.048); the negative PANSS did not show a significant correlation (r = −0.211, P = 0.347). To explore whether this association is specific in the schizophrenia smoking group, we computed the correlations between the PANSS scores and the left PCC of schizophrenia nonsmokers (r = 0.396, P = 0.076) and the RTQ scores and the left PCC of healthy smokers (r = −0.030, P = 0.894).
Associations with molecular architecture
Correlations between PET-derived neurotransmitter maps and the main effects of schizophrenia and smoking identified significant spatial relationships. Specifically, the main effect of schizophrenia was related to the serotonin 1B receptors (5HT1b_P943_HC22), cannabinoid (CB1_FMPEPd2_hc22), glutamate (mGluR5_abp_hc73), opioid (MU_CARFENTANIL_c11, MU_ carfentanil_hc39), and serotonin transporter (SERT_dasb_hc100) (Fig. 4A). And the main effect of smoking was related to the serotonin 2A and 4 receptors (5HT2a_ALT_HC19, 5HT4_sb20_hc59), cannabinoid (CB1_FMPEPd2_hc22), cerebral blood flow (CBF_ASL_ MRI), and glutamate (mGluR5_abp_hc22_rosaneto, mGluR5_abp_ hc28_dubois, mGluR5_abp_hc73_smart) (Fig. 4B).
Validation analyses
Only using the male subjects, the spatial patterns of FDG1 and FDG2 in the 4 groups were consistent with the exploratory results (Supplementary Fig. S3). We then performed the ANOVA on the FDG 1 and found that the interaction effect was found in the left temporoparietal junction (TPJ) and right superior parietal cortex, mainly involving in the attention network (Supplementary Fig. S4). But we did not find any parcel with a significant main effect of disease factor. These findings were partly similar to the exploratory results. As for the associations between the disease- and smoking-related effect on the functional dynamics with the neurotransmitter distribution, we found that these relationships were nearly similar to the exploratory results (Supplementary Fig. S5). To confirm the stability of the spatial pattern of FDGs, we used the Schaefer100 and Schaefer400 parcellations to repeat the analyses. The FDG1 and FDG2 in the 4 groups were similar to the exploratory findings in both parcellations (Supplementary Fig. S6).
Discussion
Here we show an interaction effect in the high-order association network between schizophrenia and smoking status about brain functional dynamics. The more severity of nicotine dependence in patients with schizophrenia who showed a lower dynamics gradient is associated with higher scores in PANSS measurements. In addition, the main effect of schizophrenia in FDG was found in the bilateral orbitofrontal cortices, but no main effect of smoking status. And both main effects in brain patterns were associated with the serotonin, cannabinoid, and glutamate distribution.
Notably, the influence of smoking effect was more severe in healthy controls than in patients with schizophrenia in these regions that showed significant interaction effects, which suggested that nicotine had an inverse or inconspicuous impact on the schizophrenia population than health, specifically, the left PCC, the critical hub in the default mode network (Andrews-Hanna et al. 2010), in which schizophrenia smokers had a normal trend change in FDG, accompanied by alleviated dependence on nicotine and symptoms of schizophrenia. The results are in line with studies that nicotine-induced brain deactivation reduced connectivity in the default mode network (Hahn et al. 2007; Tanabe et al. 2011), and extended the effect of nicotine on brain functional dynamics. Another notable region with significant interaction effect was the left TPJ. A clinical study found that with a 15-day repetitive transcranial magnetic stimulation treatment on the left TPJ, the clinical symptoms were improved in patients with schizophrenia (Chen et al. 2019), and the increased functional connectivity of the personalized atrophy network was positively correlated with alleviated symptom outcomes (Ji et al. 2023). Together, these findings with respect to the interaction effect support the “self-medication” hypothesis from the perspective of neural dynamics gradient in brain function.
