If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
How Discrimination Gets Under the Skin: Biological Determinants of Discrimination Associated With Dysregulation of the Brain-Gut Microbiome System and Psychological Symptoms
1 TSD and GCG contributed equally to this work as joint first authors.
Affiliations
Vatche and Tamar Manoukian Division of Digestive Diseases, University of California, Los Angeles, Los Angeles, CaliforniaDavid Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CaliforniaUCLA Microbiome Center, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CaliforniaG. Oppenheimer Center for Neurobiology of Stress and Resilience, University of California, Los Angeles, Los Angeles, CaliforniaDivision of Gastroenterology, Hepatology and Parenteral Nutrition, VA Greater Los Angeles Healthcare System, Los Angeles, California
1 TSD and GCG contributed equally to this work as joint first authors.
Gilbert C. Gee
Footnotes
1 TSD and GCG contributed equally to this work as joint first authors.
Affiliations
Department of Community Health Sciences Fielding School of Public Health, Los Angeles, CaliforniaCalifornia Center for Population Research, University of California, Los Angeles, Los Angeles, California
Department of Community Health Sciences Fielding School of Public Health, Los Angeles, CaliforniaCalifornia Center for Population Research, University of California, Los Angeles, Los Angeles, California
Vatche and Tamar Manoukian Division of Digestive Diseases, University of California, Los Angeles, Los Angeles, CaliforniaDavid Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CaliforniaG. Oppenheimer Center for Neurobiology of Stress and Resilience, University of California, Los Angeles, Los Angeles, California
Vatche and Tamar Manoukian Division of Digestive Diseases, University of California, Los Angeles, Los Angeles, CaliforniaDavid Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CaliforniaUCLA Microbiome Center, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CaliforniaG. Oppenheimer Center for Neurobiology of Stress and Resilience, University of California, Los Angeles, Los Angeles, California
Vatche and Tamar Manoukian Division of Digestive Diseases, University of California, Los Angeles, Los Angeles, CaliforniaDavid Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CaliforniaUCLA Microbiome Center, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CaliforniaG. Oppenheimer Center for Neurobiology of Stress and Resilience, University of California, Los Angeles, Los Angeles, California
David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CaliforniaDepartment of Psychiatry & Biobehavioral Sciences and Medicine, University of California, Los Angeles, Los Angeles, California
Vatche and Tamar Manoukian Division of Digestive Diseases, University of California, Los Angeles, Los Angeles, CaliforniaDavid Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CaliforniaG. Oppenheimer Center for Neurobiology of Stress and Resilience, University of California, Los Angeles, Los Angeles, California
Vatche and Tamar Manoukian Division of Digestive Diseases, University of California, Los Angeles, Los Angeles, CaliforniaDavid Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CaliforniaUCLA Microbiome Center, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CaliforniaG. Oppenheimer Center for Neurobiology of Stress and Resilience, University of California, Los Angeles, Los Angeles, California
Vatche and Tamar Manoukian Division of Digestive Diseases, University of California, Los Angeles, Los Angeles, CaliforniaDavid Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CaliforniaUCLA Microbiome Center, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CaliforniaG. Oppenheimer Center for Neurobiology of Stress and Resilience, University of California, Los Angeles, Los Angeles, California
Discrimination is associated with negative health outcomes as mediated in part by chronic stress, but a full understanding of the biological pathways is lacking. Here we investigate the effects of discrimination involved in dysregulating the brain-gut microbiome (BGM) system.
Methods
A total of 154 participants underwent brain magnetic resonance imaging to measure functional connectivity. Fecal samples were obtained for 16S ribosomal RNA profiling and fecal metabolites and serum for inflammatory markers, along with questionnaires. The Everyday Discrimination Scale was administered to measure chronic and routine experiences of unfair treatment. A sparse partial least squares-discriminant analysis was conducted to predict BGM alterations as a function of discrimination, controlling for sex, age, body mass index, and diet. Associations between discrimination-related BGM alterations and psychological variables were assessed using a tripartite analysis.
Results
Discrimination was associated with anxiety, depression, and visceral sensitivity. Discrimination was associated with alterations of brain networks related to emotion, cognition and self-perception, and structural and functional changes in the gut microbiome. BGM discrimination-related associations varied by race/ethnicity. Among Black and Hispanic individuals, discrimination led to brain network changes consistent with psychological coping and increased systemic inflammation. For White individuals, discrimination was related to anxiety but not inflammation, while for Asian individuals, the patterns suggest possible somatization and behavioral (e.g., dietary) responses to discrimination.