Orbitofrontal cortex hyperactivity of intrinsic neural dynamics is significant in patients with schizophrenia. But we did not find any statistical correlation between these alterations and clinical symptom measurements. The orbitofrontal cortices are key components in the limbic network, and is thought to be involved in sensory integration, and decision-making for emotional behaviors (Kringelbach 2005; Rushworth et al. 2007). The orbitofrontal cortex interconnected with the sensory region, amygdala, and parahippocampus (Helen Barbas and Zikopoulos 2006; Haber and Behrens 2014). Therefore, the altered functional dynamics in the orbitofrontal cortex may lead to dysfunction within the limbic system, where the abnormalities in patients with schizophrenia have been emphasized (White et al. 2008; Dugre et al. 2019).
We also noted an interesting result of the association between schizophrenia-related and smoking-related alteration patterns in brain intrinsic dynamics. Specifically, the schizophrenia-related alteration was positively correlated with the PET-derived neurotransmission maps of cannabinoid (CB1_FMPEPd2_hc22) and glutamate (mGluR5_abp_hc73). In contrast, the smoking-related alteration pattern was negatively correlated with these distribution maps. Glutamate is an excitatory neurotransmitter in the brain (Meldrum 2000) and plays a crucial role in nicotine dependence (Kenny and Markou 2004; D'Souza and Markou 2011). More importantly, the dysregulation of glutamatergic neurotransmission is critically implicated in the pathophysiology of schizophrenia (Moghaddam and Javitt 2012). We speculated that the high propensity to smoke in schizophrenia was associated with glutamatergic neurotransmission and the alteration of intrinsic dynamics. Still, we did not demonstrate the causal relationships between these factors. We also did not find any significant relationship between the main effect of disease and smoking on FDG1 with the dopamine maps. In future, we could explore the disease- and smoking-related effect on functional dynamics in corticolimbic dopaminergic pathways or other specific related circuits, rather than in the whole cortex.
Limitations
This study has several limitations. First, the sample size of this study was relatively small, which challenged reproducibility. An alternative way to confirm these findings was included in an independent dataset. Limited by the difficulty of sample collection, we have not examined the reproducibility of the findings. Second, the education duration differed in the 4 groups, with healthy controls significantly higher than schizophrenia. That is a natural phenomenon in the psychiatry population, and we have set the variable as a covariate in the group comparisons to reduce the influence. Third, the PET-derived neurotransmitter maps were from healthy participants. Future works could extend the dataset to samples from multiple brain disorders.
Conclusion
This study demonstrated the interaction effect on intrinsic neural dynamics to explain the high risk of smoking in schizophrenia, and supported the self-medication hypothesis. Moreover, schizophrenia-related and smoking-related brain alterations in functional dynamics differed in cannabinoid and glutamate neurotransmission.
Supplementary Material
Acknowledgments
We are grateful to all the participants in this study. We extend our thanks to Prof. Liao and Dr Yang of the School of Life Sciences and Technology for their guidance.
Conflict of interest statement: None of the authors has any conflict of interest to disclose.
Data availability statement
The neuroimaging data that support the findings of this study are available from the corresponding author upon reasonable request.
References
Untitled section
References
- Addicott MA, Sweitzer MM, Froeliger B, Rose JE, McClernon FJ. Increased functional connectivity in an insula-based network is associated with improved smoking cessation outcomes. Neuropsychopharmacology. 2015:40:2648–2656.
- Alexander-Bloch A, Raznahan A, Bullmore E, Giedd J. The convergence of maturational change and structural covariance in human cortical networks. J Neurosci. 2013:33:2889–2899.
- Andrews-Hanna JR, Reidler JS, Sepulcre J, Poulin R, Buckner RL. Functional-anatomic fractionation of the brain's default network. Neuron. 2010:65:550–562.
- Barbas H, Zikopoulos B. Sequential and parallel circuits for emotional processing in the primate orbitofrontal cortex. In: The Orbitofrontal Corte. New York: Oxford University Press; 2006.
- Cabral J, Kringelbach ML, Deco G. Exploring the network dynamics underlying brain activity during rest. Prog Neurobiol. 2014:114:102–131.
- Chen X, Ji GJ, Zhu C, Bai X, Wang L, He K, Gao Y, Tao L, Yu F, Tian Y, et al. Neural correlates of auditory verbal hallucinations in schizophrenia and the therapeutic response to theta-burst transcranial magnetic stimulation. Schizophr Bull. 2019:45:474–483.
- Cole DM, Beckmann CF, Long CJ, Matthews PM, Durcan MJ, Beaver JD. Nicotine replacement in abstinent smokers improves cognitive withdrawal symptoms with modulation of resting brain network dynamics. NeuroImage. 2010:52:590–599.
- Collaborators GBDT. Smoking prevalence and attributable disease burden in 195 countries and territories, 1990-2015: a systematic analysis from the global burden of disease study 2015. Lancet. 2017:389:1885–1906.
- Coustals N, Martelli C, Brunet-Lecomte M, Petillion A, Romeo B, Benyamina A. Chronic smoking and cognition in patients with schizophrenia: a meta-analysis. Schizophr Res. 2020:222:113–121.
- Damaraju E, Allen EA, Belger A, Ford JM, McEwen S, Mathalon DH, Mueller BA, Pearlson GD, Potkin SG, Preda A, et al. Dynamic functional connectivity analysis reveals transient states of dysconnectivity in schizophrenia. Neuroimage Clin. 2014:5:298–308.
- D'Souza MS, Markou A. Neuronal mechanisms underlying development of nicotine dependence: implications for novel smoking-cessation treatments. Addict Sci Clin Pract. 2011:6:4–16.
- Dugre JR, Bitar N, Dumais A, Potvin S. Limbic hyperactivity in response to emotionally neutral stimuli in schizophrenia: a neuroimaging meta-analysis of the Hypervigilant mind. Am J Psychiatry. 2019:176:1021–1029.
- Fallon J, Ward PGD, Parkes L, Oldham S, Arnatkeviciute A, Fornito A, Fulcher BD. Timescales of spontaneous fMRI fluctuations relate to structural connectivity in the brain. Netw Neurosci. 2020:4:788–806.
- Firth J, Siddiqi N, Koyanagi A, Siskind D, Rosenbaum S, Galletly C, Allan S, Caneo C, Carney R, Carvalho AF, et al. The Lancet Psychiatry Commission: a blueprint for protecting physical health in people with mental illness. Lancet Psychiatry. 2019:6:675–712.
- Fleischhacker WW, Arango C, Arteel P, Barnes TR, Carpenter W, Duckworth K, Galderisi S, Halpern L, Knapp M, Marder SR, et al. Schizophrenia--time to commit to policy change. Schizophr Bull. 2014:40(Suppl 3):S165–S194.
- Fornito A, Yoon J, Zalesky A, Bullmore ET, Carter CS. General and specific functional connectivity disturbances in first-episode schizophrenia during cognitive control performance. Biol Psychiatry. 2011:70:64–72.
- Friston KJ, Williams S, Howard R, Frackowiak RS, Turner R. Movement-related effects in fMRI time-series. Magn Reson Med. 1996:35:346–355.
- Fulcher BD, Jones NS. hctsa: a computational framework for automated time-series phenotyping using massive feature extraction. Cell Syst. 2017:5:527, e523–531.
- Fulcher BD, Little MA, Jones NS. Highly comparative time-series analysis: the empirical structure of time series and their methods. J R Soc Interface. 2013:10:20130048.
- Galvan A, Poldrack RA, Baker CM, McGlennen KM, London ED. Neural correlates of response inhibition and cigarette smoking in late adolescence. Neuropsychopharmacology. 2011:36:970–978.
- Haber SN, Behrens TE. The neural network underlying incentive-based learning: implications for interpreting circuit disruptions in psychiatric disorders. Neuron. 2014:83:1019–1039.
- Hahn B, Ross TJ, Yang Y, Kim I, Huestis MA, Stein EA. Nicotine enhances visuospatial attention by deactivating areas of the resting brain default network. J Neurosci. 2007:27:3477–3489.
- Heatherton TF, Kozlowski LT, Frecker RC, Fagerstrom KO. The Fagerstrom test for nicotine dependence: a revision of the Fagerstrom tolerance questionnaire. Br J Addict. 1991:86:1119–1127.