Conclusions
Discrimination is attributed to changes in the BGM system more skewed toward inflammation, threat response, emotional arousal, and psychological symptoms. By integrating diverse lines of research, our results demonstrate evidence that may explain how discrimination contributes to health inequalities.
). Self-reported discrimination is a risk factor for mental and physical disorders. However, the understanding of how discrimination gets embedded in biology and affects mental/physical health remains unclear.
Previously, the mechanistic understanding of the link between discrimination and disease was mainly centered on the hypothalamic-pituitary-adrenal (HPA) axis (
). However, recent studies have highlighted that the gut microbiome is extremely sensitive to stressors in influencing long-term health and inflammation (
). Stress can impact the bottom-up communication between the microbiome and the brain through several pathways, including changes in HPA activity, specific microbial metabolites, and vagus nerve and immune-inflammatory mechanisms (
Similarly, changes in brain activity can also influence the body and the gut. Neuroimaging studies suggest that discrimination rewires the brain, causing dysregulation of stress networks (
Testing whether implicit emotion regulation mediates the association between discrimination and symptoms of psychopathology in late childhood: An RDoC perspective.
The gut and its microbiome as related to central nervous system functioning and psychological well-being: Introduction to the special issue of psychosomatic medicine.
The microbiota-gut-brain axis: Pathways to better brain health. Perspectives on what we know, what we need to investigate and how to put knowledge into practice.
). Therefore, the major goal of this work is to examine how discrimination affects biology beyond the well-studied HPA axis by examining how discrimination alters the BGM system.
We propose a model highlighting the influence of discrimination on the bidirectional signaling between the brain and gut microbiome as mediated by inflammation. A novel aspect of this model reflects the dysregulation in connections between the central and enteric nervous systems (Figure 1).
Figure 1Conceptual model linking the brain-gut-microbiome system to discrimination and clinical outcomes. AAA, aromatic amino acid; EDS, Everyday Discrimination Scale; IL-1β, interleukin 1β; NFK-B, nuclear factor kappa light chain enhancer of activated B cells; PHQ, Physical Health Questionnaire; PSS, Perceived Stress Scale; SCFA, short-chain fatty acids.
Furthermore, discrimination has been predominantly studied in Black individuals as compared with White individuals, and few have studied how it may affect racial groups differently. It would seem logical that the effects of discrimination on health would be stronger among people of color than among White individuals, but the literature is mixed in this regard (
Racial/ethnic and gender differences in the association between self-reported experiences of racial/ethnic discrimination and inflammation in the CARDIA cohort of 4 US communities.
Everyday discrimination prospectively predicts inflammation across 7-years in racially diverse midlife women: Study of women’s health across the nation.
). This raises an intriguing question as to why discrimination may have varying effects across different races/ethnicities. Obesity, substance abuse, and depression have all been linked to the BGM system. Therefore, these differences may be related to differences in how dysregulation of the BGM system occurs within each race/ethnicity.
To study a holistic view of how discrimination can affect biology, we performed a detailed analysis of discrimination on the BGM system in a racially diverse population to test whether 1) experiences of everyday discrimination will be associated with increased alterations in stress brain connectivity pathways (resting-state functional magnetic resonance imaging [fMRI]) and stress-related gut microbiome (16S sequencing and fecal metabolites) as mediated by increased inflammatory processes (peripheral blood mononuclear cells [PBMCs]), 2) these associations will be related to increased adverse behavioral and psychological measures, and 3) these associations will vary across Asian, Black, Hispanic, and White individuals.
Methods and Materials
Ethics Approval and Consent to Participate
All procedures were approved by the Institutional Review Board (16-000187, 15-001591) at the University of California, Los Angeles, Office of Protection for Research Subjects. All participants provided written informed consent.
Participants
The final sample comprised 154 adults from the 165 who were initially enrolled in the study. Participants were recruited from the Los Angeles community and clinics.
Participant data included fMRI for resting-state connectivity, anthropometrics, blood samples for genetic expression of inflammation via PBMCs, stool samples (microbiome and metabolomics), and survey questionnaires, including a diet history (Table S1 in Supplement 1). Participants self-reported race/ethnicity (Asian American, Black, Hispanic, or White). Discrimination was measured using the Everyday Discrimination Scale.