- Hunter A, Murray R, Asher L, Leonardi-Bee J. The effects of tobacco smoking, and prenatal tobacco smoke exposure, on risk of schizophrenia: a systematic review and meta-analysis. Nicotine Tob Res. 2020:22:3–10.
- Isuru A, Rajasuriya M. Tobacco smoking and schizophrenia: re-examining the evidence. BJPsych Advances. 2019:25:363–372.
- Jauhar S, Johnstone M, McKenna PJ. Schizophrenia Lancet. 2022:399:473–486.
- Ji GJ, Yu Y, Miao HH, Wang ZJ, Tang YL, Liao W. Decreased network efficiency in benign epilepsy with centrotemporal spikes. Radiology. 2017:283:186–194.
- Ji GJ, Zalesky A, Wang Y, He K, Wang L, Du R, Sun J, Bai T, Chen X, Tian Y, et al. Linking personalized brain atrophy to schizophrenia network and treatment response. Schizophr Bull. 2023:49:43–52.
- Kay SR, Fiszbein A, Opler LA. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull. 1987:13:261–276.
- Kenny PJ, Markou A. The ups and downs of addiction: role of metabotropic glutamate receptors. Trends Pharmacol Sci. 2004:25:265–272.
- Khan MS, Boileau I, Kolla N, Mizrahi R. A systematic review of the role of the nociceptin receptor system in stress, cognition, and reward: relevance to schizophrenia. Transl Psychiatry. 2018:8:38.
- Kottaram A, Johnston L, Ganella E, Pantelis C, Kotagiri R, Zalesky A. Spatio-temporal dynamics of resting-state brain networks improve single-subject prediction of schizophrenia diagnosis. Hum Brain Mapp. 2018:39:3663–3681.
- Kringelbach ML. The human orbitofrontal cortex: linking reward to hedonic experience. Nat Rev Neurosci. 2005:6:691–702.
- Krystal JH, D'Souza DC, Gallinat J, Driesen N, Abi-Dargham A, Petrakis I, Heinz A, Pearlson G. The vulnerability to alcohol and substance abuse in individuals diagnosed with schizophrenia. Neurotox Res. 2006:10:235–252.
- Lawrence NS, Ross TJ, Stein EA. Cognitive mechanisms of nicotine on visual attention. Neuron. 2002:36:539–548.
- de Leon J, Diaz FJ. A meta-analysis of worldwide studies demonstrates an association between schizophrenia and tobacco smoking behaviors. Schizophr Res. 2005:76:135–157.
- Liao W, Fan YS, Yang S, Li J, Duan X, Cui Q, Chen H. Preservation effect: cigarette smoking acts on the dynamic of influences among unifying neuropsychiatric triple networks in schizophrenia. Schizophr Bull. 2019:45:1242–1250.
- Liao W, Yang S, Li J, Fan YS, Duan X, Cui Q, Chen H. Nicotine in action: cigarette smoking modulated homotopic functional connectivity in schizophrenia. Brain Imaging Behav. 2019:13:1612–1623.
- Liloia D, Brasso C, Cauda F, Mancuso L, Nani A, Manuello J, Costa T, Duca S, Rocca P. Updating and characterizing neuroanatomical markers in high-risk subjects, recently diagnosed and chronic patients with schizophrenia: a revised coordinate-based meta-analysis. Neurosci Biobehav Rev. 2021:123:83–103.
- Liu XA, Kenny PJ. alpha5 nicotinic receptors link smoking to schizophrenia. Nat Med. 2017:23:277–278.
- Liu H, Luo Q, Du W, Li X, Zhang Z, Yu R, Chen X, Meng H, Du L. Cigarette smoking and schizophrenia independently and reversibly altered intrinsic brain activity. Brain Imaging Behav. 2018:12:1457–1465.
- Meldrum BS. Glutamate as a neurotransmitter in the brain: review of physiology and pathology. J Nutr. 2000:130:1007S–1015S.
- Miller RL, Yaesoubi M, Turner JA, Mathalon D, Preda A, Pearlson G, Adali T, Calhoun VD. Higher dimensional meta-state analysis reveals reduced resting fMRI connectivity dynamism in schizophrenia patients. PLoS One. 2016:11:e0149849.
- Moghaddam B, Javitt D. From revolution to evolution: the glutamate hypothesis of schizophrenia and its implication for treatment. Neuropsychopharmacology. 2012:37:4–15.
- Organization WH . Schizophrenia. In: World Health Organization. Geneva: World Health Organization; 2022.
- Pico-Perez M, Vieira R, Fernandez-Rodriguez M, De Barros MAP, Radua J, Morgado P. Multimodal meta-analysis of structural gray matter, neurocognitive and social cognitive fMRI findings in schizophrenia patients. Psychol Med. 2022:52(4):614–624.
- Potvin S, Lungu O, Lipp O, Lalonde P, Zaharieva V, Stip E, Melun JP, Mendrek A. Increased ventro-medial prefrontal activations in schizophrenia smokers during cigarette cravings. Schizophr Res. 2016:173:30–36.
- Power JD, Barnes KA, Snyder AZ, Schlaggar BL, Petersen SE. Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. NeuroImage. 2012:59:2142–2154.
- Rashid B, Damaraju E, Pearlson GD, Calhoun VD. Dynamic connectivity states estimated from resting fMRI identify differences among schizophrenia, bipolar disorder, and healthy control subjects. Front Hum Neurosci. 2014:8:897.
- Ringin E, Cropley V, Zalesky A, Bruggemann J, Sundram S, Weickert CS, Weickert TW, Bousman CA, Pantelis C, Van Rheenen TE. The impact of smoking status on cognition and brain morphology in schizophrenia spectrum disorders. Psychol Med. 2022:52(14):3097–3115.
- Rushworth MF, Behrens TE, Rudebeck PH, Walton ME. Contrasting roles for cingulate and orbitofrontal cortex in decisions and social behaviour. Trends Cogn Sci. 2007:11:168–176.
- Schaefer A, Kong R, Gordon EM, Laumann TO, Zuo XN, Holmes AJ, Eickhoff SB, Yeo BTT. Local-global parcellation of the human cerebral cortex from intrinsic functional connectivity MRI. Cereb Cortex. 2018:28:3095–3114.
- Shafiei G, Markello RD, Vos de Wael R, Bernhardt BC, Fulcher BD, Misic B. Topographic gradients of intrinsic dynamics across neocortex. elife. 2020:9:e62116.
- Tanabe J, Nyberg E, Martin LF, Martin J, Cordes D, Kronberg E, Tregellas JR. Nicotine effects on default mode network during resting state. Psychopharmacology. 2011:216:287–295.
- Tate JC, Schmitz JM. A proposed revision of the Fagerstrom Tolerance Questionnaire. Addict Behav. 1993:18:135–143.
- White T, Cullen K, Rohrer LM, Karatekin C, Luciana M, Schmidt M, Hongwanishkul D, Kumra S, Charles Schulz S, Lim KO. Limbic structures and networks in children and adolescents with schizophrenia. Schizophr Bull. 2008:34:18–29.
- Winterer G. Why do patients with schizophrenia smoke? Curr Opin Psychiatry. 2010:23:112–119.
- Yan CG, Wang XD, Zuo XN, Zang YF. DPABI: Data processing & analysis for (resting-state) brain imaging. Neuroinformatics. 2016:14:339–351.
- Yang S, Meng Y, Li J, Fan YS, Du L, Chen H, Liao W. Temporal dynamic changes of intrinsic brain activity in schizophrenia with cigarette smoking. Schizophr Res. 2019:210:66–72.
- Yokoyama N, Sasaki H, Mori Y, Ono M, Tsurumi K, Kawada R, Matsumoto Y, Yoshihara Y, Sugihara G, Miyata J, et al. Additive effect of cigarette smoking on gray matter abnormalities in schizophrenia. Schizophr Bull. 2018:44:535–541.
Associated Data
Supplementary Materials
Data Availability Statement
The neuroimaging data that support the findings of this study are available from the corresponding author upon reasonable request.