Statistical Analysis
Group differences in demographic characteristics; brain, microbial taxa, and differential abundance testing for metabolomics; and PBMCs were determined individually (adjusting for sex, age, body mass index, and diet). For fMRI and metabolite data, sparse partial least square linear discriminant analysis (sPLS-DA) was done to analyze brain and metabolite data using the Mixomics package in R. Because the number of variables far exceeds the number of study participants, sparse multivariate nonparametric models exhibit the most robust statistical power and consistency (
Surgically induced changes in gut microbiome and hedonic eating as related to weight loss: Preliminary findings in obese women undergoing bariatric surgery.
Surgically induced changes in gut microbiome and hedonic eating as related to weight loss: Preliminary findings in obese women undergoing bariatric surgery.
). Microbiome data were analyzed using QIIME2 and DESEq2 in R. PBMC data were analyzed using a priori–specified domains of immune function that have repeatedly been found to vary in response to psychosocial risk factors (
). For PBMC, Mann-Whitney U test was performed between participants with low or high discrimination, and p values were adjusted for multiple hypothesis testing using Bonferroni correction. Data were analyzed for factors associated with discrimination as well as within-group differences associated with discrimination. Factors that were associated with high levels of discrimination were then compared across races.
Integrated analyses involving associations between different datasets were performed using Spearman’s correlation controlling for multiple hypothesis testing and presented as circos plots. Further details are provided in Supplemental Methods in Supplement 1.
Results
Participant Differences Associated With Discrimination
Of the 154 participants (80 with high discrimination and 74 with low discrimination), the high discrimination group had higher levels of anxiety (p = .009), depression (p = .009), perceived stress (p = .001), visceral sensitivity (p < .001), early-life adversity (p = .009), neuroticism (p = .01), and worse scores for mental health (p = .01) and physical health (p = .02) than the low discrimination group (Table 1). There were no significant differences in sex, age, body mass index, education, marital status, and diet. When examining across race/ethnicity within the high levels of discrimination group, Hispanic individuals reported the highest levels of early-life adversity, levels significantly higher than in Asian individuals (p = .04). Black individuals reported the highest levels of resilience, levels significantly higher than in Asian individuals (p = .02). Black individuals reported the lowest levels of neuroticism, levels significantly lower than in Asian (p = .02), Hispanic (p = .001), and White (p = .01) individuals. Black individuals had the lowest levels of trait anxiety, depression, and perceived stress and significantly lower than in Hispanic individuals (p = .03, .03, and .02, respectively). Black individuals also had the highest reported scores for mental health, scores significantly higher than in Hispanic individuals (p = .003). There were no significant differences in extraversion, socioeconomic status, and self-reported physical health in individuals who experienced high levels of discrimination across the races.
Table 1Participant Characteristics and Clinical Questionnaires
For Asian, Black, and Hispanic participants, race was the most common reason for discrimination. For White participants, gender and age were the most common reasons for discrimination (Table S2 in Supplement 1). The average Everyday Discrimination Scale score for participants who had high levels of discrimination was similar across the different races (p = .18).
Discrimination Is Associated With Altered Brain Connectivity
In the aggregated sample, high discrimination as compared with low discrimination was associated with increased connectivity between the default mode network (DMN) (supramarginal gyrus, superior temporal sulcus, lateral aspect of the superior temporal gyrus, inferior temporal sulcus, middle temporal gyrus, precuneus, and transverse temporal sulcus) and the sensorimotor network (SMN) (precentral gyrus, Heschl's gyrus, subcentral gyrus, paracentral lobule, superior frontal gyrus, posterior lateral sulcus, and inferior and superior part of the precentral sulcus). High discrimination was also associated with increased connectivity between the central autonomic network (CAN) (medial orbital gyrus, gyrus rectus, frontomarginal gyrus, and orbital sulcus) and the emotion regulation network (ERN) (several subregions of the anterior cingulate cortex, orbital part of the inferior frontal gyrus, and parahippocampal gyrus), salience network (SAL) (anterior insula, anterior midcingulate cortex), and occipital network (OCC) (middle and superior occipital sulcus, superior occipital gyrus, and occipital pole). High discrimination as compared with low was associated with decreased connectivity between regions of the central executive network (CEN) (intraparietal sulcus, subparietal sulcus, superior parietal lobule, and middle frontal gyrus) to various regions of the CAN, SMN, and OCC. (Loadings and variables of importance from the sPLS-DA are listed in Tables S3–S7 in Supplement 2.)
More specific discrimination-based differences were observed in brain connectivity when stratified by race/ethnicity (Figure 2). For Black participants, high levels of discrimination as compared with low were related to higher connectivity within the CEN (orange) and with the DMN (blue). For Hispanic participants, high discrimination as compared with low was associated with higher connectivity within regions of the DMN and between regions of the DMN with regions from the CEN, SAL (yellow), CAN (red), and OCC (purple). For Asian participants, high discrimination, as compared with low, was associated with higher connectivity within regions of the SMN (green) and within the OCC. In addition, there was higher connectivity between regions of the CEN with CAN and DMN. For White participants, high discrimination as compared with low was associated with higher connectivity involving various brain regions within the ERN (pink) and reward network (RN, gray), which was unique to this group and within the SMN, DMN, SAL, CAN, CEN, and OCC, which was also seen in the other groups.
Figure 2Brain regions associated with discrimination by race/ethnicity. Sparse partial least square linear discriminant analysis (sPLS-DA) plots, resting-state pairwise differences by levels of discrimination, and anatomical diagram of brain regions associated with discrimination across the different races/ethnicities: Black (A, B), Hispanic (C, D), Asian (E, F), and White (G, H). AngG, angular gyrus; AoCS, anterior occipital sulcus; ATrCos, anterior transverse collateral sulcus; CcS, calcarine sulcus; CgSMarp, marginal branch of the cingulate sulcus; Cun, cuneus; EDS, Everyday Discrimination Scale; FMarG/S, fronto-marginal gyrus and sulcus; FuG, fusiform gyrus; HG, Heschl’s gyrus; InfCirInS, inferior segment of circular sulcus of the insula; InfFGOpp, opercular part of the inferior frontal gyrus; InfFS, inferior frontal sulcus; InfOcG/S, interior occipital gyrus/sulcus; InfPrCS, inferior part of the precentral sulcus; IntPS/TrPS, intraparietal sulcus (interparietal sulcus) and transverse parietal sulci; JS, sulcus intermedius primus (of Jensen); LinG, lingual gyrus; LORs, lateral orbital sulcus; MACgG/S, middle-anterior part of the cingulate gyrus and sulcus; MFG, middle frontal gyrus; MoCG, middle occipital gyrus; MOcS/LuS, middle occipital sulcus and lunatus sulcus; MTG, middle temporal gyrus; OcPo, occipital pole; OrG, orbital gyri; PaCL/S, paracentral lobule and sulcus; PerCaS, pericallosal sulcus; PrCun, precuneus; RG, gyrus rectus; SbCG/S, subcallosal gyrus/sulcus; SbPs, subparietal sulcus; SuMarG, supramarginal gyrus; SupFG, superior frontal gyrus; SupFS, superior frontal sulcus; SupPL, superior parietal lobule; SupTG, superior temporal gyrus; SupTGLp, lateral aspect of the superior temporal gyrus; Thal, thalamus; TPo, temporo-parieto-occipital; TrFPoG/S, transverse frontopolar gyri and sulci.
Similar patterns were observed when investigating differences across races in the high discrimination group only: Black individuals had greater connectivity in the CEN, DMN, and OCC but decreased connectivity in the CAN compared with White individuals. Hispanic individuals, compared with Asian individuals, had greater connectivity in the DMN and OCC, but decreased connectivity in the SMN, and compared with White individuals, they had greater connectivity in the CEN and DMN. Asian individuals, compared with White individuals, had greater connectivity in the SMN (details in Table S8 in Supplement 2).
Discrimination Is Associated With Gut Microbiome and Metabolite Changes
Microbiome and metabolite differences related to discrimination were only apparent when stratified by race/ethnicity (Figure 3).
Figure 3Microbiome and fecal metabolites associated with discrimination by race/ethnicity. (A) Differential abundance testing by DESEq2 of bacterial taxa associated with discrimination in Black individuals. (B) Taxonomic plot of genera with a relative abundance ≥ 1% by discrimination in Black individuals. Similar analysis represented for Hispanic (D, E) and White (G, H) individuals. Fecal metabolites by discrimination in Black (C), Hispanic (F), and Asian (I) individuals. Asian individuals had no microbiome differences by discrimination level, and White individuals had no metabolites that were different by discrimination level. EDS, Everyday Discrimination Scale.
Black participants had 9 bacterial species that were different by discrimination. High levels of Prevotella, Coprococcus, and Tyzzerella and lower levels of species belonging to Bacteroides, Parabacteroides, and Ruminococcaceae were observed in the high discrimination group compared with those found in the low discrimination group. For Black participants, high discrimination was also associated with a reduced level of hydroxy-N6,N6,N6,-trimethyllysine as compared with low discrimination.
In Hispanic individuals, high discrimination as compared with low was associated with a higher level of Bacteroides stercosis and lower levels of valerate, levulinate, 3-(4-hydroxyphenyl)lactate, pregnanolone/allopregnanolone sulfate, and isovalerate.
For Asian participants, 11 fecal metabolites were higher in participants with high discrimination than in partipants with low discrimination. These include 3-beta-hydroxy-5-cholestenoate, beta-sitosterol, campesterol, cholesterol, desmosterol, fucosterol, palmitoyl-sphinganine, and palmitoyl-sphingosine.
For White participants, a high level of discrimination as compared with low was associated with changes in 7 bacterial species (reduction in Prevotella copri, Bacteroides salyersiae, Blautia stercosis, Faecalibacterium prausnitzii, and unknown species of Prevotella and Ruminococcaceae and an elevation in Catenibacterium mitsuokai).
Bacteria and metabolite differences across races in individuals experiencing high levels of discrimination are summarized in Tables S9 and S10 in Supplement 1. In this analysis of individuals who experienced high levels of discrimination, Pcopri was the highest in Black and Hispanic individuals and was the lowest in White individuals (p = .04). Similarly, of the metabolites analyzed in individuals who experienced high levels of discrimination, only isovalerate, valerate, and fucosterol were statistically different between the races. Isovalerate and valerate were significantly lower in Hispanic than in White individuals (p = .04 and .04, respectively), and fucosterol was significantly higher in Asian than in White individuals (p = .03).
Effects of Discrimination on Inflammatory Markers
Of the a priori set of immune markers, which included 19 genes involved in inflammation and 32 genes involved in type I interferon responses, 4 markers were associated with high discrimination as compared with low discrimination (Figure 4).
Figure 4Expression levels of several inflammatory markers extrapolated from peripheral blood mononuclear cells for Black (A), Hispanic (B), and White (C) participants. ∗p value < .05. EDS, Everyday Discrimination Scale.
In Black participants, high discrimination as compared with low was associated with elevated levels of prostaglandin-endoperoxidase synthase 1 (PTGS1). In Hispanic participants, high discrimination as compared with low was associated with elevated levels of interferon-induced protein 35 (IFI35) and interleukin 1β (IL1B). In White participants, high discrimination as compared with low was associated with a reduction in interferon regulatory factor 8 (IRF8). There were no inflammatory markers that were different in Asian participants.
Examining only individuals who experienced high levels of discrimination across races, Black individuals had the highest level of PTGS1, which was significantly higher than in White individuals (p = .03). Hispanic individuals had the highest level of IL1B, which was significantly higher than in White individuals (p = .001). There was no statistical difference in the expression level of IFI35 or IRF8 in individuals who experienced high levels of discrimination across races.
Association Networks Differ by Discrimination and by Race/Ethnicity
The association networks within the BGM system with high discrimination levels and by race/ethnicity are represented in the connectograms (Figure 5).
Figure 5Networks depicting high discrimination associated with brain-gut microbiome immune factors. Networks relating brain, peripheral blood mononuclear cells, microbiome, and clinical questionnaire data by race in those individuals experiencing high discrimination. Red lines are positive associations, and blue lines are negative associations.
In Black participants, hydroxy-N6,N6,N6,-trimethyllysine was negatively associated with the DMN and CEN, and inflammatory markers IRF8 and IL-1β were positively associated with resilience, which was also positively associated with the DMN and CEN. There were positive connections between stress and anxiety with several bacterial species (Parabacteroides and Bacteroides).
Among Hispanic participants, inflammatory marker IRF8 was associated with higher levels of anxiety and neuroticism and with the lipid metabolite valerate. Higher levels of physical symptoms (Patient Health Questionnaire and Short Form Survey 12 Physical Component Score) were positively associated with the DMN and with the anterior midcingulate cortex (key region of the SAL), but high socioeconomic status was also positively associated with the DMN.
In Asian participants, there were several positive associations between metabolites related to cholesterol (lipid pathway), microbial species (Prevotella, Ruminococcaceae), and anxiety (state and trait), neuroticism, depression, and physical symptoms. Asian individuals who experienced high discrimination also had several connections to the SMN and OCC.
White participants had several associations between the gut microbiome (Coprococcus, Ruminococcus, Ruminococcaceae, Parabacteroides, Alistipes, Bacteroides) and widespread brain networks (including the ERN and RN) and with anxiety, depression, neuroticism, early-life adversity, stress, visceral sensitivity, and physical symptoms. It was the only group that demonstrated negative associations between resilience and other variables.
Discussion
We examined the association between everyday experiences of discrimination and alterations in the BGM system. Generally, discrimination was associated with anxiety, depression, and worse measures of physical and mental health. However, these associations varied across race/ethnicity, with Black individuals having no association between discrimination and mental health.
A history of discrimination was associated with widespread connectivity differences in several networks, but with race/ethnic differences contributing to the greatest variance. The differences seen in the BGM system between the different racial/ethnic groups could be due to the varying types of discrimination experienced by the different groups. A key feature of our work is the inclusion of multiple forms of discrimination. While racial discrimination is important, our study allowed participants to report on discrimination based on other factors (e.g., gender, age, religion). This was critical for capturing the spectrum of experiences for our diverse sample. For minorities, skin color and race were the most common reason for discrimination. For White individuals, gender and age were the most common reasons. Discrimination based on race and skin color can occur as early as in childhood, a period of time that is critical for the development of the BGM system (
). One of the most common reasons suggested to explain the cause of negative physical health outcomes associated with discrimination is an increase in allostatic load and the involvement of multiple biological systems (
). Discrimination based on race and skin color, which can occur in early childhood, could lead to a longer period of stress and allostatic load than other forms of discrimination (
). While self-generated thoughts can be a source of creative insight and introspection, they can lead to distress and negatively impact performance of specific tasks (
). These results suggest that alterations within the DMN and SMN may reflect difficulties with cognitive and affective appraisal of pain in relation to discrimination. Discrimination was also associated with heightened emotion regulation (ERN), autonomic (CAN), alertness (SAL), and attention toward salient stimuli (OCC). Experiences of discrimination are typically stressful (
). However, the discrimination-related patterns in the brain showed strong differences when examined by racial/ethnic groups.
In Black participants, discrimination was associated with higher connectivity within the DMN and CEN. The DMN is important in self-recollection of past experiences (
Default mode network, meditation, and age-associated brain changes: What can we learn from the impact of mental training on well-being as a psychotherapeutic approach?.
), possibly using this as a coping mechanism to regulate distress accompanying discrimination.
In Hispanic participants, discrimination was associated with greater connectivity within the DMN, CEN, SAL, CAN, and OCC. Psychosocial stress is associated with cognitive impairment (
), and the SAL might be involved in regulation of heightened vigilance associated with discrimination. Hispanic individuals who experienced more discrimination were more likely to have early-life trauma, higher levels of anxiety, depression, stress, and visceral sensitivity than those who experienced less discrimination, similar to the hyperactivity of SAL regions observed in patients with anxiety, depression, and posttraumatic stress disorder (
) and may indicate altered regulatory functions via viscerosensory mechanisms. These responses are not only critical for adapting to internal or external challenges, but also initiate signals that trigger emotion, affect decision making, and promote social behavior (
). Alterations in CAN highlight a risk factor for both mental and physical health problems. The alterations in the OCC suggest enhanced attention toward threatening stimuli experienced during discrimination (
In Asian individuals, discrimination was associated with higher connectivity within the SMN. The SMN is engaged in interoceptive, autonomic, sensory, motor, and reward processing (
). Increased SMN connectivity indicates disrupted sensory functions associated with somatization, similarly observed in stressed individuals and patients with major depressive disorder (
). The anterior insula, together with the SMN, has been implicated in social pain and, as a result, in physical pain due to distress associated with exclusion (
White participants who experienced high levels of discrimination displayed a chaotic enhanced resting-state connectivity in numerous large-scale networks including the ERN and RN, which may underscore the hypersensitivity and inability to cope with discrimination in comparison with other races/ethnicities. Disruptions in the communication between large-scale networks may reflect difficulty with reacting and coordinating efficiently to experiences of discrimination (
Discrimination-Associated Gut Microbiome and Metabolite Changes
Pcopri was the only bacterium species that was significantly different across races. Pcopri was the highest in Black and Hispanic individuals who experienced discrimination as compared with White individuals who experienced discrimination. Pcopri produces a superoxide reductase and phosphoadenosine phosphosulfate reductase (
When examining metabolites, Black individuals with high discrimination had lower levels of hydroxy-N6,N6,N6-trimethyllysine, and Hispanic individuals had lower levels of branched-chain fatty acids as compared with Black and Hispanic individuals with low discrimination, respectively. Hydroxy-N6,N6,N6-trimethyllysine is a by-product of carnitine biosynthesis. Carnitine has anti-inflammatory and cardioprotective properties (
). Hispanic individuals had the lowest levels of branched-chain fatty acids and had significantly lower levels than White individuals who also experienced similar levels of discrimination. Branched-chain fatty acids can have anti-inflammatory and anticancer properties and are important to colonic motility and health (
Unlike the patterns observed in Black and Hispanic individuals, the microbiome and metabolite panel of Asian and White individuals are less related to inflammation. In Asian individuals, high discrimination was associated with higher levels of metabolites that have been implicated in lipid metabolism. This profile may suggest dietary preference for foods high in fat in Asian individuals who experience high levels of discrimination. In White individuals, discrimination was associated with the lowest levels of Pcopri.
Discrimination-Associated Inflammatory Changes
In Black individuals, discrimination was associated with higher levels of PTGS1, and in Hispanic individuals, discrimination was associated with higher levels of IL1B. Both of these were higher in Black and Hispanic individuals who experienced high levels of discrimination, as compared with White individuals who experienced similar levels of discrimination. PTGS1 is also known as cyclo-oxygenase 1 (COX1) and is the enzyme that catalyzes the conversion of arachidonate to prostaglandins. High levels of prostaglandins are produced in response to injury or infection and are major drivers of inflammation (
). These findings suggest that discrimination may lead to a chronic state of inflammation, specially in Black and Hispanic individuals.
Discrimination-Related Changes Within the BGM System and Clinical Implications
The BGM patterns highlight that high levels of discrimination in Black participants are associated with higher levels of inflammatory biomarkers as compared with Black participants with lower levels of discrimination. Despite the increase in inflammatory markers, the group as a whole showed the lowest levels of anxiety and depression, irrespective of discrimination. Black participants as a group had the highest resilience scores of any race. These findings suggest that the effect of discrimination on mental health in this group is likely being buffered by top-down processes related to resilience and cognitive flexibility (
In Hispanic participants, high discrimination was associated with peripheral markers (inflammation-IRF8 and gut microbes) and several clinical behaviors (anxiety, physical health symptoms), but socioeconomic status and DMN activity related to better coping strategies and cognitive control could be overriding these negative effects. Similarly, some studies have demonstrated that socioeconomic status can be protective against discrimination (
Race and other sociodemographic categories are differentially linked to multiple dimensions of interpersonal-level discrimination: Implications for intersectional, health research.
In Asian participants with high discrimination, there were positive associations between metabolites related to cholesterol and to several clinical measures (anxiety, depression, physical symptoms) and with SMN activity (social pain and visceral somatosensory processes) (
). This suggests that Asian individuals with high discrimination are possibly eating foods that are high in fat to deal with the associated feelings of anxiety, depression, and somatosensory/visceral signals, which is consistent with studies demonstrating the emphasis on physical symptoms as a way to deal with painful emotional and stressful situations.
In White participants with high discrimination, there were several widespread associations within the BGM system, and it was the only group that included connections to the ERN and RN (a network associated with emotional stress). This pattern, together with the negative association with resilience, highlights the decreased regulatory deficiency in reacting and coordinating efficiently to novel and stressful experiences of discrimination in White participants.
Limitations
While this is the first study to examine discrimination across different racial groups in relation to the BGM system, there are several limitations to the current study. Black individuals were underrepresented in the study. This low sample size could make the analysis for Black individuals underpowered to discern small effect sizes as well as raise the possibility of sampling bias. However, we do not present any data that conflict with previously published works regarding Black individuals and discrimination, but rather expand on their relation to the BGM system. Future studies looking into the BGM system and discrimination should attempt to increase the representation of Black individuals. Finally, while a major strength of this article is the incorporation of multiple biological systems, we did not examine other systems that are likely involved in discrimination such as the HPA axis and the autonomic nervous system. But this body of work shows that discrimination has a holistic effect on the body and the mind, and therefore, discrimination’s effect on health is complex and multifactorial.
Conclusions
Unfair treatment is experienced by all people. Our findings provide a preliminary framework for understanding how unfair treatment is perceived and processed in the brain and how these are, in turn, related to inflammation, gut microbiome, and psychological symptoms. Of course, much more work remains, but it provides an initial step toward understanding how social inequalities become a whole-body experience and gives some understanding of how expressions of “racism makes me sick to the stomach” might have an actual manifestation in the body.
Acknowledgments and Disclosures
This research was supported by the National Institutes of Health (Grant Nos. R01 MD015904 [to AG], K23 DK106528 [to AG], R03 DK121025 [to AG], T32 DK07180 [to TD], ULTR001881/DK041301 [University of California, Los Angeles CURE/Digestive Diseases Research Core Center/Clinical and Translational Science Institute Pilot and Feasibility Study] [to AG], P50 DK064539 [to EAM], R01 DK048351 [to EAM], and P30 DK041301) and pilot funds provided for brain scanning by the Ahmanson-Lovelace Brain Mapping Center. These funders played no role in study design, or the collection, analysis, and interpretation of the data.
Funding was acquired by AG. TSD, GCG, and AG were involved in conceptualization of the study; TSD and AG developed the methodology; and TSD, AG, ZC, VS, YZ, YG, and SC performed the formal analysis. AG was responsible for resources/data curation. TSD, GCG, HB-S, MW, VO, LAK, JSL, BN, XZ, SC, EAM, and AG contributed to writing and original draft preparation. TSD, YZ, and AG contributed to visualization. The entire study was supervised by AG. All authors read and approved the final manuscript.
We acknowledge the assistance of the Neuroimaging Core, Bioinformatics and Statistics Core, Microbiome Core, and the Biorepository Core of the University of California, Los Angeles Microbiome Center for their assistance with various processing, storage, and analyses assistance of the current manuscript. We acknowledge Dr. Steve Cole for his assistance in processing the PBMC samples.
The datasets generated during and/or analyzed during the current study are not publicly available due to an ongoing collaboration with multiple principal investigators involving participant identifiers at the G. Oppenheimer Center for Neurobiology of Stress and Resilience. However, data are available from the corresponding author on reasonable request.
Participants or the public were not involved in the design, conduct, reporting, or dissemination plans of this research.
AG is a scientific consultant to Yamaha. EAM is a scientific advisory board member of Danone, Axial Biotherapeutics, Amare, Mahana Therapeutics, Pendulum, Bloom Biosciences, Seed, and APC Microbiome Ireland. All other authors report no biomedical financial interests or potential conflicts of interest.
Testing whether implicit emotion regulation mediates the association between discrimination and symptoms of psychopathology in late childhood: An RDoC perspective.
The gut and its microbiome as related to central nervous system functioning and psychological well-being: Introduction to the special issue of psychosomatic medicine.
The microbiota-gut-brain axis: Pathways to better brain health. Perspectives on what we know, what we need to investigate and how to put knowledge into practice.
Racial/ethnic and gender differences in the association between self-reported experiences of racial/ethnic discrimination and inflammation in the CARDIA cohort of 4 US communities.
Everyday discrimination prospectively predicts inflammation across 7-years in racially diverse midlife women: Study of women’s health across the nation.
Surgically induced changes in gut microbiome and hedonic eating as related to weight loss: Preliminary findings in obese women undergoing bariatric surgery.
Default mode network, meditation, and age-associated brain changes: What can we learn from the impact of mental training on well-being as a psychotherapeutic approach?.
Race and other sociodemographic categories are differentially linked to multiple dimensions of interpersonal-level discrimination: Implications for intersectional, health research.
01703-6&id=gr1.jpg){kind=link}
01703-6&id=gr2.jpg){kind=link}
01703-6&id=gr3.jpg){kind=link}
01703-6&id=gr4.jpg){kind=link}
01703-6&id=gr5.jpg){kind=link}